Ecological assessment of riparian forests in Benin:
Phytodiversity, phytosociology, and spatial distribution of tree species
Armand Kuyéma NATTA
Promotor:
Prof. Dr. Ir. L.J.G. van der Maesen
Hoogleraar in de Plantentaxonomie, Wageningen Universiteit,
Nederland
Prof. Dr. Ir. B. Sinsin
Professor of Ecology, Department of Environment Management, Faculty of
Agronomic Sciences, University of Abomey-Calavi, Benin.
Co- promotor: Dr. A. Akoègninou
Department of Vegetal Biology, Faculty of Science and Technique, University
of Abomey-Calavi, Benin.
Promotiecommissie:
Prof. Dr. Ir. A. de Gier (ITC – Enschede, Nederland)
Prof. Dr. F.J.J.M. Bongers (Wageningen Universiteit, Nederland)
Prof. Dr. J. Lejoly (Free University of Brussels, België)
Prof. Dr. S. Porembski (University of Rostock, Germany)
Ecological assessment of riparian forests in Benin:
Phytodiversity, phytosociology, and spatial distribution of tree species
Armand Kuyéma NATTA
Proefschrift
ter verkrijging van de graad van doctor
op gezag van de rector magnificus
van Wageningen Universiteit
Prof. Dr.Ir. L. Speelman
in het openbaar te verdedigen
op woensdag 26 november 2003
des namiddags te vier uur in de Aula
NATTA Armand Kuyéma (2003).
Ecological assessment of riparian forests in Benin: Phytodiversity, phytosociology, and
spatial distribution of tree species
Ph.D. Thesis Wageningen University, with summaries in English, French and Dutch.
ISBN 90-5808-954-1
Key words: Riparian forests, flora, diversity, endangered species, structure, plant community,
ordination, classification, rivers, streams, spatial distribution, ecological factors, sampling
designs, Benin, West Africa.
To
My mother,
Gwendoline,
Kévin.
CONTENTS
Chapter 1
General introduction
1
Chapter 2
Study area
7
Chapter 3
Riparian forests, a unique but endangered ecosystem in Benin
15
Chapter 4
Assessment of riparian forest fragments plant diversity in West African
savanna regions: an overview from Benin
27
Chapter 5
Structure and ecological spectra of riparian forests in Benin
41
Chapter 6
A phytosociological study of riparian forests in Benin (West Africa)
53
Chapter 7
Spatial distribution and ecological factors determining the occurrence
of Pentadesma butyracea Sabine (Clusiaceae) in Benin (West Africa)
73
Forest structural parameters and floristic composition spatial variation
and modelling across rivers in Benin
83
Ouémé and Comoé: forest-savanna border relationships in two riparian
ecosystems in West Africa
105
Chapter 8
Chapter 9
Chapter 10
Assessing the density of Khaya species through Simple Random Sampling,
Stratified Sampling and Systematic Sampling
117
Chapter 11
General discussion
129
Chapter 12
General conclusion
143
Summary
149
Résumés
155
Samenvatting
161
References
167
Riparian forests plant species list
193
Acknowledgements
211
Curriculum vitae
213
List of publications
215
Chapter 1
GENERAL INTRODUCTION
A.K. Natta
Chapter 1: General Introduction
Chapter 1
GENERAL INTRODUCTION
1.1. RESEARCH BACKGROUND
The loss and fragmentation of tropical forest is the single greatest threat to the world’s
biological diversity (Whitmore 1990, Huston 1994). In 1992, the Convention on Biological
Diversity highlighted that measures must be implemented for the conservation of natural
ecosystems, especially for tropical forests, which are famous for being the most species rich
ecosystems on earth. Although many species have been described, very little is known about
their ecology (Sayer & Wegge 1992). In Africa the most species-rich forests, at least for
woody plants, are in the wetter areas of West Central Africa (from the base of Mt Cameroon
South East into Gabon), and in the North East of Madagascar (Gentry 1992). Unfortunately
this diversity is characterised by a deforestation rate of 0.7 %, more than twice the world
average of 0.3 % (FAO 2000).
In West and Central Africa, the rain forest block is interrupted between South East
Ghana and South West Nigeria. Biogeographers call this discontinuity the Dahomey-Gap.
Innumerable animals and plants are common to Central Africa and upper Guinea, showing
that the two forests blocks have been connected in the past. However, this does not mean that
the forests are identical. The two rain forests blocks are characterised by special endemic
species to confirm that they have not only been isolated but are intrinsically different
(Kingdon 1990). On the other hand, the West-African countries located in the Dahomey-Gap,
such as Benin, are characterised by the absence of tropical rain forests and a low percentage
of dense semi-deciduous forests. In consequence the need to conserve the biological diversity
of remaining fragmented forests, represented by sacred forests, few protected dense semideciduous upland forests and riparian forests, in this dry wedge cannot be underestimated.
All over the world, and particularly in tropical savanna, the natural vegetation
associated with waterways and mostly represented by riparian forests (or RFs) is credited to
be among the most species-rich ecosystems (Meave & Kellman 1994, PGRN/IUCN 1994,
Roggeri 1995, Nilsson et al. 1997). Riparian forests are important areas for global
biodiversity (Sala et al. 2000), because they protect key resources for mankind, such as water
sources and quality, and stream environment (Vought et al. 1994, Lowrance et al. 1997,
Montgomery 1997, Trimble 1999), and harbour a diversified flora and physical structure
(Gibbs & Leitão 1978, Gregory et al. 1991, Bersier & Meyer 1994, Kellman et al. 1994,
Meave & Kellman 1994, Woinarski et al. 2000, Kokou et al. 2002). Ecologists have also
examined the value of riparian forests as habitats for many animals, and recognised them as a
priority area for conservation of terrestrial mammals (Doyle 1990, Darveau et al. 1998, de
Lima & Gascon 1999, Darveau et al. 2001), as well as birdlife (Stauffer & Best 1980, Gates
& Giffen 1991, McGarigal & McComb 1992, Larue et al. 1995, Murray & Stauffer 1995,
Darveau et al. 1995, Whitaker & Montevecchi 1997, Saab 1999, Woinarski et al. 2000).
As tropical forests become more fragmented due to deforestation, riparian forests play
a crucial role in providing habitat corridors between forest patches to increase landscape
connectivity (Forman & Godron 1986, Forman 1997, Machtans et al. 1996). In proportion to
their area within a watershed, they perform more ecological and productive functions than do
adjacent uplands (NRC 2002). The vital ecological, hydrological and biogeochemical vital
functions of these forests bordering waterways are now largely acknowledged. In recognition
of these roles, they have been increasingly protected by policies and legislation (DNRE 1996,
de Lima & Gascon 1999, NRC 2002, Natta et al. 2002). Unfortunately, they are under severe
1
Chapter 1: General Introduction
threat worldwide (Sparovek et al. 2002), and current management strategies, particularly in
the tropics, seem to have limited effects.
A large body of work has demonstrated that riparian forests play a critical role
regulating interactions between terrestrial and aquatic components of temperate zones
landscapes (Gregory et al. 1991, Gilliam 1994, Naiman & Décamps 1997), however there
have been relatively few studies dedicated to them in the tropics (Bowden et al. 1992,
McDowell et al. 1992, McClain et al. 1994, Chestnut & McDowell 2000, Groffman et al.
2001) in general, and Benin in particular.
Flora and fauna reserves represent 11% of Benin surface area, and the flora is
estimated to have 3,000 species (PFB 1997), which is relatively poor in comparison with
neighbouring humid countries. The natural wooded vegetation represented by savanna and
open forests occupy 72.2 % of the country, but dense semi-deciduous forests and riparian
forests cover only 3.42 % of the country (CENATEL 2001). During the preparation of
Benin’s Environmental Management Plan, it was noted that the lack of scientific knowledge
of the country’s environment constituted a major obstacle to the realisation of the plan and
measurement of the effectiveness of the proposed actions. In 1994, following the spirit of the
Rio conference, the sustainable development agreement between Benin and the Netherlands
emphasised on the knowledge and valorisation of the biological diversity. In 1995 this
agreement has retained, as emergency programs, scientific surveys of plant species and the
compilation of Benin’s flora. The flora is known to be a major component of a region’s or a
country’s biodiversity. Scientific research on species diversity is of great importance in
providing accurate information and can enhance the management and sustainable use of
phytogenetic resources, especially for endangered ecosystems (PFB 1997), such as riparian
forests, which are among the most important forest ecosystems in the majority of degraded
woodlands and mosaic of savanna landscapes (Natta 2000).
In Benin the rapid changes in land use have led to the progressive destruction and
fragmentation of riparian forests, which provide fertile soil for cultivation, give an
opportunity for irrigation, and shelter a wide range of valuable and scarce plants and animals.
Therefore, they are systematically targeted for illegal selective tree cutting, hunting and
conversion to agriculture. Also, when the riparian forests degrade, forest species become
sparse, leaving the vegetation open for the savanna species to invade. Their rich biological
resources, especially plant species, are disappearing before they can be inventoried and
assessed (Natta 2000). We see that riparian forests can be classified among the endangered
ecosystems of Benin because they are marginal among the wooded vegetation and because of
their advanced state of degradation. In spite of the fact that there is growing recognition of
the ecological, hydrologic, biogeochemical, socio-cultural, economic and aesthetic
importance of RFs in Benin, they have remained insufficiently studied (Mondjannagni 1969,
Paradis 1988, Sokpon et al. 2001) and managed as key vegetation formation for biodiversity
protection (Natta et al. 2002). Riparian forests have often been ignored, or excluded from
vegetation studies in favour to upland forests. RFs flora is poorly known, and no published
account concerning the diversity, ecology and spatial distribution of species in riparian
forests is available in Benin. We often don’t even know which species are present, which one
are abundant, rare and what are the underlying ecological factors that govern their presence
along waterways.
1.2. OBJECTIVES AND
BIODIVERSITY IN BENIN
APPROACH
FOR
THE
ASSESSMENT
OF
RIPARIAN
FORESTS
The overall objective of this research is to contribute to a better knowledge of the flora,
diversity and ecology of riparian forests in Benin. The specific objectives are to (a) compile a
2
Chapter 1: General Introduction
preliminary riparian forests plant species list, (b) assess plant species and ecosystem
diversities, (c) investigate plant communities, (d) clarify the structural and floristic
relationship of riparian forests with adjacent plant communities, and (e) assess the ecology of
certain endangered tree species in riparian forests.
‘Biological diversity’, in short ‘Biodiversity’ is defined as “the variability among
organisms from all sources including, inter alia, terrestrial, marine and other aquatic
ecosystems and the ecological complexes of which they are part. This includes diversity
within species, between species and of ecosystems”. Also Biological resources include
genetic resources, organisms or parts thereof, populations of species or any other component
of ecosystems with actual or potential use or value for humanity (UNCEB 1992).
In this thesis, the term ‘Riparian Forests (or RFs)’ does not refer exclusively to dense
vegetation dominated by woody species in which grasses are virtually absent (i.e. moist
evergreen or dense semi-deciduous tropical forests). Instead, it stands for any topographic
(lowest parts of a landscape occupied by waterways) and edaphic (moist soil dependent)
woodland at stream/river edges, which is made of a mosaic of plant communities, dominated
by woody species in open or closed canopy. Under Benin conditions, open canopies and
degraded woodlands at streamside are dominant. Also, light coming from aside allows many
heliophytic and pioneering species to grow in riparian forests.
The compilation of the RF plant species list is made through the collection of
flowering or fruiting plant growing inside RFs. Field observations coupled with empirical
knowledge about the ecology of each RF species allow to find out which species are most
frequent, rare, valuable, endangered, endemic, or show a specific adaptability to riparian
habitat. Plant species diversity (or phytodiversity) is assessed using several parameters:
species richness and abundance, diversity index (Shannon), Equitability index (Pielou),
species abundance models, differences in site diversities (comparison of Shannon index),
basal area, stem density, life forms and geographical affinity. The diversity of plant
communities was assessed through the Braun-Blanquet phytosociological approach combined
with multivariate analysis of floristic data.
1.3. ORGANISATION OF THE THESIS
Chapter 2 introduces the study area that covers about 70% of the total area of Benin. Chapters
3 to 10 are self-contained and cover introduction, methods used for data collection and
analysis, results, discussion and conclusions. Chapter 3 presents an overview of riparian
forests in Benin, as unique but endangered ecosystem. Chapter 4 and 5 highlight the floristic
diversity and variability of structural parameters within riparian forests. Chapter 6 assesses
plant communities’ diversity using the Braun-Blanquet phytosociological approach, and
ordination and classification of floristic relevés. Chapter 7 deals with the ecology of
Pentadesma butyracea Sabine (Clusiaceae), a tree species rare in Benin. Chapter 8 and 9
focus on the variability of species composition and structural parameters across riparian
forests. In chapter 10, sampling methods are compared to estimate the density of two Khaya
species. A general discussion and conclusions are presented in chapters 11 and 12,
respectively.
3
Chapter 2
STUDY AREA
A.K. Natta
Chapter 2: Study Area
Chapter 2
STUDY AREA
2.1. LOCATION OF THE STUDY AREA IN BENIN
Benin is located at the so-called Dahomey Gap, in which Sudan-type savanna vegetation
extends as far as the sea, through a hiatus in the West African rain forest over some 200 km
from South East Ghana to South East Benin (Guillaumet 1967, Schnell 1976, Onochie 1979,
Whitmore 1990, Martin 1991, Maley 1996). In the entire Dahomey Gap, the present day
pattern of vegetation is very much obscured by the omnipresent impact of dense human
settlements and ever developing agriculture and fishing (Jenik 1994, Kokou & Caballé 2000).
Nevertheless forest islands and riparian forests stretching along the rivers contain many
humid forest elements (Adjanohoun 1968, Kokou 1998, Kokou et al. 1999, Sokpon et al.
2001). The conservation of forest patches and forest species is this dry wedge is of great
ecological and economic importance (Natta et al. 2002). Forest patches of varying size,
including riparian forests, and large trees of Antiaris toxicaria, Diospyros mespiliformis,
Milicia excelsa, on firm ground indicate the potential climax of the semi-deciduous closedcanopy forest (Hall & Swaine 1981, Sokpon 1995). Figures 2.1 and 2.2 present Benin
republic in West Africa and in the Gulf of Guinea, respectively.
An extensive preliminary field reconnaissance of riparian forests in all the ecological
regions of Benin was made. It appears that South of 7º N (i.e. Littoral, Atlantic, Ouémé,
South-Plateau, Mono, and South-Couffo provinces) riparian forests are very degraded and
sites at least 1 ha large (i.e. in which we can installed > 20 plots each of 500 m2) are
uncommon. Therefore to fit with the present-day distribution pattern of riparian forests, and
for practical reasons, several large and least degraded sites North of 7º N were selected for
intensive data collection. This study area stretches from 7º to 12º20 N, and covers about 70%
of the total Benin area. It is subdivided into 3 ecological zones:
1 - The Guinean region: from 7º N (South of Abomey latitude) to about 8º N (at Savè
latitude);
2 - The Sudano-Guinean zone: from about 8º N to about 9º10 N (at PénéssoulouBétérou latitude);
3 - The Sudanian region: from about 9º10 N to the extreme North of the country
(12º20 N at Karimama latitude).
In comparison with the Sudano-Guinean and Sudanian regions, the vegetation in the
Guinean region has been surveyed many more times:
- Coastal zone (Adjanohoun 1966, 1968, de Souza 1979, Paradis 1975a, 1976, 1979, 1981,
1983, FAO 1980, Akoègninou 1984, Adjanohoun et al. 1989, Djego 2000, etc.);
- The complex of lakes, lagoons and inundation valleys (Paradis & Adjanohoun 1974, Paradis
1975a, 1975b, 1979, 1980, 1983, FAO 1980, Paradis & Rabier 1979, Texier et al. 1980,
Adjanohoun et al. 1989, Guinko 1974, Essou 1991, Sinadouwirou 1997, etc.);
- The Niaouli forest (Mondjannagni 1969, Akoègninou 1984, Hountondji 1998, etc.);
- The Lama forest (Paradis & Houngnon 1977, FAO 1980, Adjanohoun et al. 1989, Agbani
2002, etc.);
- The Pobè forest (Akoègninou 1984, Adjanohoun et al. 1989, Sokpon 1995, etc.);
- The Ouémé, Dogo and Kétou regions (apart from the Pobè forest) (Paradis et al. 1978, FAO
1980, Paradis 1983, Akoègninou 1984, Adjakidjè 1984, Adjanohoun et al. 1989, Bossou
2001, etc.);
- The plateau of Allada (Paradis 1983, Adjakidjè 1984, Adjanohoun et al. 1989, etc.);
7
Chapter 2: Study Area
- The Kouffo zone (FAO 1980, Paradis 1983, Akoègninou 1984, Adjakidjè 1984,
Adjanohoun et al. 1989, Amétépé 1997, etc.).
Figure 2.1: Benin in West Africa
Figure 2.2: Benin with its rivers
8
Chapter 2: Study Area
Until recently, riparian forests in the three zones of the study area have rarely been
studied as compared to other vegetation types. This study is concerned with edaphic and
hygrophile forests along fresh water bodies. Therefore the study area does not include the
coastal complex, which consists of several littoral cordons, marshy shoals, mangroves,
lagoons (e.g. Porto-Novo and Ouidah), and lakes (e.g. the Nokoué, Ahémé, Toho, Togbadji,
Azili, Sélé, etc.). Likewise, the ‘barre’ region that adjoins the coastal plain through an uneven
embankment where there is a depression referred to as the ‘Lama forest’ is excluded from the
study area.
2.2. GEOMORPHOLOGY
Geologically, the study area can be divided in three parts: the Northern plateau of South
Benin, the crystalline plateau in the Centre and North, and the Niger River basin in the
extreme North.
The Southern part of the study area corresponds to the Northern plateau (Aplahoué,
Abomey and Kétou) of Southern Benin, which corresponds to the West African continental
terminal. It reaches 100-150 m in altitude, with red sandy argillaceous soil and lateritic crusts
(Paradis 1983).
The major part of the study area, from the North of Bohicon to Kandi, is occupied by
the crystalline plateau, which has tropical ferruginous soils. Within this vast peneplain there
are many residual inselberg landscapes (300 - 400 m high with convex sides) including the
hills of Dassa-Zoumé, Savalou, Savè, Yaoui, Ouari-Maro, Bembèrèkè and Sinendé. The
North West part corresponds to the Atacora region dominated by a series of mountain chains
that reaches an elevation of 600 m.
The Northern part of the study area corresponds to the Niger River basin, which
ranges from the Goungoun village latitude to the Niger River. It is a sandy plateau with an
average altitude of 200 to 250 m. Sedimentary formations predominate, with here and there
cuirass hills of about 270 m high. There are also important valleys of variable lengths with
alluvial soils along the Sota, Alibori and Niger Rivers.
2.3. CLIMATE
Benin is a hot tropical country under two influences: the humid maritime wind that blows
from April to November in South-North direction and the dry continental trade wind, the
Harmattan, that blows from the Sahara desert in the North-South direction. The Harmattan is
the main cause of drought. The mean temperatures are constantly high (25° C) with daily
amplitude below 5° C in the South and 10° C in the North. The largest temperature variations
occur during the days of Harmattan. Nowadays, the country is characterised by a great
irregularity of annual rainfall from one year to another and within each year. In the SouthNorth direction annual rainfall decreases (1300 to 800 mm) while evapotranspiration and
temperatures increase. The climate is of two types (Guinean in the South and Sudanian in the
North) separated by a transitional Sudano-Guinean zone (from about 8° to 9° N).
The South of the study area, (7° to 8° N), has a sub-equatorial (or tropical humid)
climate subdivided into four seasons of unequal length: two rainy seasons (from April to July,
and September to October), and two dry seasons (from November to March, and end of July
to August).
The North, from about 9° to 12° N, is the domain of Sudanian (i.e. dry tropical)
climate with two seasons: a rainy season from May to September-October, and a dry season
from October to April. The annual rainfall varies from 1100 to 800 mm. The rainy season is
shortened as latitude increases toward the Niger River basin in the extreme North. In the
9
Chapter 2: Study Area
Sudanian region higher rainfalls are linked to relief (e.g. Atacora region, Bembèrèkè and
Nikki with 1200 to 1300 mm). The extreme North (Malanville - Karimama) is the domain of
the North-Sudanian climate, with an average annual rainfall of 800 mm, and up to 2000 mm
of evapotranspiration, aggravated by the Harmattan winds.
Between the Guinean and Sudanian regions (about 8° to 9° N), there is a gradual
climatic change characterised by the disappearance of the short rainy season and the fusion of
the two peaks of rainfall that are typical for the sub-equatorial climate of Southern Benin.
Annual rainfall varies from 1100 to 1300 mm and evapotranspiration between 1400 to 1500
mm.
2.4. HYDROGRAPHY
The study area includes three hydrographic basins: 1) the medium and upper parts of the
Ouémé basin in South and Central Benin; 2) the Niger River and its tributaries (i.e. the
Mékrou, Alibori and Sota) in North Benin; and 3) the upper part of the Pendjari basin in
North West Benin.
The river network is presented in Figure 2.2. Every year the rivers are at their highest
level and cause inundation from August to December in the South, and August to November
in the Centre and North; while streams are characterised by daily submersion following
storms or heavy rains. Streams in Benin are generally temporary; meanwhile some have a
permanent regime linked to the relief (e.g. headwaters in the Atacora mountains region). In
valleys, depending on the topography, and volume of water in rivers and streams, inundation
and emersion alternate. Along the Pendjari and Niger rivers there are vast flood plains.
2.5. HUMAN INFLUENCE
From 1979 to 2002, the population of Benin has doubled from 3.3 to 6.7 millions of
inhabitants. The average annual population growth is 3.5 %. The South of the country has the
highest population density with 150 inhabitants per km2, while the Centre and the North have
only 20 inhabitants per km2 (INSAE 1979, 2002).
Farming, tree cutting, as well as a high incidence of fire have reduced the original
woodland to their current state. In recent years the deterioration of vegetation formations
within Benin has increased at an alarming rate (CENATEL 2000). Each year about 100,000
ha of natural vegetation are degraded and the tendency is not decreasing (CENATEL 1995).
In the whole country, farming on marginal lands, deforestation of riparian forests and
savanna woodlands for cotton and yam cultivation, excessive pruning of valuable trees and
uncontrolled bush fires are on the increase. In the South, the present-day pattern of vegetation
is very much obscured by the omnipresent impact of dense human populations.
There are many small relict groves (i.e. sacred forests) and remnant semi-deciduous
forests (e.g. Lama, Pobè, Itchédé, Dangbo, etc.) of varying size scattered all over the area
(Sokpon 1995). In Central Benin, original forests have been cleared and replaced by a mosaic
of savanna and dry forests, while the North with a much drier climate, is unfavourable for
forest establishment, and is therefore dominated by Sudanian savanna. The process of natural
resources degradation leads to the impoverishment of biodiversity due to the loss of wooded
vegetation and their replacement by shrub savanna, which is known to have much less
ecological potential.
10
Chapter 2: Study Area
2.6. VEGETATIONS
As for the whole country, the study area vegetation is characterised by a great variety and a
fragmentation of phytocenoses caused, on the one hand, by climatic, topographic and edaphic
factors and, on the other hand, by human influence on the environment.
The South of the study area prolongs the derived savannah (Jones 1945) and Southern
Guinean zone of Nigeria (Keay 1959). It is the domain of:
- Humid semi-deciduous forests with Antiaris toxicaria, Strombosia glaucescens,
Triplochiton scleroxylon and Terminalia superba (Northern part of the Guinean sub-zone IIA
of Adjanohoun et al. 1989);
- Dry semi-deciduous forests and derived savanna with Anogeissus leiocarpus, Afzelia
africana and Lonchocarpus sericeus (Northern part of the Guinean sub-zone IIB & C of
Adjanohoun et al. 1989);
- Dry forests with Anogeissus leiocarpus and Daniellia oliveri (Guineo-Sudanian transitional
sub-zone IIIA of Adjanohoun et al. 1989);
- Woodland savanna (with Ceiba pentandra, Milicia excelsa and Daniellia oliveri), and tree
and shrub savanna (with Daniellia oliveri, Elaeis guineensis and Lophira lanceolata) (FAO
1980);
- Savanna on hills (with Afrotrilepis pilosa, Ficus abutifolia, Hildegardia barteri and
Aeollanthus pubescens) (Yédomonhan 2002).
- Tree and shrub savanna (with Anogeissus leiocarpus, Vitellaria paradoxa, Daniellia oliveri,
Isoberlinia doka and Parkia biglobosa) with high incidence of cultivation (FAO 1980).
The Centre of the study area is a prolongation of the Northern Guinean zone of
Nigeria (Keay 1960) and a small part of the dry semi-deciduous forest, the fire sub-type of
Hall & Swaine (1981). It is the domain of:
- Patches of semi-deciduous forest with a few big stems of semi-deciduous trees (e.g. Antiaris
toxicaria, Milicia excelsa, Cola spp., Khaya senegalensis and Celtis spp.) in the Bantè,
Ouèssè and Bassila districts (FAO 1980). Particularly, a facies of dry semi-deciduous forest
is seen in the Sudano-Guinean region of Bassila with Khaya grandifoliola, K. senegalensis,
Albizia spp., Cola gigantea, Antiaria toxicaria, Milicia excelsa, Erytrophleum guineensis,
Anogeissus leiocarpus, etc. (variant of sub-zone IIB of Adjanohoun et al. 1989);
- Open forests with Isoberlinia spp., Pterocarpus erinaceus, Khaya senegalensis and Monotes
kerstingii (Centre and North of the sub-zone IIIB of Adjanohoun et al. 1989);
- Woodland, tree and shrub savanna (with Anogeissus leiocarpus, Vitellaria paradoxa,
Daniellia oliveri, Isoberlinia doka and Parkia biglobosa) (FAO 1980).
- Tree and shrub savanna with high incidence of cultivation (FAO 1980).
- Savanna on hills South of Alafiarou, and in the Kouffé Mountains reserve forest (FAO
1980).
The North of the study area is a prolongation of the Sudanian zone of Nigeria (Keay
1949) and the plains of Northern Togo (Ern 1979). The Sudanian region (or dry continental
zone, FAO 1980) is the domain of typical savannas, or dry forests on deep or rich soils:
- Open forests with Isoberlinia spp., Pterocarpus erinaceus, Khaya senegalensis and Monotes
kerstingii (Northern part of the sub-zone IIIB of Adjanohoun et al. 1989);
- Sudanian savannas with Afzelia africana, Burkea africana, Anogeissus leiocarpus, Parkia
bigblobosa, Vitellaria paradoxa, Terminalia spp., Daniellia oliveri, Combretum spp.,
Crossopteryx febrifuga, Acacia spp., Balanites aegyptiaca, Sclerocarya birrea, etc. and
numerous grasses (sub-zone IVA and IVB of Adjanohoun et al. 1989, FAO 1980); The subzone (IVB) includes the Atacora mountains chain;
- Savannas and dry forests with Combretaceae of the Gourma (or Pendjari) plain in North
West Benin, along the Togo and Burkina-Faso borders (sub-zone IVC for Adjanohoun et al.
11
Chapter 2: Study Area
1989). In the West of the Pendjari complex, there is a facies of tree and shrub savanna with
Anogeissus leiocarpus, Combretum spp., Acacia spp., Balanites aegyptiaca and Ziziphus
mauritiana (FAO 1980);
- Tree and shrub savanna with high incidence of cultivation (FAO 1980).
The extreme North of the study area (very dry continental zone with 6-7 dry months,
FAO 1980) is dominated by:
- Woodland, tree and shrub savanna (with Anogeissus leiocarpus, Combretum spp., Acacia
spp., Balanites aegyptiaca and Ziziphus mucronata).
- Tree and shrub savanna with high incidence of cultivation.
- Saxicolous tree and shrub savanna (with Adansonia digitata, Combretum spp. and Acacia
spp.).
- Shrub savanna and prairie (with Echinochloa stagnina, Vetiveria nigritana, etc.) in the
Niger River inundation plain.
The annual crops (maize, cassava, yam, millet, sorghum, cotton), the perennial plantations
(oil-palm from the last two centuries onwards in the South), the gathering of firewood,
production of charcoal, selective cutting of valuable trees and uncontrolled bush fire, have
brought about the destruction of forests and woodlands and replaced these by tickets and
grassy phytocenoses. The result is a mosaic of savanna, bush fallow and cultivated areas in
which remnant forest and savanna species are frequently seen. Meanwhile, the important
hydrographic network still allows the presence of many riparian forests, widely distributed all
over the country.
2.7. SITES OF DATA COLLECTION
Representative sites of the largest and least degraded riparian forests were chosen along
several rivers and streams in the Guinean region (Samiondji and Bétécoucou), the SudanoGuinean zone (Toui-Kilibo, Idadjo, and Pénéssoulou) and the Sudanian region (Bétérou,
Onklou, Daringa, Ouaké, Affon, Yarpao, Boukombé, Toukountouna, Pendjari Biosphere
Reserve, Kandi, Gbèssè and Malanville; see Figure 4.1).
12
Chapter 3
RIPARIAN FORESTS, A UNIQUE BUT ENDANGERED ECOSYSTEM IN BENIN
Published as:
NATTA A.K., SINSIN B. & VAN DER MAESEN L.J.G.: Riparian forests, a unique but
endangered ecosystem in Benin. Notulae Florae Beninensis 4. – Bot. Jahrb. Syst.
124: 55-69. 2002. ISSN 0006-8152.
Chapter 3: Riparian forests, a unique but endangered ecosystem in Benin
Chapter 3
RIPARIAN FORESTS, A UNIQUE BUT ENDANGERED ECOSYSTEM IN BENIN
Notulae Florae Beninensis 4
NATTA A.K., SINSIN B. & VAN DER MAESEN L.J.G.: Riparian forests, a unique but endangered
ecosystem in Benin. – Bot. Jahrb. Syst. 124: 55-69. 2002. ISSN 0006-8152.
ABSTRACT
Riparian forests are often small in area, but are of extreme ecological and economic value for
local people. The interest of riparian forests lies in their resources: basically fertile and moist
soils, water, wood and non-timber forest products that are utilised by neighbouring
populations to satisfy their basic needs and as source of income. Their abusive overuse is
widespread and growing in Benin. As a result rich biological resources, especially plant
species, are disappearing before they can be inventoried and assessed. Most recent
management plans for protected areas barely consider the conservation of riparian forests.
The forest law in Benin recognises their uniqueness, but several problems arise in the
implementation of the rules in these particular areas. The most common and widespread tree
species in riparian forests in Benin are Pterocarpus santalinoides, Cola laurifolia, Syzygium
guineense, Berlinia grandiflora, Elaeis guineensis, Manilkara multinervis, Xylopia
parviflora, Dialium guineense, Diospyros mespiliformis and Parinari congensis. Waterways
and their forested banks are rich in birdlife, and serves as a focal point for primates and
animals of many kinds. Site-specific studies at ground level are essential to assess riparian
systems because their narrow linear shapes generally require data on plant community
structure, floristic composition and animal presence. Details of vegetation layers under the
dominant trees cannot yet be detected by remotely sensed data. This paper presents an
overview of the biological diversity of riparian forests in Benin and discusses several issues
associated with their protection and conservation.
Key words: riparian forests, flora and fauna diversity, biodiversity protection, Benin.
3.1. INTRODUCTION
The current state of knowledge about species and ecosystems in the tropics is far from
complete. Detailed knowledge about the ecology of plants, floristic composition and diversity
of plant communities, that would be the obvious first step in understanding and conserving
them, is still incomplete or lacking. It is generally agreed that species extinction is largely
related to the reduction and fragmentation of their habitats (SWAMINATHAN 1990). Therefore
the protection of species can best be done through protecting habitats. In order to formulate
policies or take measures to tackle the decline of biodiversity it is important to identify the
ecosystems, which have a high diversity value, and are endangered. It is needed to investigate
their ecology, assess their current diversity and understand the factors and processes of their
impoverishment. In Benin, savanna is the dominant wooded vegetation. The narrow strips of
often fragmented woodland that border streams and rivers, - the riparian forest - are good
examples of such shrinking and endangered ecosystems.
The rapid changes in land use in Benin have led to the destruction and fragmentation
of riparian forests. They are classified as endangered ecosystems because they are marginal
among the wooded vegetation and of their high degradation state. Their importance in the
functioning of river/streams ecosystems and biodiversity protection is recognised by several
15
Chapter 3: Riparian forests, a unique but endangered ecosystem in Benin
authors (ADJANOHOUN 1965; FORMAN & GODRON 1986; BAKER 1990; TABACCHI et al. 1990;
POLANSKY 1994). Although small in size, riparian forests are important in the conservation of
numerous plants and animals. Usually situated on fertile and moist soils near water, their
economical interest lies in the supply of wood and non-timber forest products.
Despite their wide distribution in Benin and large economic and ecological value,
riparian forest ecosystems have, until recently, remained largely ignored and unmanaged as
key ecosystems in biodiversity protection. This is partially due to the insufficient scientific
knowledge about species occurring in riparian forests, which hinders appropriate planning,
and conservation initiatives dedicated to them. Although the government of the Republic of
Benin decreed a new forest law in 1993, the implementation of the law related to the
protection of riparian vegetation, especially in non-protected areas, leaves to be desired.
3.2. FORESTS IN BENIN
Centuries of intense human activity accompanied by a drying climate resulted in the loss of
most of Benin’s closed forests long ago. Small patches of moist forest remain today only on
moist soils, or as numerous sacred forests and riparian forests. FAO estimated in 1980 that
only 470 km2 (only 0.4% of Benin’s land area) remained under natural cover of closed broadleaved forest. Of this, only 140 km2 was considered to be undisturbed. Flora and fauna
reserves occupy 21,440 km2 (PGRN/IUCN 1994). Benin’s flora is estimated to have close to
3,000 species (PFB 1997), which is relatively poor in comparison with more humid
neighbouring countries.
Benin is said to be a non-forestry country because tropical rain forests are absent and
dense semi-deciduous forests are rare among wooded vegetations (SOKPON 1995). In fact,
Benin is located at the discontinuity of the tropical forests zone in West Africa, ‘the Dahomey
gap’, which is the product of topographic, oceanographic and climatic interactions (JENIK
1994). For ERN (1988), this gap includes essentially the drier types of the Guineo-Congolian
forest belt in South-Eastern Ghana, Southern Togo and parts of Southern Benin. As a result
there are no evergreen tropical forests or rain forests in Benin. At the same time there is an
increasing demand of wood material to satisfy the need for fuel and construction. In the South
of Benin, increasing population pressure on natural ecosystems has led to rapid degradation of
woodlands. The remaining forest patches include the Lama and Pobè relict forests (ERN 1988;
SOKPON 1995). In Central Benin, original forests have been cleared and replaced by a mosaic
of savannas and dry forests. The North, with a much drier climate, is unfavourable for forest
establishment, and is therefore dominated by more thinly wooded Sudanian savanna.
Meanwhile, the important hydrologic network allows the presence of many riparian
forests widely distributed all over the country. The hydrographic network includes the
Atlantic Ocean watershed with the Ouémé, Couffo and Mono Rivers and their tributaries,
while the Niger River, takes in the waters of the Sota, Alibori and Mékrou Rivers. In the
North West of Benin, the Pendjari stream starts in the Atacora mountains and ends in the
Volta River in Ghana. In the South there is an important complex of coastal lagoons, lakes
and swamps areas. Most streams are seasonal.
3.3. DEFINITION AND DESCRIPTION OF RIPARIAN FORESTS
Riparian (from the Latin ripus, bank; as of a river) points to the banks and other terrestrial
areas adjacent to watercourses, fresh-water bodies, and surface-emergent aquifers whose
transported waters provide soil moisture in excess of that otherwise available locally,
sufficient to support a mesic (moist soil dependent) vegetation distinct in structure and/or
floristics from that of the contiguous and more xeric uplands (WARNER 1979). The regional
16
Chapter 3: Riparian forests, a unique but endangered ecosystem in Benin
and temporal dimensions of climate, geomorphology, topography and biogeography mould
both structure and floristics of the riparian systems (WARNER & KATIBATH 1980).
Different authors have used several terms to designate riparian forests: ripicole forests
(GREEN 1979a, DEVINEAU 1984), gallery forests (ADJANOHOUN 1965; DEVINEAU 1975, 1976,
1984, GREEN 1979b; KELLMAN et al. 1994; NATTA 2000; SOKPON et al. 2001); riverbank
forests (GREEN 1979a), riverine forests (ADJANOHOUN 1965; DEVINEAU 1984; VARTY 1990;
MONNIER 1990; MEDLEY 1992; SOKPON 1995; NATTA 2000; SOKPON et al. 2001), riverine
woodlands (VAN ETTEN 1999). Other authors (e.g. BAKER 1990; TABBACHI et al. 1990;
LEINARD et al. 1999) have used riparian vegetation in the sense of any plant community
occurring along rivers or streams. The inter-African agreement on the definition of tropical
vegetation types at Yangambi (ref. 1956) classified riparian forests among edaphic (soil
dependent) forest formations (TROCHAIN 1957).
Riverine forests are dependent on river processes including inundation and transport
of sediments. They occur in narrow strips along river courses, and are created and maintained
by groundwater seepage from the river and by periodic flooding. In the Jubba Valley
(Southern Somalia) riverine forests occur in small isolated patches rarely more than 300 m
wide (VARTY 1990). In the same region, MEDLEY (1992) described the Tana riverine forests
(in Kenya) as fragmented patch-mosaic forests within a narrow corridor defined by the
groundwater hydrologic regime, floods and meandering pattern of the stream.
In the West-African region, MONNIER (1990) described the riverine forests as
deciduous forest characterised by a linear shape, a structural originality and a microclimatic
specificity. In the forest-savanna ecosystems of Côte d’Ivoire, ADJANOHOUN (1965) qualified
riverine forests at both riversides as strips dominated by big trees and liana species, belonging
almost exclusively to dense forest. They appear to be one of the most important ecosystems
in the majority of savanna landscapes where they underline the course of rivers.
In the Benin context, where savanna is the most dominant ecosystem, we suggest to
use the term ‘riparian forest’ for any forest type occurring at river banks and along streams.
This may include semi-deciduous forest, dry deciduous forest and woodland savanna located
at river or stream banks with more or less sharp limits in respect to adjacent plant
communities. Although these plant communities, dominated by trees, are often fragmented
and degraded, there are still good examples of undisturbed riparian forests (e.g. at Yarpao,
Pénéssoulou and the Bondjagou forest in the Pendjari National Park). In many regions high
human pressure has reduced the width of riparian forests up to a single tree, so they are often
extremely narrow and quite linear (strip-like) in configuration. Floristic data and the
distinction between core and edge species may be important to describe the limit of riparian
forests when the structure is uniform beyond the riparian forest.
The main characteristics of typical riparian forest ecosystems can be summarised as:
- Location along rivers and streams in close contact or not with the water flow;
- Combination of mesic terrestrial vegetation, dependent animal life, and local microclimate
and dynamic water and soil processes;
- Multi-layered vegetation structure dominated by moist-dependent trees, which form a
distinctive physiognomy in the landscape, particularly in the savanna region. Associated with
these mesic trees is a host of mesic understorey, shrub and ground cover species;
- Floristic composition often includes the plant species of the region that require most
moisture.
We do not consider wetland vegetations occurring in swamp areas (shrubs, grass
savanna, marsh, mangrove vegetation, etc.) as riparian forests. The distinction between
several riparian forest stands can be expressed in plant communities on the basis of their
physiognomic, floristic, and structural characteristics through phytosociological surveys.
17
Chapter 3: Riparian forests, a unique but endangered ecosystem in Benin
These are compiled on the basis of abundance-dominance by allocating a coefficient to each
plant species according to the ratio of species density to the area surveyed.
3.4. ECOLOGICAL AND ECONOMIC IMPORTANCE OF RIPARIAN FOREST ECOSYSTEMS
Biodiversity is often seen as an indicator of the well-being of ecological systems, as well as a
useful tool in environmental monitoring and assessment. The role played by biodiversity is
strongly related to natural ecosystems, which are often the only viable locations for the
conservation of certain natural variability (SALLEY & NIENG 1997). Water and vegetation are
central components of tropical ecosystems (OTTO 1993), which in turn are key repositories of
global biodiversity. In the sub-Sahara savanna region, where small patches and fragmented
woodlands often represent remnant forests, riparian forest stands are sites of important
ecological and economic values. Also, resource managers understand the interrelationships
between the four components of an ecological site: landscape position (physiography),
climate, soils and vegetation. These basic relationships must still be considered for riparian
areas, as must landform, fluvial geomorphology and hydrology. Thus soils, fluvial
geomorphology and climate are the most reliable indicators of riparian vegetation potential
(LEINARD et al. 1999).
More than other vegetation types, riparian forests are well-known for their role in
controlling water sources, watersheds, water runoff and quality, mineral nutrient flows
(PIEGAY 1997), stream bank erosion (VAN ETTEN 1999). They influence sedimentation, bank
stabilisation, flood and hydrologic regimes, pollution, water shading and cooling, sediment
filtering, aquifer recharge, and transport of sediments (MEDLEY 1992). According to THOMAS
(1996), typical riparian forest tree species are dependent on river flows, a shallow aquifer,
and the community and population structure of riparian forests are related to spatio-temporal
patterns of flooding. They are under the influence of fluvial geomorphology processes
(BAKER 1990). Also they usually act as routes for movement of terrestrial plants and animals
across the landscape (FORMAN & GODRON 1986) and serves as the most suitable area for
plant species adapted to a moist climatic regime (MEDLEY 1992).
On riparian systems in general, the United States Council on Environmental Quality
stated ‘no ecosystem is more essential to the survival of fishes and wildlife’ (WARNER &
KATIBATH 1980). Although small in size (VARTY 1990; MONNIER 1990), riparian forests are
important in the conservation of a large range of plants and animals. Not only do they
constitute a natural habitat or the last refuge for many species, but also they contain many
endemic species and extinction-menaced species (ROGGERI 1995). Some authors have
discussed the necessity to preserve plant diversity in riparian forests as a means to protect
certain animals. In the Tana River National Primate Reserve of Kenya, Medley (1992)
showed that the preservation of key resources in riverine forest, such as the endangered Tana
River red colobus (Colobus badius rufomitratus) and crested mangabey (Cercocebus
galeritus galeritus), should be coupled with protection of forest heterogeneity that
characterises this dynamic landscape.
3.5. DESTRUCTION AND LOSS
Misuses of riparian systems are global, chronic and accelerating (WARNER & KATIBATH
1980). Decades or centuries of human influence are usually seen as the main causes that have
reduce riparian forests in both size and structural complexity. The result is a poorer
biodiversity and landscape impoverishment. Their rich biological resources, especially plant
species, are disappearing before they can be inventoried and assessed (NATTA 2000). As a
consequence, this affects ecological equilibrium especially at local level (MONNIER 1990).
18
Chapter 3: Riparian forests, a unique but endangered ecosystem in Benin
The interest of riparian forests lies in their resources: basically fertile and moist soils,
water, wood and non-timber forest products that are utilised by neighbouring populations to
satisfy their basic needs and as source of income. The fertile soil, the presence of water and
moist soils, the supply of sediments from inundation, allow certain human activities such as
shifting cultivation, irrigation and cattle ranching. Also the presence of a variety of plants and
animals related to water regime, such as valuable and rare tree and animal species, enables
selective tree cutting (for timber, traditional fuel gathering for sale, honey collection) and
hunting. The collection of all the seeds of valuable tree species (e.g. Pentadesma butyracea,
SINADOUWIROU 2000) is an additional threat to the survival of their population. Also
disruption of watershed vegetation, damming of watercourses and/or diversion of stream flow,
excessive lowering of aquifer levels through pumping, channelisation and levee construction
on watercourses often cause significant damage to riparian systems. As a result indigenous
multilayered plant communities have been completely removed in many parts and replaced by
weeds, open field, shrubs or grass savanna with much less ecological value and economic
potential.
3.6. PLANT COMMUNITIES AND FLORISTIC COMPOSITION OF RIPARIAN FORESTS IN BENIN
Following the work of ADJANOHOUN et al. (1989), HOUINATO et al. (2000) divided Benin in
ten phytogeographic districts with distinctive geomorphologic, plant communities and flora.
For each district an overview of riparian forest flora is presented.
The coastal region is the domain of marshlands, lakes, lagoons, coastal thicket or
scrub and a number of small relict groves. The present-day pattern of vegetation is very much
obscured by the omnipresent impacts of dense human populations. This flat region lacks
typical riparian forests, as defined above. The Pobè district is the domain of dense semideciduous forests and is a continuation of the lowland rain forest zone of Nigeria. SOKPON
(1995) studied the structure and floristic composition of plant communities in the protected
forest of Pobè, typical of this South-Eastern part of the country. He identified a riverine forest
dominated by Cleistopholis patens, Ficus mucuso, Elaeis guineensis and Cercestis mirabilis.
The Ouémé and Kouffo districts prolong respectively the derived savanna and the
Southern Guinea zones of Nigeria. Here Pterocarpus santalinoides, Cola laurifolia, Parinari
congensis, Antidesma venosum, Napoleonaea vogelii, Syzygium guineense, Dialium
guineense and Cynometra megalophylla are the dominant tree species in riparian forests. The
Zou district is a transition zone between the Guinean and Sudanian climates. Riparian forest
physiognomy is similar to the dense semi-deciduous forests. The most frequent species are
Berlinia grandifolia, Elaeis guineensis, Hexalobus crispiflorus, Pouteria alnifolia, Cola
gigantea, C. millenii, Lecaniodiscus cupanioides, Napoleonaea vogelii, Pterocarpus
santalinoides and Uvaria chamae.
The South Borgou district corresponds to the North Guinean zone of Nigeria and the
riparian forest flora here contains numerous tree species of lower and much wetter latitudes
such as: Berlinia grandifolia, Parinari congensis, Detarium senegalense, Diospyros
mespiliformis, Dialium guineense, Khaya grandifoliola, K. senegalensis, Millettia thonningii,
Albizia zygia, Albizia glaberrima, Trilepiseum madagascariense. The North Borgou district,
which prolongs the Sudanian zone of Nigeria, is characterised by the omnipresence of
riparian forests similar in physiognomy with those of South Borgou but with a different flora.
Dominant species are Syzygium guineense, Uapaca togoensis, Berlinia grandifolia, Brenadia
salicina, Khaya senegalensis, Elaeis guineensis, Manilkara multinervis, Vitex doniana,
Mimusops andongensis, Diospyros mespiliformis, Synsepalum passargei, Fadogia agrestis,
Ficus spp., Celtis toka, Borassus aethiopum and Raphia sudanica.
19
Chapter 3: Riparian forests, a unique but endangered ecosystem in Benin
The Atacora chain oriented North-North-East/South-South-West is characterised by
hygrophile riparian forests at the foot of the hills. Their structure and composition present
some similarities with the ones of North Borgou. Common species are Syzygium guineense,
Uapaca togoensis, Berlinia grandifolia, Brenadia salicina, Pentadesma butyracea,
Chrysobalanus atacoriensis, Eriocoelum kerstingii, Khaya senegalensis, Ficus spp.,
Diospyros mespiliformis, Anogeissus leiocarpus and Vitex doniana.
The Pendjari district corresponds to the North West of the Atacora province. It
includes the major part of the Pendjari National Park and the Western plains up to the Togo
Border. In the Pendjari National Park, GREEN (1979b) identified several plant associations
among riparian forests. The most important are Cola laurifolia, Morelia senegalensis and
Syzygium guineense association at its maximum development in the unique and species rich
riparian forest of Bondjagou; Khaya senegalensis, Diospyros mespiliformis and Anogeissus
leiocarpus along the Pendjari stream in the Atacora chain; Celtis integrifolia, Kigelia
africana and Diospyros mespiliformis, Garcinia livingstonei and Mitragyna inermis along the
Pendjari stream North of the Bondjagou forest; Anogeissus leiocarpus, Combretum nigricans
sometimes with Terminalia glaucescens along several streams in the park; Anogeissus
leiocarpus and Crossopteryx febrifuga along small streams around the Bondjagou forest, in
the centre of the park and at the foot of the hills. In the Western plains towards the Togo
border, riparian forests are dominated by Anogeissus leiocarpus, Mitragyna inermis, Elaeis
guineensis, Ficus spp., Sarcocephalus latifolius, Vitex doniana and Terminalia macroptera
(NATTA 1997).
In the Centre and North of the W National Park in Northern Benin, riparian forests
along streams are the most important ecosystems in the majority of shrub savanna dominated
by Mimosoideae and Combretaceae. Here, riparian forests have the structure of woodland
savanna less than 15m high along the Mékrou, Alibori and Pako streams. They are extremely
narrow and quite linear with often only a single tree at stream banks. The most frequent
species are: Cola laurifolia, Syzygium guineense, Morelia senegalensis, Pterocarpus
santalinoides, Combretum lecardii, Crateva adansonii, Borassus aethiopum, and Vitex
chrysocarpa. Oxytenanthera abyssinica is sometimes seen in the Mékrou and Alibori beds
(PGRN/IUCN 1994). GREEN (1979a, 1979b) also identified several plant associations along
the Niger river and streams of the W National Park such as Khaya senegalensis, Diospyros
mespiliformis and Borassus aethiopum; Anogeissus leiocarpus, Combretum nigricans with
sometimes Sclerocarya birrea in the North of the Park along the Mékrou stream; Anogeissus
leiocarpus and Crossopteryx febrifuga along the Niger river, the Mékrou stream and small
streams; Cola laurifolia, Morelia senegalensis, Syzygium guineense, Anogeissus leiocarpus,
Feretia apodanthera, Anogeissus leiocarpus and Balanites aegyptiaca.
Trilepiseum madagascariense, Cleistopholis patens, Albizia glaberrima, Antidesma
venosum, Sterculia tragacantha, Cynometra megalophylla, Drypetes floribunda and
Syzygium guineense are common species of semi-deciduous forests on ferralitic soils or
vertisols in the Pobè and the Ouémé districts South of 7º10 N, (AKOÈGNINOU 1984; SOKPON
1995). These trees are also found almost exclusively along rivers and streams at higher
latitudes. Such is the case in the Toui-Kilibo, Pénéssoulou, Ouari-Maro, and Ouémé
Supérieur protected areas. Dialium guineense, a typical species of the plateau and tops of
slopes in dense semi-deciduous forests (SOKPON 1995), is found exclusively in riparian
forests between 7º20 and 11º N. Also Elaeis guineensis is widespread in the South, but is
found in gallery forests in the Centre and North of the country. Syzygium guineense is always
found in gallery forests as far as in the W National Park. These examples tend to confirm the
hypothesis of FORMAN & GODRON (1986) that riparian forests usually act as routes for
movement of plants across the landscape, especially at less moist latitudes.
20
Chapter 3: Riparian forests, a unique but endangered ecosystem in Benin
Despite decades of neglect, the Mounts Kouffé reserve still contains dozens of small
patches of riparian forests and dense moist forests. These remnants appear undisturbed
internally and therefore have a very high biodiversity value (PGRN/IUCN 1994). GREEN
(1979a) and PGRN/IUCN (1994) have recommended to extend the South-Eastern limit of
Pendjari National Park beyond the Atacora chain to protect the riverine forest and adjacent
mountains, thereby increasing biodiversity by adding rich habitats and beautiful landscapes to
the park. Also a buffer zone should be created along the Ouémé River, including its forested
banks to conserve the riverine forest and to maintain a physical corridor between the Ouémé
Supérieur and Ouari-Maro forest reserves.
Summing up, the most common and widespread tree species in riparian forests in
Benin are Pterocarpus santalinoides, Cola laurifolia, Syzygium guineense, Berlinia
grandiflora, Elaeis guineensis, Manilkara multinervis, Xylopia parviflora, Dialium guineense
Diospyros mespiliformis and Parinari congensis.
3.7. WILDLIFE IN RIPARIAN FORESTS
The Ouémé river and its forested banks are rich in birdlife, and serves as a focal point for
primates and animals of many kinds. The Mounts Kouffé with its contiguous Ouari-Maro and
Ouémé Supérieur protected forests are said to have the highest ranking of biodiversity as
compared with all the other protected areas in Benin (PGRN/IUCN 1994). It is very likely
that the 215 bird species and the potential 150 additional ones, reported by the PGRN/IUCN
mission of 1994, have riparian forests as their main habitat.
In the North of Benin and the South of Burkina Faso, flood plains, shallow lakes and
riparian forests are the principal habitats of the 278 bird species found in the Pendjari and Arli
National Parks (ROGGERI 1995). Riparian forests are reportedly the main habitat of certain
animals such as the bushbuck (Tragelaphus scriptus), red-flanked duiker (Cephalophus
rufilatus) and vervet monkey (Cercopithecus aethiops tantalus) (LEVAUX 1990). Also in the
protected areas, certain animals depend on the dispersion of fruits of riparian forest trees. It is
the case e.g. of elephants and baboons that use Borassus aethiopum seeds in the Pendjari
National Park.
3.8. THE USE OF REMOTE SENSING IN DETECTING AND MAPPING RIPARIAN FORESTS
Remotely sensed data and geographical information systems (GIS), which deal with spatial
data, have considerable utility in acquiring vital information especially on vegetation. They
have a great potential in providing up-to-date and accurate information that can be used for
plant communities management and monitoring activities.
We have conducted a study in the Toui-Kilibo protected area (Central Benin) to assess
qualitatively and quantitatively the usefulness of satellite imagery (Spot XS image) for
identification and classification of land cover types in riparian forests and neighbouring
vegetation types (see NATTA 2000). It was possible to differentiate between several land
cover types from the aerial photographs and satellite images along the river, but not along
streams. Although results seem promising in terms of spectral separability of gallery (along
streams) and riverine (along rivers) forests, differences in tree species structure and floristic
composition, as observed in the field, could not always be made out from the remotely sensed
data.
The coarse spatial resolution of SPOT XS image (20 m) and the unknown spectral
signatures of species or group of species have lowered the accuracy records. The great spatial
resolution of predictive vegetation maps obtained from digital image classification exclude
small patches and narrow strips of vegetation such as riparian forests, and therefore limit the
21
Chapter 3: Riparian forests, a unique but endangered ecosystem in Benin
use of the current civilian satellite imagery in riparian forests study. WARNER & KATIBATH
(1980) have already found that satellite imagery allowing vegetation mapping with a scale
range between 1:80,000 to 1:35,000 was only of limited use; meanwhile they found that an
aerial photograph at 1:6,000 is the most useful for detailed riparian system inventory. In our
case study we found that even aerial photographs with a scale of 1/20,000 or larger are useful
to identify and map broad stands of riparian forests in a savanna landscape.
Because data on plant communities’ structure (dominant trees, understorey, shrub
cover and ground cover), floristic composition, as well as associated animal life, are generally
required ground level surveys are an essential component of any detailed riparian system
assessment. Site-specific studies in the field are required to assess riparian forests because
their narrow linear shapes generally require data on plant community structure, floristic
composition and animal presence. Details of vegetation layers under the dominant trees
cannot yet be detected by remotely sensed data.
3.9. PROTECTION, AND RESTORATION OF RIPARIAN ECOSYSTEMS RESOURCES
In the new Benin forest law (no. 93-009 of July the 2nd of 1993, Ref. RB 1993), clearance of
wood and shrubs is not allowed within 25 m at both sides of any water course and stretch of
smooth water (article 28). This limit seems to be too short, in reality stands of riparian forests
can be wider than 50 m. A reasonable limit, without any clearance or selective tree cutting
from watercourses, of 100 m will provide sufficient guarantee for the protection of
watercourses and riparian resources.
In the management plans of Goun-Goun, Sota and Goroubi, Toui-Kilibo, Ouémé
Supérieur and Pénéssoulou forest reserves, gallery forests will be left uncut, and rare species
outside the gallery forests will not be cut either. In the meantime, foresters and rural
extensions workers must give the good example and promote local people’s rights and duties
specified by the forest law, especially to help local people respect the restrictions concerning
protected or valuable trees (e.g. Milicia excelsa, Khaya senegalensis, and K. grandifoliola at
stream sides).
Still several questions remain with regard to the implementation of the forest law: Do
local people respect the restrictions on riparian forest utilisation? Are alternatives provided to
neighbouring people, particularly farmers, not to overuse or destroy riparian forests (e.g.
when seeking for fertile soils or water for irrigation)? Are private forestry developers aware
of protection of riparian forests in unprotected areas? How far can foresters enforce the law in
applying the punishments to those who illegally destroy riparian forest resources? Can any
suggestions be given to improve the efficiency of the law in protecting riparian forests
resources in the terrain?
To protect, improve, and restore in a sustainable way the riparian vegetation
resources, several objectives have to be addressed:
- identify the major riparian forests in the country and determine their historical extent;
- make inventories of their floristic and faunistic resources;
- identify causes of riparian forests destruction and threats to their resources maintenance,
with special attention to local people’s views;
- implement activities for the sustainable conservation of riparian forests in partnership with
neighbouring inhabitants.
Information needed
To achieve these objectives, data should be collected on different topics related to riparian
forest such as:
22
Chapter 3: Riparian forests, a unique but endangered ecosystem in Benin
- location, extent, climate, soils types, river/stream bank conditions;
- vegetation structure, floristic composition, and vegetation condition (species distribution,
abundance, size, reproduction strategies, evolution trends, etc.);
- fauna with their characteristics and ecological requirements;
- river and /stream water quality, hydrologic characteristics;
- adjacent land uses and their impacts on riparian resources (effects of farming, tree cutting,
fire, hunting, channelisation, hydrologic manipulations and other discernible threats);
- measures to be taken for riparian resource protection, improvement, and restoration
(possible solutions to the major threats at local, district, provincial and national levels,
restoration potential for the riparian systems).
Stakeholders
Several stakeholders have to be involved in this task: the forest and natural resource
management authorities, foresters, projects, research institutes and Non-Governmental
Organisations (NGO’s) dealing with natural resources protection, local population and forest
resources users (e.g. timber gatherers, hunters), etc. Effective conservation actions can best
be achieved if the local residents are either collaborating or in control. Therefore initiatives
have to be taken to rise local people’s awareness concerning their interests/benefits in
protecting riparian forests (landscape beauty, watershed protection, water sources protection,
conservation of valuable or rare floristic and faunistic species). This can be done through the
empowerment of local forest management committees through participatory approaches.
The implementation and improvement of the forest law are primarily the task of
foresters in the terrain. The application of the law and good collaboration with local natural
resources management committees, if effective, will surely provide positive effects on
forested areas including riparian systems.
Role of the Forest and Natural Resource Management (NRM) Department and
Research Institutes in the implementation of the forest law concerning riparian forests
Their attributions could be:
- Definition of thematic objectives (e.g. riparian forests fauna, flora, river and stream
hydrology, reforestation, socio-economic aspects, etc.);
- Setting up of criteria for the assessment and monitoring of riparian forests resources;
- Creation of a data-base on riparian forests in the framework of any NRM program;
- Monographs of specific species (e.g. Chrysobalanus atacoriensis, Pentadesma butyracea,
etc.) or resource.
Role of research projects and NGOs dealing with natural resources protection
Projects and NGOs dealing with natural resources protection are certainly very important for
the protection, improvement, and restoration of riparian vegetation resources. Their actions
could focus on:
- Inventory of natural resources in riparian forests;
- Vulgarisation of the forest law regulations on riparian forests and effective actions to fight
against the depletion of riparian forest resources in collaboration with local people;
- Empowerment of local forests management committees through participatory approach;
23
Chapter 3: Riparian forests, a unique but endangered ecosystem in Benin
- Alternatives to reduce the depletion of riparian forest resources (e.g. cheap and easy to use
irrigation systems for farmers, agroforestry in uplands, appropriate techniques to improve soil
fertility, beekeeping, etc.);
- Experimentation on husbandry of rare or native riparian forest species;
- Participatory reforestation of degraded riparian forests and revegetation of stream banks
with rare or native riparian forest species (e.g. Khaya grandifoliola, K. senegalensis,
Pentadesma butyracea). Exotic plants (e.g. teak, Gmelina arborea, Anacardium occidentale
and Mangifera indica) should not be avoided;
- Valorisation of riparian forests in National Parks and forest reserves for ecotourism,
recreation, research and sport fishing (e.g. construction of a viewing tower near the
Bondjagou forest);
- Protection and valorisation of rich riparian forests outside protected areas, especially around
human settlement. Indigenous riparian forests around cities should not to be converted to
settlement, instead be managed (delineated and protected) as natural parks for recreation,
research, education, etc.;
- Improvement to the forest law.
Some improvements can be given to the forest law. It is urgent to put under protection or
extend the protected areas of Ouémé Supérieur, Mounts Kouffé, Ouémé-Boukou, Dogo,
Kétou to the second side of the Ouémé River that is not currently included in the reserves.
This disposition can be extended to any other stream or river harbouring important riparian
vegetation (e.g. Yarpao). Also the principle of regulated tree harvest in riparian forests might
be accepted (e.g. up-rooted trees, old or dying trees at river/stream edges) followed by special
care to rare or useful tree species regeneration.
3.10. CONCLUSIONS
In line with the Convention on Biological Diversity (UNCED 1992), which emphasised the
need for every country to take responsibility for conserving its own biodiversity, a prudent
approach to biodiversity conservation necessitates preserving all potential rich or endangered
ecosystems. Therefore Benin’s relict forests, especially riparian forests, are important
conservation sites that need more care than currently available. In the majority of savanna
ecosystems, efforts toward the conservation of riparian forests will increasingly become a
priority in land use planning (NATTA 2000), this will allow a wide range of plants and animals
to benefit from the protection of forests along rivers and streams. Riparian systems are at the
same time so important as a collection of natural resources and so limited in individual size
that they both merit and require higher resolution inventory procedures than those deployed
for other larger ecosystems (savanna and upland forests). The responsibility of different
actors, (e.g. forest division, research institutes, projects and NGO’s dealing with the
conservation of biodiversity in Benin in collaboration with local residents), should be drawn
to seek for appropriate initiatives to preserve the remaining rich and diverse riparian forests,
especially in non-protected areas.
24
Chapter 4
ASSESSMENT OF RIPARIAN FOREST FRAGMENTS PLANT DIVERSITY IN
WEST AFRICAN SAVANNA REGIONS: AN OVERVIEW FROM BENIN
Submitted to the Journal of Biogeography
Natta, A.K. and van der Maesen, L.J.G.
Chapter 4: Assessment of riparian forest plant diversity in Benin
Chapter 4
ASSESSMENT OF RIPARIAN FOREST FRAGMENTS PLANT DIVERSITY IN
WEST-AFRICAN SAVANNA REGIONS: AN OVERVIEW FROM BENIN (*)
Natta, A.K. (1) and van der Maesen, L.J.G. (2)
(1) Faculty of Agronomic Sciences, University of Abomey-Calavi, Benin 01 BP 526 Cotonou,
aknatta@yahoo.com; (2) Biosystematics Group, Wageningen University, Gen. Foulkesweg 37, 6703 BL
Wageningen, The Netherlands, jos.vandermaesen@wur.nl.
(*) Submitted to the Journal of Biogeography
ABSTRACT:
Aim: To assess the flora and plant diversity of riparian forests. To compare the diversity of
several riparian forest sites.
Location: From 7º10 to 12º20 N in Benin (West Africa).
Methods: The largest and least disturbed forest fragments along rivers and streams were
surveyed in the three ecological (i.e. Guinean, Sudano-Guinean and Sudanian) regions of
Benin. In total 373 phytosociological relevés (each of 500 m2) were made from 1999 to 2002,
using the Braun-Blanquet method. A plant species list was made from these relevés.
Biodiversity was assessed through species richness, Shannon index (H’), Equitability index
of Pielou (E), and species abundance models.
Results: In all 1003 species (about 1/3 of the estimated number of species in the flora of
Benin), 120 families, 513 genera, and 224 tree species (i.e. dbh ≥ 10 cm) were identified. The
richest (sub) families were the Papilionioideae, Poaceae, Rubiaceae and Euphorbiaceae.
Riparian forests are characterised, on the one hand, by a flood-dependent flora, and on the
other hand by many species from semi-deciduous forests and savannas. Shannon index varied
from 2.4 to 5.8 bits and Pielou’s equitability from 0.51 to 0.86. Species richness varied from
129 to 358 while tree species richness ranged from 27 to 99 per ha.
Main conclusions: This study showed that in fire-prone environments, such as Benin,
relatively large numbers of species still are maintained in small forest fragments along
waterways. Riparian forests are characterised by a diversified flora, and specific plant species
that should be used for the restoration of degraded stands.
Key words: riparian forests, flora, diversity, species abundance model, Benin.
4.1. INTRODUCTION
Benin is located in ‘the Dahomey gap’, which is a break of the West African rain forest belt
between Nigeria and Ghana. This interruption permits the incursion of savanna towards the
coast as a result of the dry conditions prevailing in Southern Benin, Togo and South-Eastern
Ghana (Onochie, 1979; Ern, 1988; Jenik, 1994; Kokou et al., 1999; Kokou & Caballé, 2000).
As a result there are no evergreen tropical forests in Benin (Sokpon, 1995; Tossou, 2002).
High incidences of farming as well as tree cutting and fire have reduced the original
woodland to their current state.
Flora is a major component of a region or a country’s biodiversity. It is historically the
result of a long process of natural selection (Braun-Blanquet, 1972) in relation with climatic
and edaphic conditions as well as human interference. In the majority of degraded woodlands
and mosaic of savanna landscapes, riparian forests (hereafter RFs) belong to the most
27
Chapter 4: Assessment of riparian forest plant diversity in Benin
important forest ecosystems, which follow the outline of rivers and streams (Natta, 2000).
The flora of Benin is incompletely known (PFB, 1997), and the diversity, relative rarity and
threats of many hot-spots in vegetation, such as riparian forests, have remained insufficiently
studied (Mondjannagni, 1969; Paradis, 1988; Sokpon et al., 2001). This lack of information
to the scientific community and managers hinders appropriate planning and conservation
initiatives dedicated to them (Natta et al., 2002).
The aim of this paper is to investigate the flora of riparian forests through several
parameters: plant species richness and abundance, endemic, valuable and threatened plants,
adaptability to a specific habitat, Diversity Index (Shannon), Equitability Index (Pielou),
differences in sites diversities (comparison of Shannon indices), and species abundance
models. This research also intends to provide accurate floristic data for riparian forests
biodiversity protection and restoration in different ecological regions.
4.2. METHODS
Surveys were made in all ecological regions of Benin excluding the coasts, beaches, coastal
lagoons, lakes and marshlands. The study area (Figure 4.1) covered the latitudes 7º10 to
12º20 N. The largest and least disturbed forest fragments along rivers and streams were
surveyed in the Guinean region (Samiondji and Bètèkoukou), Sudano-Guinean zone (TouiKilibo, Idadjo, Bétérou and Pénéssoulou) and Sudanian region (Yarpao, Pendjari Biosphere
Reserve, Gbèssè). In total 373 plots (each of 500 m2) were installed in several RF stands
using the phytosociological approach of Braun-Blanquet, but tree height and diameter at
breast height (i.e. dbh) were measured in 304 plots (Table 4.1). About 19 ha of RFs were
surveyed for the compilation of a plant species list. Rectangular plots, with variable length
and width, were preferred to circular ones to fit with the shape of waterways and width of
riparian forests. The identity of each plant was recorded in the field and specimens taken as
vouchers for the National Herbarium of Benin and the National Herbarium of the
Netherlands, Wageningen University Branch. Species names mainly follow Keay & Hepper
(1954-1972), Brunel et al. (1984), Berhaut (1967), and Lebrun & Stork (1991-1997). Stem
diameter at breast height (dbh) was measured at 1.3 m above the ground, or above buttresses
when present for trees (dbh ≥ 10 cm).
In the present paper the assessment of plant species diversity in RFs encompasses
species richness, Shannon diversity index (H’), Equitability index of Pielou (E), and species
abundance models (also called Dominance-diversity curves). Here species abundance models
test the statistical difference between observed pattern within the floristic data (tree species
abundance and families richness) and expected pattern as described by each model. Visual
inspection of rank-abundance figures suggested two predictors of species abundance models:
the geometric and log series. The χ2 test compared the two distributions, the expected and
observed number of each species in each abundance class.
We compared the diversity of RF sites (1 ha per site), using a t-test based on the
Shannon Index and its variance (cf. Hutcheson, 1970; Magurran, 1988; Kent & Coker, 1992).
This method was already successfully used on RFs in Benin and Ivory Coast (cf. Natta, 2000;
Porembski, 2001; Natta & Porembski, in press, see chapter 9). The usefulness of multivariate
methods (e.g. classification and indirect ordination), in detecting trends and subdivisions
within riparian forests floristic data, was assessed in another paper (cf. Natta et al., in press,
see chapter 6).
28
Chapter 4: Assessment of riparian forest plant diversity in Benin
N
Gbèssè
*
Pendjari
Reserve
*
Towns
Yarpao
*
Pénéssoulou
Bétérou
*
*
* TouiKilibo
Idadjo
*
Bétékoukou
*
Samiondji
1 : Grand-Popo
2 : Cotonou
3 : Ouidah
4 : Porto-Novo
5 : Niaouli
6 : Aplahoué
7 : Pobè
8 : Bohicon
9 : Kétou
10 : Savalou
11 : Savè
12 : Bantè
13 : Tchaourou
14 : Bassila
15 : Djougou
16 : Bembèrèkè
17 : Natitingou
18 : Parakou
19 : Kouandé
20 : Tanguiéta
21 : Toui
22 : Kandi
23 : Banikoara
* Sites of data collection
(with > 20 plots)
States limits
Major waterways
*
Isohyets
Figure 4.1: Study sites where data were collected in riparian forests in Benin
29
Chapter 4: Assessment of riparian forest plant diversity in Benin
4.3. RESULTS
4.3.1. Floristics and stand characteristics of riparian forests throughout Benin
In Benin, RFs vary in their width along the watercourses, which they fringe. Some are no
more than a few meters wide and have an open canopy, while others are much more
extensive and have a closed canopy and a well-developed structure. The floristic and stand
characteristics of riparian forest are summarised in Table 4.1. Species richness (SR) per ha
varied from 129 to 358. Tree density and basal area per ha varied from 253 to 785, and from
23 to 59 m2, respectively.
Table 4.1: Floristic and stand characteristics of riparian forests throughout Benin
Regions
Sites
Relevés
Species Richness Species Richness
Trees
(SR) per site
(SR) per ha
density/ha
Basal area
(BA/ha)
Guinean
Samiondji
36
306
249
726
41.7±17.4
Sudano
Guinean
Pénéssoulou
Idadjo-Bétérou
Okpara
Streams at Toui- Kilibo
40
40
56
62
450
183-195
-
343
129-195
-
544
748-785
315
257
38.9±15.3
45.6±15.8
-
All Sudano-Guinean sites
198
556
129-343
257-785
42.6±15.8
RFs on plateau
RFs at hill foot
43
27
435
467
182-278
358
538-627
732
35.6±16.5
42.7±9.9
All Sudanian sites
70
591
182-358
538-732
40±20.6
304
1003
129-358
253-785
41.5±17.9
Sudanian
All RFs sites in Benin
SR = Species richness per site (SR/site) or per ha (SR/ha); BA/ha = Basal area per ha in m2; Trees = stems with
dbh ≥ 10 cm; RFs = Riparian forests.
4.3.2. Diversity of riparian forest flora in Benin
From 373 floristic relevés altogether 1003 species (about 1/3 of the estimated number of
species in the Benin flora) from 120 families and 513 genera were recorded. The ten (sub)
families richest in species are the Fabaceae-Papilionoideae, Poaceae, Rubiaceae,
Euphorbiaceae, Cyperaceae, Asteraceae, Acanthaceae, Fabaceae-Caesalpinioideae, Moraceae
and Malvaceae. In all, 224 tree species (i.e. dbh ≥ 10 cm) were identified. Tree species with
the highest abundance are Pterocarpus santalinoides, Cola laurifolia, Syzygium guineense,
Dialium guineense, Berlinia grandiflora, Cynometra megalophylla, Elaeis guineensis,
Diospyros mespiliformis, Uapaca togoensis (see Table 4.2). The list of plant species collected
in several riparian forests sites is presented at the end of this dissertation. This flora contains
species from various forest types (dense semi-deciduous, dry, open), savannas and various
degraded woodlands and fallow communities. Species such as Pterocarpus santalinoides,
Cola laurifolia, Syzygium guineense, Culcasia scandens and Morelia senegalensis are seen at
all latitudes along waterways.
On the other hand, some species are confined to certain regions showing that RFs
present niches to conserve dense forest species in various landscapes. In the humid Guinean
region of Southern Benin and along larger waterways (rivers), typical species are Cynometra
megalophylla, Salacia pallescens, Parinari congensis, Drypetes floribunda, Cassipourea
congoensis, and Garcinia livingstonei. Along small waterways (streams), mainly in the
Sudanian region of North Benin, typical RF species are Raphia sudanica, Khaya
30
Chapter 4: Assessment of riparian forest plant diversity in Benin
senegalensis, Uapaca togoensis, Vitex doniana, Oxytenanthera abyssinica, Berlinia
grandiflora, Brenadia salicina, Garcinia ovalifolia, Pentadesma butyracea and
Chrysobalanus atacoriensis.
Table 4.2: Riparian forests tree species dominance: number of individuals and basal area
no.
1
2
3
4
5
6
7
8
9
Species
Pterocarpus santalinoides
Cola laurifolia
Syzygium guineense
Dialium guineense
Berlinia grandiflora
Cynometra megalophylla
Elaeis guineensis
Diospyros mespiliformis
Uapaca togoensis
Total of the 9 most abundant tree species
Total of the remaining (215) tree species
Overall total for 224 tree species
Absolute
abundance
1194
631
499
271
249
248
222
204
201
Abundance
(%)
14.86
7.86
6.21
3.37
3.1
3.08
2.76
2.54
2.50
3719
4311
8030
46.3
53.6
100
Total BA BA(%)
(m²)
1038
12.8
707
8.7
585.2
7.2
264.6
3.25
502.7
6.18
460.7
5.66
240.7
2.96
463.2
5.7
237.4
2.92
4499.52
3630.82
8130.34
55.3
44.6
100
BA= Basal Area in m²
Many species from forest, regrowth vegetation or pioneer species in the Guinean
region and Sudano-Guinean zone of Southern and Central Benin (e.g. Elaeis guineensis,
Dracaena arborea, Dialium guineense, Holarrhena floribunda) are found in streamside
vegetations in the savanna region of North Benin. Also, certain pioneer or secondary forest
species (e.g. Leea guineensis, Rauvolfia vomitoria, Phyllanthus muellerianus, Anthocleista
nobilis, Cleistopholis patens, Spathodea campanulata, Ceiba pentandra, Milicia excelsa and
Ricinodendron heudelotii) of the tropics (cf. Whitmore 1990) are usually seen in RFs in
Benin. Other species (e.g. Khaya senegalensis, Afzelia africana, Milicia excelsa and Antiaris
toxicaria) that have completely disappeared on plateaus in densely populated areas are found
in RFs at higher latitudes or far from settlements. We also find some either endemic or
valuable species almost exclusively in RFs. Thunbergia atacorensis (Acanthaceae) is one of
the rare endemic species in Benin mainly found in RFs at hills feet. Among valuable species,
Dissotis anthennima is known for its importance in medicine, Pentadesma butyracea and
Xylopia aethiopica as major Non Timber Forest Products in West Africa.
Summing up, RFs in Benin are characterised, on the one hand by a flood-dependent
flora widely distributed all over the country, on the other hand by many forest (pioneer and
typical semi-deciduous) and savanna species.
4.3.3. Riparian forest sites diversity
The Shannon index varies from 2.43 to 5.4 bits respectively along the Ouémé river at
Samiondji (South Benin) and along streams in Toui-Kilibo reserve forest (Central Benin).
The overall RFs diversity is 5.8 bits, and the Equitability index of Pielou is 0.74 (Table 4.3).
RFs along streams (Toui-Kilibo, Pénéssoulou, Sudanian RFs on plateau and at hill feet) have
higher diversities than those along rivers (Okpara, Ouémé, Pendjari and Sota): t = 31.57; df =
7502; p < 0.001.
31
Chapter 4: Assessment of riparian forest plant diversity in Benin
Table 4.3: Plant species diversity in several riparian forests sites in Benin
REGION
Sites
Plots Tree species Abundance Shannon index Evenness
(n) richness (SRt)
(N)
(H')
(E)
Guinean
Samiondji (Sa)
36
27
1307
2.43
0.51
Sudano
Guinean
Pénéssoulou (Pe)
Idadjo-Bétérou (ID-BE)
Okpara River (Okp)
Streams at Toui-Kilibo (Str TK)
40
40
56
62
99
64
59
78
1086
1537
883
1003
5.17
3.91
4.91
5.4
0.78
0.65
0.84
0.86
Sudanian RF on plateau (Sud Plat)
Sudanian RF at hill foot (Sud Hill)
43
27
84
65
1226
988
4.97
4.27
0.78
0.71
Rivers (large waterways)
Streams (small waterways)
156
148
111
144
4291
3739
4.44
5.87
0.65
0.82
304
224
8030
5.81
0.74
Sudanian
Riparian forests of BENIN
n = Plots; SRt = Tree species richness per site; N = Abundance; H’ = Shannon index; E = Equitability index of
Pielou.
The comparison of RF sites diversities (Table 4.4), based on the t-test adapted to
Shannon index, gave the ranking:
Streams at Toui-Kilibo (Str TK) > Pénéssoulou (PE) > Sudanian RF on plateau (Sud plat) >
Okpara River (Okp) > Sudanian RF at Hills feet (Sud Hills) > Idadjo-Bétérou (ID-BE) >
Samiondji (SA).
Table 4.4: Comparison of the diversity of riparian forest sites
SITES
Pénéssoulou Idadjo/Bétérou Okpara River Streams at TouiKilibo forest
Samiondji (SA)
PE>SA (***)
Pénéssoulou (PE)
Idadjo-Bétérou (IDBE)
Okpara River (Okp)
Streams at TouiKilibo (Str TK)
Sudanian RF on
Plateau (Sud plat)
ID-BE>SA (***) Okp>SA (***) Str TK>SA (***)
PE>ID-BE (***) PE>Okp (***) Str TK>PE (***)
Okp>ID-BE Str TK>ID-BE (***)
(***)
Str TK>Okp (***)
-
-
-
Sudanian RF on
plateau
Sudanian RF at Hill
feet
Sud plat>SA (***)
PE>Sud plat (**)
Sud plat>ID-BE (***)
Sud Hills>SA (***)
PE>Sud Hills (***)
Sud Hills>ID-BE
(***)
Sud plat : Okp (NS) Okp>Sud Hills (***)
Str TK>Sud plat (***)
Str TK>Sud Hills
(***)
Sud plat>Sud
Hills (***)
NS = Not statistically significant at p=0.05, i.e. similar in terms of tree species diversity; (>) More diverse than;
** Significance at 0.01; *** Significance at < 0.001; Details of the calculations are shown in Table 4.5 in the
Annex.
4.3.4. Species and families abundance models
Species abundance model, which describes the distribution of species richness frequency,
shows that not all species are equally abundant (Figure 4.2). 45 % of the tree species have
less than 4 individuals. Tree species with 1 or 2 individuals are very common (left part of the
curve) while those with larger number of individuals (>10) are uncommon (right part forming
a tail). The best abundance distribution model for RF tree species data set is the log series:
χ2calculated = 12.15 < χ2table = 15.5 at p = 0.05 df = 8. There is no significant difference, at p =
0.05, between the observed and expected distributions of tree species abundance as predicted
by the log series model.
32
Chapter 4: Assessment of riparian forest plant diversity in Benin
Number of tree species
40
36
32
28
24
20
16
12
8
4
0
0
100
200
300
400
500
600
700
800
900
1000
1100
1200
Number of individuals per tree species
Figure 4.2: Tree species abundance distribution of riparian forests tree species in Benin
Also, the distribution of family richness frequency shows that obviously not all
families have equal numbers of species (Figure 4.3). Out of 120 families, 33 (i.e. 27.5%) are
represented by 616 species (about 62% of the total RF flora) and 55% families have each less
than 3 species. Families with 1 or 2 species are very common (left part of the curve), while
families with large number of species (i.e. > 10) are fairly uncommon (right part forming a
tail). The fitting of models shows that a log series is the best fit for the family richness:
χ2calculated = 5.7 < χ2table = 11.07 at p = 0.05 df = 5. We conclude that there is no significant
difference, at p = 0.05, between the observed and expected distributions of family richness as
predicted by the log series model.
40
36
Number of families
32
28
24
20
16
12
8
4
0
0
10
20
30
40
50
60
70
80
90
Number of species per family
Figure 4.3: Family richness abundance distribution of riparian forests in Benin
33
100
Chapter 4: Assessment of riparian forest plant diversity in Benin
4.4. DISCUSSION
4.4.1. Flora of riparian forests
In our large-sized data set, we enumerated more species than for the RFs in South Togo
(West of Benin Republic), where Kokou et al. (2002) found 499 species grouped into 41
families and 108 genera. Many species (e.g. Pterocarpus santalinoides, Cola laurifolia,
Cynometra megalophylla, Manilkara multinervis and Parinari congensis) typical of RFs in
Benin are dominant and characteristics of RFs in Southern Côte d’Ivoire (see Devineau,
1975; 1976). Many common species in RFs are found to be encroachments of upland forests.
Elaeis guineensis, for example is a heliophytic and pioneer species abundant in all West and
central African forests (Sowunmi, 1999; Maley & Chepstow-Lusty, 2001). In the Benin
context, as RFs combine plants from various ecosystems, it is likely that they are more
diverse than vegetation formations consisting of only one ecosystem. Meave & Kellman
(1994) already found that most riparian forest flora in a Neotropical savanna environment of
Belize is not specialised to the riparian environment and there is a significant supplement of
species having other habitat affinities (i.e. rain forest and non-rain forest taxa). A similar
result was found in Northern Australia, where riparian strips contain vegetation dominated by
diverse floristic elements unrelated to monsoon rain forests (Woinarski et al., 2000). For
Bersier & Meyer (1994), RFs are mosaics composed of patches of different vegetation types,
which cover the entire range of forest succession. Thus RFs potentially provide more
different categories of food items for a variety of animals than other forest types.
Little is known about the relative rarity and threats facing most RF species in Benin.
More is known from Khaya senegalensis, Afzelia africana, Milicia excelsa and Antiaris
toxicaria (Eyog Matig et al., 2001; Sinsin et al., 2002a; Eyog Matig et al., 2002; Sokpon &
Ouinsavi, 2002). The absence of numerous rare, endemic or disjunct plant species in the RFs
of Benin, may suggest that these forests have not geographically been isolated for a very long
time, instead the historical patterns of migration resulted in an accumulation of various taxa
from surrounding vegetations in, and at the edge of RF corridors.
Yarpao (Atacora hilly region) and Pénéssoulou reserve forest (semi-deciduous forest
region) have the highest plant richness of riparian forests, 358 and 343 species per ha
respectively (see Table 4.1). Therefore we might say that protection of the dry semideciduous forests and the difficult accessibility of genuine RFs at hills feet have a positive
impact on the conservation of plant diversity. Species richness of RFs (per ha) in Benin
varied from 129 to 358. This is within common limits of certain tropical dense forests: 156 in
the Taï National Park in Côte d’Ivoire (Dengueadhe, 1999), and 100-300 in South America
(Loizeau, 1992). The total vascular plant list for a hectare of forests at Kade (Ghana) is nearly
300 species (Swaine et al., 1987). In a 2.5 ha survey of primary forest in Dja Fauna Reserve
(Cameroon) Sonké & Lejoly (1998) found 125 to 138 species. On the contrary, SR per ha of
RFs in Benin is far lower than values from several neotropical rain forests (cf. Whitmore,
1990).
4.4.2. Family richness in riparian forests
There are some similarities, with regards to dominant families, between the RFs of Benin and
various forest types in Benin, Africa and more generally in the tropics. Akoègninou (1984)
found 93 families with the Rubiaceae, Papilionoideae, Poaceae, Euphorbiaceae and
Apocynaceae as the most pluriform in several dense semi-deciduous forests in Southern
Benin. In the humid zones of Southern Benin, Sokpon & Adjakidjè (1999) found 100 families
with 364 species, the richest families being the Poaceae, Rubiaceae, Cyperaceae, Fabaceae,
34
Chapter 4: Assessment of riparian forest plant diversity in Benin
Euphorbiaceae, Moraceae and Apocynaceae. For these authors, the Leguminosae
(Papilionoideae, Mimosoideae and Caesalpinioideae) are often the most species-rich family
in forest formations. Likewise, the Cyperaceae are usually abundant in humid zones, while
the Poaceae, the second richest family, show the effects of degradation or drier Sudanian
climate.
In Burkina-Faso (North of the Republic of Benin), the Papilionaceae, Rubiaceae,
Caesalpiniaceae and Poaceae were the most species-rich families in gallery forests of the
Hippopotamus Pond Biosphere Reserve (Bélem & Guinko, 1998). In the Dja Fauna Reserve
(Cameroon), the Euphorbiaceae, Rubiaceae, Annonaceae, Meliaceae, Caesalpiniaceae,
Sapindaceae and Sapotaceae have the highest tree species richness (Lejoly et al., 1996; Sonké
& Lejoly, 1998). On a 2-ha dense forest at Kade (Ghana) the Leguminosae was the best
represented family (Swaine et al., 1987). Meave & Kellman (1994) found the same result in a
1.6 ha of RF in the Mountain Pine Ridge (Belize). According to Whitmore (1990) the three
tropical rain forest blocks have abundant Leguminosae (subfamily Caesalpinioideae),
Annonaceae, Euphorbiaceae, Lauraceae, Myristicaceae, Rubiaceae and Sapotaceae. For
Gentry (1988) three families (Rubiaceae, Annonaceae and Euphorbiaceae) are always among
the ten most species-rich families on all continents.
Summing up, the Leguminosae, Euphorbiaceae, Rubiaceae and Annonaceae are the
most species-rich families always found in Benin RFs and in most tropical forest formations.
Meanwhile the Cyperaceae show the influence of humid sites, while the high richness of the
Poaceae is indicative of open canopy, degradation of RFs as well as edge effects.
4.4.3. Tree species richness in riparian forest sites
Tree species richness per ha varied from 27 to 99 (see Table 4.1), and these values are quite
similar to the mean value (52) found by Meave & Kellman (1994) in a RF of Belize. In view
of the large interregional variability in tree species richness per ha, figures of RFs in Benin
are comparable to values (65-77) of some dense forests in Côte d’Ivoire at Taï National Park
and Yapo (Corthay, 1996; Dengueadhe, 1999). Lejoly (1994) and Lejoly et al. (1996) found
92 and 84 tree species per 0.5 ha respectively in Dja Fauna Reserve (Cameroon) and Ngotto
forest (Central Africa Republic). Swaine et al. (1987) enumerated 120 tree species in a two 1ha samples of moist semi-deciduous forest at Kade (Ghana).
On the other hand, the species richness values found in Benin are lower than those
(86-122) of Dogbo, Guiroutou and Djapadji forests in Côte d’Ivoire (Dengueadhe, 1999); 69131 for Gabon forests (Reitsma, 1988); and in some neotropical wet forests (e.g. Bajo
Calima, Western Columbia), which are among the most species-rich in the world, with over
250 tree species (Faber-Langendoen & Gentry, 1991). For Whitmore (1990), tree species
richness per ha of forests vary from 23 in Nigeria to 283 at Yanamomo (Peruvian Amazon),
the richest forest found so far.
4.4.4. Comparison of riparian forest sites diversities
A species diversity index, which takes into account the total number of species and their
relative abundance offers a good description of communities and allows comparison between
them. The overall RFs diversity in Benin (5.8 bits) represents quite a high value among forest
formations. This figure is similar to the one (5.4 bits) of Dja Fauna Reserve in Cameroon
(Sonké & Lejoly, 1998). Meanwhile it is higher than the values given by (Sokpon et al.,
2001) and (Kokou et al., 2002) for certain RFs in Benin and Southern Togo, respectively.
Sites along streams have higher diversity values than sites along rivers, and the reason
is the lower species richness and uneven distribution of the dominant trees along rivers where
35
Chapter 4: Assessment of riparian forest plant diversity in Benin
4.5% of trees are represented by 58% of total individuals, while along streams 4.8% of trees
are represented by only 35.4% of the total abundance. Tree species along streams are more
evenly distributed (E = 0.82) than along rivers (E = 0.65). This difference in diversities
between streams (smaller waterways) and rivers (larger waterways) is confirmed by the
multivariate analysis (DCA and TWINSPAN) of the RFs data set (cf. Natta et al., in press,
see Chapter 6).
The comparison of tree species diversity between the two RF types (along streams and
rivers) is an indirect way of comparing the effect of flood frequency, duration or intensity,
and importance of waterways on species richness and abundance. The absence or short
periods of inundation along streams allow plants with wider ecological range (forest pioneers
and savanna species) to grow. At the same time a specialised (thus less diversified) flora,
which supports 4 to 6 months of submersion per year, can only grow along rivers. Therefore
at this stage of the investigations on RFs, the duration of submersion or flood seems
negatively correlated with tree species diversity. This result substantiates the hypothesis that
frequent or long environmental stresses, such as floods along rivers, will lead to a decrease in
the diversity of species (at least for tree species), and a loss of diversity will lead to the
increased abundance and dominance of those species that can successfully maintain
themselves on isolated or harsh sites. Hanson et al. (1990) described a similar phenomenon in
RF patches along the Upper mid-West of Iowa river (USA). Meanwhile, as species diversity
is sometimes indicative of the diversity, stability and well-being of ecological systems, and
used for environmental monitoring (Magurran, 1988; Withmore, 1990; Huston, 1994), we
might say that any dense and least disturbed RF is in equilibrium with the prevailing
ecological conditions.
4.4.5. Species and families abundance models
Log series adequately described tree species and family richness patterns of the RFs data set.
In an assessment of forest plant diversity in Nepal, Acharya (1999) found that the log series
describes species abundance pattern better than the geometric series, meanwhile the fit was
not so good for plots with low disturbance. Also a gradual decrease in disturbance was well
explained with the progression of species abundance models from the geometric series to a
log series. From a 50-ha forest plot on Barro Colorado Island (Panama), the pattern of total
species abundance deviates from the log-normal in having too many rare species (Hubbell &
Foster, 1986). In the Peruvian Amazon, only 15% of species had more than 2 individuals and
63 % had only one individual (Whitmore, 1990).
Log series, that are intermediate between geometric series and log normal series, often
characterise communities in which species of intermediate abundance become more common.
Theoretically, the geometric series provides the best fit to the observed species abundance in
species-poor communities under a harsh environment, while in large species-rich or mature
communities the distribution of species abundance is usually log normal. As succession
proceeds, or as conditions ameliorate, species abundance patterns grade from geometric
series into those of the log series (Whittaker, 1972; Magurran, 1988). Therefore RFs, under
the Benin conditions, are made of moderately species-rich or mature communities in which
families with intermediate species richness and tree species with intermediate abundance are
fairly common.
In large assemblages (as it is for our RFs data set), empirical patterns predict log
normal series. However, new insights raised by Magurran & Henderson (2003), suggest that
an ecological community can be separated into two components:
1 - core (i.e. typical) species, which are persistent, abundant, and grow naturally under
good/optimal ecological conditions. They are log normally distributed.
36
Chapter 4: Assessment of riparian forest plant diversity in Benin
2 – occasional (i.e. non-typical) species, which occur infrequently, have low abundance and
different habitat requirements. They follow a log series distribution.
4.5. CONCLUSIONS
This paper has made an attempt to assess the flora, plant species diversity, as well as species
abundance models that best fit RFs in Benin. The flora of RF contains about 1/3 of all species
in Benin, 120 families and 513 genera. The Leguminosae, Euphorbiaceae, Rubiaceae and
Annonaceae are the most species-rich families found time and again in Benin’s RFs as well
as in tropical rain forests. RFs can be recognised and characterised by exclusive, frequent and
species typical to certain regions. Endemism is low compared to rain forests, yet in fire-prone
environment the RFs of Benin harbour plant species that prefer the wettest conditions. They
are a refugee ecosystem for forest plants, and act as route or corridor for movement of certain
species across the landscape.
As RFs along streams have higher tree species diversity than those along rivers, we
conclude that flood frequency, duration and intensity are somehow negatively correlated with
tree flora diversity. A challenge is to assess the spatio-temporal effects of flood variables on
plant species diversity. This study showed that relatively large numbers of species still are
maintained in small forest fragments along waterways. Although RFs are often strip-like and
fragmented ecosystems, their rich and varied flora invites protection and restoration. Also this
phytodiversity, should be used in selecting the most suitable plant species for the
rehabilitation of degraded RF corridors, in each ecological region of the country. This will
allows them to fulfil their vital ecological, socio-economic and cultural functions.
37
Chapter 4: Assessment of riparian forest plant diversity in Benin
ANNEX
Table 4.5: Details of the comparison of tree species diversity between seven riparian forests sites
Comparison of riparian forest sites
H'1
H'2
df
t
calculated
t-table
Significance Hypothesis
accepted
Pénéssoulou (40 relevés) / Samiondji (36 relevés)
Idadjo-Bétérou (40 relevés) / Samiondji (36 relevés)
Okpara (56 relevés) / Samiondji (36 relevés)
Streams Toui-Kilibo (62 relevés) / Samiondji (36
relevés)
Sudanian RF on plateau (43 relevés) / Samiondji (36
relevés)
Sudanian RF at foot of hills (27 relevés) / Samiondji
(36 relevés)
Idadjo-Bétérou / Pénéssoulou
Okpara/Pénéssoulou
5.17
3.91
4.91
5.4
2.43
2.43
2.43
2.43
2392
2679
2186
2211
34.08
18.82
32.36
40.48
p=0.001; t=3.29
p=0.001; t=3.29
p=0.001; t=3.29
p=0.001; t=3.29
***
***
***
***
H1
H1
H1
H1
4.97
2.43
2460
32.76
p=0.001; t=3.29
***
H1
4.27
2.43
2210
21.96
p=0.001; t=3.29
***
H1
3.91
4.91
5.17
5.17
2498
1969
17.09
3.68
p=0.001; t=3.29
p=0.001; t=3.29
***
***
H1
H1
Streams Toui-Kilibo(TK streams) / Pénéssoulou
5.4
5.17
2015
3.39
p=0.001; t=3.29
***
H1
Sudanian RF on plateau / Pénéssoulou
4.97
5.17
2263
2.79
p=0.01; t=2.57
**
H1
Sudanian RF at foot of hills / Pénéssoulou
4.27
5.17
1980
10.85
p=0.001; t=3.29
***
H1
Okpara / Idadjo-Bétérou
4.91
3.91
2318
14.34
p=0.001; t=3.29
***
H1
TK streams / Idadjo-Bétérou
5.4
3.91
2538
22.58
p=0.001; t=3.29
***
H1
Sudanian RF on plateau / Idadjo-Bétérou
4.97
3.91
2747
15
p=0.001; t=3.29
***
H1
Sudanian RF at foot of hills / Idadjo-Bétérou
4.27
3.91
2105
4.43
p=0.001; t=3.29
***
H1
TK Streams / Okpara
5.4
4.91
1826
7.78
p=0.001; t=3.29
***
H1
Sudanian RF on plateau / Okpara
4.97
4.91
2071
0.88
p=0.2;
NS
H0
t=1.28
Sudanian RF at foot of Hills / Okpara
4.27
4.91
1778
8.03
p=0.001; t=3.29
***
H1
Sudanian RF on plateau / TK streams
4.97
5.4
2224
6.72
p=0.001; t=3.29
***
H1
Sudanian RF at foot of Hills / TK streams
4.27
5.4
1722
14.84
p=0.001; t=3.29
***
H1
4.27
1954
8.7
p=0.001; t=3.29
***
H1
Sudanian RF on plateau / Sudanian RF at foot of Hills 4.97
Conclusion
PE>SA: Pénéssoulou is more diverse than Samiondji
Id-Be>SA: Idadjo-Bétérou is more diverse than Samiondji
Okp>SA: Okpara is more diverse than Samiondji
Tk>SA: Streams of Toui-Kilibo forest are more diverse than
Samiondji
Sud Plat>SA: Sudanian RF on plateau are more diverse than
Samiondji
Sud Hill>SA: Sudanian RF at foot of hills are more diverse than
Samiondji
PE>Id-Be: Pénéssoulou is more diverse than Idadjo-Bétérou
PE>Okpara: Pénéssoulou is more diverse than Okpara
TK Streams>PE: Streams at Toui-Kilibo are more diverse than
Pénéssoulou
PE>Sud Plat: RF of Pénéssoulou are more diverse than
Sudanian RF on plateau
PE>Sud Hill: Pénéssoulou are more diverse than Sudanian RF
at foot of hills
Idadjo-Bétérou>Okpara: Idadjo-Bétérou is more diverse than
Okpara
TK>Idadjo-Bétérou : Streams of Toui-Kilibo forest are more
diverse than Idadjo-Bétérou
Sud RF on plateau : Sud RF on plateau are more diverse than
Idadjo-Bétérou
TK>Idadjo-Bétérou : Streams of Toui-Kilibo forest are more
diverse than Idadjo-Bétérou
TK>Okpara : Streams of Toui-Kilibo forest are more diverse
than Okpara
Sudanian RF on plateau and RF along Okpara are similar in
terms of tree species diversity
Okpara>Sudanian RF at foot of Hills: Okpara is more diverse
than Sudanian RF at foot of hills
Streams TK>Sudanian RF on plateau: TK streams are more
diverse than Sudanian RF on plateau
Streams TK>Sudanian RF on plateau: TK streams are more
diverse than Sudanian RF on plateau
Sud RF plateau>Sud RF Hills: Sud RF on plateau are more
diverse than Sud RF foot of Hills
NS = Non Statistically significant at p=0.05 ; ** Significant at 0.01 ; *** Significant at < 0.001 ; H’1& H’2 = Shannon indices.
38
Chapter 5
STRUCTURE AND ECOLOGICAL SPECTRA OF RIPARIAN FORESTS IN BENIN
A.K. Natta
Chapter 5: Structure and ecological spectra of riparian forests in Benin
Chapter 5
STRUCTURE AND ECOLOGICAL SPECTRA OF RIPARIAN FORESTS IN BENIN
Natta A.K. (1)
(1)
Department of Environment Management, Faculty of Agronomic Sciences. FSA/UAC 01 BP 526 Cotonou,
Benin; aknatta@yahoo.com.
Abstract
This paper is a preliminary investigation of the structural characteristics of riparian forests in
Benin. So far such information is not available in Benin. The structure was analysed through
different methods: life forms, phytogeographic types, diameter class distribution, basal area,
and stem density. Riparian forests displayed a physiognomy that is highly variable in terms of
vertical stratification. Riparian forests in Benin were similar to many dense tropical forests,
and to the West African ones in particular, in terms of phytogeographical types (large
contribution of Guineo-Congolian basin species), life forms (high abundance of therophytes
and medium size trees, low percentage of mega-phanerophytes and woody lianas), diameter
class distribution (reverse J type), basal area (23-59 m2/ha) and stem density (253 to 785 trees
10 cm dbh /ha). Species dominance, in terms of abundance and basal area, was a major
characteristic of riparian forests in Benin. Many stands of riparian forests are facing various
levels of structural and floristic simplification, which include fundamental transformations in
vegetation physiognomy, from dense pristine stands to riparian scrub, or bare land.
Key words: riparian forests, structure, life forms, phytogeographic types, diameter, basal
area, stem density, species dominance, Benin.
5.1. INTRODUCTION
Pattern in vegetation structure is a four-dimensional phenomenon. It can be studied at
different scales, in space and time. Generally, vegetation structure is concerned with
distribution of individuals along the vertical and horizontal axes, as well as spatial
arrangement of physiognomical, taxonomic, morphological and functional characteristics of
the elements building the vegetation (Popma et al. 1988, Sterck & Bongers 2001). Various
methods have been used to assess the structure of tropical forests (Whitmore 1990), among
which riparian forests are a major component.
In savanna regions, several authors have documented the structural characteristics of
riparian forests (abbreviated as RFs), which contrast strongly with the open forest and
savanna types which otherwise dominate the landscape (Adjanohoun 1965, Monnier 1990,
Woinarski et al. 2000). However, in the pre-forest zone of Côte d’Ivoire (West Africa), the
physiognomy of riparian forests is very much similar to the one of dense semi-deciduous
forest, with which they share similar features such as presence of buttresses, cauliflory,
presence of numerous epiphytes and lianas (Devineau 1975, 1976, 1984).
The floristic composition and plant community diversities of riparian forests in Benin
have been studied (cf. Natta & van der Maesen, see Chapter 04), but an account of the
general structure is lacking. The scale of the study (19 ha of riparian forests) and sites
conditions (various climatic conditions, floristic compositions and successional states), imply
41
Chapter 5: Structure and ecological spectra of riparian forests in Benin
that the selected structural parameters should give a general picture of riparian forests in
Benin. Life forms are frequently used as a means of ecological characterisation of vegetation
formations or plant communities, while the geographic affinity of a flora contains
information about evolution patterns of species within the region of occurrence. Also basal
area, stem density and diameter class distribution are intrinsic characteristics of a given plant
community, and are often used to compare vegetation types. This paper investigates the
structural characteristics of riparian forests in Benin, in terms of life forms, phytogeographic
types, stem density and basal area. Two research questions will be answered:
1 - What are the most prominent life forms and phytogeographic types of riparian forests, and
what are the implications of their occurrence in Benin?
2 - What are the average basal area and stem density of riparian forest sites in Benin, and are
these figures comparable to those from other tropical forests?
5.2. METHODS
The study was conducted from 7º10 to 12º20 N in Benin (West Africa). About 19 ha of the
least disturbed forest stands along rivers and streams were surveyed in the Guinean region
(Samiondji and Bètèkoukou), Sudano-Guinean zone (Toui-Kilibo, Idadjo, Bétérou and
Pénéssoulou) and Sudanian region (Yarpao, Gbèssè, Pendjari Biosphere Reserve), see Figure
4.1. In total 373 rectangular plots (each of 500 m2), were installed in several RF sites using
the Braun-Blanquet phytosociological method of floristic data collection. Tree height and dbh
were measured in 304 plots (i.e. dendrometric relevés, see Table 5.1). Species names follow
Keay & Hepper (1954-1972), Brunel et al. (1984), Berhaut (1967) and Lebrun & Stork
(1991-1997).
Life forms followed Raunkaier (1934), Schnell (1971) and Keay & Hepper (19541972). They were: Phanerophytes (Ph) subdivided into mega-phanerophyte (MPh > 30 m),
meso-phanerophyte (mPh: 8 to 29 m), micro-phanerophyte (mph: 2 to 7 m) and nanophanerophyte (nph < 2m); Therophytes (Th); Hemicrytophytes (He); Chamaephytes (Ch);
Lianas (L); Geophytes (Ge); Epiphytes (Ep) and Parasites (Par).
The phytogeographic types were named after White (1986) and Keay & Hepper
(1954-1972). They were:
- Species widely distributed in the tropics (Cosmopolitan-Cosmo, Pantropical-Pan, AfroAmerican-AA and Paleotropical-Paleo).
- Species widely distributed in Africa (Tropical Africa-TA, Pluri Regional in Africa-PRA)
- Regional species in Africa (Sudanian-S, Guinean-G, Sudano-Guinean-SG, SudanoZambesian-SZ, Guineo-Congolian-GC).
The family, life form and phytogeographic affinity of each plant species were added
to the species list. Stem diameter at breast height (dbh) was measured for trees (i.e. dbh ≥ 10
cm). Several variables were measured in each plot and on individual trees. Plot variables
were: centre co-ordinates, riparian forest width, species richness and abundance. We plotted
frequency distribution curves for the life forms, phytogeographic types, diameter class, and
calculated basal areas and tree densities of riparian forest sites per ha.
5.3. RESULTS
Floristic and stand characteristics of riparian forests throughout Benin
In Benin, riparian forests vary in their width along the watercourses, which they fringe. Some
are no more than a few meters wide and have an open canopy, while others are much more
42
Chapter 5: Structure and ecological spectra of riparian forests in Benin
extensive and have a closed canopy and a well-developed structure. The floristic and stand
characteristics of riparian forest in Benin are presented in Table 5.1.
Table 5.1: Floristic and stand characteristics of riparian forests throughout Benin
Regions
Sites
Floristic Dendrometric SR/site
relevés (n1) relevés (n2)
Guinean
Samiondji
38
36
Sudano
Guinean
Pénéssoulou
Idadjo-Bétérou
Okpara
Streams at Toui- Kilibo
41
48
56
78
40
40
56
62
All Sudano-Guinean sites
223
198
556
RFs on plateau
RFs at hill foot
76
36
43
27
All Sudanian sites
112
All riparian forest sites in Benin
373
Sudanian
306
SR/ha SR of trees Trees
Basal area
/ site
density/ha (BA/ha)
249
27
726
41.7±17.4*
99
34-54
59
78
544
748-785
315
257
38.9±15.3
45.6±15.8
-
129-343
34-99
257-785
42.6±15.8
435
467
182-278
358
36-44
65
538-627
732
35.6±16.5
42.7±9.9
70
591
182-358
36-65
538-732
40±20.6
304
1003
129-358
224
253-785
41.5±17.9
450
343
183-195 129-195
-
n1 = number of plots where floristic data were collected using the phytosociological approach; n2 = number of
plots where data were collected on individual trees; SR = Species richness per site (SR/site) or per ha (SR/ha);
BA/ha = Basal area per ha in m2; Trees = stems with dbh ≥ 10 cm; (*) ± Standard Deviation; (-) data not
available.
Life forms variability and frequency distribution
Phanerophytes (Ph), lianas (L) and therophytes (Th) are the richest life forms (Figure 5.1a).
Field data confirm that trees are on average 14 to 18 m tall, with occasional heights of 20 to
25 m. If present, layers (i.e. vertical distribution of tree crowns) occur at variable heights, and
are stand dependent. In the least disturbed stands, the top layer of uneven and discontinuous
crowns is composed of a few emergent trees such as Ceiba pentandra, Parinari congensis
and Cynometra megalophylla. Within phanerophytes, medium size trees (meso and microphanerophytes) are more frequent (75.6 %) than mega-phanerophytes (3 %) (Figure 5.1b).
Also, liana richness in the Benin RFs is high (236 species, i.e. 23.6% of the overall RF
species). Meanwhile, woody lianas (LMPh and LmPh, see Figure 5.1c) are only represented
by 21 species (i.e. 8.9 % of the liana richness and 2 % of the total RF flora).
Among the six epiphyte species (0.6% of RF richness), the most frequent are the
stem-tuber epiphytes Calyptrochilum christyanum and C. emarginatum. Less frequent is the
humus-collecting epiphytic fern with clustered roots, Platycerium angolense.
43
Chapter 5: Structure and ecological spectra of riparian forests in Benin
Species richness
35 0
30 0
25 0
20 0
15 0
10 0
50
0
Ph
L
Th
Ch
He
Ge
Ep
Pa r
Life fo r m s
Species Richness
(a): Overall life forms frequency distribution. Ph = Phanerophytes; L = Lianas; Th = Therophytes;
Ch = Chamephytes; He = Hemicryptophytes; Ge = Geophytes; Ep = Epiphytes; Par = Parasites.
150
120
90
60
30
0
MPh
m Ph
m ph
nph
Ph an e r o p h yte life fo rms
(b): Phanerophytes life forms frequency distribution. MPh = Mega-phanerophytes; mPh = mesophanerophytes; mph = micro-phanerophytes; nph = nano-phanerophytes.
Species richness
100
80
60
40
20
0
LMPh
Lm P h
Lm p h
Ln ph
Lge
Lth
Lian a life fo r ms
(c): Liana life forms frequency distribution in riparian forests. LMPh = Megaphanerophyte lianas;
LmPh=mesophanerophyte lianas; Lmph=microphanerophyte lianas; Lnph=nanophanerophyte
lianas.
Figure 5.1: Life form frequency distribution in riparian forests
44
Chapter 5: Structure and ecological spectra of riparian forests in Benin
Phytogeographical affinity of riparian forest species
Beside the large contribution of species from the Guineo-Congolian basin (33 % for GuineoCongolian and Guinean types), species of wide distribution all over the tropics (Pantropical
16.7 %), and in tropical Africa (13.6 %) in particular, are the richest phytogeographic types
of riparian forests in Benin (Figure 5.2).
250
Species richness
200
150
100
Cosmopolitan
Sudanian
Afro-American
Paleo-Tropical
Sudano-Zambesian
Sudano-Guinean
Guinean
Tropical Africa
Pantropical
Guineo-Congolian
0
Pluri Regional in Africa
50
Phytogeographic types
Figure 5.2: Phytogeographic types frequency distribution in riparian forests
Diameter class distribution
The size-class distribution (or stand table) of trees with a dbh ≥10 cm is drawn from 7765
stems. Stems with dbh < 50 cm are on the left part while bigger tree with dbh > 70 cm form
the tail at the right part of the curve (Figure 5.3). Here higher abundance is found in the
smaller size classes with a more or less logarithmic decline in numbers with increasing size.
76.4% of stems have less than 30 cm dbh.
45
Chapter 5: Structure and ecological spectra of riparian forests in Benin
2500
2000
1500
1000
180 to 189
160 to 169
140 to 149
120 to 129
100 to 109
90 to 94
80 to 84
70 to 74
60 to 64
50 to 54
40 to 44
10 to 14
0
30 to 34
500
20 to 24
Abundance per DBH class
3000
DBH class (cm)
Figure 5.3: Diameter class distribution for riparian forest trees (dbh ≥ 10 cm) in Benin
Species dominance in riparian forests
Nine (i.e. 4%) tree species out of 224 contribute 46.3 % of the total abundance and 55.3 % of
the total basal area. They are Pterocarpus santalinoides, Cola laurifolia, Syzygium guineense,
Dialium guineense, Berlinia grandiflora, Cynometra megalophylla, Elaeis guineensis,
Diospyros mespiliformis and Uapaca togoensis.
5.4. DISCUSSION
Life forms variability
The life forms found in riparian forests are similar to many riparian forest sites in tropical
Africa, in general and the savanna zones in particular. Akoègninou (1984) found similar
ranking of life forms types in several dense semi-deciduous forests of Southern Benin. In
gallery forests of the Hippopotamus Pond Biosphere Reserve (Burkina-Faso), phanerophytes
and therophytes accounted for 64.8 % and 13.6 % of the total species richness respectively
(Bélem & Guinko 1998). Phanerophytes are known to be the major life form in forest
ecosystems, but in Benin’s RFs the overall percentage is lower (56 %) than in other forest
formations (e.g. 80-90 % given by Mangenot (1955) for dense tropical forests). Also at
Lamto (South Côte d’Ivoire), Devineau (1984) found that RFs have more medium-size trees
(meso- and micro-phanerophytes) than large ones (i.e. mega-phanerophytes). Along the Tana
river (Kenya), riparian forests have a disturbed physiognomy characterised by a low mean
height (14.4 m) (Medley 1992). In contrast, in tropical rain forests canopy tree heights
fluctuate between 30 to 35 m and emergent trees may reach heights above 40 m (Popma et al.
1988, Whitmore 1990).
In Southern Togo, Kokou et al. (2002) found that gallery forests supported the
greatest liana richness with 150 species (i.e. 30.8 % of the overall gallery forest richness), and
the highest Shannon diversity index. In riparian forests along the Tana river (Kenya), 27% of
the woody species are lianas (Medley 1992). Akoègninou (1984) found 150 liana species (i.e.
26.1% of the plant richness) in dense semi-deciduous forests in Southern Benin. In several
rain forests in Côte d’Ivoire, the density of lianas was negatively correlated with the age of
46
Chapter 5: Structure and ecological spectra of riparian forests in Benin
logged areas, and it ultimately reaches the maximum value of undisturbed patches (Kouamé
& Traoré 2001). Lianas (woody climbers) and vines (herbaceous climbers) are an intrinsic
and important component of tropical forests (Hall & Swaine 1981, Hegarty & Caballé 1991,
Kokou et al. 2002) and are considered a major component for forest reconstitution (Kouamé
& Traoré 2001, Parren 2003). Degradation (e.g. selective tree cutting or pruning, tree fall),
lateral lighting as well as the multiplication of edge effects and ecotones through
fragmentation, are favourable conditions for high abundance of lianas in RF.
The high abundance of vines and climbing shrubs (Lmph, Lnph, Lge and Lth, see
Figure 5.1c) over woody lianas (LMPh) in RFs is indicative of relatively low forests having
an irregular canopy (Medley 1992). On the other hand, the high species richness of
therophytes (Figure 5.1a) indicates that many RFs in Benin are either degraded or influenced
by neighbouring open plant communities. Like woody lianas, epiphytes reach their fullest
development in humid tropical forests (Braun-Blanquet 1972).
Phytogeographical affinity of riparian forest species
Guinean species (Guineo-Congolian and typical Guinean) are abundant in dense semideciduous forests in Southern Benin (Akoègninou 1984), and in riparian forests at Bantè
(Central Benin) (Akoègninou et al. 2001). In several gallery forests at Lamto (South Côte
d’Ivoire), Devineau (1975) found 70 to 75% of Guineo-Congolian species against 15% of
Sudano-Zambesian distribution. In gallery forests of the Hippopotamus Pond Biosphere
Reserve (Burkina-Faso), Guineo-Congolian and Sudano-Zambesian species accounted for
61.7% and 38.3% of the total plant richness respectively (Bélem & Guinko 1998). This
justifies the classification of RFs among humid vegetations. The floristic affinity towards the
Guineo-Congolian rain forests is indicative for supply of water. The continuous and adequate
availability of ground water and shallow water table, which originate from waterways, allow
for the establishment and long-term existence of these species.
The geographical affinities (Guineo-Congolian, Pantropical and Sudanian species) of
RFs from the West-African savanna regions support three conclusions: (a) The presence of
numerous pan-african species with heavy seeds and fruits (e.g. Diospyros mespiliformis,
Oncoba spinosa) suggest that the dense or evergreen forests in Africa were once continuous
(Medley 1992); (b) the high contribution of Guineo-Congolian basin species substantiates the
thesis of an early period of continuous forest block in West and Central Africa (see also
Akoègninou 1984, Tossou 2002); (c) on the other hand, the incursion of savanna taxa and
Sudano-Guinean transitional zone species (11.3 and 9.4 %, respectively) in RFs indicates the
lower water balance along waterways in the dry season, particularly in North Benin.
Diameter class distribution
Similar results in diameter class distribution were found in RFs of Belize (Meave & Kellman
1994), where about 78% of trees were in the smallest dbh class (10-20 cm). Following these
authors we might characterise RFs in Benin as low-biomass community with many smallstemmed trees, compared to the high-biomass continuous upland tropical rain forests.
The decreasing curve has a reverse-J (or J-shaped, see Figure 5.3) distribution typical
of natural forest regenerating from seed. It suggests a stable size and age class distribution
(Swaine et al. 1987). This shape is characteristic of climax species (Whitmore 1990). For
Faber-Langendoen & Gentry (1991) this distribution type describes a mature stand, with
many small individuals and few large ones. We hypothesise that the most abundant species
among the flood-tolerant species of RFs are certainly at a stage of equilibrium with the
climatic and edaphic conditions.
47
Chapter 5: Structure and ecological spectra of riparian forests in Benin
Basal area and stem density of riparian forests
The overall mean basal area (± S.D.) of RFs in Benin, obtained from 7765 stems (dbh ≥10
cm), was 41.5 ± 17.9 m2 ha-1 and the mean density 586 ± 192 stems ha-1 (see Table 5.1).
Sokpon et al. (2001) found a similar range for stem density ha-1 (312 to 665) for some
edaphic forests in Benin with basal area ranging from 24.8 to 41 m2 ha-1. Goudiaby (1998)
found a density of 507 trees/ha in a gallery forest of South East Senegal. Along the Tana river
(Kenya), riparian forests have a low tree density and coverage, 409 stems ha-1 and 23.1 m2/ha
respectively (Medley 1992). The mean tree density of a 1.6 ha RF at the Mountain Pine Ridge
(Belize) was 766 ± 241 stems ha-1 (Meave & Kellman 1994) that exceeds the values from
Benin, obviously the mean basal area (21.9 ± 8.8 m2 ha-1) in Belize is far lower.
Values of basal area of RFs in Benin (23-59 m2/ha) are similar to those of some West
African upland dense forests (Taï, Dogbo, Guiroutou, Djapadji) in Côte d’Ivoire with 25-59
m2/ha (Dengueadhe 1999) and humid forests in Ghana (25.5 to 33 m2/ha) (Hall & Swaine
1976, Swaine et al. 1987). In Gabonese dense forests, Hladik (1982) and Reitsma (1988)
found 35 and 35.7-42.9 m2/ha respectively. In Cameroon, Sonké & Lejoly (1998) found a
mean value of 30.5 m2/ha in Dja fauna reserve and Gartlan et al. (1986) found 27.6 m2/ha in
the Korup forest reserve. At Parana high forest (Paraguay) the mean basal area is 39.6 m2/ha
(Stutz De Ortega 1987), while trees in the Los Tuxtlas rain forest plot (Mexico) had an
average basal area of 34.9 m2/ha (Bongers et al. 1988). For Gentry (1988) values range from
35 to 45 m2/ha in Neotropical and Asian forests. Basal area figures found in Benin fit well
with those given by Mori & Boom (1987) for tropical forests (21.3 - 53 m2/ha).
Tree densities of RFs in Benin are similar to those in Taï, Dogbo, Guiroutou, Djapadji
and Yapo upland forests in Côte d’Ivoire with 376–649 stems/ha (Corthay 1996, Dengueadhe
1999); while Swaine et al. (1987) obtained 552 ± 13 trees/ha in a 2-ha moist semi-deciduous
forest at Kade (Ghana). In the Dja fauna reserve (Cameroon), Lejoly et al. (1996) and Sonké
& Lejoly (1998) found respectively 603 and 461 trees/ha. The mean tree density was 471 ha-1
in the Korup reserve forest, Cameroon (Gartlan et al. 1986).
Forest structure at the Los Tuxtlas rain forest (Mexico) was characterised by a low
density, i.e. 346 trees/ha (Bongers et al. 1988). Whitmore (1990) found a density of 580
stems/ha in Peru. For Neotropical forests the values ranged from 167 to 1947 trees/ha (Gentry
1982). Studies of many riparian forest fragments from Belize and Venezuela have shown that
they contain tree species densities comparable to continuous forests, and a non-specialised
tree flora comprised of species characteristic of continuous forest (Meave et al. 1991).
Species dominance in riparian forests
The phenomenon of species dominance, in which one or a few species contribute much to the
total abundance and basal area, is not new in riparian systems as well as tropical forests in
general. Along the Tana river (Kenya), riparian forests were characterised by a high species
importance attributable to a few trees (Medley 1992). In a RF of Belize, Meave & Kellman
(1994) found that the seven most abundant species (2.4% of the total species richness)
account for 1/3 of the total number of stems. According to Wolter (1993) cited by Sonké &
Lejoly (1998), dominant species can comprise up to 58 % of the total individuals and 75 % of
basal area.
Tropical forests dominated by a few species have been considered to result from
edaphic conditions unfavourable for a large majority of species (Richards 1952, Hartshorn
1988) such as poor nutrient status, presence of some minerals in toxic concentrations, and
temporary or permanent flooding (Martijena & Bullock 1994, Martijena 1998). The latter
48
Chapter 5: Structure and ecological spectra of riparian forests in Benin
factor seems more important in the present case, and this indicates the superior adaptation of
such dominant tree species to a regime with vertical and lateral harsh disturbance. Therefore
these species can be termed ‘edaphic and climatic climax species’ of RFs in the prevailing
ecological conditions of Benin. They are keystone species and intrinsic components in the
functioning of complex and self-organising RF systems. Moreover, they are essential for
ecosystem resilience (i.e. resistance to disturbance and speed to return to a stable equilibrium)
(Folke et al. 1996).
Process of riparian forests degradation
The impact of natural events and human activities on the structure and functioning of riparian
areas in Benin include changes in the hydrology of waterways, alteration of the geomorphic
structure of riparian areas, and the removal of riparian vegetation. From field observations,
we have recognised three major non-successive phases of riparian forests degradation:
a) Structural simplification of dense and undisturbed stands, which is caused either by
selective removal of big stems, destruction of the understorey (e.g. due to domestic livestock
grazing or fire in the savanna region), or localised landslides that induce canopy opening and
gaps at lower heights.
b) Floristic richness depletion, which involves constant and abusive harvest (e.g. systematic
cutting, collection of all fruits or seeds of certain valuable plants), reduction in stands size
(too frequent fires, shifting cultivation, dam or bridge construction), etc.
c) Complete disappearance consecutive to severe or irreversible transformation of the
vegetation to other land and water-use (e.g. farm, dam, road, bridge construction, housing,
etc.).
5.5. CONCLUSIONS
Riparian forests in Benin display a physiognomy (i.e. canopy height and vertical
stratification) that is highly variable, though the understorey is generally dense. Tree size
(stem diameter and height) is generally small. Most RF sites in Benin are degraded, highly
influenced by adjacent more open ecosystems (edge effect) and characterised by open
canopies. This is shown by a high amount of grasses (particularly annual species), a relatively
low percentage of phanerophytes (particularly mega-phanerophytes and woody lianas as
compared to tropical dense forests), and a high abundance of medium size and small trees.
Therefore they can be termed as relatively low hygrophile forests with irregular canopy.
The most prominent phytogeographic types of riparian flora range from the GuineoCongolian ones to species of wide distribution all over the tropics, and in tropical Africa.
This substantiates the hypotheses of an early period of continuous dense forest block in West
and Central Africa, and the influence of edge effects.
In terms of diameter class distribution, basal area and stem density RFs in Benin are
very similar to many tropical dense forests and in particular to the West African ones. The
phenomenon of species dominance, whereby few species contribute much to the total
abundance and basal area, is also a major characteristic of RFs in Benin. Many riparian
forests stands are facing various levels of structural and floristic simplification. This includes
fundamental transformations in vegetation physiognomy, from dense pristine riparian forests
to riparian scrub, or bare land. Action should be taken to overcome these threats to
biodiversity by maintaining the existing legal protection and/or durable exploitation.
49
Chapter 6
A PHYTOSOCIOLOGICAL STUDY OF RIPARIAN FORESTS IN BENIN
(WEST AFRICA)
In press, Belgian Journal of Botany
NATTA A.K., SINSIN B. and VAN DER MAESEN L.J.G.
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
Chapter 6
A PHYTOSOCIOLOGICAL STUDY OF RIPARIAN FORESTS IN BENIN
(WEST AFRICA)*
NATTA A.K. (1), SINSIN B. (2) and VAN DER MAESEN L.J.G. (3)
(1)
(2)
(3)
Department of Environment Management, Faculty of Agronomic Sciences. University of Abomey-Calavi
FSA/DAGE/UAC 01 BP 526 Cotonou Benin; aknatta@yahoo.com; natta@bj.refer.org
Department of Environment Management, Faculty of Agronomic Sciences. University of Abomey-Calavi
FSA/DAGE/UAC 01 BP 526 Cotonou Benin; bsinsin@bj.refer.org
Biosystematics group, Building no. 351, Generaal Foulkesweg 37 6703 BL, Wageningen University, the
Netherlands; jos.vandermaesen@wur.nl.
(*) In press, Belgian Journal of Botany
Abstract. - Floristic ordination and classification of riparian forests in Benin were derived
from a comprehensive floristic inventory. TWINSPAN classification and DCA analysis of a
data set of 818 plant species and 180 relevés yielded 12 plant communities. Importance of
waterways, relief, topography, latitude and longitude were the five major environmental
gradients that best differentiated riparian plant communities. A syntaxonomic classification
of the identified riparian forests plant communities is presented. Riparian forests in Benin
belong to the Mitragynetea Schmitz 1963, which is the phytosociological class of hygrophile
fresh water forests of tropical Africa. Based on similarities of ecological conditions and
floristic composition, we classified the 12 plant communities into 3 orders: Alchornetalia
cordifoliae Lebrun 1947, Lanneo-Pseudospondietalia Lebrun & Gilbert 1954 and
Pterygotetalia Lebrun & Gilbert 1954.
Key words: Riparian forests, classification, ordination, syntaxonomy, Benin.
6.1. INTRODUCTION
The recognition and definition of natural groupings of plant species in vegetation
formations have always been a challenging field of study for many researchers.
Phytosociology is concerned with the detection and characterisation of distinct plant
assemblages as social units (groups of similar relevés), which repeat themselves over space
(BRAUN-BLANQUET 1972, KENT & COKER 1992). Some authors have shown the applicability
of the phytosociological approach in forests of the humid tropics (HOMMEL 1990). Plant
communities provide a far better means of analysing the relation between vegetation and
climate, edaphic, topographic and human factors (LAWSON et al. 1970, GARTLAN et al. 1986,
RUSSEL-SMITH 1991) than the physiognomic approach does (HOMMEL 1990).
Multivariate methods for vegetation classification and ordination have been applied to
tropical West African forest types (SWAINE & HALL 1976, GARTLAN et al. 1986, SOKPON
1995, KOKOU 1998). These methods have shown to be useful in extracting meaningful
gradients and vegetation types or plant communities in tropical humid forests, which seems
too complex for satisfactory analysis by more traditional methods (SWAINE & HALL 1976). In
Benin, apart from the work of SOKPON et al. (2001), no attempt was made so far in the field
53
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
of the phytosociology of riparian vegetation in general, and in particular riparian forests
(hereafter abbreviated as RFs).
Control over internal structure of riparian ecosystems, in particular vegetation, has
been attributed to fluvial dynamics, frequent perturbations (e.g. flooding) and soil moisture
(MEDLEY 1992, THOMAS 1996, PIEGAY 1997, LEINARD et al. 1999), all of which are processes
related to topography (HANSON et al. 1990).
In this paper we report the results of a multivariate analysis of RF based on
comprehensive floristic data collected throughout Benin, a country located at the
discontinuity of the tropical West African rain forest block. The aim of the paper is to assess
whether meaningful ecological factors and environmental gradients can be derived from
numerical analysis of RF samples and to provide a floristic classification (i.e. a typology of
RF plant communities). A related aim is to document the environmental relations of the
identified plant communities and their characteristic species.
6.2. MATERIAL & METHODS
6.2.1. Study area
All types of ecological regions of Benin, excluding the coast, beaches, coastal
lagoons, lakes and marshlands, were surveyed. The study area (Figure 6.1) covered the
latitudes 7º10 to 12º20 N. Representative sites along rivers and streams were surveyed in the
Guinean region (Samiondji and Bétékoukou), the Sudano-Guinean zone (Toui-Kilibo,
Ouèssè, Bétérou, Onklou-Daringa and Bassila-Pénéssoulou) and the Sudanian region (Ouaké,
Affon, Natitingou, Koussoukouingou, Toukountouna, Pendjari Biosphere Reserve, Kandi,
Ségbana and Malanville).
6.2.2. Phytosociological relevés in riparian forests
Floristic data were collected in 12 riparian forests sites (Table 6.1) using the BRAUNBLANQUET (1972) method for vegetation analysis. Rectangular plots, each of 500 m2, with
variable length and width, were preferred for practical reasons to fit with any shape of the
waterway and the structural uniformity (physiognomy of the RF). When all terrestrial plants
in all stages are included, many authors found smaller plot sizes (500 to 1000 m2) acceptable
to allow the classification of forest plot data by means of tabular comparison (HOMMEL
1990). Field plots (50m by 10m) were established in fragmented forests in Southern Togo
(KOKOU et al. 2002). Rectangular plots of 500 or 1000 m2 were used to collect floristic data
in several edaphic forests in Benin (SOKPON et al. 2001). In the strip-like forest fragments
that are RF, homogeneous plots of 500 m2, with varying length and width, seem to be of a
reasonable size for the numerical ordination and classification of phytosociological relevés
(see also Chapter 8).
At each site plots were established, in homogeneous stands, at about 100 m from each
other, from a random starting point. In each plot all vascular plants were recorded and
specimens taken for the herbarium. As we are dealing with a particular vegetation formation
defined by certain hydrographic, soil and topographic characters, relevés taken into account
in the present paper were selected from a larger data set with the greatest possible uniformity
or homogeneity in regard to their physiognomy, species composition, and dominant species.
Plots highly disturbed or less than 10 m wide, and outlier relevés were removed from the
original 232 plots. This yielded 180 relevés and 818 plant species. These relevés are taken as
the most representative of the ecological conditions occurring in the studied sites.
Plant nomenclature follows KEAY & Hepper (1954-1972), BRUNEL et al. (1984),
BERHAUT (1967) and LEBRUN & STORK (1991-1997). The conventional six-part abundancedominance scale of BRAUN-BLANQUET (1972) is adopted in the present work. This scale was
54
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
applied for each species within each of the three strata: Dominant (height > 10 m), Medium
(height between 3 and 10m) and Understorey (< 3 m high) species.
N
Gbèssè
*
Pendjari
Reserve
*
Towns
Yarpao
*
Pénéssoulou
Bétérou
*
*
*
Idadjo
*
Bétékoukou
*
Samiondji
1 : Grand-Popo
2 : Cotonou
3 : Ouidah
4 : Porto-Novo
5 : Niaouli
6 : Aplahoué
7 : Pobè
8 : Bohicon
9 : Kétou
10 : Savalou
11 : Savè
12 : Bantè
13 : Tchaourou
14 : Bassila
15 : Djougou
16 : Bembèrèkè
17 : Natitingou
18 : Parakou
19 : Kouandé
20 : Tanguiéta
21 : Toui
22 : Kandi
23 : Banikoara
*Sites of data collection
(with > 20 plots)
States limits
Major waterways
*
Isohyets
FIG. 6.1.- Study sites where data were collected in riparian forests in Benin
55
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
Table 6.1: Floristic and stand characteristics of riparian forests throughout Benin
Climatic
region
Sites
Type of riparian
forests
Relevés
(n) *
Species
Trees
Basal area
richness/ha density/ha (BA/ha)
Guinean Samiondji (Sa)
Ouémé river
31
249
726
41.7±17.4
Sudano
Streams in
Pénéssoulou forest
Ouémé river
Ouémé river
34
343
544
38.9±15.3
17
30
129
195
748
785
35.8±10.7
41.6±15.8
8
22
358
540
732
33.1±7.9
42.7±9.9
Pénéssoulou (Pe)
Idadjo (Id)
Guinean Bétérou (Be)
Sota (So)
Yarpao (Ya)
Sudanian Bétérou (Be), Batia (Ba)
Sota river
Streams at hills feet
Streams on plateau
with inundations
Bétérou (Be), Toucountouna
Streams on plateau
(Tc), Gbèssè (Gb), Péhunco (Ph) seldom inundated
10
-
602
40.5±8.9
16
-
531
25.9±4.6
Porga (Po), Konkombri (Ko)
12
-
571
33.1±17.5
180
129-358
253-785
41.5±17.9
Pendjari river
Total
* n = number of phytosociological relevés; Trees = stems with dbh ≥ 10 cm.
6.2.3. Data analysis
For data analysis of the 818 plant species and the 180 representative RF samples, both
ordination and numerical classification methods were used. No species of the representative
relevés was omitted in the analysis. Detrended Correspondence Analysis (DCA) (HILL &
GAUCH 1980, TER BRAAK 1995, TER BRAAK & SMILAUER 1998) was used as indirect gradient
method to detect the underlying ecological gradients/factors within the RF relevés set. The
definition of plant communities was achieved through Two-Way Indicator Species
(TWINSPAN) (HILL 1979). Dendrograms were obtained with STATISTICA® 5.1 (1998)
using Euclidean distance as distance measure and Ward’s minimum variance as aggregation
method.
The classification of the individualised plant communities followed the work of
LEBRUN (1947), LEONARD (1952), SCHNELL (1952), LEBRUN & GILBERT (1954), SCHMITZ
(1963, 1971, 1988), and SOKPON et al. (2001).
6.3. RESULTS
Environmental gradients/factors of riparian forests samples partition
The stand characteristics of the surveyed riparian forests are presented in Table 6.1.
Detrended Canonical Analysis (DCA) of 180 relevés and 818 plant species (Figure 6.2)
showed a major indirect floristic factor (axis 1) correlated with the type or importance of
waterways: small (streams) or large (rivers). There are three groups of plant communities: RF
plant communities of the Pénéssoulou forest (group 1), along streams all over the country
(group 2) and along rivers (group 3). This result is in accordance with the TWINSPAN
classification output. The TWINSPAN output table (three A0 format sheets) is available on
request from the authors. Each one of the 3 groups is further analysed in a partial analysis.
Group 1: Riparian Forests (RF) of Pénéssoulou protected forest
The partial DCA analysis of 34 samples and 282 plant species showed a floristic
gradient (axis 1) of two plant communities correlated with topography (Figure 6.3). This
56
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
result is in accordance with the dendrogram (Figure 6.4). The denomination of each plant
community is derived from two characteristic species, which are also the most indicative of
the ecological conditions, topography in this case. They are:
1 - Plant community of Isolona thonneri and Callichilia barteri (10 relevés) along streams in
the centre of Pénéssoulou protected forest. This plant community is located at the lowest
parts of the forest along streams with frequent inundation.
2 - Plant community of Motandra guineensis and Pararistolochia goldieana (24 relevés)
along streams at the East and West parts of Pénéssoulou reserve forest. This community is
located in drained sites (i.e. seldom inundated).
Group 2: Riparian forests along streams all over the country
DCA of 48 relevés and 284 plant species (Figure 6.5) showed a major floristic factor
(axis 1) correlated with land form variation, namely relief (RF at hill foot versus RF on
plateau) and topography (RF on plateau regularly inundated versus RF on plateau seldom
inundated). In general, this result is in accordance with the TWINSPAN output and the
dendrogram (Figure 6.6). Four plant communities are discriminated from the ordination and
classification analysis:
3 - Plant community of Chrysobalanus icaco subsp. atacoriensis and Pentadesma butyracea
along streams at hill feet in the Atacora mountain chain (22 relevés).
4 - Plant community of Alchornea cordifolia and Ficus trichopoda along streams on regularly
inundated plateaus all over the country (10 relevés).
5 - Plant community of Berlinia grandiflora and Khaya senegalensis along streams on
drained plateaus (i.e. seldom inundated), mainly in the Sudanian region of the country (8
relevés).
6 - Plant community of Raphia sudanica and Oxytenanthera abyssinica along streams on
drained plateaus, mainly in the Sudanian region (8 relevés).
Group 3: Riparian forests along rivers all over the country
DCA of 98 relevés and 291 plant species (Figure 6.7) showed a major gradient
correlated with latitude (Axis 1: Guinean region of South versus Sudano-Guinean and
Sudanian regions in the Central and North Benin) and longitude (Axis 2: extreme West
versus Centre and East Benin). 3 sub-groups of relevés are shown:
(a) - Bétérou, Idadjo and Sota in the Centre;
(b) - Porga and Konkombri in the North-West and
(c) - Samiondji in the South of the country.
Each subgroup is further analysed in a partial DCA ordination and a classification.
Only the DCA of sub-group (a), (Figure 6.8) showed a floristic gradient correlated
with latitude (Axis 1): Centre of the Sudano-Guinean zone (Idadjo), North of the SudanoGuinean zone (Bétérou), and Sudanian region (Sota) at Ségbana latitude. These results are in
accordance with the dendrogram (Figure 6.9) and TWINSPAN classification (Figure 6.10).
Therefore five plant communities are discriminated in RF along rivers:
7 - Plant community of Cynometra megalophylla and Parinari congensis along the Ouémé
river in the Guinean region of Southern Benin (31 relevés).
8 - Plant community of Capparis thonningii and Crateva adansonii along the Ouémé river in
the Sudano-Guinean region of Central Benin (30 relevés).
9 - Plant community of Lepisanthes senegalensis and Drypetes floribunda along the Ouémé
river in the Sudano-Guinean region of Central Benin (17 relevés).
10 - Plant community of Uapaca heudelotii and Irvingia smithii along the Sota river in the
North-East of the country (8 relevés).
57
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
11 - Plant community of Garcinia livingstonei and Combretum acutum along the Pendjari
river in the North-West of the country (12 relevés).
6.4. DISCUSSION
6.4.1. Environmental gradients and factors partitioning riparian forest plant
communities
In total, 5 environmental gradients/factors were identified from the ordination process:
1 - Type or importance of waterways: RF along streams / RF along rivers;
2 - Relief: RF at hill feet / RF on plateau;
3 - Topography: RF on regularly inundated plateau / RF on seldom inundated plateau;
4 - Latitude: RF in the Guinean region of South-Benin / RF in the Centre of the SudanoGuinean region (Idadjo) / RF in the North of the Sudano-Guinean region (Bétérou) / RF in
the Sudanian region (Sota) at Ségbana latitude;
5 - Longitude: RF in the extreme West / RF in the Centre and East of the country.
The three sets of the RF data: Pénéssoulou, rivers and streams (see Figures 6.2, 6.10
and 6.11), somehow match the vegetation formations and phytogeographical districts of
Benin. The flora of RF along rivers is much more dependent on frequent floods than the flora
along streams, where the surrounding vegetation greatly influences stream side vegetation.
On the other hand the Pénéssoulou region in West-Central Benin is the easternmost point of
the dry semi-deciduous forest fire subtype (HALL & SWAINE 1981). This vegetation also
covers Central Togo, and Central and South Ghana. In this hygrophile enclave surrounded by
savannas, RFs occupy thalweg bottoms and share numerous species (e.g. Cleistopholis
patens, Pierrodendron kerstingii, etc.) with the Pobè region, which is the wettest region of
Benin in the South-East.
The first 2 DCA axes, which have the highest contribution to the total variation in the
data set, were linked to environmental gradients/factors: importance of waterway, relief,
topography, latitude and longitude. Depending on forest types and sites characteristics,
various authors have also documented the major environmental gradients/factors that explain
the grouping of floristic relevés. In an application of ordination and classification to closedcanopy forest in Ghana, rainfall was not the only factor with a large effect on forest
composition. Also rock type and occasional ground fires were important in drier forests.
Altitude was important in wetter forest (HALL & SWAINE 1976, SWAINE & HALL 1976).
RUSSEL-SMITH (1991) identified a primary latitude-moisture gradient and a subsidiary
topographic-drainage gradient as the two major environmental gradients underlying the
grouping of 1219 sites and 559 rain forest taxa in Northern Australia. According to HANSON
et al. (1990), the most important processes related to topography are flooding, soil moisture
and fluvial dynamics. For various tropical forests, altitudinal zonation, rainfall variability, as
well as the influence of rock type, occasional ground fires, play an important role in the
typology of forest communities (NAKASHIZUKA et al. 1992).
The interpretation of the DCA gradients is broadly in accordance with valuable work
on the phytosociology of edaphic and hygrophile vegetations of fresh water (e.g. LEBRUN &
GILBERT 1954, SCHMITZ 1988, SOKPON et al. 2001). LEBRUN & GILBERT (1954) enumerated
the ecological factors that play a role in the discrimination of edaphic forests linked to
hydromorphic soils: a) variation of the water table (i.e. level and periodicity) above the soil
surface and in the edaphic profile; b) the degree of alluviation; and c) the intensity of soil
drainage. They further mentioned that physiographic characters of the site, the nature of
water and substrate might play a role in the typology of hydromorphic vegetations, but
marginal as compared to the precedent factors. Generally, the influence of flood and variation
of the water table is considerable (MANDANGO & NDJELE 1986). SOKPON et al. (2001) found
58
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
that latitude, drainage and duration of flooding were the best discriminating factors for 51
wetland forest relevés in Benin. Meanwhile, their typology did not go beyond the formation
level (i.e. gallery, swamp, periodically inundated and riverine forests). A limitation of that
work is that no plant communities were identified, as in the present study. Also the
differentiation in such broad vegetation types, based on the floristic composition, does not
always conform to the reality of the terrain.
The amount of variation explained by the DCA analysis is fairly low (14 to 21%). A
reason might be the high number of relevés (180), species (818) and uncontrolled variables.
GANGLO (2000), and HOUINATO (2001) found almost the same range of values for the
explained variance, respectively 16 to 22% for plantations understorey in Central and
Southern Benin, and 18 to 25% for savanna in Central Benin. BOCARD et al. (1992) pointed
out that it may not be feasible to measure all the environmental variables (in the broad sense:
biological interactions and external environmental factors) that are relevant in an ecological
study. Given these constraints, and at this stage of the study of RFs in Benin, the underlying
factors found through the interpretation of the DCA axes can nevertheless be considered as
important in the structuring of these plant communities.
6.4.2. Classification of RF relevés
The apparently ‘homogeneous’ site characteristics of the narrow RF on continually
moist, sometimes flooded and with sandy shores, hides in fact more complex fluvial
processes moulded by the regional climate, relief, and human activities. This allows a variety
of groupings of plant species. The floristic classification of RF relevés allows us to identify
11 constant plot assemblages, each one referring to a plant community (i.e. plant association).
An additional riparian forest plant community (Mimosa pigra and Ficus asperifolia) regularly
cited in the literature (LEBRUN & GILBERT 1954, MANDANGO 1982, MANDANGO & NDJELE
1986, SCHMITZ 1988) is widely distributed on sandy banks of rivers in tropical Africa.
However it could not be clearly distinguished and classified as the 11 previous types through
the DCA and TWINSPAN analysis. Therefore we can say that there are at least 12 plant
communities, each having its distinct and well-developed stands, in the Benin riparian
forests. This classification, based on comprehensive phytosociological relevés, is both readily
interpretable and repeatable.
The two species forming the denomination of the plant community were carefully
selected among the characteristic (exclusive, selective, preferential) species according to the
degrees of fidelity defined by BRAUN-BLANQUET (1972). These species are among
characteristic species having a high ecological indicator value. Each plant community may
occur in many stations (or sites) but they all exist in only one well-defined and ecologically
characteristic habitat, e.g. RF along rivers in the Guinean region of Southern Benin, RF along
streams at hill feet in the North West of the country, and RF along streams on plateau in the
Sudanian region.
In general, the results from DCA, TWINSPAN classification and dendrograms are
similar and they appear to be complementary techniques in the partition of groups of relevés
based on their floristic composition affinity.
6.4.3. Relation plant communities/environmental factors
Although some environmental factors were found to induce a certain grouping of RF
samples based on presence/absence of species, the interpretation of causal relationships
between the identified plant communities and these gradients should be taken with care
(KENT & COKER 1992). From detailed understanding of the relationship between riparian
vegetation and ecological factors, we might say that the importance of waterways, latitude,
longitude, relief and topography act indirectly through climatic, geomorphologic and edaphic
59
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
variables. HOMMEL (1990) already found that both physiognomy and floristic composition
clearly reflect the relation between climate and vegetation.
On the one hand, the most important climatic conditions in Benin are rainfall,
temperature, hygrometry, Potential Evapotranspiration (PET) and seasons duration; on the
other hand, soil texture, structure and nutrient status, and bedrock type are the most important
edaphic variables. The combination of these variables modulated by the relief and topography
not only control the volume, velocity of water and the duration of inundation but also specific
requirements of individual species in these edaphic and hygrophyle plant formations.
Generally for plant communities, the effective external factors are numerous and variable,
and the possible combinations so manifold. Also the overlapping is frequent and the relation
of habitat to plant community is not a simple and reversible function. In consequence, a clear
and unequivocal delimitation of the identified plant communities’ habitats according to
operative external factors appears quite unattainable (BRAUN-BLANQUET 1972).
As plant species experience the conditions provided by many environmental variables
(TER BRAAK 1987), from the field data we can draw preliminary conclusions concerning the
response of certain species to the combination of the ecological variables cited above. Species
that seem to be influenced by climate and edaphic conditions along waterways in:
- the Guinean region and Sudano-Guinean zone: Cynometra megalophylla, Lepisanthes
senegalensis, Drypetes floribunda, Cassipourea congoensis, Cleistopholis patens,
Napoleonaea vogelii, Pierreodendron kerstingii;
- the Sudano-Guinean zone: Pararistolochia goldieana, Callichilia barteri, Detarium
senegalense, Motandra guineensis, Pandanus candelabrum, Leea guineensis, Khaya
grandifoliola, Pseudospondias microcarpa;
- the Sudano-Guinean and Southern Sudanian zones: Pentadesma butyracea, Isolona
thonneri, Raphia sudanica, Celtis toka, Garcinia ovalifolia, Ixora brachypoda, Uapaca
togoensis, Ficus trichopoda;
- the Sudanian region: Chrysobalanus icaco subsp. atacoriensis, Brenadia salicina,
Chionanthus niloticus, Eriocoelum kerstingii, Thunbergia atacoriensis, Oxytenanthera
abyssinica, Synsepalum passargei;
- the Northern Sudanian zone: Irvingia smithii, Garcinia livingstonei, Combretum acutum;
- the whole country: Pterocarpus santalinoides, Mimosa pigra, Ficus asperifolia, Dialium
guineense, Elaeis guineensis, Cola laurifolia, Berlinia grandifolia, Parinari congensis,
Kigelia africana, Crateva adansonii, Morelia senegalensis, Syzygium guineensis var.
guineensis, Alchornea cordifolia, Khaya senegalensis, Uapaca heudelotii.
Biotic factors such as intra and inter species relationships and human induced
disturbance, which play a role in the grouping and layering plant species are not explored in
the present study. This preliminary information about the occurrence of individual species
might be useful in drawing distribution maps of these species and the plant communities in
which they reach their optimal development in Benin.
6.4.4. Syntaxonomic relationships of the 12 plant communities
We present a syntaxonomic classification of the identified plant communities under
the hierarchical structure (orders and suitable alliances) of edaphic forest formations of fresh
water occurring in West and Central Africa. Thus far, in Benin such a general system of RF
plant communities classification is not available.
All tropical African RF belongs to the Mitragynetea SCHMITZ 1963, the class of
hygrophile forests of fresh water (SCHMITZ 1971, 1988). As for gallery forests of the Congo
basin (see LEBRUN & GILBERT 1954), RF of Benin have either a Guinean or SudanianZambesian tendency. In Benin the most frequent plants of the Mitragynetea SCHMITZ 1963
60
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
are Pterocarpus santalinoides, Cola laurifolia, Syzygium guineense, Dialium guineense,
Morelia senegalensis, Elaeis guineensis, Parinari congensis, Manilkara multinervis,
Phaulopsis barteri, P. ciliata, Taccazea apiculata, Achyranthes aspera, Afzelia africana,
Xylopia parviflora, and Antidesma venosum.
Based on field data, similarities of ecological conditions and floristic composition, we
classified the 12 RF plant communities into 3 orders that are Alchornetalia cordifoliae
LEBRUN 1947, Lanneo-Pseudospondietalia LEBRUN & GILBERT 1954 and Pterygotetalia
LEBRUN & GILBERT 1954 (Figure 6.11).
* Alchornetalia cordifoliae LEBRUN 1947:
It is the order of shrubby and pre-forest plant communities following herbaceous
semi-aquatic formations along larger waterways (Ouémé, Sota rivers), with relatively
important alluvial deposits. This order also occurs in larger valleys along streams with a long
period of inundation. The most frequent species are Mimosa pigra, Ficus asperifolia,
Alchornea cordifolia, Ficus spp., etc. Within this order we can distinguish 3 alliances each
containing one plant community:
1 – Mimosion pigrae MANDANGO 1982, alliance with a Guinean tendency although it
can be found in the Sudano-Zambesian region (LEBRUN & GILBERT 1954). The Mimosa pigra
and Ficus asperifolia plant community is classified in this alliance. It occurs as dense
shrubby but isolated stands always found on sandy bars close to the lowest level of the water
along rivers. Therefore, every year it is flooded for 3 to 4 months. MANDANGO & NDJELE
(1986) defined it as an association in the Congo basin.
2 – Alchorneion cordatae LEBRUN 1947, alliance of shrubby and pioneer plant
communities of fresh water. The Alchornea cordifolia and Ficus trichopoda plant community
is the only that can be classified in this alliance. In fact this community is widely distributed
over the country in sites where the substrate is either constantly humid or periodically
inundated or muddy with clayey soil. Several authors (e.g. LÉONARD 1952, KALANDA 1981,
LEJOLY & MANDANGO 1982, MANDANGO & NDJELE 1986) have described a variety of plant
communities with Alchornea cordifolia as characteristic species.
3 – Uapacion heudelotii LEBRUN & GILBERT 1954. This alliance harbours the plant
community of Uapaca heudelotii and Irvingia smithii that was already defined as association
by LÉONARD (1947) and KALANDA (1981). It is widely distributed throughout tropical Africa
from Congo to Senegal (SCHMITZ 1988). Meanwhile in Benin the two characteristic species
are only seen associated in the North, showing the association’s Northern Sudanian tendency.
* Lanneo-Pseudospondietalia LEBRUN & GILBERT 1954:
This order groups all riverine forest with a seasonally alternation of heavy inundation
and short period of drainage. It is typical to the Guinean region (e.g. Samiondji latitude),
meanwhile penetrates the Sudano-Guinean zone (Ouèssè, Bétérou and Pénéssoulou latitudes).
The most frequent species of this order are Napoleonaea vogelii, Cassipourea congoensis,
Salacia pallescens, Synsepalum brevipes, Cleistopholis patens, Pseudospondias microcarpa,
Pierreodendron kerstingii, Spathodea campanulata, Trilepiseum madagascariense,
Parquetina nigrescens. The associated plant communities are:
a) - In the Guinean region of Southern Benin at Samiondji latitude: Cynometra megalophylla
and Parinari congensis along the Ouémé river.
b) - In the Sudano-Guinean region of Central Benin at Ouèssè and Bétérou latitudes:
Lepisanthes senegalensis and Drypetes floribunda; Capparis thonningii and Crateva
adansonii along the Ouémé river.
c) - In the Pénéssoulou protected forest: Isolona thonneri and Callichilia barteri; Motandra
guineensis and Pararistolochia goldieana.
61
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
* Pterygotetalia LEBRUN & GILBERT 1954:
This order groups all riparian forests with a Sudanian-Zambesian tendency occurring
in valleys or on drained plateau with short periods of inundation (few days during the rainy
season) and long periods of dryness. For LEBRUN & GILBERT (1954), this order is not only
edaphic but also physiographic as steep banks of hills can hem in streams. The most frequent
species are Berlinia grandiflora, Khaya senegalensis, Raphia sudanica, Oxytenanthera
abyssinica, Uapaca togoensis, Diospyros mespiliformis, Vitex doniana, Strychnos nigritana,
Oxyanthus unilocularis, Chionanthus niloticus, Garcinia ovalifolia, etc. This order includes:
a) - gallery forests of Central and North Benin on drained plateau (Berlinia grandiflora and
Khaya senegalensis; Raphia sudanica and Oxytenanthera abyssinica plant communities),
b) - gallery forests at hill feet (Chrysobalanus icaco subsp. atacoriensis and Pentadesma
butyracea plant community), and
c) - gallery forests along the Pendjari streams and its tributaries in the North-West of the
country (Garcinia livingstonei and Combretum acutum plant community).
Sites descriptions, floristic composition, physiognomy, ecological spectra and
structural characteristics and phytosociological tables for each plant community will be
discussed in separate papers (Natta et al., in preparation, for association of Garcinia
livingstonei and Combretum acutum).
62
+5.0
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
Pe11
Pe7
Pe12
Pe9
Pe19
Pe8
Pe13
Pe5
Pe2
Pe25
Pe21Pe6
Pe15
Pe20
Pe16
Pe1
Pe22 Pe14
Pe17
Pe27Pe49 Pe18
Pe54 Pe47
Pe24 Pe55
Pe4
Pe53
Pe28Pe23Pe51
Pe48
Pe52
Pe26
Pénéssoulou (Group I)
Sa29Sa28
Sa26
Sa25
Id7 Sa8
Id8
Id9 Id4
Id1
Id15
Id3
Be7
Id16
Id2 Id14
Be30
Streams all over the
country (Group II)
Ya1
Ya10
Ya5
Ya17 Ya8 Ya3
Ya14
Di4 Di3
Ya11 Ya15
Ya6
Ya13
Ya18
Ya7 Ba1
Di1
Di2
Ya19
Ou1
Be27Be20
Ph2
tk2
Gb5
Gb3
So8
Be29
Tc4
Tc7
Tc1Tc2
Tc5
Tc6
Gb4
Tc3
Gb2
Wa2
Ba7
Id12
Be23 Be26
Be1
Be22Be18
So7
Be13
Be14
So4So6
So5
So3
So1
Sa35
Sa19 Sa37
Sa30
Sa31
Sa9
Sa11 Sa23
Sa15
Be24
Be3
Be21
Be5
Sa36
Sa14
Sa33
Sa13
Sa7
Sa24Sa4
Sa22
Sa12
Id5
Id10
Be2
Be4
Id13
So2
Ph1
Sa34
Id11
Id6
tk1
Ya4
Ya9
Sa27
Sa38
Sa18
Sa32
Sa6
Be28 Sa16
Sa17
Sa10 Sa20
Sa3
Sa21
Rivers (Ouémé, Sota,
Alibori, Pendjari)
Group III
Be11Be10
Be15
Be16 Be9 Be17
Be12
Be25Be6
Be19
Pj9
Tc8
Be8
Pj8
Po2
Ko5
Ba5
Ba4
Pj10
Ko7
Wa1
Ba3
Po3
Ko6
Po1
Po4
Pj11
Ba2
Ba8
Relevés along streams
Relevés along rivers
Axis 1 symbolises the type or importance of water ways:
Relevés of RF along Streams versus those along Rivers
FIG. 6.2. - Detrended Correspondence Analysis ordination of 180 riparian forests plots in Benin, showing 3 groups of plant
communities. This result is in accordance with the TWINSPAN analysis. Relevés of RFs along streams versus those along rivers.
-2.5
63
-1.5
+5.0
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
Axis 2
Plant community 1:
west & east of Pénéssoulou forest
(higher and medium parts)
Plant community 2:
centre of Pénéssoulou forest
(lowest part of the forest
with frequent Inundation)
Pe53
Pe4
Pe2
Pe51 Pe47
Pe52
Pe55
Pe48
Pe15
Pe14
Pe49 Pe23
Pe20 Pe22
Pe24Pe27 Pe25
Pe13
Pe21
Pe26
Pe16
Pe12Pe54
Pe6
Pe18
Pe17
Pe8 Pe1Pe5
Pe28
Pe19
Pe7
Pe11
Pe9
Axis 1
Axis 1 symbolises the topography: bottom versus medium & centre of valleys
FIG. 6.3. - DCA ordination of 34 plots of Group I (RF of Pénéssoulou reserve forest; see Fig. 6.2.) showing 2 plant communities
correlated with topography. (
) subdivisions obtained with TWINSPAN.
64
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
Dendrogram of 34 releves of Penessoulou (WARD, Euclidian distance)
110
100
90
Dissimilarity (%)
80
70
60
50
40
30
PE28
PE25
PE27
PE23
PE22
PE55
PE47
PE4
PE54
PE17
PE19
PE14
PE12
PE9
PE13
PE5
PE2
PE26
PE49
PE24
PE21
PE20
PE48
PE53
PE52
PE51
PE16
PE15
PE18
PE11
PE7
PE6
PE8
PE1
Plant community 1
Plant community 2
FIG. 6.4. - Dendrogram of 34 plots in the Pénéssoulou reserve forest showing 2 plant communities.
(
) subdivisions obtained with TWINSPAN.
65
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
Axis 1 symbolises land form variation:
Relief (hills feet/plateau) and
Topography of plateau (inundation/no inundation)
Axis 2
Axis 1
RFs on plateau with
no inundation
Riparian forests (RFs) at hill feet
RFs on plateau regularly inundated
FIG. 6.5. - DCA ordination of 48 plots of RF along streams (see Group II in Fig. 6.2) showing 4
plant communities. (
) divisions obtained with TWINSPAN.
Tree of 48 releves of riparian forests along streams (WARD, Euclidian distance)
110
RFs on
inundated
plateau
RFs at hills feet
100
RFs on plateau
seldom inundated
RFs on plateau seldom
inundated
90
80
60
50
40
30
20
FIG. 6.6. - Dendrogram of 48 plots of RF along streams showing 4 plant communities.
(
) subdivisions from TWINSPAN.
66
GB2
GB3
GB4
GB5
WA1
SO8
WA2
TK1
PH1
PH2
TK2
OU1
BE29
TC1
BE30
TC7
TC5
TC6
TC2
TC3
BA2
TC4
BA8
BA3
BA4
BA5
BA7
DI1
BA1
DI2
DI3
DI4
YA1
YA9
YA10
YA8
YA11
YA3
YA15
YA14
YA5
YA4
YA6
YA7
YA13
YA17
YA18
10
YA19
Dissimilarity (%)
70
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
Axis 2
Centre &
East > 20 N
(c)
(a)
(b)
West < 10 50’ N
Axis 1
RFs along rivers in the Centre and North
RFs along rivers in the South
FIG. 6.7. - DCA ordination of 98 plots of RF along rivers all over the country (see Group III in
Fig. 6.2) showing 3 sub-groups according to Latitude (axis 1) and Longitude (axis 2).
Axis 2
Axis 1
Sota river at Ségbana
latitude (North)
Ouémé river at Bétérou
latitude (Centre)
Ouémé river at Ouèssè
latitude (South)
FIG. 6.8. - Partial DCA ordination of sub-group (a) of Fig. 6.7. (i.e. 55 plots of RF along rivers at
Ouèssè, Bétérou and Ségbana) showing 3 plant communities according to Latitude (axis 1).
67
Dissimilarity (%)
120
100
80
60
40
20
0
Centre
North west
South Sudanian region
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
South of the country
68
North
Sudanian
Guinean region: Samiondji = Sa; Sudano-Guinean zone: Pénéssoulou = Pe; Bétérou = Be; Idadjo = Id; Toui-Kilibo = Tk; Sudanian region: Yarpao = Ya; Péhunco =
Ph; Dimari = Di; Ouaké = Ou; Gbèssè = Gb; Sota = So; Warrana = Wa; Toucountouna = Tc; Batia = Ba; Pendjari = Pj; Porga = Po; Konkombri = Ko.
Name of riparian forest sites in Figs 6.2, 6.3, 6.4, 6.5, 6.6, 6.7, 6.8 and 6.9:
FIG. 6.9. - Dendrogram of 98 relevés of riparian forests along rivers in Benin, showing 5 plant communities.
(
) divisions from TWINSPAN
SA34
SA26
SA25
SA38
SA18
SA32
SA28
SA29
SA27
SA6
SA30
SA19
SA9
SA8
SA37
SA36
SA35
SA24
SA23
SA14
SA16
SA15
SA17
SA22
SA13
SA12
SA11
SA4
SA33
SA31
SA7
SA21
SA10
SA20
SA3
ID13
ID6
ID12
ID5
ID3
ID2
ID14
ID9
ID11
ID7
ID8
ID10
ID4
ID16
ID15
ID1
PO4
PJ8
PJ9
PO3
PO2
PO1
TC8
KO7
PJ11
PJ10
KO6
KO5
BE17
BE19
BE25
BE8
BE15
BE13
BE11
BE16
BE9
BE12
BE6
BE23
BE22
BE20
BE27
BE18
BE14
BE28
BE24
BE21
BE10
BE3
BE7
BE2
BE26
BE5
BE4
BE1
SO7
SO5
SO4
SO3
SO6
SO2
SO1
Riparian Forests (RF) data base
Twinspan
cut levels
Identified
gradients
1
Importance
of waterway
2
Latitude
3
Topography
Relief
Latitude
4
Topography
Latitude
5
Longitude
No.
Plant community
Number of plots
(180 relevés and 818 plant species)
RF along STREAMS (82 relevés)
RF along RIVERS (98 relevés)
Centre
(inundated)
Sudano-Guinean &
Sudanian regions
Streams all over
the country
Sudano-Guinean
zone (Pénéssoulou)
East & West
(drained)
Plateau
With inundation
Hill feet
(Yarpao)
South SudanoGuinean zone
(Idadjo)
drained
Guinean
region
(Samiondji)
North Sudano-Guinean &
Sudanian regions
North Sudano-Guinean
North-East Sudanian
North-West
Sudanian (Porga)
North SudanoNorth-East
Guinean (Bétérou) Sudanian (Sota)
1
Is. Th. & Ca. ba
10
2
Mo. gu. & Pa. go.
24
3
Al. co. & Fi. tr.
10
4
Be. gr. & Kh. se.
8
5
6
Ra. su. & Ox. ab. Ch. at. & Pe. bu.
8
22
7
Le. se. & Dr. Fl.
17
8
Ca. th. & Cr. ad.
30
9
Ua. he. & Ir. sm.
8
10
Ga. li. & Co. ac.
12
= Plant community of Isolona thonneri and Callichilia barteri along streams in the centre of Pénéssoulou reserve forest.
1 - Is. Th. & Ca. ba.
2 - Mo. gu. & Pa. go. =Plant community of Motandra guineensis and Pararistolochia goldieana along streams at the East and West parts of Pénéssoulou reserve forest.
3 - Al. co. & Fi. tr.
= Plant community of Alchornea cordifolia and Ficus trichopoda along streams on plateau regularly inundated all over the country.
=Plant community of Berlinia grandiflora and Khaya senegalensis along streams on drained plateau, mainly in the Sudanian region of the country.
4 - Be. gr. & Kh. se.
5 - Ra. su. & Ox. ab. =Plant community of Raphia sudanica and Oxytenanthera abyssinica along streams on drained plateau, mainly in the Sudanian region of the country.
=Plant community of Chrysobalanus icaco subsp. atacoriensis and Pentadesma butyracea along streams at hill feet in the Atacora mountain chain.
6 - Ch. at. & Pe. bu.
= Plant community of Lepisanthes senegalensis and Drypetes floribunda along the Ouémé river in the Sudano-Guinean zone of Centre Benin.
7 - Le. se. & Dr. Fl.
= Plant community of Capparis thonningii and Crateva adansonii along the Ouémé river in the Sudano-Guinean zone of Centre Benin.
8 - Ca. th. & Cr. ad.
= Plant community of Uapaca heudelotii and Irvingia smithii along the Sota river in the North-East of the country.
9 - Ua. he. & Ir. sm.
10 - Ga. li. & Co. ac.
=Plant community of Garcinia livingstonei and Combretum acutum along the Pendjari stream in the North-West of the country.
11 - Cy. me. & Pa. co. = Plant community of Cynometra megalophylla and Parinari congensis along the Ouémé river in the Guinean region of Southern Benin.
12 - An additonal plant community, Mi.pi & Fi.as. = Mimosa pigra and Ficus asperifolia, from the literature exist but could not be discriminated through the numerical classification proccess.
Fig. 6.10. - TWINSPAN classification hierarchy and relationship of the initial 180 relevés to 11 plant communities
69
11
Cy. me. & Pa. co.
31
Chapter 6: A phytosociological study of riparian forests in Benin (West Africa)
Riparian forests of Benin (7-12ºN)
Class of Mitragynetea Schmitz 1963
Typical Sudanian and Sudanian tendency riparian forests
Typical Guinean and Guinean tendency riparian forests
Order: Alchorneetalia
cordifoliae Lebrun 1947
Order: Pterygotetalia Lebrun & Gilbert 1954
Order: Lanneo-Pseudospondietalia Lebrun & Gilbert 1954
Order: Alchorneetalia cordifoliae
Lebrun 1947
Alliance: Alchorneion Alliance: Uapacion
heudelotii Lebrun
cordatae Lebrun 1947
Alliance: Mimosion
pigrae Mandango 1982
& Gilbert 1954
11
12
Association: Mimosa pigra
& Ficus asperifolia
Mandango & Ndejele 1986
7
8
Pl. com.: Lepisanthes
senegalensis &
Drypetes floribunda
Pl. com.: Cynometra
megalophylla &
Parinari congensis
2
Pl. com.: Motandra
guineensis &
Pararistolochia goldieana
Pl. com.: Capparis
thonningii & Crateva
adansonii
6
1
4
Pl. com.: Chrysobalanus
icaco subsp atacoriensis
& Pentadesma butyracea
Pl. com.: Isolona thonneri
& Callichilia barteri
5
10
Pl. com: Raphia
sudanica &
Oxytenanthera
abyssinica
Pl. com: Berlinia
grandiflora &
Khaya senegalensis
3
9
Pl. com.: Alchornea Association: Uapaca
heudelotii & Irvingia
cordifolia &
smithii Léonard 1947 &
Ficus trichopoda
Kalanda 1981
Pl. com.: Garcinia
livingstonei &
Combretum acutum
FIG. 6.11.- Synthesis of the 12 Riparian forests plant communities identified in Benin and their synsystematic relationships.
Pl. com.= Plant community. Numbers 1 to 12 refer to FIG. 6.10.
70
Chapter 7
SPATIAL DISTRIBUTION AND ECOLOGICAL FACTORS DETERMINING
THE OCCURRENCE OF PENTADESMA BUTYRACEA SABINE (CLUSIACEAE)
IN BENIN
Submitted to Acta Oecologica
Notulae Florae Beninensis 6
Natta A. K., Sinadouwirou Th.A., Sinsin B. and van der Maesen L.J.G.
Chapter 7: Spatial distribution of Pentadesma butyracea in Benin
Chapter 7
SPATIAL DISTRIBUTION AND ECOLOGICAL FACTORS DETERMINING
THE OCCURRENCE OF PENTADESMA BUTYRACEA SABINE (CLUSIACEAE)
IN BENIN (*)
Notulae Florae Beninensis 6
Natta A. K.(1), Sinadouwirou Th.A.(2), Sinsin B.(3) and van der Maesen L.J.G.(4)
(1)
Agronomic Engineer, MSc. Department of Environment Management, Faculty of Agronomic Sciences;
University of Abomey-Calavi FSA/UAC, 01 BP 526 Cotonou, Benin. Tel/Fax: + 229 303084;
aknatta@yahoo.com
(2)
Agronomic Engineer, MSc. Centre National de Gestion des Réserves de Faune (CENAGREF) - Parc du W.,
BP 75 Kandi, Benin. Tel/Fax: + 229 303084; tsina@yahoo.com
(3)
Professor of Ecology, Department of Environment Management, Faculty of Agronomic Sciences. University
of Abomey-Calavi. FSA/UAC, 01 BP 526, Cotonou, Benin. Tel/Fax: + 229 303084; bsinsin@bj.refer.org
(4)
Professor of Plant Taxonomy, Biosystematics group, Wageningen University; National Herbarium of the
Netherlands, Wageningen University Branch, Gen. Foulkesweg 37, 6703 BL Wageningen, The Netherlands.
Tel: + 31(0)317 483170; Fax: + 31(0) 317 484917, jos.vandermaesen@wur.nl
(*) Submitted to Acta Oecologica
ABSTRACT
Sample surveys were conducted in all ten ecological districts recognised in Benin (West
Africa) to assess the spatial distribution and ecological factors determining the occurrence of
Pentadesma butyracea, a multipurpose tree species found exclusively in some gallery forests.
Natural stands of the species occur in two regions at Bassila (Pira to Bodi) and in the Atacora
mountain chain (Perma to Tandafa). Isolated trees are found around Agbassa village
(between Ouèssè and Alafiarou in Central Benin) and as far as the latitude of Gbèssè village
(Ségbana district) in the North East of the country. The presence of Pentadesma in the
Bassila region was linked to the presence of dry semi-deciduous forests in the SudanoGuinean region of Central Benin. In the Sudanian region only riparian forests, with
permanent water at hills foot, provide a water balance of sufficient level: the most important
causal factor for the persistence of this species in a relatively dry and fire-prone landscape.
Key words: Pentadesma butyracea, Non Timber Forest Products, indicator species, gallery
forest, Benin.
7.1. INTRODUCTION
Every country needs to undertake monographical research on biological diversity with special
care to the multipurpose values of keystone or indicator ecosystems and species (UNCED
1992). Conservation agencies and scientists have recognised the requirement for spatial
knowledge of biodiversity for purposes of planning, management and conservation
evaluation (Acharya 1999).
73
Chapter 7: Spatial distribution of Pentadesma butyracea in Benin
Pentadesma butyracea, tallow tree (Clusiaceae), is known to be a common species of
dense evergreen forest (Aubreville 1959, Vivien & Faure 1985, Hawthorne 1996), and
natural stands are found in Africa from Sierra-Leone to the Democratic Republic of Congo
(Bamps 1971, Ouattara 1999). Therefore its presence in Benin, located at the discontinuity
(the Dahomey gap) of the belt of closed tropical rain forest from Elmina (near Cape Coast,
Ghana) to Porto-Novo, is an important issue for Beninese phytogeography. In Benin savanna
is the dominant ecosystem, and there is an increasing interest in the study and conservation of
Pentadesma butyracea, a multipurpose species found exclusively in some gallery forests (i.e.
riparian forests) (Houngbedji 1997, Sinadouwirou 2000). The species is well known for its
edible butter produced from its nuts that resemble those of Vitellaria paradoxa (Adomako
1977, Guelly 1994, Baumer 1995), and the fat is used for tallow production (Whitmore 1990,
Abbiw 1990, Schreckenberg 1996). The wood of Pentadesma butyracea is of good quality
(Purba & Sumarua 1987, Rachman et al. 1987, Tuani et al. 1994). Young stems are cut and
sold as vegetal toothbrush in the region of Natitingou (North West Benin). In Ghana,
Pentadesma oil is used for candle-making and margarine, its stems serve as chewing sticks
and are made into for hair combs, its roots decoction is taken as anthelmintic and its bark
combats diarrhoea and dysentery (Abbiw 1990).
Meanwhile, a few authors have studied (e.g. Gunasekera et al. 1977, Tuani et al.
1994) the chemical, cosmetic and pharmacological properties of its butter, leaves, bark and
roots. The ecology of Pentadesma butyracea upon which it is useful to base conservation and
management activities (e.g. reforestation of gallery forests, agroforestry and valorisation of
its various uses) is yet to be fully investigated. So far no information has been available on
the spatial distribution of Pentadesma butyracea under the less favourable climatic conditions
of Benin, as compared with the two West-African rain forest blocks. This paper maps the
current distribution of the species and assesses the ecological factors of its presence and
development in Benin.
7.2. MATERIAL AND METHODS
7.2.1. Study area
Riparian forests were surveyed in the ten ecological districts of Benin (Adjanohoun et al.
1989, Houinato et al. 2000). Benin is located at the discontinuity of the tropical rain forest
zone in West Africa, the ‘Dahomey Gap’, which is the product of topographic,
oceanographic, climatic and human interactions (Jenik 1994). This gap includes essentially
the drier types of the Guineo-Congolian forest belt in South-Eastern Ghana, Southern Togo
and parts of South-Eastern Benin (Ern 1988). As a result, there are no evergreen tropical
forests or rain forests in the country. Centuries of intense human activity accompanied by a
drying climate resulted in the loss of most of Benin closed forests long ago. Small patches of
moist forest remain today only on moist soils or as numerous sacred groves and riparian
forests.
The important hydrographic network allows the presence of many riparian forests
widely distributed all over the country. The hydrographic network includes the Atlantic
Ocean watershed with the Ouémé, Couffo and Mono Rivers and their tributaries and the
Niger River, which is the end point of the Sota, Alibori and Mékrou Rivers. In the North
West of Benin, the Pendjari River starts in the Atacora Mountains and ends in the Volta River
in Ghana. In the Southern part there is an important complex of coastal lagoons, lakes and
swamps. Table 7.1 summarises the climatic and edaphic characteristics of the two main areas
of occurrence (Bassila and Natitingou) of Pentadesma butyracea in Benin.
74
Chapter 7: Spatial distribution of Pentadesma butyracea in Benin
7.2.2. Data collection in riparian forests
Several riparian forests were surveyed using the Braun-Blanquet method for vegetation
analysis based on phytosociological relevés. 500 m2 plots were installed in different sites in
the Guinean region (Samiondji and Bétécoucou), the Sudano-Guinean region (Toui-Kilibo,
Ouèssè, Agbassa, Alafiarou, Bétérou, Daringa, Onklou, Bassila and Pénéssoulou) and the
Sudanian region (Ouaké, Affon, Boukombé, Toucountouna, Natitingou, Matéri, Tanguiéta
and all the Pendjari Biosphere Reserve, Kouandé, Péhonko, Kérou, Banikoara, Kandi,
Malanville, Karimama, Ségbana). For other parts of the country, a review of scientific work
on the flora of Benin and the National Herbarium was made to check the presence of the
species. Additional information was gathered as well from local people about the presence of
this species on their land.
Table 7.1: Climatic and edaphic characteristics of the two main regions (Bassila and
Natitingou) of occurrence of Pentadesma butyracea in Benin
Major factors (1)
Region of Bantè-Bassila
(Sudano-Guinean)
Region of Natitingou
(Sudanian)
Climate type
Sudano-Guinean climate (transitional
Typical Sudanian climate
zone between the Guineo-Congolian
and the Sudanian climate characterised
by progressive fusion of the two peaks
of rainfall typical of the GuineoCongolian climate)
Seasons
2 seasons:
*rainy season: mid-April to endOctober (6 months)
*dry season: End-October to mid-April
(6 months)
2 seasons:
*rainy season: May to mid-October
(5 months)
*dry season: mid-October to April
(7 months)
Humid season (P>½PET)
from April to October
from May to October
Period of humidity excess
(P>PET)
from June to September
from June to September
Annual rainfall (mm)
1100 – 1300
1000 –1100
Average temperature (oC)
26 – 32
28 - 38
Relative humidity (%)
15 (dry season) to 99 (rainy season)
16 (dry season) to 98 (rainy season)
Mean annual insolation (hours)
2420
2660
Average annual potential
evapotranspiration (mm)
1536
1510
Hydromorph soils at hill feet, or
Hydromorph and deep soils of
lowlands, or along streams surrounded along streams surrounded by
plateaus, or stony stream banks.
by plateau.
silt-clayed texture
Clayed-sandy or silt-clayed texture
with shallow water table
Often presence of sand bars
(1)
Climatic data observed over the last 40 years (1956-1995); P = Rainfall (mm); PET = Potential EvapoTranspiration
Edaphic conditions at stream
banks
75
Chapter 7: Spatial distribution of Pentadesma butyracea in Benin
In total 373 plots were installed all over the country and their coordinates taken by GPS. The
presence of Pentadesma butyracea was checked and measurements taken from individual
trees (dbh ≥ 5 cm, height, crown width, and distance to river or stream edge). The presence of
the species was also checked outside plots when walking along rivers and streams. All
Pentadesma butyracea trees detected outside plots were measured and their coordinates
taken. Special care was made to detect the regeneration (dbh < 5 cm).
Co-ordinates of plot centres where the species is found and coordinates of trees (≥ 10
cm) allow mapping the distribution of the species all over the country using the ArcView GIS
3.2 software.
7.3. RESULTS AND DISCUSSIONS
7.3.1. Pentadesma butyracea is a core species of some gallery forests in Benin
Pentadesma butyracea grows in multi-species stands and is always found along some streams
with more or less permanent water all year round. It is characterised by a barochore
distribution of its diaspores. This species has not yet been seen along larger waterways such
as the Ouémé, Zou, Sota, Alibori and Niger Rivers. We might say that the species somehow
compensates the hydric deficit of the Sudanian region and Sudano-Guinean zone while
avoiding long periods of floods (4 to 6 months) that characterises most riparian forests along
rivers in Benin. Pentadesma butyracea has a high indicative value of gallery forests, as it is
always tied to water. Moreover the distribution of the individual trees show that they are
always found close to stream beds inside the gallery forests, near trees such as Pterocarpus
santalinoides, Syzygium guineense var. guineense and Berlinia grandiflora. Pentadesma only
occurs at gallery forest edges when the vegetation is degraded. Thus it can be termed as a
core gallery forest species in Benin.
Although it is an evergreen tree, it is frequent along waterways in the pre-forest,
Sudano-Guinean and savanna zones of Central Côte d’Ivoire (Ouattara 1999). It has a
gregarious distribution pattern shown by clusters of individuals in several segments of the
streams (Sinadouwirou 2000). In Benin typical stands are found in the Pénéssoulou protected
forest and along the Yarpao stream (Natitingou district).
7.3.2. Distribution of Pentadesma butyracea in Benin
Based on recorded presence of the species, there are four separate areas of occurrence of
Pentadesma butyracea in Benin (Figure 7.1). The first and second areas of occurrence are
located in the Sudano-Guinean zone of Central Benin (Bassila region and around Agbassa
village, mid-way between Idadjo and Alafiarou villages in Cental Benin). The third area of
occurrence is in the Sudanian region of North West Benin (Natitingou region), while the
fourth one is a small area near Gbèssè village far in the North East (Ségbana district). The
Southern limit is around 8o15’ N in the district of Bantè. The Northern limit is the axis
Tanougou-Séri (10o45’ N) and Gbèssè village (11o N). The Western limit follows the Togo
border (around 1o22’ E at Pénéssoulou latitude), while the Eastern limit extends to (3o16’E).
So far the species has not been seen elsewhere, but isolated trees might be found along
streams in potential zone of occurrence that connect the four mentioned areas.
The distribution of Pentadesma butyracea in Benin, based on intensive data
collection, is a subset, of its larger-scale African distribution map (Bamps 1971). The species
occurs in corresponding ecological areas: the mountains and plains of North and Central
Togo (Zepernick & Timler 1984). In Ghana, the species is found in wet evergreen forests
(Swaine & Hall 1976).
76
Chapter 7: Spatial distribution of Pentadesma butyracea in Benin
While the Atacora mountain chain and the dry semi-deciduous forest (fire sub-type)
of Bassila region common to Benin, Togo and Ghana do harbour stands of Pentadesma
butyracea, the absence of the species in Southern Benin (below 8o N) is questionable.
0°30
1°00
1°30
2°00
2°30
3°00
3°30
4°00
NIGER
12°00
12°00
BURKINA
- FASO
Benin
N
AFRICA
11°30
11°30
Kandi
N
11°00
4
Tanougou
11°00
Gbèssè
I
Tandafa
Guilmaro
Tanguiéta
G
10°30
10°30
Kouandé
Natitingou
E
3
R
Nikki
10°00
Djougou
10°00
I
9°30
A
Parakou
9°30
Pénéssoulou
Legend
Bassila
9°00
T
8°30
Agbassa
1
Ouèssè
Bantè
O
9°00
1
Bassila region
2
Agbassa
3
Natitingou
region
Gbèssè
8°30
State limit
8°00
2
4
Savè
G
8°00
Presence of
Pentadesma
butyracea
Potential area of
occurrence of
Pentadesma
butyracea
O
7°30
Abomey
7°00
7°30
7°00
Lokossa
Porto-Novo
Cotonou
6°30
0°30
1°00
1°30
2°00
2°30
6°30
Scale : 1/3,300,000
3°00
3°30
4°00
Figure 7.1: Distribution map of Pentadesma butyracea in Benin
77
Chapter 7: Spatial distribution of Pentadesma butyracea in Benin
One might expect the presence of Pentadesma in the Pobè region, the wettest region of Benin
in terms of rainfall (1300 to 1400 mm per year) and relative humidity. In fact this region is
dominated by remnant dense semi-deciduous forests that are contiguous to the relatively
”dry” forest of South-Western Nigeria (Onochie 1979). According to Swaine & Hall (1976)
Pentadesma butyracea occurs usually in wet evergreen forest with at least >1750 mm mean
annual rainfall. It is obvious that Pentadesma occurs in Benin in a less suitable climate
environment than elsewhere. Therefore we cannot expect large stands in fire-prone and drier
landscapes. Centuries of intense land clearance, vegetation fragmentation through fire,
destruction of riparian forests, selective tree cutting, over-collection of nuts, as well as a
drying climate have probably contributed to the fragmentation and isolation of distribution
areas of Pentadesma butyracea. It is likely that the current non-contiguous areas of
occurrence of this species are remnant and refugee zones in Benin that need to be protected
and managed for their conservation.
7.3.3. Causal factors of the presence and persistence of Pentadesma butyracea in Benin
Several authors have documented the influence of abiotic and biotic factors on the presence
and spatial repartition of plants. For natural stands, climatic and edaphic conditions are the
most important factors, meanwhile intense disturbance may determine a particular
distribution pattern or cause complete disappearance of a species.
* Expanse of the dry semi-deciduous forest: the probable cause of the presence of
Pentadesma butyracea in the region of Bassila.
The region of Bassila is the easternmost point of the dry semi-deciduous forest, fire subtype
(Hall & Swaine 1981). This hygrophile enclave surrounded by savannas stretches out to Togo
(Southern part of Togo mountains) and to the South and Centre of Ghana, where the species
is present (Bamps 1971, Zepernick & Timler 1984). From this entry point Pentadesma
butyracea probably spread to the East around the Alafiarou village and South around Bantè
district.
Another viewpoint is that the species has always been there in the past and due to
certain factors (e.g. human-induced degradation or drying climate) the current occurrence
area in the Sudano-Guinean zone of Central Benin is remnant of a previously larger one.
* High water balance: the most important causal factor for the persistence of Pentadesma
butyracea in the Sudanian region of Natitingou.
The Sudanian region of Natitingou (North West of Benin), is dominated by the Atacora
mountains chain continuing into the mountains of Northern Togo; mountains at the foot of
which Pentadesma butyracea is present (Zepernick & Timler 1984). The niches for
Pentadesma in this region are gallery forests at the foot of hills, because they provide an
ample water supply: a combination of medium rainfall and evapotranspiration, high relative
humidity, lower temperatures, resurgence of water at the foot of hills, presence of water in
streams all year round. Here the relief and topography counterbalance the less suitable
climatic conditions.
According to Ouattara (1999), in Côte d’Ivoire Pentadesma butyracea disappears
when annual rainfall falls below 1000 mm. Therefore the viability of natural stands of the
species depends on the existence of high to medium quality habitats, such as the gallery
forests in the Bassila and Natitingou region. On the other hand the presence of a few
individuals as far as at the latitude of Kandi-Ségbana (with an average annual rainfall of 900
78
Chapter 7: Spatial distribution of Pentadesma butyracea in Benin
to 1000 mm) shows the adaptation of the species to sub-optimal habitats in terms of annual
rainfall and relative humidity. Under natural conditions, this matches the thesis that the longterm survival of populations can be strongly affected by the spatial and temporal distribution
of both suitable and unsuitable habitat patches (Pulliam et al. 1992, Carrol et al.
1996). Gallery forests of the Sudano-Guinean zone and Sudanian region Benin, therefore act
as refugee ecosystem for Pentadesma butyracea.
In general, little is known about the geographical patterns of genetic diversity for most
tropical forests species (Aide & Rivera 1998); and about Pentadesma butyracea populations
in West Africa in particular. The distribution of genetic diversity within and among the four
non-contiguous populations in Benin, and those in Togo and Ghana will give insight in the
origin and evolution of this species in Benin.
As Pentadesma butyracea in Benin is exclusively present in gallery forests, some
hypotheses on the ecological importance of these hygrophile and edaphic forest formations
are substantiated.
* Gallery forests act as refugee ecosystem for forest tree species in a fire prone environment
In many seasonally dry regions, but also in formerly forested areas, gallery forests
have the character of a refuge for plants and animals (Porembski 2001). The extent and
impact of fire in a system of gallery forests were evaluated in the Mountain Pine Ridge
savanna of Belize, (Kellman & Tackaberry 1993, Kellman et al. 1994) and such systems
represent plausible refugia for forest species in fire-prone landscapes. In Benin, it is obvious
that Pentadesma butyracea cannot survive but inside the wettest ecosystem that exists in the
majority of fire-influenced landscapes.
* Gallery forests harbour the tree species requiring moist habitats in the landscape
Fresh-water bodies, surface-emergent aquifers and soil moisture in excess in gallery
forests are the underlying factors that support mesic vegetation distinctive in structure and/or
floristics from that of the contiguous more xeric uplands (Warner 1979). Due to their elevated
moisture regime (compared to the surroundings), they enable the establishment and
persistence of plant species, which under zonal conditions are bound to wet forest types, often
evergreen rain forest (Porembski 2001). Tropical riparian forest fragments are known for
their potential, albeit limited, to maintain large numbers of species and may act as safe sites
for tropical rain forest species (Kellman et al. 1994, Meave & Kellman 1994). Likewise
gallery forests are the most suitable vegetation for plant species adapted to a moist climatic
regime (Medley 1992). Therefore, Pentadesma finds suitable ecological niches in gallery
forests that border streams in savanna dominated landscape.
* Gallery forests act as route for movement of certain species across the landscape
Riparian forests are important as routes for movement of plants and terrestrial animals
across the landscape (Forman & Godron 1986). They play an important role as migration
channels, which provide opportunities for genetic exchange between geographically isolated
populations (Porembski 2001). This seems to be the case with Pentadesma butyracea in
gallery forest corridors at hills feet along the Atacora mountain chain and in the semideciduous forest stretching from Ghana to Benin.
79
Chapter 7: Spatial distribution of Pentadesma butyracea in Benin
7.3.4. Major causes of gallery forests degradation in Benin and their consequences on
natural stands of Pentadesma butyracea
Natural stands of Pentadesma butyracea are disappearing along with the destruction of
gallery forests in Benin.
* Destruction of gallery forests
Rapid changes in land use in Benin have led to the destruction and fragmentation of
riparian forests. Riparian forests provide fertile soil for cultivation and provide an opportunity
for irrigation (Natta 2000). Riparian forests form a highly endangered ecosystem, and their
destruction could have severe consequences for the genetic exchange between disjunct
vegetation stands (Natta et al. 2002). In the Sudano-Guinean region of Côte d’Ivoire, agropastoral management is the underlying factor of the disappearance of the species from its
natural habitat (Ouattara 1999). As a core species of gallery forest, destruction of gallery
forest implies disappearance of Pentadesma butyracea individuals.
* Harvest of Pentadesma nuts
The regeneration of Pentadesma is already very difficult under natural conditions as
seeds have a short germinative power (Ouattara 1999). Therefore the collection of all or most
nuts of this species from the wild by local people is a major threat to its survival in Benin.
* Cutting of young stems for sale as vegetal teeth brushes
As in Ghana (see Abbiw 1990), young stems of Pentadesma are used as chewing
sticks in the North West of Benin. Therefore, the growth of young stems is seriously affected
and many cannot reach the stage of fructification, particularly in the Natitingou region.
* Selective tree cutting of the largest stems for timber
The good wood quality and the facility of working induce people to harvest the
biggest stems for the local wood industry.
* Disturbance due to fire
Uncontrolled bush fires often affect badly the regeneration of Pentadesma butyracea
in the dry season. Effects are more visible in isolated patches or narrow gallery forests than in
fair-sized forests.
7.4. CONCLUSION
Pentadesma butyracea is found in Benin in four non-contiguous areas. Its distribution is
principally influenced by the presence of numerous gallery forests in the semi-deciduous
forest of Bassila region and at the foot of hills in the Atacora mountain chain that continue in
Togo and Ghana. These four current areas appear to be remnant and refugee zones of
Pentadesma butyracea, and the assessment of their genetic diversity will provide new insight
on this enigma. In the meantime, these remnant populations need to be protected and
managed with and by local people.
80
Chapter 8
FOREST STRUCTURAL PARAMETERS AND FLORISTIC COMPOSITION
SPATIAL VARIATION AND MODELLING ACROSS RIVERS IN BENIN
A.K. Natta
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
Chapter 8
FOREST STRUCTURAL PARAMETERS AND FLORISTIC COMPOSITION
SPATIAL VARIATION AND MODELLING ACROSS RIVERS IN BENIN
Natta A.K.(1)
(1)
Department of Environment Management, Faculty of Agronomic Sciences. FSA/UAC 01 BP 526 Cotonou,
Benin; aknatta@yahoo.com.
ABSTRACT
Forest structural parameters, floristic composition and spatial distribution of tree species were
assessed across several rivers in Benin through multivariate analysis, distribution curves and
model fitting. The variation of tree stems from riverside up to 100 m away reveals an uneven
distribution of abundance, while tree height and basal area variations at riverside do not show
any clear pattern. Results from the numerical analysis show differences in the floristic
composition at riverside, across riparian forests and neighbouring plant communities. The
gradual change in species composition and relative abundance, which corresponds to a
gradual environmental change, indicates that tree species have different affinities to water
and humidity from the river. Typical riparian forest tree species are Syzygium guineense,
Pterocarpus santalinoides, Parinari congensis, Cola laurifolia, Napoleonaea vogelii,
Cynometra megalophylla, Drypetes floribunda and Manilkara multinervis. This research not
only confirms empirical knowledge about the ecology of certain riparian forest and
surrounding vegetations plant species, but also makes a distinction between river front, the
central portion and edge species of riparian forests. Criteria (e.g. average distance range and
best fit models) are also provided for the most common tree species along river. These results
suggest that riparian forests little disturbed by men can be partitioned in three different
habitats among tree species, along the horizontal gradient of wetness. Meanwhile due to lack
of sufficient data we cannot yet draw conclusions about the distribution behaviour of all tree
species encountered along the riverside. An implication for plant species diversity assessment
in riparian forest is that sampling design (i.e. plot size, shape and layout in the terrain) should
take into account the three major parts of the floristic composition at riverside (river front,
central portion and edge). Therefore rectangular plots with varying width and length, and
covering the whole riparian forest cross section are the most suitable sampling units for
savanna regions.
Keys words: Riparian forest, adjacent vegetation, structure, floristic composition, spatial
distribution, modelling, Benin.
8.1. INTRODUCTION
The riparian zone, functionally defined as a three dimensional zone of direct interaction
between aquatic and terrestrial environments, has boundaries that extend laterally from the
channel to the limits of flooding and vertically into the vegetative canopy (Swanson &
Lienkamper 1982, Gregory et al. 1991). Riparian forests are at the same time so important as
a collection of natural resources and so limited in individual size that they both merit and
require higher resolution inventory procedures than those deployed for other larger
83
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
ecosystems such as savanna and upland forests (Natta et al. 2002). In both predominantly
forested areas and savanna regions, they sustain a type of vegetation that is distinct from the
surrounding area in species composition and vegetation structure. Their location between
river and adjacent landscape contribute to their ecotone character (Brinson & Verhoeven
1999). Due to their elevated moisture regime, they enable the establishment of plant species
which under zonal conditions are bound to wet forest types (Porembski 2001).
The study of West-African riparian forests has hitherto been neglected and the
published studies dealt mainly with their structural and floristic characteristics and dynamics
(Devineau 1975, Bonkoungou 1984, Bélem & Guinko 1998, Goudiaby 1998, Lykke &
Goudiaby 1999, Natta 2000, Natta et al. in preparation). Few publications have been devoted
to the spatial variation of species and forest structural parameters across riparian forests
(Porembski 2001, Natta & Porembski, in press, see chapter 9). Classification and ordination
of riparian forests floristic data in Benin resulted in three major groups of plant communities:
riparian forest along rivers, along streams and the one of Pénéssoulou protected forest
(Sudano-Guinean zone of Central Benin) (Natta et al. in press, see Chapter 6).
This paper deals with the spatial variation of forest structural parameters and floristic
composition across rivers. The objective is to investigate the floristic composition variation
and assess the spatial distribution of tree species at riverside, across riparian forests and
adjacent vegetation formations. We also explore the effect of distance from water on the
composition and structure of tree species, so as to detect riverside, interior and edge species
of riparian forests. The research questions were:
Do structural parameters (i.e. abundance, height and basal area) vary across riparian
forests and neighbouring plant communities at riverside?
Is there any relationship between riparian forest width and tree species richness?
Does the floristic composition vary across riparian forests and neighbouring plant
communities?
Is there any recognisable spatial distribution of tree species, across riparian forests
and neighbouring plant communities, and how can it be modelled?
This contribution aims to provide insights into the ecology of certain tree species at
riversides in West African riparian systems and provide recommendations for sampling
designs.
8.2. Material and Methods
To assess the floristic composition variation and capture distribution patterns of tree species,
a cross section or crown cover intersection method was applied. This method consists of a
100 m long transect, which runs through the riparian forest and beyond its edge through the
adjacent (i.e. neighbouring) plant communities. At every point, a tape was laid on the ground,
perpendicular to the river direction, beyond the riparian forest edge. The position of each tree
stem (dbh ≥ 10 cm) is measured from the river bed to the base of each tree having its crown
intercepting the tape. This method was applied every 100 m from a random starting point,
perpendicularly to river direction, so as to cover at least all the riparian forest and the ecotone
(i.e. transitional) zone with adjacent vegetation. Five riparian forest sites were surveyed from
7º20 to 11º30 N in Benin, along the Ouémé, Alibori, Sota and Pendjari rivers. One riverside
was surveyed at each site. In total 56 transects were done.
We measured the distance from the river, dbh, height and crown width of each stem.
Riparian forest width, ecotone zone and the type of adjacent vegetation formation were
recorded as well. The position of trees permitted to compile floristic lists at different distance
intervals (e.g. every 1 m, 2m, etc.) from the river bed up to 100 m. The floristic composition
of each distance class has been taken taken as a relevé. The variability and similarity of the
84
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
floristic composition from the river bed were assessed through Detrended Correspondence
Analysis (DCA) (Hill & Gauch 1980, ter Braak & Smilauer 1998) and Two-Way Indicator
Species Analysis (TWINSPAN) (Hill 1979). Clusters are obtained with Statistica® (1998)
using Euclidean distance as distance measure and Ward’s minimum variance as aggregation
method. The numerical analysis will detect any gradual change of the floristic composition in
relation with the distance to river bed. Break lines, if detected, within this gradual floristic
change will indicate types of natural grouping of species linked to landscape units at
riverside. The upper limits of the detected distance intervals will be taken as threshold values
to classify each tree species among these natural groupings. We tested the presence of
riverside species (RS), species of riparian forest middle (MS), riparian forest edge species
(ES), and adjacent ecosystem species (AES), through a t-test, which compared the
distribution of stem positions to each class threshold value.
The distribution pattern of each tree species, in the direction perpendicular to river, is
obtained in plotting the relative abundance of stems (Y) versus their distance to river (X).
Scatter plots are drawn and several models fitted. The best model, in terms of the coefficient
of correlation (R2), significance of overall regression equation (F value) and regression
coefficients (t-tests) was selected. The illustration of such typology (see Figure 8.1), which
substantiates the presence of a riparian vegetation width effect (or patch width effect, Forman
& Godron 1986), is summarised as follow:
a - Riverside species (RS) are those close to river bed and only present in riparian
forests. This group includes species that have constantly decreasing number of individuals
within riparian forest, from the river.
b - Species of the middle of riparian forest (MS) have a more or less bell shape
abundance distribution centred at the middle (i.e. central portion) of riparian forest.
c - Riparian forest edge species (ES) have their highest abundance at the external end
of riparian forests.
d - Adjacent ecosystem species (AES) are generally not seen inside riparian forests
and belong to mature neighbouring natural vegetation formations. Under the study area
conditions this group includes species typical of savanna woodland, and open, dry or dense
semi-deciduous upland forests.
The response of plant species expressed in individual plant growth, frequency or
cover to the varying conditions along an ecological gradient often takes the form of bellshaped curves (van Groenewoud 1975). Meanwhile, the theoretical species distribution
curves presented in Figure 8.1 are consistent with field evidence. In effect, along larger
waterways (i.e. rivers) with more or less extensive flood plain the riparian vegetation is
usually distinct from that of surrounding uplands. Moreover within riparian forests, the
abundance of certain tree species decreases dramatically a short distance from the river while
other shows the reverse trend (Ericsson & Schimpf 1986, MacDougall & Kellman 1992,
Stromberg et al. 1996, Natta 2000, Porembski 2001, Natta & Porembski, in press see chapter
9).
85
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
10
Abundance (%) per species
9
(a ) - R ive r s ide s pe c ie s (R S )
(b) - S pe c ie s a t the c e ntra l po rtio n o f ripa ria n fo re s t (M S )
(c ) - R ipa ria n fo re s t e dge s pe c ie s (ES )
(d) - Adja c e nt e c o s ys te m s pe c ie s (AES )
8
7
Maximum riparian forest w idth
Adjacent vegetation formation
6
5
Ecotone
4
3
2
1
0
0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
Distance from river bed (m)
Figure 8.1: Simplified theoretical distribution curves of tree species at riverside, across
riparian forests and neighbouring vegetation formations (riparian vegetation width effect)
8.3. RESULTS
8.3.1. Characteristics of the cross sections at each site
We present here results of the least disturbed transects along the Ouémé, Sota, and Pendjari
rivers. Table 8.1 summarises the characteristics of the cross sections at riverside at each site.
Some 106 tree species were identified from 56 cross sections through riparian forests and
adjacent plant communities. The average width of undisturbed riparian forest (35.31 ± 7.97
m) includes three units generally found at riverside: the riverside close to river bed, the
middle and edge of riparian forest.
Table 8.1: Characteristics of the studied sites
Sites
Rivers
Samiondji (Guinean region)
Idadjo (Sudano-Guinean zone)
Bétérou (Sudano-Guinean zone)
Sota & Porga (Sudanian region)
Total of all rivers
Ouémé
Ouémé
Ouémé
Sota & Porga
No. of
transects
21
16
11
08
56
No. of
individuals
496
426
420
228
1571
No. of tree
species
44
48
55
39
106
Mean (± SD) RF
width (m)
29.14 ± 9.63
41.12 ± 19.5
38.27 ± 8.96
27.25 ± 9.60
35.3 ± 7.97
SD = Standard Deviation; RF = Riparian forest
8.3.2. Variation of the number of stems, tree height and basal area at riverside
The abundance of tree species stems over a width of 100 m is unequally distributed (Figure
8.2), and is best explained by an exponential function. There is a constant decrease of tree
individuals in dense riparian forests (up to 50 m from the river). After 50 m, in the adjacent
plant communities, the further decrease is less pronounced. Eventually a slight increase of
stem abundance is seen from 90 to 100 m. On the contrary, tree height (Figure 8.3) and basal
area (Figure 8.4) variations at riverside do not show any clear patterns, and the coefficients of
86
30
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
Abundance of stems (%)
determination (R2) were very low. This shows the weak relationships between these
variables. Meanwhile the first distance class (0 - 2 m), that has the highest abundance (see
figure 8.2) and lowest average height (see figure 8.3), gives the highest basal area value (see
figure 8.4).
10
9
8
7
6
5
4
3
2
1
0
y = 0.0011x2 - 0.1489x + 5.8331
R2 = 0.8619
0
10
20
30
40
50
60
70
80
90
100
Distance from river bed (m)
Average height of stems (m)
Figure 8.2: Distribution of tree stems across the riverside
20
18
16
14
12
10
8
6
4
2
0
y = -0.0004x 2 + 0.0486x + 9.7428
R2 = 0.0963
0
10
20
30
40
50
60
Distance from river bed (m)
70
80
90
100
Figure 8.3: Variation of the average tree height across the riverside
Average basal area (m2/ha)
8
y = 0.0001x 2 - 0.0185x + 2.1511
R2 = 0.0859
7
6
5
4
3
2
1
0
0
10
20
30
40
50
60
Distance from river bed (m)
70
80
90
100
Figure 8.4: Variation of the average basal area of trees across the riverside
87
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
8.3.3. Relationship between tree species richness and riparian forest width
Tree species richness and riparian forest width are positively correlated (Figure 8.5), but R2 is
very low (0.21 for linear and polynomial curves). Therefore we expect more tree species to
accumulate more or less linearly as riparian forest width increases.
16
y = 0.0871x + 3.1726
14
R2 = 0.2166
Tree species richness
12
10
8
6
4
2
0
0
10
20
30
40
50
60
Riparian forests w idth (m)
70
80
90
100
Figure 8.5: Relationship between tree species richness and riparian forest width along rivers
8.3.4. Floristic composition variation at riverside
Variation of tree species richness and floristic relevés distribution from the river bed up to
100 m are assessed. Tree richness seems to decrease with increasing distance from the river,
but the fit is not good (see Figure 8.6).
25
y = -0.0005x 2 - 0.0209x + 15.437
R2 = 0.3045
Number of tree species
20
15
10
5
0
0
10
20
30
40
50
60
70
Distance from river bed (m)
80
90
100
110
Figure 8.6: Distribution of tree species richness at riverside
The DCA and TWINSPAN analyses (see Figures 8.7 and 8.8, respectively) show a
clear pattern in the floristic relevés distribution. The gradual change in the floristic
composition is illustrated by the succession of the relevés from the river to about 60 m, this
includes the whole riparian forest width, the transitional zone and parts of the neighbouring
vegetations. Beyond 60 m the floristic composition seems to exhibit no variation. The first
DCA axis (Figure 8.7) is correlated with distance from the river bed. It symbolises the
88
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
variation of the floristic composition at riverside, across riparian forest and adjacent
vegetations.
The gradual change in species composition at riverside is also clearly shown by the
TWINSPAN output (available on request from the authors), and is summarised in Figure 8.8.
It shows the existence of an ecotone between 30 and 57 m, which separates the dense riparian
forests and the adjacent plant communities. The interpretation of the DCA (Figure 8.7),
TWINSPAN (Figure 8.8) outputs, the average RF width (27.34 to 43.28 m), and the
knowledge about the ecology of each species occurring at river edges, suggest three break
lines (or intervals) within riparian forests:
- [0 - 16] m from the river bed: for species mostly found close to the river bed;
- [17 - 29] m from the river bed: for species mostly found in the centre (i.e. middle
portion) of the riparian forest;
- [30 - 57] m from the river bed: for species in the transitional zone with adjacent
plant communities.
These results confirm the theoretical subdivision of species composition, at least for
tree species, into riverside, middle and edge species of riparian forest, and species typical of
adjacent vegetations (see Figure 8.1). Table 8.2 is derived from Figures 8.7 and 8.8. It
summarises the floristic variation at riverside and the natural grouping of tree species within
riparian forests. The distance intervals that are obtained from the TWINSPAN output indicate
that the floristic composition of undisturbed riparian forests (i.e. on average 27.34 to 43.28 m
wide), include those species always found close to the river bed, in the middle and edge of
riparian forest respectively.
Table 8.2: Floristic composition variation across riparian forests and adjacent vegetations
Distance interval (1)
0 to 16 m
17 to 29 m
30 to 57 m
58 to 101 m
Landscape units
Riverside close to river
bed
Middle of riparian
forest
Edge of riparian forest Adjacent vegetation
and ecotone zone
formations
Pattern of species
RS
MS
ES
AES
Threshold distances
Most faithful tree
species in each
landscape unit
16 m
17 and 29 m
30 and 57 m
More than 57 m
Syzygium guineense,
Pterocarpus santalinoides,
Parinari congensis,
Cola laurifolia,
Napoleonaea vogelii
Cynometra
megalophylla,
Drypetes floribunda,
Manilkara
multinervis
Dialium guineense,
Diospyros mespiliformis,
Elaeis guineensis,
Cola gigantea,
Ceiba pentandra
Albizia ferruginea,
Combretum collinum,
Lonchocarpus sericeus,
Millettia thonningii
Anogeissus leiocarpus
(1)
Distance intervals are obtained from TWINSPAN analysis (see Figure 8.8); RS = Riverside Species; MS =
Middle of riparian forest Species; ES = Ecotone zone Species; and AES = Adjacent Ecosystem Species.
The position of tree individuals from river compared to threshold values of each
riparian forest portion (see Table 8.2) allow the detection of distribution patterns, and the
investigation of the issue of core and edge species in riparian forests. For each tree species,
the t-test compared the positions of all stems from river bed to riparian forest portions
threshold distances: 16 m for river front (or riverside); 17 and 29 m for species always in the
middle of riparian forest; 30 and 57 m for ecotone zone species; and more than 57 m for
adjacent plant communities species. Table 8.3 (annex) summarises the outputs from t-tests.
Therefore the following typology and definitions are given for tree species found at
riverside in the Benin context:
- Riverside tree species (RS) are those that always occur between 0 and 16 m from the
river (i.e. river front species);
- Tree species in the middle of riparian forest (MS) are those that occur always
between 17 and 29 m from the river (i.e. central portion of the riparian forest);
89
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
- Tree species, that always occur between 30 and 57 m, are either riparian forest edge
species or ecotone species (ES); and
- Tree species typical of adjacent plant communities (AES) are those that always
occur outside riparian forests.
- We also define typical RF tree species along rivers as those always occurring
between 0 and 35.3 m away from the river bed.
This typology is only valid for rivers in Benin. The typology related to small
waterways (i.e. streams) is not investigated in the present chapter.
The abundance distribution models for 18 tree species are shown in Figures 8.9; 8.10;
8.11 and 8.12 (see Annex). We present here results for tree species having at least 20
individuals recorded at riversides across riparian forests. In general the distribution curves
follow the theoretical trends of Figure 8.1 for each species type.
90
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
Group I
Group II
Relevés of the
river front
(0 to 16 m)
Relevés in
the middle
of riparian
forest (17
to 29 m)
Group III
Group IV
Relevés of
riparian
forest
edges and
ecotone
zone (30 to
57 m)
Relevés outside riparian forests (> 57 m)
89
91
87
79
75
93
23
55 99
25
1
3
5
9 13
7
11 15
19
43
33
17 21
41
3935
31
59
49
51
53
65
27
61
85
101
45
83
81
73
57
95 696367
77
47
29
37
97
71
Axis 1 is correlated with distance from the river bed (0 to 100 m)
-2.0
+7.0
Figure 8.7: Detrended Correspondence Analysis (i.e. ordination) of the floristic composition of tree species at riverside. It shows a
gradual variation of tree species composition at riverside. (
) subdivisions and distance intervals are obtained from TWINSPAN
(see Figure 8.8) and show the partition for riparian flora in three groups (river front, middle and forest edge).
91
Floristic releves from 0 to 100 m away from the river bed
(51 releves every 2 m interval across riparian forest and adjacent plant communities)
Cut levels
1
2
Releves
Group (as in Figure 8.7)
Interpretation of each group
Distance interval (m) of each group
36 releves between 30 and 100 m from the river bed
21 releves mainly between
58 and 100 m
from the river bed
15 releves between 0 and 29 m from the river bed
15 releves mainly between
30 and 57 m
from the river bed
5 8 9 7 8 7 9 8 6 7 1 4 4 6 6 8 8 9 6 7 7
9 9 1 5 7 9 9 1 7 7 0 9 5 3 9 3 5 3 5 3 1
0
4 5 6 9 5 5 5 9 3 4 3 3 3 3 4
7 1 1 7 5 7 3 5 9 1 1 5 3 7 3
IV
III
Adjacent plant communities
Ecotone zoen between riparian forest
and adjacent plant communities
] 57 - 100]
[30 - 57]
8 releves between
17 and 29 m
from the river bed
2 1 2 0 2 1 2 2
9 7 5 9 7 9 1 3
II
Middle of riparian forest
[17 - 29]
7 releves between
0 and 16 m
from the river bed
1 0 1 0 1 0 0
5 5 1 7 3 1 3
I
River side
[0 - 16]
NB: Numbers 1, 3, …, 99 represent floristic releves at [0-2[, [2-4[, ….,[98-100[ m distance range
Figure 8.8 : TWINSPAN analysis of 51 releves across riparian forests and adjacent plant communities. It shows 4 groups of releves correlated
with distance from the river bed: (I) at river side, (II) middle of riparian forest, (III) ecotone zone, and (IV) adjacent plant communities.
92
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
8.4. DISCUSSION
8.4.1. Partitioning riparian forests in three: river front, middle and forest edge
The DCA, TWINSPAN and Clusters analysis outputs, and frequency distribution curves
showed interesting features regarding plant species ecology at riverside under the study area
conditions, and probably for the whole Dahomey Gap region. The horizontal structure of
species exhibit complex patterns across the studied rivers, and the underlying factors are still
poorly understood. Rivers and riverbanks constitute very complex environmental gradients,
with few (if any) factors having an overriding influence on species richness and composition.
Probably, the relative importance of any factor varies greatly over the geographic scale
chosen (Nilsson et al. 1989). Also, the predicted covariation in species composition and
diversity, and the heterogeneity of environmental factors, which is consistent with the
intermediate disturbance hypothesis, are documented from several riparian systems (Minshall
et al. 1985, Tabacchi et al. 1990, Kellman & Tackaberry 1993). It is generally assumed that
interior and edge areas are affected by patch size and shape (Forman & Godron 1986) and a
gradient of environmental variables (e.g. moisture, light) will influence the distribution and
frequency of tree and shrub species within and across riparian forests (Porembski 2001).
Apart from a few emergent species (e.g. Ceiba pentandra, Cola gigantea, Parinari
congensis and Manilkara multinervis) the average tree height is often less than 18 m. The
more or less distinct forest layers, each receiving progressively less light, are not always seen
across riparian forests and adjacent plant communities. This is in accordance with the climax
formations of Benin that are open, dry, or dense semi-deciduous forests with a relatively low
height. Generally, riparian forests display a physiognomy that is highly variable, though the
understorey is generally dense. Therefore they can be termed as relatively low edaphic and
hygrophile forests with irregular canopy (Natta et al. in press, chapter 9), compared to upland
tropical rain forests where canopy tree heights fluctuate between 30 to 45 m, and emergent
trees may reach heights above 50 m (Popma et al. 1988, Whitmore 1990).
Our results reveal that when vegetation formations are undisturbed at the riverside,
the floristic composition varies gradually across riparian forest and surrounding plant
communities (see Figures 8.7 & 8.8). For Forman & Godron (1986) a gradual change is
present where the major controlling environmental factors vary evenly (linearly) with
distance. This suggests that, over short distance, water availability, humidity and flood
frequency (July to December), as well as soil and micro-climatic variables may vary and
differently affect both groups of species and individual species at the riverside. Across a Lake
Superior tributary in Minnesota (USA), tree composition was correlated with stream gradient,
and at least two effects of floods, oxygen depletion in the root zone and soil modification,
could influence riparian vegetation (Ericsson & Schimpf 1986). In riparian ecosystems of
South-Western USA, ordinations analyses provided evidence that riparian forests are
structured along gradients relating to moisture, salinity, disturbance from fire, and community
maturity (Busch & Smith 1995).
Certain tree species have a greater fidelity for riparian habitats along rivers. The
decreasing affinity of tree species (see Figures 8.9; 8.10 and Table 8.3) to water availability
from river to riparian forest edges is as follow: Syzygium guineense, Pterocarpus
santalinoides, Parinari congensis, Cola laurifolia, Napoleonaea vogelii, Drypetes floribunda,
Cynometra megalophylla, Manilkara multinervis and Dialium guineense. Pterocarpus
santalinoides, Parinari congensis, Cola laurifolia and Manilkara multinervis were found to
be the most frequent species along the Ouémé, Mono, Couffo, Alibori, Sota and Mékrou
rivers in Benin (Sokpon et al. 2001). However, these authors have not made any distinction
or ranking among riparian forest species. Typical riparian forest species are those always
93
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
present at river front and in central portions of riparian forests, (i.e. the first eight species in
Table 8.2). Therefore they are well adapted to high soil moisture and periodic flooding.
Meanwhile, the effects of water-related factors across rivers in Benin are not fully elaborated.
Apart from the variation in floristic composition with water-related variables,
attention has been focused upon the extent of edaphic and micro-climatic effects in riparian
fragments flora because of the large perimeter to area ratio that these patches
characteristically possess. In Nigeria, Nye (1954) found that the composition of riverain
forest species in the immediate neighbouring of streams varies with the soils of the catena. In
Belize, understorey light intensity was highest at riparian forest edge, but rapidly decreased
towards the forest interior, where light levels were comparable to other tropical rain forest
understoreys. Also the distribution of six species across riparian forest indicates that the lightcorrelated spatial patterning of seedlings persists to adulthood (MacDougall & Kellman
1992). These results suggest that the light regime in riparian forests has sufficient variation to
support several regeneration and adult strategies, despite their small patch sizes.
8.4.2. Transition zone between riparian forests and neighbouring plant communities
In the study area, the boundary or ecotone between riparian forest and surrounding savanna
or forest is on average between 30 and 57 m from the river bed. Under low human
disturbance, this overlap zone is more characterised by a gradual change in species
composition (Figure 8.7) than a change of structural parameters (i.e. stem density, average
height, and basal area; see Figures 8.2, 8.3, and 8.4 respectively), and species richness (see
Figure 8.6). The most frequent tree species in this interval, with regards to their abundance,
are Dialium guineense, Diospyros mespiliformis, Elaeis guineensis, Cola gigantea and Ceiba
pentandra. Interestingly, statistical analysis shows that these five species are not typical to
riparian forest found along rivers (see Table 8.3 in Annex). Our results match with empirical
knowledge from various floras (e.g. Keay & Hepper 1954-1972, Brunel et al. 1984, Berhaut
1967). Dialium guineense and Diospyros mespiliformis reach their optimal growth inside
dense deciduous upland forests. Elaeis guineensis, a heliophytic and pioneering species, is
naturally abundant in all West and Central African forests (Sowunmi 1999, Maley &
Chepstow-Lusty 2001). Also, Ceiba pentandra is a late secondary pioneer species widely
distributed in the tropics (Whitmore 1990, Meave & Kellman 1994), while Cola gigantea,
typical of the Guineo-Congolian basin, is among the largest tree species of dense semideciduous forests.
Another issue revealed by species lateral distribution patterns (Figures 8.9 to 8.12 in
Annex) is that the ecotone zone is composed mainly of species from both sides (i.e. riparian
forest and adjacent savanna or upland forest). Numerous authors (e.g. Williamson 1975,
Simberloff & Gotelli 1984, van der Maarel 1990, Kent & Coker 1992) have shown how one
vegetation type grades into another through a transitional or ecotone zone, which
concentrates a certain percentage of the two sides’ species. A sharper ecotone often occurs
where the amount of an environmental factor changes abruptly. In the West-African savanna
region, abrupt borders seem to be controlled by an interplay of large herbivores, fires (mostly
lit by humans), deforestation, shifting cultivation and edaphic factors (Porembski 2001).
Between 8º30 and 9º N in Central Benin and North East Côte d’Ivoire, the gallery
forest/savanna contact is made through a belt either of light demanding species (e.g.
Anogeissus leiocarpus, Mitragyna inermis, Alchornea cordifolia) or species from dense semideciduous and open forest (Pouteria alnifolia, Ceiba pentandra, Albizia spp. and Antiaris
toxicaria) (Poilecot et al. 1991, Natta & Porembski, see chapter 9). At the Mountain Pine
Ridge savanna site in Belize (Central America), the transition zone between riparian forest
and savanna is narrow and often dominated by a tall light-loving grass (Tripsacum latifolium)
94
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
that grows in dense swards up to the edge of the forest (MacDougall & Kellman 1992).
Fringing (i.e. edge specialist) species, such as species of savanna affinity, may act as a
resilient buffer preventing fire intrusion into the riparian forest interior thus conserving
species at the riparian habitat core (Meave & Kellman 1994). Forman & Godron (1986)
hypothesise that the interdigitation zone is an area both of high total species diversity, since
two vegetation types are in close contact, and of low diversity of interior species. Our own
data on tree species at riversides could not yet confirm or reject this hypothesis.
Beyond the ecotone zone, in adjacent plant communities, the five most frequent tree
species are either of large distribution in the Guineo-Congolian basin, Sudano-Guinean, or
Sudano-Zambesian regions, but reach their optimal growth in upland forest or savanna
woodland. Albizia ferruginea is a common semi-deciduous forest tree. Lonchocarpus
sericeus and Millettia thonningii are deciduous tree species usually seen in fringing and semideciduous forest near water. Anogeissus leiocarpus extends from the driest savanna to the
borders of the forest zone, usually in moist situations but also in relatively dry situations,
while Combretum collinum is a common savanna tree (Keay et al. 1964, Keay & Hepper
1954-1972).
8.4.3. Modelling tree species distribution across rivers
Tree stems are not evenly distributed across riparian forests in Benin. The decrease of the
number of individuals at the riverside to 50 m away (Figure 8.2) suggests a decrease in tree
species density within riparian forests. This is linked to the biology and adaptation of typical
riparian forest tree species to recurrent floods and shallow water table. Such natural selection
and adaptation are revealed by the fact that all river front species (i.e. Syzygium guineense,
Pterocarpus santalinoides, Cola laurifolia and Cynometra megalophylla) are multi-stemmed
species.
Provided sufficient data, it is eventually possible to get best fitting models for the
lateral distribution patterns of plant species at the riverside. Modelling deals with the
construction of a manageable system, which is simpler than the reality that is modelled, but
which nevertheless shares interesting features and behaviour with the real systems (Tongeren
& Prentice 1986). In general polynomial and linear functions are the best-fit models, and low
values of R2 should be linked to complex relationships of tree species to environmental
variables under natural conditions. These results accredit the thesis of the presence of a
riparian vegetation width effect for tree species, which is a pattern where species distributions
are related to the width of vegetation formations at riverside. Nevertheless, the effects of
riparian forest width, and disturbance on the width, species composition and community
structure along waterway types (i.e. streams and rivers) need further investigation in Benin.
8.4.4. Plot size and shape for plant diversity assessment in riparian forests
The variation of tree species composition at riverside according to environmental factors has
several implications upon sampling designs, designated at assessing riparian plant diversity.
The primary criteria of site selection should be appearance (i.e. physiognomy), species
composition and dominant species encountered at riverside. According to the research
objectives, random or systematic sampling designs are suitable in continuous and dense
vegetation belts by waterways. On the contrary, stations for data collection have to be
purposely selected in dense riparian forest stands to avoid bare land or degraded plots. Also,
sites designated for drawing species-area curve of riparian forests should be carefully selected
in stands uniform with regards to physiognomy and species composition, and avoid ecotone
95
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
zones where edge species are frequent. Including the transitional (i.e. ecotone) zone at a
riverside could cause the curve to rise with increasing plot size.
When all terrestrial plants in all stages are included, plot sizes of 500 to 1000 m2 are
acceptable for classification of forest relevés (Hommel 1990). Field plots (50 m by 10 m)
were established in forests fragment forests in Southern Togo (Kokou et al. 2002). In Benin,
rectangular plots of 500 or 1000 m2 were used to collect floristic data in several edaphic
forests (Sokpon et al. 2001). In Central Benin, a 500 m2 plot size was taken as the minimal
plot size of riparian forest (Natta 2000). From the results presented above, circular and square
plots may not always cover the whole riparian forest width (i.e. in average 27 to 43 m wide).
Circular plots in particular in riparian forests, are characterised by an unequal probability of
sampling species, because our results show that species and stems are not evenly, nor
randomly distributed in riparian forests (see also Ericsson & Schimpf 1986, MacDougall &
Kellman 1992, Stromberg et al. 1996, Natta 2000, Porembski 2001, Natta & Porembski, in
press). Therefore circular plots will tend to oversample stems and species of the central
portion of riparian forests compared to those at the river front and riparian forest edge. They
also do not guarantee a total coverage of the riparian forest width at every stand. In
homogeneous strip-like riparian forest fragments, rectangular plots, with varying length and
width, fit with any shape of the waterway (see also Begon et al. 1986, Goudiaby 1998).
8.4.5. Optimal riparian forest width to be protected under Benin conditions
Riparian forest width varies from river to river, and along a single river system. This has
several vital functional implications. The Benin forest law (no. 93-009 of July the 2nd of
1993), does not allow clearance of wood and shrubs within 25 m from both sides of any water
course and stretch of smooth water (article 28). This limit was not set up based on scientific
data. From field observations, Natta et al. (2002) suggested a limit of 100 m. Can we answer
the important question: How wide should riparian forest corridors be to fulfil their vital
ecological functions? A prudent approach for riparian biodiversity conservation is to preserve
not only the full range of river front, interior and edges species of riparian forests (i.e. at least
57 m from the river bed), but also parts of the adjacent upland forest or savanna as buffer
zone. Moreover dense and continuous canopy at the riverside can extend to more than 100 m
(e.g. Samiondji, Idadjo, Pénéssoulou, Yarpao, etc). This could guide legislators to update the
optimal threshold distance to be protected at each riverside at 100 m (i.e. four time the actual
distance).
Overseas studies have shown the importance of the issue of optimal distance to be
protected at both waterway sides. In South East Brazil, the optimal width that allows RFs to
fulfil its multifunctional roles (e.g. sediment retention and improvement of water quality) was
found to be 52 m, which was actually wider and more efficient than to the official legal
recommendation of 30 m (Sparovek et al. 2002). In South East Australia, Dignan & Bren
(2003) found that the fixed minimum non-disturbance width of 20 m along all permanent
streams does not provide maintenance of the light regime that could protect riparian forests
values, instead they suggest forested buffers 70 to 100 m wide.
8.5. CONCLUSION
The present study investigates the structure, floristic composition variation, and spatial
distribution of tree species across riparian forests and their adjacent plant communities at
riverside. Horizontal and vertical structures of tree species exhibit complex patterns at
riversides. On the one hand, tree stems are characterised by an uneven distribution across
riparian forests, on the other hand height and basal area variations at riverside do not show
96
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
any clear patterns. The numerical analysis confirms a gradual variation of the floristic
composition at riverside, across riparian forests and neighbouring plant communities. This
floristic and abundance gradient may correspond to a change of water-related, as well as soil
and microclimate variables. This research not only confirms empirical knowledge about the
ecology of certain riparian forest and surrounding vegetations plant species, but also makes a
distinction between river front, central portion and edge species of riparian forests. Criteria
(e.g. average distance range and models) are provided for the most common tree species
along rivers. These results suggest a partitioning of riparian forests in three habitats among
tree species along the horizontal gradient.
In the present paper we could only draw conclusions about the distribution pattern of
18 tree species out of 106 encountered at the riverside. Investigations on the remaining
species should continue. An implication of our results for riparian forest phytodiversity
assessment is that plot size, shape and layout in the terrain should take into account the river
front, central portion and riparian forests edge species. Due to the non-coverage of the whole
riparian forest width and unequal chance of species and stems to be sampled, circular and
square plots are not suitable for structural parameters and phytodiversity assessment in
riparian forests. Instead rectangular plots with varying length and width, and covering the
whole cross section of riparian forest appear to be the most suitable under the study area
conditions, and probably for savanna regions. The present study also provides scientific
guidelines for an improvement of the forest law regarding the distance to be protected at
riverside, i.e. 100 m instead of 25 m.
97
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
Annex
Table 8.3: Summary of the t-test outputs. Abundance distribution curves are plotted in Figures 8.9, 8.10 and 8.11.
Species
1-Syzygium guineense
2-Pterocarpus santalinoides
3-Parinari congensis
4-Cola laurifolia
5-Napoleonaea vogelii
6-Drypetes floribunda
7-Cynometra megalophylla
8-Manilkara multinervis
9-Dialium guineense
10-Diospyros mespiliformis
11-Elaeis guineensis
12-Cola gigantea
13-Ceiba pentandra
14-Albizia ferruginea
15-Combretum collinum
16-Lonchocarpus sericeus
17-Millettia thonningii
18-Anogeissus leiocarpus
Abundance Position (m) from
river bed (± SD)
72
213
32
181
25
39
70
25
107
58
21
21
35
22
23
32
40
114
3.66 ± 4.97
7.43 ± 8.6
8.28 ± 5.94
14.15 ± 9.8
14.27 ± 6.51
22.55 ± 12
22.07 ± 20.12
27.39 ± 18.16
30.6 ± 22.59
52.54 ± 35.61
29.93 ± 18.49
39.53 ± 19.18
49.39 ± 24.41
49.41 ± 19.93
53.17 ± 14.31
60.93 ± 24.34
61.36 ± 22.77
71.11 ± 22.96
Compared with
Riverside limit (16m)
Riverside limit (16m)
Riverside limit (16m)
Riverside limit (16m)
Riverside limit (16m)
Middle limits (17-29m)
Middle limits (17-29m)
Middle limits (17-29m)
Riparian forest limit (38m)
Riparian forest limit (31m)
Riparian forest limit (26m)
Riparian forest limit (36m)
Riparian forest limit (30m)
Riparian forest limit (29m)
Riparian forest limit (32m)
Riparian forest limit (32m)
Riparian forest limit (43m)
Riparian forest limit (34m)
t-calculated
21.05
14.53
7.34
2.53
1.32
3.35
2.87
2.41
2.81
4.05
0.75
0.77
4.04
4.21
3.87
5.81
3.99
14.87
df
142
424
62
360
48
76
138
48
212
114
40
40
68
42
44
62
78
226
Probability
p<0.001***
p<0.001***
p<0.001***
p = 0.0117*
p = 0.19 NS
p = 0.0012**
p = 0.004**
p = 0.019*
p = 0.005**
p < 0.001***
p = 0.45 NS
p = 0.44 NS
p = 0.001***
p = 0.001***
p < 0.001***
p < 0.001***
p = 0.0014**
p < 0.001***
Results
Typical riverside species
Typical riverside species
Typical riverside species
Typical riverside species
Typical riverside species
Riparian forest middle species
Riparian forest middle species
Riparian forest middle species
Riparian forest edge species
Ecotone zone outside riparian forest
Ecotone zone outside riparian forest
Ecotone zone outside riparian forest
Ecotone zone outside riparian forest
Ecotone zone outside riparian forest
Ecotone zone outside riparian forest
Adjacent vegetation species
Adjacent vegetation species
Adjacent vegetation species
SD = standard deviation; *** highly significant (p<0.001); ** significant at p=0.01; * significant at p=0.05; NS = non significant at p = 0.05
98
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
Abundance of stems (%)
50
(a) Syzygium guineense : river side species
y = 0.0044x 2 - 0.5564x + 15.01
R2 = 0.3629
40
30
20
10
0
0
10
20
30
40
50
60
70
80
90
100
Distance from river bed (m)
Abunbance of stems (%)
28
(b) Pterocarpus santalinoides: river side species
y = 0.0036x 2 - 0.4723x + 13.586
R2 = 0.6475
24
20
16
12
8
4
0
Abundance of stems (%)
0
10
20
30
40
50
60
70
Distance from river bed (m)
32
28
24
20
16
12
8
4
0
10
20
20
Abundance of stems
(%)
90
100
80
90
100
80
90
100
80
90
100
(c) Parinari congensis: river side species
y = 0.0031x 2 - 0.4131x + 12.223
R2 = 0.4304
0
30
40
50
60
70
Distance from river bed (m)
(d) Cola laurifolia: river side species
y = 0.0019x 2 - 0.2723x + 9.3742
R2 = 0.7428
16
12
8
4
0
0
10
28
Abundance of stems (%)
80
20
30
40
50
60
70
Distance from river bed (m)
(e) Napoleonaea vogelii : river side species
y = 0.0016x 2 - 0.2469x + 8.951
R2 = 0.2787
24
20
16
12
8
4
0
0
10
20
30
40
50
60
70
Distance from river bed (m)
Figure 8.9: Typical riverside tree species (RS) mainly found less than 16 m from the river bed
99
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
Abundance of stems (%)
28
(a) Cynometra megalophylla : middle of riparian forest species
y = 0.0011x 2 - 0.1738x + 7.0832
R2 = 0.2437
24
20
16
12
8
4
0
0
10
Abundance of stems (%)
20
20
30
40
50
60
70
Distance from river bed (m)
80
90
100
(b) Drypetes floribunda: middle of riparian forest species
y = 0.0002x 2 - 0.079x + 5.4468
R2 = 0.2142
16
12
8
4
0
0
10
20
30
40
50
60
70
80
90
100
Distance from river bed (m)
Abundance of stems (%)
12
(c) Manilkara multinervis : middle of riparian forest species
y = -0.0524x + 4.6312
R2 = 0.2177
10
8
6
4
2
0
0
10
20
30
40
50
60
Distance from river bed (m)
70
80
90
100
Figure 8.10: Typical species of the middle of riparian forests (MS) mainly found
between 17 and 29 m away from the river bed
100
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
Abundance of stems
(%)
10
(a) Dialium guineense : riparian forest edges species
y = 0.0001x 2 - 0.0565x + 4.4502
R2 = 0.3536
8
6
4
2
0
0
10
20
Abundance of stems
(%)
10
90
100
6
4
2
0
10
20
20
Abundance of stems (%)
80
(b) Diospyros mespiliformis : ecotone zone species
y = 0.0013x 2 - 0.1257x + 3.9463
R2 = 0.2164
8
0
30
40
50
60
70
Distance from river bed (m)
80
90
100
90
100
90
100
90
100
(c) Elaeis guineensis : ecotone zone species
y = -0.0004x 2 - 0.0042x + 3.602
R2 = 0.0943
16
12
8
4
0
0
10
20
30
40
50
60
Distance from river (m)
70
80
(d) Cola gigantea : ecotone zone species
y = -0.0013x 2 + 0.1049x + 1.0386
R2 = 0.1403
16
Abundance of stems (%)
30
40
50
60
70
Distance from river bed (m)
12
8
4
0
0
10
20
30
40
50
60
70
80
Abundance of stems
(%)
Distance from river bed (m)
12
10
8
(e) Ceiba pentandra : ecotone zone species
y = -0.0009x 2 + 0.093x + 0.4664
R2 = 0.0702
6
4
2
0
0
10
20
30
40
50
60
70
Distance from river bed (m)
80
Figure 8.11: Typical species at riparian forest edges (ES) or in the ecotone zone with
neighbouring plant communities (i.e. between 30 and 56 m away from river bed)
101
Chapter 8: Structural parameters and floristic composition variation across rivers in Benin
Abundance of stems
(%)
16
(a) Albizia ferruginea : adjacent plant community species
y = -0.0016x 2 + 0.1618x - 0.7031
R2 = 0.1177
12
8
4
0
0
Abundance of stems
(%)
16
Abundance of stems
(%)
30
40
50
60
70
Distance from river bed (m)
80
90
100
8
4
0
0
16
10
20
30
40
50
60
70
Distance from river bed (m)
80
90
100
(c) Lonchocarpus sericeus : adjacent plant community species
y = -0.0006x 2 + 0.0871x - 0.3127
R2 = 0.0709
12
8
4
0
0
16
Abundance of stems
(%)
20
(b) Combretum collinum : adjacent plant community species
y = -0.0012x 2 + 0.1275x - 0.396
R2 = 0.0726
12
10
20
30
40
50
60
Distance from river bed (m)
70
80
90
100
(d) Millettia thonningii : adjacent plant community species
y = -0.0008x 2 + 0.1069x - 0.6855
R2 = 0.1019
12
8
4
0
0
16
Abundance of stems
(%)
10
10
20
30
40
50
60
70
Distance from river bed (m)
80
90
100
(e) Anogeissus leiocarpus : adjacent plant community species
y = 0.0003x 2 + 0.0187x + 0.0413
R2 = 0.4249
12
8
4
0
0
10
20
30
40
50
60
70
Distance from river bed (m)
80
90
100
Figure 8.12: Typical species of plant communities adjacent to riparian forests (AES), and beyond
the ecotone zone (i.e. found constantly beyond 57 m from the river bed).
102
Chapter 9
OUÉMÉ AND COMOÉ: FOREST-SAVANNA BORDER RELATIONSHIPS IN TWO
RIPARIAN ECOSYSTEMS IN WEST AFRICA (*)
Notulae Florae Beninensis 8
In press, Botanische Jahrbücher.
Natta A.K. and Porembski S.
Chapter 9: Ouémé and Comoé: forest-savanna border relationships in two riparian ecosystems in West Africa
Chapter 9
OUÉMÉ AND COMOÉ: FOREST-SAVANNA BORDER RELATIONSHIPS IN TWO
RIPARIAN ECOSYSTEMS IN WEST AFRICA (*)
Notulae Florae Beninensis 8
Natta A.K.(1) and Porembski S.(2)
(1)
(2)
Department of Environment Management, Faculty of Agronomic Sciences. University of Abomey-Calavi,
FSA/DAGE/UAC 01 BP 526 Cotonou, Benin; aknatta@yahoo.com;
Institut für Biodiversitätsforschung, Allgemeine und Spezielle Botanik, Universität Rostock, Wismarsche Str.
8, D-18051 Rostock, Germany; stefan.porembski@biologie.uni-rostock.de
(*) In press, Botanische Jahrbücher.
ABSTRACT
The floristic composition, species richness and structure at river edges are compared between
two gallery forests/savanna ecosystems along the Ouémé (Central Benin) and Comoé (NorthEast Côte d’Ivoire) rivers. Although the overall physiognomy of the two gallery forest sites
seems similar and they share the most prominent families, there are marked differences in
terms of canopy density and height, herb layer density, number of individuals, tree richness
and diversity (H’), and species composition. Gallery forest width, top canopy density and
height were more developed along the Comoé than along the Ouémé, but the herb layer is
more luxuriant at the latter site. Concerning absolute density and basal area Cynometra
megalophylla accounted for 9 to 16 %, and 26 to 33 %, respectively in the two sites. At the
two sites this species was time and again the most frequent and dominant one at both
riverside and in the middle of the gallery forest.
A detailed comparison of the three gallery forest plot types (riverside, middle and
savanna edge) revealed considerable differences and contrasting results concerning the
dominant trees, species composition, the number of species and individuals, average height
and diameter, and basal area. Species characteristic for the river front and the central portion
of gallery forests were Cynometra megalophylla, Dialium guineense, Cassipourea
congoensis, Syzygium guineense and Parinari congensis at both sites, with Pterocarpus
santalinoides being particularly abundant along the Ouémé river. The savanna edge has the
most distinct floristic composition with Pouteria alnifolia at both sites; Antidesma venosum,
Fagara zanthoxyloides along at the Comoé site; and Ceiba pentandra, Albizia spp. and
Antiaris toxicaria at the Ouémé site. The difference between the plot types concerning
species diversity (H’) was not statistically significant. The present study shows the variability
and complexity of ecological processes between and within gallery forests sites.
Key words: gallery forest, forest edge, savanna, core species, Benin, Côte d’Ivoire, West
Africa.
9.1. INTRODUCTION
Gallery forest (also riparian forest) is one of the major vegetation formations, which
underlines the outline of waterways in savanna or in forest-savanna mosaic ecosystems
(MONNIER 1990, NATTA 2000). They sustain a type of vegetation that is distinct from the
105
Chapter 9: Ouémé and Comoé: forest-savanna border relationships in two riparian ecosystems in West Africa
surrounding areas in species composition and vegetation structure (POREMBSKI 2001).
Although linear in shape and small in area, as compared to large extents of dense forests, the
role played by tropical gallery forests in fluvial processes, and animal and plant species
conservation is well documented (FORMAN & GODRON 1986, DECAMPS et al. 1988, BAKER
1990, LEVAUX 1990, MEAVE et al. 1991, ROGGERI 1995, MEDLEY 1992, THOMAS 1996,
PIEGAY 1997, LEINARD et al. 1999, VAN ETTEN 1999, NATTA et al. 2002).
The study of West-African gallery forests has hitherto been limited (POREMBSKI
2001). So far, most of the published work concerns floristic and structural characteristics of
genuine or single gallery forests along specific waterways and the gallery forest/savanna edge
relationships, both in species composition and structure, were little investigated.
The Ouémé and Comoé, which are among the largest rivers in Benin and Côte
d’Ivoire respectively, with their forested banks harbour a rich birdlife and serve as major
habitat for primates and other animals of many kinds. Moreover, they are important
extrazonal corridors, acting as refugee ecosystem for forest tree species in a fire-prone
environment, harbouring species requiring moist habitats, and allowing the migration of
species far beyond their zonal distributional area. At the two sites, more than 600 km apart, a
field survey shows the presence and dominance of some typical gallery forests tree species,
such as Cynometra megalophylla, Dialium guineense, Cassipourea congoensis and Drypetes
floribunda. The present paper aims at comparing the floristic composition and structure of
two genuine gallery forest/savanna ecosystems, at the same latitude, along the most important
rivers in two different West-African countries, Benin and Côte d’Ivoire. The spatial
distribution of certain species inside and along the gallery forest edge is investigated as well.
9.2. MATERIAL AND METHODS
9.2.1. Study sites
The two gallery forest ecosystems (hereafter GF), which are approximately at the same
latitude, are located within protected areas, the Comoé National Park and the Kouffé
Mountains (Figure 9.1). They are representative of GFs occurring between 8°30 and 9° N
along the Ouémé (Central Benin) and the Comoé (North East Côte d’Ivoire) rivers, which are
among the largest waterways in the two countries. At this latitude, the belts of GFs are almost
continuous with only a few local gaps. Apart from bush fire traces, the two sites showed no
signs of human impact. Their climates are similar in many points, particularly average
rainfall (1100 - 1200 mm) and a pronounced seasonality with a dry season from November to
March/April and a rainy season from May to October. From January to March the dry-hot
Harmattan results in daytime temperatures in excess of 40° C whilst air humidity drops below
20%. During the night temperature can fall below 20° C, and the sudden decrease is
occasionally accompanied by dew. The hydrological regime of the Comoé and Ouémé rivers
is of tropical transition type, characterised by a unique rise in the water level in AugustSeptember-October. At the end of the dry season, in March/April, the two rivers usually stop
to flow but large water holes are still present.
The Ouémé river GFs at the latitude of Ouèssè are located at the Eastern limit of the
Kouffé Mountains forest reserve that lies along the boundary between the Sudanian regional
centre of endemism and the Guinea-Sudanian regional transition zone. It is moister than the
protected areas of Northern Benin, so its plant and animal species are more diverse. Despite
many decades of neglect, the reserve still contains dozens of small patches of GF and dense
moist forest. These remnants appear undisturbed internally, and are therefore of very high
biodiversity value (PGRN/IUCN 1994). GF ecosystems and the surrounding vegetation are
among the least degraded in the Sudano-Guinean zone. This zone is characterised by the
106
Chapter 9: Ouémé and Comoé: forest-savanna border relationships in two riparian ecosystems in West Africa
decrease of rain and merging of the two peaks of rainfall typical of the Guinean region of
Southern Benin. The mean annual temperature is 28° C and the elevation ranges from 190 to
390 m a.s.l. Dense semi-deciduous and dry forests represent climax vegetation formations at
this latitude. The maximum GF width may reach 200 to 300 m, but the dense and continuous
canopy is generally less than 100 m wide.
The investigation of GF along the Comoé river focused on the South-Western part of
the Comoé National Park (CNP) where semi-deciduous forest represents a large proportion
the natural vegetation. Meanwhile, due to annual burning, the vegetation consists mainly of
different savanna types (>85%), forests islands (4.4%) and gallery forests (2.3%) (FGU
KRONBERG 1979). The climate of the Central and Southern areas of the CNP can be assigned
to the ‘secteur sub-soudanais’, which is of tropical sub-humid type (ADJANOHOUN & AKÉ
ASSI 1967, GUILLAUMET & ADJANOHOUN 1971, ELDIN 1971). Mean annual temperature is in
the range of 26-27° C. March is the hottest month with a mean daily temperature of 37° C.
The lowest temperature (15° C) is recorded in January. The river Comoé runs through the
park for about 200 km and drains with its tributaries some 78,000 km2 (POILECOT et al.
1991). Permanent and semi-permanent water occurs in many places. The width of GFs may
reach up to 600 m, but at the study site, they attain a maximum width of 130 m (POREMBSKI
2001).
NIGER
MALI
14°
Niamey
BURKINA FASO
Bamako
Ouagadougou
12°
BENIN
OuèssèBétérou
NIGERIA
10°
Comoé
TOGO
GUINEA
COTE D’IVOIRE
8°
GHANA
Yamoussoukro
Lomé
PortoNovo
6°
Accra
LIBERIA
GUINEA GULF
States limits
4°
Capital
8°
Major rivers
Scale : 1/ 4 147 000
Study areas
0
6°
125
250 Km
2°
4°
2°
0°
2°
4°
6°
Figure 9.1: The Comoé National Park and Kouffé Mountains gallery forests sites
in Côte d’Ivoire and Benin in West Africa.
107
Chapter 9: Ouémé and Comoé: forest-savanna border relationships in two riparian ecosystems in West Africa
9.2.2. Data collection and analysis
The investigation and comparison of the two sites are based on 9000 m2 of GF (i.e. 4500 m2
per site). Floristic data were collected through phytosociological survey in plots (300 m2 to
500 m2), laid out along a gradient from the stream-side to the border with the savanna. Three
plot types were considered: riverside, middle of GF and savanna edge. Per site and at each
plot type 1500 m2 were surveyed. All trees and lianas of a diameter at breast height (dbh, at
1.3 m above the ground level) of at least 5 cm were recorded and identified. Nomenclature of
the species recorded follows KEAY & HEPPER (1954-1972), BRUNEL et al. (1984), BERHAUT
(1967), and LEBRUN & STORK (1991-1997).
Tree (i.e. dbh ≥ 5 cm) species diversity for each plot type was calculated using the
Shannon-index H’. A t-test based on the Shannon index (H’) and its variance was used to test
for differences in alpha-diversity (including evenness) between the plot types, and the two GF
sites. This method was successfully used on riparian forests of Central Benin (NATTA 2000)
and in the CNP (POREMBSKI 2001). For further details see MAGURRAN (1988). The floristic
similarity of the two GF sites was assessed through the Jaccard and SØrensen coefficients.
9.3. RESULTS
9.3.1. General characterisation of the Ouémé and Comoé GF sites
GFs investigated along the Comoé and Ouémé rivers have generally a well-marked boundary
with the surrounding savanna, which is regularly burnt in the dry season. Elsewhere the
contact is more abrupt, with a grass layer bordering the GF. The two GF sites show no clear
evidence of stratification, with the different strata transgressing into each other. Here and
there, there is a continuum of the top layer canopy with adjacent dense semi-deciduous or dry
forests, but ground cover of the dominant trees is always lower at GF fringe than inside.
At the Comoé river site, the herbaceous stratum at GF/savanna border is dominated by
Vetiveria fulvibarbis, Loudetia simplex, and among the important trees there are Anogeissus
leiocarpus, Mitragyna inermis, Holarrhena floribunda, Tamarindus indica, Acacia
sieberiana. The interior of this GF is made up of species such as Cymonetra megalophylla,
Cola cordifolia, Manilkara multinervis, and Diospyros mespiliformis at a height of 25-30 m.
Occasionally, emergent trees such as Ceiba pentandra, reach a height of up to 40 m. The
lower storeys consist of smaller trees (e.g. Dialium guineense, Drypetes floribunda,
Cassipourea congoensis, Diospyros abyssinica, Celtis brownii) and shrubs (e.g. Oxyanthus
racemosus, Rinorea kibbiensis, Tapura fischeri, etc.). Lianas are abundant, with Cnestis
ferruginea, Dioscorea spp., Cissus spp., Strophanthus spp., and Strychnos spp. Both towards
the Comoé river as well as towards the savanna border, light demanding trees (e.g.
Pterocarpus santalinoides, Cassia sieberiana, Christiana africana) can be found. A
continuous herb layer is only developed in some gaps.
On the contrary, at the Ouémé site Andropogon spp. and Pennisetum spp. are among
the dominant herbaceous species at GF/savanna border whilst the most important tree species
are Millettia thonningii, Anogeissus leiocarpus, Albizia spp., Combretum spp., Pouteria
alnifolia, Acacia sieberiana and Lonchocarpus sericeus. The top canopy tree species inside
the GF are Cynometra megalophylla, Parinari congensis, Cola laurifolia, and Pterocarpus
santalinoides with a maximum height of 16-20 m. Here, Dialium guineense is among the top
layer trees. The emergent tree Cola gigantea, reaches a maximum height of 26 m, which is
much lower than those along the Comoé river. The lower storeys consist of smaller trees such
as Drypetes floribunda, Lepisanthes senegalensis, Cassipourea congoensis, Salacia
pallescens, Vitex chrysocarpa and Morelia senegalensis. Lianas are also abundant: Strychnos
108
Chapter 9: Ouémé and Comoé: forest-savanna border relationships in two riparian ecosystems in West Africa
spp., Loeseneriella spp., Alafia spp., Taccazea apiculata, Connarus africana, Paullinia
pinnata, Lonchocarpus cyanescens, Cremaspora triflora and Canthium horizontale. The herb
layer consists of Achyranthes aspera, Phaulopsis spp., Polygonum salicifolium, Heliotropium
indicum, Ruspolia hypocrateriformis, Ruellia praetermissa, and Cyperus spp., and, in
contrary to the CNP, is continuous even under dense canopy.
9.3.2. Floristic diversity at the Ouémé and Comoé GF sites
The two sites share the most prominent families Leguminosae, Rubiaceae and Sapotaceae,
but they have only 16 tree species, out of 88, in common. The Jaccard and SØrensen similarity
coefficients are therefore very low, 13.3 and 23.5% respectively. The GFs along the Comoé
have higher tree species richness and abundance than those along the Ouémé (see Table 9.1).
The t-test, based on the Shannon index that combines species richness and abundance, reveals
that the two sites differ highly significantly (t = 8.48, df = 1305, p < 0.001) in terms of tree
species diversity. At the same latitude, the GFs along the Comoé are thus more diverse than
those along the Ouémé. In absolute density (individuals/ha), Cynometra megalophylla ranks
first in the two sites, followed by Cola laurifolia along the Ouémé and Dialium guineense
along the Comoé.
Table 9.1: Major characteristics of gallery forests along the Ouémé and Comoé rivers.
Major characteristics
GF along the Ouémé
(Central Benin)
50
Tree species richness
624
Number of individuals
Most prominent families Leguminosae (8); Moraceae (4);
Rubiaceae (4); Annonaceae (3);
(species number)
GF along the Comoé (CNP)
(North East Côte d’Ivoire)
54
906
Leguminosae (8); Euphorbiaceae (7);
Rubiaceae (3) and Sapotaceae (3)
Sapindaceae (3) and Sapotaceae (3)
Most frequent trees
Cynometra megalophylla (102); Cola
laurifolia (65); Lepisanthes
senegalensis (58); Pterocarpus
santalinoides (56); Drypetes
floribunda (53) and Xylopia
parviflora (43)
Cynometra megalophylla (84); Dialium
guineense (74); Cassipourea congoensis
(63); Ouratea affinis (55); Tapura fischeri
(54) and Drypetes floribunda (47)
Shannon index (H’)
Pielou index (E)
3.97
0.70
4.87 (most diverse site)
0.85
The most frequent tree species is Cynometra megalophylla, which accounts for 9 to 16
% of the total abundance, and 26 to 33 % of the total basal area at the Comoé and Ouémé
sites respectively (Table 9.2).
Table 9.2: Measured variables for Cynometra megalophylla at the Ouémé and Comoé sites.
Variables
Abundance (%)
Basal Area (%)
Mean diameter (cm)
Mean height (m)
Ouémé (Central Benin)
Comoé (North East Côte d’Ivoire)
16.3
33.06
19.53
9.48
9.3
26.1
18.6
16.2
109
Chapter 9: Ouémé and Comoé: forest-savanna border relationships in two riparian ecosystems in West Africa
9.3.3. Floristic and structural variability at the riverside, inside GF and at GF edge
A detailed comparison of the three GF plot types (riverside, middle of GF and savanna edge)
revealed considerable differences concerning the number of individuals and dominant tree
and shrub species. At the Comoé GF site, the number of individuals increased from
‘riverside’ (263) and ‘middle’ (317) to ‘savanna’ (326), whilst it decreased (345, 176 and 103
respectively) at the Ouémé site (see Table 9.3). ‘Savanna edge’, ‘middle of GF’ and
‘riverside’ were clearly differentiated when considering the most important species at the two
sites and certain species showed particular preferences for specific plot types.
Table 9.3: Dominant tree species (with regards to number of individuals, dbh ≥ 5 cm) in
different GF plot types at the Ouémé and Comoé sites.
(a) The Ouémé GF site at Ouèssè latitude
GF sites
Plot type
Trees abundance
Number of tree species
Mean height (m)
Average diameter (cm)
2
Total basal area (m )
Shannon index (H’)
Dominant tree species
Ouémé (Central Benin)
Riverside
345
32
7.44
13.5
169.6
3.9
Middle of GF
176
26
7.47
13.32
83.2
3.8
Savanna edge
103
22
8.1
16.37
74.1
3.5
Cynometra megalophylla (57)
Pterocarpus santalinoides (49)
Cola laurifolia (48)
Syzygium guineense (36)
Morelia senegalensis (22)
Drypetes floribunda (20)
Vitex chrysocarpa (18)
Xylopia parviflora (15)
Dialium guineense (11)
Cynometra megalophylla (29)
Lepisanthes senegalensis (28)
Xylopia parviflora (27)
Cola laurifolia (18)
Drypetes floribunda (10)
Dialium guineense (9)
Manilkara multinervis (7)
Morelia senegalensis (6)
Vitex chrysocarpa (6)
Drypetes floribunda (23)
Lepisanthes senegalensis (19)
Cynometra megalophylla (17)
Manilkara multinervis (7)
Ceiba pentandra (5)
Albizia ferruginea (3)
Antiaris toxicaria (3)
Dialium guineense (3)
Pouteria alnifolia (3)
(b) The Comoé GF site in the South-West of the Comoé National Park
GF sites
Plot type
Trees abundance
Mean height (m)
Average diameter (cm)
2
Total basal area (m )
Number of tree species
Shannon index (H’)
Dominant tree species
Comoé (North East Côte d’Ivoire)
Riverside
263
8.82
14.2
101.4
27
2.69
Middle of GF
317
9.21
14.7
112.3
23
2.41
Savanna edge
326
8.44
13.3
98.5
34
2.5
Cynometra megalophylla (25)
Syzygium guineense (15)
Parinari congensis (15)
Cassipourea congoensis (14)
Manilkara multinervis (14)
Drypetes floribunda (12)
Mallotus oppositifolius (11)
Dialium guineense (9)
Lecaniodiscus cupanioides (9)
Diospyros abyssinica (8)
Cynometra megalophylla (48)
Dialium guineense (42)
Cassipourea congoensis (37)
Drypetes floribunda (35)
Tapura fisheri (35)
Ouratea affinis (32)
Mallotus oppositifolius (26)
Holarrhena floribunda (17)
Diospyros abyssinica (17)
Lecaniodiscus cupanioides (17)
Pouteria alnifolia (25)
Dialium guineense (23)
Antidesma venosum (18)
Tapura fischeri (17)
Fagara zanthoxyloides (15)
Lannea kerstingii (15)
Alchornea cordifolia (14)
Anogeissus leiocarpus (13)
Christiana africana (13)
Diospyros mespiliformis (13)
At the Comoé site, species characteristic for the central portion of the GF were
Cynometra megalophylla, Dialium guineense and Cassipourea congoensis. Cynometra
110
Chapter 9: Ouémé and Comoé: forest-savanna border relationships in two riparian ecosystems in West Africa
megalophylla was also abundant along the Comoé river front where Syzygium guineense and
Parinari congensis were likewise frequent. At the Ouémé site the difference between ‘river
edge’ and ‘middle of GF’ is best shown by a decrease of abundance of certain species such as
Cynometra megalophylla, Lepisanthes senegalensis, Cola laurifolia, Drypetes floribunda,
Morelia senegalensis, Vitex chrysocarpa, Dialium guineense, Xylopia parviflora and
Manilkara multinervis. Meanwhile, Pterocarpus santalinoides and Syzygium guineense are
particularly abundant along the Ouémé river. ‘Savanna edge’ has the most distinct floristic
composition with Pouteria alnifolia at both sites; Antidesma venosum, Fagara zanthoxyloides
along the Comoé river and Ceiba pentandra, Albizia spp. and Antiaris toxicaria at the Ouémé
site.
Contrasting results were obtained concerning the tree species richness and diversity
(H’) between the three plot types. For example, the number of species was highest at
‘savanna edge’ and ‘riverside’ plots respectively for both Comoé and Ouémé. On the
contrary, species diversity was at a maximum in the ‘riverside’ plots, but minimum diversity
was reached in the ‘middle’ plots at the Comoé site and at ‘savanna edge’ at the Ouémé site.
However, the difference between the plot types concerning species diversity (H’) was not
statistically significant (at p = 0.05) in both GF sites.
Concerning the structure variability at the three plot types, contrasting results were
obtained as well. Along the Comoé, both ‘middle’ and ‘riverside’ attained higher values for
tree height, diameter and basal area (m2/ha) than ‘savanna edge’, where tree height was
clearly lower. Likewise, maximum tree height (30 m) was found for an individual of Ceiba
pentandra occurring in ‘Central’ plots. On the contrary, at the Ouémé site, ‘savanna edge’
plots had the highest average diameter and height. Total basal area decreased from the
‘riverside’ to ‘savanna edge’ plots. The tallest tree, Cola gigantea (26 m), was found at the
GF/savanna boundary.
9.4. DISCUSSION
With regard to the most prominent families of the Comoé and Ouémé GFs, there are some
similarities with several tropical GFs and dense forests. Leguminosae, Euphorbiaceae,
Rubiaceae and Annonaceae are the most species-rich families always found in Benin riparian
forests (NATTA et al., in prep.). In the Dja fauna reserve (Cameroon) Euphorbiaceae,
Rubiaceae, Annonaceae, Meliaceae, Caesalpiniaceae, Sapindaceae and Sapotaceae have the
highest tree species richness (LEJOLY et al. 1996, SONKÉ & LEJOLY 1998). In a dense 2-ha
forest at Kade (Ghana) Leguminosae was the best represented family (SWAINE et al. 1987).
MEAVE & KELLMAN (1994) found similar results in a 1.6 ha of riparian forest in the
Mountain Pine Ridge (Belize). They further found similarities between this riparian forest
flora and the flora of neotropical continuous rain forests. The three tropical rain forest blocks
are abundant in Leguminosae, Annonaceae, Euphorbiaceae, Lauraceae, Myristicaceae,
Rubiaceae and Sapotaceae (WHITMORE 1990).
The floristic dissimilarity between the two West African sites is illustrated by the
comparison of the Shannon indices, which combine species richness and abundance. The
higher diversity of the Comoé site over the Ouémé site can be attributed to the regional flora
and vegetation diversity. Benin is located in ‘the Dahomey gap’, a discontinuity of the West
African rain forest belt, which is the product of erratic rainfall, topographic and
oceanographic interactions (JENIK 1994). This gap includes essentially the drier types of the
Guineo-Congolian forest belt in South-Eastern Ghana, Southern Togo and parts of Southern
Benin (ERN 1988). So in Benin, savanna and agro-ecosystems are widespread, and gallery
forests are always among the most dense and diverse vegetation formations (NATTA et al.
2002). Côte d’Ivoire is a major rain forest country in West Africa with a much more
111
Chapter 9: Ouémé and Comoé: forest-savanna border relationships in two riparian ecosystems in West Africa
diversified vegetation and flora than Benin. Due to their elevated moisture regime (compared
to adjacent regions), GFs in the South of the CNP enable the establishment and persistence of
plant species from wetter zones such as evergreen rain forests from lower latitudes in Côte
d’Ivoire (POREMBSKI 2001). At both sites, species characteristic for river front and central
portion of GF were Cynometra megalophylla, Dialium guineense, Cassipourea congoensis,
Syzygium guineense and Parinari congensis, while Pterocarpus santalinoides is particularly
abundant along the Ouémé. For the Benin riparian forests, Pterocarpus santalinoides is the
most common species at all latitudes (NATTA et al., in prep.).
The Ouémé and Comoé GFs act as outpost for certain species (e.g. Dialium
guineense, Drypetes floribunda, Diospyros abyssinica, Holarrhena floribunda, Fagara
zanthoxyloides, Ceiba pentandra, Celtis spp., Mallotus oppositifolius, Spathodea
campanulata, Elaeis guineensis, Albizia spp., etc.) which occur in humid forest on plateau
between 8° and 9º N or which can be found under more humid climatic conditions in the rain
forest region of Southern Côte d’Ivoire. The presence of these species substantiates some
hypotheses on the ecological importance of these hygrophilous and edaphically controlled
forest formations. They play an important role as migration channels, which provide
opportunities for genetic exchange between geographically isolated populations. In many
seasonally dry regions, but also in formerly forested rain forest areas, gallery forests possess
the character of a refugium for plants and animals (POREMBSKI 2001). FORMAN & GODRON
(1986) have documented the importance of gallery forests as routes for movement of plants
and terrestrial animals across the landscape. For MEDLEY (1992), gallery forests are the most
suitable vegetation for plant species adapted to a moist climatic regime. Tropical riparian
forest fragments represent plausible refugia for forest species in fire-prone landscapes, and
are known for their potential, albeit limited, to act as safe sites for tropical rain forest species
(KELLMAN & TACKABERRY 1993, KELLMAN et al. 1994, MEAVE & KELLMAN 1994,
KELLMAN & MEAVE 1997).
Cynometra megalophylla, a typical riverine forest species, is the most frequent and
dominant species at the riverside and in the middle of GF in the two sites. It may occur
occasionally in other forest types (KEAY & HEPPER 1964, HALL & SWAINE 1981). In Benin,
apart from riparian forests, this species is dominant only in a periodically flooded forest on
vertisol (dark-clayey soil), the Lama forest (AGBANI 2002). Concerning absolute density and
basal area Cynometra megalophylla accounted for 9 to 16 %, and 26 to 33 %, respectively in
the two sites. In fact, the dominance of a group or a single species is not a new phenomenon
within tropical forest in general, and GFs in particular. According to WOLTER (1993) cited by
SONKÉ & LEJOLY (1998), dominant species can comprise up to 58 % of the total individuals
and 75 % of the basal area. In Benin, 9 tree species (Pterocarpus santalinoides, Cola
laurifolia, Syzygium guineense, Dialium guineense, Berlinia grandiflora, Cynometra
megalophylla, Elaeis guineensis, Diospyros mespiliformis and Uapaca togoensis) out of 224
contribute to 46.3 % of the total abundance and occupy 55.3 % of the total basal area.
Pterocarpus santalinoides alone contributes to 14.8 % and 12.8 % of total abundance and
basal area respectively (Natta et al., in prep). In a riparian forest of Belize, MEAVE &
KELLMAN (1994) found that the seven most abundant species (2.4% of the species richness)
account for 1/3 of the total number of stems.
At least three reasons might explain the dominance of Cynometra megalophylla at the
two GF sites. Firstly, mono-dominance may develop if disturbance does not occur over long
periods and if the regional pool of shade tolerant species is limited (HART 1990). Secondly,
according to POREMBSKI (2001), this species may possess particular competitive traits in the
specific abiotic conditions of a GF. Thirdly field data show that Cynometra megalophylla
occurs along larger waterways South of 9º10’ N (latitude of Bétérou in Central Benin, see
NATTA et al., in press Chapter 6) and as far North as Ferkéssédougou (9º30’ N) in Northern
112
Chapter 9: Ouémé and Comoé: forest-savanna border relationships in two riparian ecosystems in West Africa
Côte d’Ivoire (AKÉ ASSI, pers. comm.). This species seems to reach its optimal ecological
range in the Guinean zone and in sub-humid areas along rivers. It is hypothesized that
frequent or lasting environmental stresses, such as floods, result to a decrease in species
diversity (at least for trees), which in turn will lead to the increased abundance and
dominance of those species that can successfully maintain themselves in isolated or harsh
sites. HANSON et al. (1990) described a similar phenomenon in GF patches along rivers in the
upper mid-West of the Iowa river (USA). The dominant species along waterways are seen as
keystone species and intrinsic components in the functioning of complex and self-organizing
riparian ecosystems. They are generally essential for ecosystem resilience (FOLKE et al.
1996). The phenomenon of dominance by a few or single species should also be investigated
from the viewpoint of micro-climatic and soils conditions.
The plots situated at the forest/savanna boundary were distinguished from the middle
of GF and riverside by their physiognomy and structure (i.e. height, diameter, basal area of
trees), floristic composition and species richness; but with contrasting trends between the two
sites. Along the Ouémé river there is a constant decrease of the number of individuals, tree
species, basal area and Shannon indices (H’) from the “riverside’ to ‘middle’ of GF ending at
the ‘savanna edge’. At the Comoé site, the number of individuals and species are highest at
savanna edges whilst average tree height, diameter and basal area are highest in the ‘middle
of GF and at ‘river edge’. Surprisingly, the difference between the three plot types concerning
tree species diversity (H’) was not statistically significant at both sites.
At both sites, the GF/savanna contact is made through a belt of either light demanding
species such as Anogeissus leiocarpus, Mitragyna inermis, Alchornea cordifolia (POILECOT et
al. 1991, POREMBSKI 2001) or species from dense semi-deciduous and open forest (Pouteria
alnifolia, Ceiba pentandra, Albizia spp. and Antiaris toxicaria). Light regime and flood
frequency somehow play an important role for GF edge communities. From the trends
detected here, the assumption that a gradient of environmental variables (e.g. moisture, light)
influences the distribution and frequency of tree and shrubs species within GF hold true.
More detailed data about microclimate and edaphic conditions, as well as seedling
establishment, growth, light demand, and other ecological factors for typical GF and savanna
boundary species, would enable to draw more definitive conclusions. The present study
illustrates the complex variability of ecological processes within gallery forest and between
gallery forest and the adjacent savanna.
ACKNOWLEDGEMENTS
We gratefully acknowledge the assistance and financial support from several institutions and
persons. Professor L. Aké Assi (Centre National Floristique de l’Université d’Abidjan) kindly
provided plant determinations and information about the ecology of typical gallery forest
species in Côte d’Ivoire. The BIOTA-West staff supported additional field check at the
Comoé National Park. We would like to thank Professor K.E. Linsenmair (Würzburg,
Germany), Dr. F. Fischer (Würzburg, Germany) and Dr. K. Hahn-Hadjali (Frankfurt.,
Germany) for field assistance in the Comoé National Park. We are also indebted to Professor
L.J.G. van der Maesen (Wageningen University, the Netherlands), co-ordinator of the Benin
Flora Project, which provides financial assistance for data collection in Benin. We also wish
to thank Professor B. Sinsin (Faculty of Agronomical Sciences / University of AbomeyCalavi, Benin) for his guidance during field work and data analysis in Benin.
113
Chapter 10
ASSESSING THE DENSITY OF KHAYA SPECIES THROUGH SIMPLE RANDOM
SAMPLING, STRATIFIED SAMPLING AND SYSTEMATIC SAMPLING
Natta A.K., de Gier A. and van der Maesen L.J.G.
Chapter 10: Assessing the density of Khaya species through SRS, StRS, and SS in Central Benin
Chapter 10
ASSESSING THE DENSITY OF KHAYA SPECIES THROUGH SIMPLE RANDOM
SAMPLING, STRATIFIED SAMPLING AND SYSTEMATIC SAMPLING
Natta A.K.(1), de Gier A.(2) and van der Maesen L.J.G.(3)
(1) MSc. DAGE/FSA/UAC 01 BP 526, Tel/fax: 00 229 303084 Cotonou, Benin; aknatta@yahoo.com
(2) Professor of sampling design and forest management at ITC - Department of Natural Sciences, Enschede;
The Netherlands. degier@itc.nl
(3) Professor of Plant Taxonomy, Biosystematics Group, Wageningen University, The Netherlands;
jos.vandermaesen@wur.nl
ABSTRACT
Three sampling designs (i.e. Simple Random, Stratified and Systematic) were used to
estimate the density of Khaya senegalensis and K. grandifoliola trees in the Pénéssoulou
forest reserve (Central Benin). For both Khaya species, stratified random sampling provided
the lowest variance, coefficient of variation, standard error, and sampling error, thus was
taken as the most precise and reliable design over simple random and systematic samplings.
Densities of trees per ha (dbh ≥ 10 cm) were 5.4 ± 1.9, and 2.3 ± 1.1 for Khaya senegalensis
and K. grandifoliola respectively. Stratification with proportional allocation of sampling units
has proven valuable in providing precise and reliable density estimates in the study area. On
the contrary, systematic sampling gave the least precise and reliable estimates. Other things
being equal we recommend stratified random sampling for the assessment of population
estimates of Khaya trees in the Pénéssoulou reserve forest. Results have confirmed empirical
knowledge about the ecology of Khaya species and shown that the selection of the most
precise sampling design, with regards to estimating a given parameter, can be useful for the
sustainable management of forest trees in the study area. Meanwhile the assessment of the
properties of more complicated sampling designs, such as adaptive sampling and distance
sampling, could bring out issues not covered or revealed by the three conventional methods
that were applied here.
Key words: sampling designs, endangered species, Khaya senegalensis, Khaya
grandifoliola, density, Pénéssoulou forest, Benin.
10.1. INTRODUCTION
Khaya senegalensis (Desv.) A. Juss. and K. grandifoliola C. DC. (Meliaceae) are among the
most best-known endangered species in the West African savanna and pre-forest regions.
They are very important for the local economy in the whole West African region. They are
well-known for their good timber, fuel wood and charcoal quality (Eyog Matig et al. 2002),
medicinal values (Kerharo & Adam 1974, Berhaut 1979, Adjanohoun et al. 1989, Keita et al.
1999, Sokpon & Ouinsavi 2002), as well as for the high nutritive value of their leaves as
livestock fodder (Sinsin 1991, Touré 1991, Meurer 1994). Overexploitation of natural stands,
weak recruitment of seedlings, and habitats degradation have resulted in a high level of threat
and genetic erosion. These two species of Khaya are classified among those threatened most
on a world scale (Oldfield et al. 1998). They are mostly found in riparian forests in the
savanna regions, but also in semi-deciduous forest and forest outliers, and Northern limits of
the Guinea and Congo rain forest blocks (Keay & Hepper 1958). In the Ibadan region and in
117
Chapter 10: Assessing the density of Khaya species through SRS, StRS, and SS in Central Benin
Southern Kaduna State (Nigeria), the two Khaya species were once dominant in riparian
forests (Onochie 1979).
K. senegalensis (also called dry-zone mahogany, or Acajou wood), with a shining
foliage, has a wide range up to the Sahelo-Sudanian region in West Africa. In Benin, it
spreads from 8°N to the far Northern part, but has disappeared from many natural stands in
the guineo-congolese zone (see Sinsin et al. 2002a). It usually grows on clayey-silty or
sandy-silty soils. K. senegalensis constitutes the first timber species for Sudano-Sahelian
countries (Eyog Matig et al. 2002), and has the highest conservation priority ranking for five
West African countries: Benin, Togo, Ghana, Niger and Chad (Adjanohoun et al. 1989, Eyog
Matig et al. 2001). It also contributes to the cure of 55 diseases in Benin (Sokpon & Ouinsavi
2002). Meanwhile, missing regeneration indicate that K. senegalensis stands are jeopardised
in several sites, particularly in North Benin (Sinsin et al. 2002a). Also the practice of roadside
plantation of K. senegalensis, promoted during colonial times for shade and landscape
beautification, has all but ceased in West Africa. The remaining trees are heavily lopped for
fodder and damaged for the medicinal bark. On the other hand, K. grandifoliola is a large
forest tree, and an important timber tree occurring more frequently in dry semi-deciduous
forest and forest outliers (Guillaumet & Adjanohoun 1971). It does not extend beyond the
Sudano-Guinean zone (Aubreville 1950). This species seems to reach its optimal growth in
riparian and deciduous forests in the Bassila and Djougou regions (Central Benin).
Current management activities of the endangered species of Benin in general, and
Khaya species in particular, is hampered by lack of knowledge about the spatial distribution,
and structure related parameters. The Project Flora of Benin has set as major objectives to
contribute in filling the gap of knowledge on vegetation formations, as well as the
distribution, structure and threats of most Benin’s plants. Although endangered, little is
known about the distribution and occurrence of Khaya species in Benin. Apart from ethnobotanical studies, there are only a few publications on Khaya species in Benin. Sinsin et al.
(2002a) have documented the spatial distribution of Khaya senegalensis in Benin, and linked
the distribution map to levels of human disturbance. So far no distribution map have been
published for Khaya grandifoliola. Not only the spatial distribution of the two species still
needs to be fully investigated, but also there is no comprehensive and reliable study on
density estimation of Khaya species in various ecological regions in Benin.
To estimate a given parameter, it is now acknowledged by scientists and forest
resource managers that it is desirable to select an efficient sampling design that provides
precise estimates. This can eventually be used in conservation or sustainable harvest
planning. Likewise the selection of candidate sampling methods must fulfil certain
requirements related to the target species and study area characteristics. To study Khaya
species we have selected one of the least degraded protected forests, which provides good
ecological conditions for the two species. The driving mechanisms of the spatial organisation
of the two Khaya species are mostly unknown in the country, and as yet unpredictable in the
selected study area.
If we assume that the distribution of these species is random, simple random sampling
can be applied to be representative of the whole population. This design is the base of most
field survey designs, and is also commonly used as a basis of comparison. Also it is
sometimes convenient to apply sampling designs that are easy to draw and execute, and that
ensures a uniform (i.e. even) coverage of sampling points throughout the whole study area.
Therefore systematic sampling was applied. As the study area is relatively flat, we assume
that the systematic layout of sampling points is not related to a periodic variation of
ecological factors or vegetation types. On the other hand, the vegetation map of the study
area allowed us to divide the population into homogeneous and non-overlapping vegetation
types (riparian forests, open and dry forests, and savanna), each one having its own
118
Chapter 10: Assessing the density of Khaya species through SRS, StRS, and SS in Central Benin
characteristics. The assumption here is that the distribution of the target species may be
related to one or more strata, hence stratified random sampling could be more efficient. The
stratification of the study area is also suggested by empirical knowledge about optimal
vegetation type for each Khaya species (Keay & Hepper 1958, Aubreville 1950, personal
observations).
This study aims at estimating the density and assessing the precision of population
estimates of two Khaya species through three conventional sampling designs, simple random,
systematic and stratified random samplings. A research question is put: what are the densities
of Khaya species, and which sampling design provides the most precise population
estimates? Results from this study are intended to improve knowledge of the ecology of
Khaya species, and eventually to sustainably manage the remaining natural stands in the area.
10.2. METHODS
10.2.1. Study area
The Pénéssoulou forest (Figure 10.1) was selected for several reasons. It is located in the
Sudano-Guinean zone of the Bassila district (Centre West Benin), which is the westernmost
part of the semi-deciduous forest, fire-subtype in the sense of Hall & Swaine (1981).
According to Adjanohoun et al. (1989), the climate of the region is transitional between the
subequatorial climate of the Guinean region (South Benin), and the Sudanian climate (North
Benin). It is characterised by progressive fusion of the two peaks of rainfall typical of the
subequatorial climate. There are two seasons: a rainy season from mid-April to end-October,
and a dry season from end-October to mid-April. Annual rainfall ranges between 1100 and
1300 mm/year, and average temperature from 26 to 32º C. The relative humidity varies from
15 to 99 % in the dry and rainy seasons respectively. The mean annual insolation is 2420
hours, and the average annual Potential Evapotranspiration is 1536 mm. The relief is flat with
gentle slopes. This hygrophile enclave stretches out to Central Togo and Central and South
East Ghana (Adjanohoun et al. 1989). Forests in the Bassila region are floristically diverse
with a emergent stratum, 20 to 25 m high, where species such as Khaya grandifoliola, K.
senegalensis, Zanha golungensis, Cola gigantea, Albizia spp., Holoptelia grandis, Antiaris
toxicaria, Milicia excelsa, Vitex doniana, Diospyros mespiliformis, Clesitopholis patens,
Dalbergiella welwitschii and Anogeissus leiocarpus are frequent (PRRF 1998).
Until recently, the whole district was among the least disturbed forest zones of Benin,
but now deforestation due to timber gathering is alarming (PRRF 1998, personal
observations). Selective tree cutting has resulted in the disappearance of most large
specimens of timber trees (e.g. Khaya spp., Antiaris toxicaria, Milicia excelsa, Afzelia
africana, Prosopis africana). People now have turned to less valuable savanna trees, such as
Isoberlinia doka and I. tomentosa. Within the Bassila region, the Pénéssoulou reserve forest
is well known for its dense riparian forest network, which occupies 13% of the total area of
the forest, and is the most dense vegetation formation (PRRF 1998). The two Khaya species
seem to reach their optimal growth in the Sudano-Guinean zone where this protected area is
situated. Also, since 1998 the Pénéssoulou reserve forest experiences a management plan in
partnership with local populations. We noticed that current management activities appear to
have proven efficient in controlling poaching, trapping, fire, and encroaching cultivation.
Other things being equal, this protected forest, which provides good conditions for the
survival of Khaya stands, was selected for the present study.
119
Chapter 10: Assessing the density of Khaya species through SRS, StRS, and SS in Central Benin
1°
2°
3°
NIGER
12°
AFRICA
BURKINA
FASO
12°
Malanville
N
11°
11°
Natitingou
10°
10°
Pénéssoulou forest
NIGERIA
Parakou
9°
9°
Pénéssoulou
reserve forest
TOGO
Major cities
8°
8°
States limits
Major waterways
Bohicon
7°
7°
0
100 km
Porto - Novo
Cotonou
1°
2°
3°
Figure 10.1: The Pénéssoulou forest reserve in Benin.
The Pénéssoulou forest covers some 5470 ha, from which a rectangular area (Figure 10.2), 1
km large and 3.6 km long (i.e. 360 ha or 6.6% of the protected forest), was surveyed in that is
known to be the densely wooded and least disturbed part.
10.2.2. Sampling designs and field procedures
Three conventional sampling designs (Simple Random, Stratified and Systematic) were used
to estimate tree densities. The population consisted of 9,000 plots from which 180 were
actually selected and surveyed. Plot centres were located in the terrain through a GPS, a
clinometer, and tapes. Slope-corrected rectangular plots of 400 m2 were used. In each plot,
the diameter of all Khaya trees (i.e. dbh ≥ 10 cm) was measured. Field procedures were
specific to each sampling design.
Simple random sampling (SRS) is the base of most sampling designs and is widely
applied in forest resources inventory. Most other sampling procedures that are designated to
achieve greater economy and precision are modification of SRS. The central idea behind SRS
is the equal probability selection, i.e. every possible combination of n units is equally likely
to be the sample selected. The estimates of population mean and variance are designunbiased, as they do not depend on any assumptions about the population itself (Cochran
120
Chapter 10: Assessing the density of Khaya species through SRS, StRS, and SS in Central Benin
1977, Freese 1990, Thompson 1992, Ott & Longnecker 2001). 180 out of 9000 plots were
randomly selected without replacement from their x and y coordinates on the study area map,
and then located accurately in the field. The recorded characteristic in each sampling unit is
the number of stems of Khaya trees.
Units in a systematic sampling (SS) are selected according to a pre-specified pattern
throughout the population. Here the study site is divided in 90 blocks, each containing 100
plots. Two starting points were randomly selected in the first block, resulting in two sets
(primary units) of systematically selected plots. Therefore 180 plots (secondary units) were
surveyed under the systematic design. The chosen design permits the calculation of means
and standard deviations. All properties of estimates are obtained based on the design by
which the sample of primary units is selected (Thompson 1992).
In stratified random sampling (StRS), the population is divided in strata on the basis
of similarity of some characteristics, here vegetation types. The underlying assumption was
that the variable of interest (tree density) was related to the different strata. The vegetation
map of the Pénéssoulou forest (Figure 10.2), which is derived from satellite images from
1998, was used to determine the size of each vegetation type. The three vegetation types were
proportionally sampled with regards to their size (Table 10.1). Within each stratum, a random
sample without replacement has been independently selected, and estimates computed per
stratum. The strata estimates are combined to give a population estimates. StRS usually offers
two major advantages over SRS: it provides separate estimates of the mean and variance of
each stratum, and for a given sampling intensity, it often gives more precise estimates of the
population parameters than would a SRS of the same size.
Towards Djougou
N
Nioro
Pénélan
Pénéssoulou
Deciduous and open forests
Riparian forests
Tree and shrub savanna
Plantations
Towards Bassila
Boundary of the protected forest
Villages
Paved road
Secondary road
0
0.5
1 Km
Study site in the forest
Figure 10.2: Location of the study site in the Pénéssoulou forest (From PRRF/Bassila, 1998)
121
Chapter 10: Assessing the density of Khaya species through SRS, StRS, and SS in Central Benin
Table 10.1: Vegetation types and plot characteristics in the study area
Vegetation types Plots in Percentage Simple Random Stratified Random
(or strata)
the site
(%)
Sampling (SRS) Sampling (StRS)
Riparian forests
1620
18
32
Dry & open forests 1800
20
36
Savanna
5580
62
112
Total
9000
100
180
180
Systematic
Sampling (SS)
180
10.2.3. Assessment of the precision of population estimates
Many authors have used or suggested various estimators (e.g. variance, variance ratio,
standard error, sampling error, coefficient of variation, etc.) to compare sampling designs and
assess their relative precision, reliability, accuracy or efficiency. An unbiased estimate of
variability can be used to assess the reliability of a survey result. Along with getting unbiased
estimates come goals of precise or low variance and procedure that are convenient or costeffective to carry out. Standard error, or standard error of the means is considered as the
standard deviation among sample means. It gives an indication of the reliability of the mean:
the smaller its value, the more reliable the mean. As pointed out by Cochran (1977) and Ott
& Longnecker (2001), standard error of the population mean and total can be used to estimate
the precision actually attained in a survey that has been completed, compare the precision
obtained by SRS with that given by other sampling methods, or indicate how accurate any
estimate should be. Likewise confidence intervals (t*standard error) are used to assess the
accuracy of the computed estimate (Thompson 1992).
Likewise, the ratio of the variance, also called design effect by Kish (1965), obtained
from any sampling design over the one obtained from SRS for the same number of units, is
used in appraising the efficiency of sampling designs. The coefficient of variation is the
standard deviation expressed as a percentage of the mean. It is often used as an index of
process or population variability (Ott & Longnecker 2001). According to Snedecor &
Cochran (1989), one may judge the success of a statistics, evaluate experiments or compare
populations by looking at the coefficient of variation.
For Cochran (1977) better precision or lower variance is sometimes reached while
minimising the cost, expressed as time or money to locate and enumerate sampling points. As
the objective of study was not to assess the efficiency (i.e. lowest variance for a given cost) of
the three sampling designs, time involved in data collection was not measured. However, we
have used the same number of sampling units (i.e. 180) for the three designs, and have
assumed that locating and enumerating these sampling units for the 3 designs cost the same
time.
Following the authors cited above, we used five estimators (i.e. variance, variance
ratio, standard and sampling errors, and coefficient of variation) to compare the precision and
reliability of the three designs with regards to density estimation. They were selected because
of their well-known utility. Ample explanations and more complicated issues related to the
five estimators are provided by classical and specialised statistical publications. Generally the
lower (or smaller) the variance, variance ratio, standard and sampling errors, or coefficient of
variation, obtained from a given sampling design, the more reliable and precise is the
estimated parameter. The combination of two or more lowest values of the selected
estimators is much sough-after, and indicates the potential of a sampling design as being the
most reliable and/or precise.
The formulae that were used to estimate population mean, density, total, variance,
variance ratio, standard error of the mean, sampling error, coefficient of variation (see Table
122
Chapter 10: Assessing the density of Khaya species through SRS, StRS, and SS in Central Benin
10.2) are given by Cochran (1977), Snedecor & Cochran 1989, Freese 1990, Thompson
(1992), and Ott & Longnecker (2001).
10.2.4. Data processing
Data were processed through Excel, using the formulae given in Table 10.2.
Table 10.2: Unbiased estimators of population parameters from Simple Random Sampling
(SRS), Stratified Random Sampling (StRS) and Systematic Sampling (SS).
Estimators
Simple Random
Sampling (SRS)
X
1
n
=
Mean/plot
n
¦
xi
i=1
Stratified Random Sampling
(StRS)
1 L
X st =
¦ N hxh
N h =1
Systematic Sampling (SS)
y=
1
Mn
Variance
Variance ratio
(design effect)
Standard error
Sampling error (*)
S2 =
i
− X )2
i =1
n −1
S 2 S 2 =1
S2
n
SX =
Density/ha
Total (i.e. 360 ha
for the study area)
2
h
L
S
1
S (yst ) = 2 ¦Nh (Nh − nh )
nh
N h=1
2
(t * S ) / x * 100 %
¦( y − My)
i
N −n
var y = 2 * i=1
n −1
M Nn
S 2 ( y st ) S 2
2 2
L
n·
§
* ¨1 − ¸ S = 1 * ª N h S h
¦«
© N ¹ X st
N 2 h=1 ¬« nh
X
Coefficient of
variation
i
i =1
n
n
¦ (x
n
¦y
(t * S )/ x
X st
st
var y S 2
§ nh ·º
¨¨1−
¸¸»
© N h ¹¼»
* 100
s
* 100 %
CV =
x
s
* 100 %
CV =
x
x ± t * S x ( ha )
x ± t * S x (360 ha )
x st ± t * S x st ( ha )
x st ± t * S x st (360 ha )
SX =
S2
n
n·
§
* ¨1 − ¸
N
¹
©
(t * S ) / x * 100 %
X
CV =
s
* 100 %
x
x ± t * S x ( ha )
x ± t * S x (360 ha )
(*) All confidence intervals are calculated at probability level of 0.05.
10.3. RESULTS
Outputs from the three sampling designs are summarised in Tables 10.3 and 10.4 for Khaya
senegalensis and K. grandifoliola, respectively. For Khaya senegalensis, stratification
resulted in the lowest variance (0.001), ratio variance (0.004), coefficient of variation (18.55
%), standard error (0.040), and sampling error (36.36 %). Therefore it gave the most precise
estimate of the population density (i.e. 5.4 ±1.9 trees/ha), and total (i.e. 1945 ± 707 trees/360
ha). Mean and variance per stratum show the absence of Khaya senegalensis in riparian
forests in the study area.
With Khaya grandifoliola, stratification gave also the lowest variance (0.0006), ratio
variance (0.001), coefficient of variation (25.62 %), and standard error (0.024). The sampling
error from StRS was a bit higher (50.2 %) than the one obtained from SRS (47.6 %), and this
is related to a smaller value of the mean (i.e. 0.095) obtained with StRS compared to 0.172
from SRS. A close look at the formula of sampling error (see Table 10.2) shows that for a
given t-value and standard error: the smaller the mean, the higher the sampling error. This is
actually the case as: sampling error (StRS) = 1.059*sampling error (SRS). The lowest
123
2
Chapter 10: Assessing the density of Khaya species through SRS, StRS, and SS in Central Benin
stratified mean results from the fact that K. grandifoliola is absent from the savannas, which
occupy 62 % of the study area. Therefore the mean/plot in savanna is null and this has
lowered the overall stratified mean/plot. Because it provided the lowest values for variance,
ratio variance, coefficient of variation and standard error (4 estimators over 5), StRS was
taken as the most precise sampling for the estimation of K. grandifoliola density. Therefore
the most precise estimate of population density and total for K. grandifoliola were 2.3 ± 1.1
trees/ha, and 855 ± 429 trees/360ha respectively.
SS appeared to be, in both cases, the least precise sampling design, as in all cases it
gave the highest variance, variance ratio, standard error, sampling error and coefficient of
variation.
Table 10.3: Results from the three sampling designs for Khaya senegalensis
Estimated parameters
Simple Random
Sampling (SRS)
Mean/plot (0.004 ha)
0.211
Variance
0.379
Ratio of variance (design effect)
Density/ha (*)
Total abundance (360 ha)
Coefficient of variation in %
Standard error of the mean
Sampling error in % (*)
1
5.2 ± 2.2
1900 ± 802
291.9
0.045
42.21
Stratified Random
Sampling (StRS)
0.216 Stratified mean
0.000 (riparian forests)
0.250 (dry, open forests)
0.267 (savanna)
0.001
0.000
0.364
0.360
Systematic
Sampling (SS)
0.270
Stratified variance
(riparian forests)
6.396
(dry, open forests)
(savanna)
0.004
5.4 ± 1.9
1945 ± 707
18.55
0.040
36.36
16.844
6.8
2450
929.09
1.770
1274.72
* 95% confidence interval
Table 10.4: Results from the three sampling designs for Khaya grandifoliola
Estimated parameters
Simple Random
Sampling (SRS)
Mean/plot (0.004 ha)
0.172
Variance
0.3221
Ratio of variance (design effect)
Density/ha (*)
Total abundance (360 ha)
Coefficient of variation in %
Standard error of the mean
Sampling error in % (*)
1
4.3 ± 2.05
1550 ± 738
329.55
0.041
47.66
Stratified Random
Systematic
Sampling (StRS)
Sampling (SS)
0.095 Stratified mean
0.160
0.250 (riparian forests)
0.250 (dry, open forests)
0.000 (savanna)
0.0006
0.3226
0.2500
0.0000
Stratified variance
(riparian forests)
2.0897
(dry, open forests)
(savanna)
0.001
2.3 ± 1.1
855 ± 429
25.62
0.024
50.21
(*) 95% confidence interval
124
6.487
3.9
1400
929.3
1.011
1275
Chapter 10: Assessing the density of Khaya species through SRS, StRS, and SS in Central Benin
10.4. DISCUSSION
Considerable study is often required to be able to select the proper design for a given problem
in a given situation. So far such a study, comparing several sampling designs, is the first one
done in the Pénéssoulou reserve forest.
Own data collected in several riparian forest sites throughout the country showed that
densities of Khaya species varied from 1.6 to 18 trees/ha, and 5 to 13 trees/ha for Khaya
senegalensis and K. grandifoliola, respectively. Khaya grandifoliola is meanly encountered
in the Bassila region, where the Pénéssoulou forest is located. We could not compare our
results with other similar studies in various vegetation types in Benin, as published data are
non-existent.
Field observations confirm empirical knowledge about the ecology of the two Khaya
species in the study area. Khaya senegalensis is not observed in riparian forests, the wettest
vegetation type in the study area. This suggests that K. senegalensis favours less humid
conditions such as savannas and open forests. On the contrary Khaya grandifoliola is not
observed in savanna because it favours wetter vegetation types. Therefore, values for mean
densities of Khaya species can be misleading if they are not linked to the ecological
preferences of each species. Although we have not collected data on the regeneration of these
species, observations throughout the whole country suggest that the study area with its
current management activities, is comparatively well-off, and has a great potential for the
long term conservation of Khaya species.
For both species, StRS gave the lowest values for all the five estimators for K.
senegalensis, and four out of five (i.e. variance, variance ratio, coefficient of variation and
standard error) for K. grandifoliola. Considering variance ratio, stratification has reduced the
variance obtained from SRS to about 235 and 545 times for K. senegalensis and K.
grandifoliola, respectively. Generally the reduction of sampling variation is obtained by subdividing the population into relatively homogeneous strata. But the efficacy of stratification
depends on foreknowledge of the behaviour of experimental material (Snedecor & Cochran
1989.), i.e. an expected relationship between stem density and vegetation type for the species
concerned. We have used a recent vegetation map (from 1998), and stratification of the study
area in three vegetation types appeared to be effective, in comparison to SRS and far better
than SS, for the same number of sampling units. If we assume that 180 sampling units of the
three designs cost the same in time to locate and enumerate, then the designs can be
compared for efficiency. Under this assumption, we might say that StRS is also more
efficient than SRS and SS. Additional field investigations will provide new insights in this
issue.
Although relatively easy to draw, to execute, and to control, the coverage of the whole
population in a systematic way is not a guarantee for a better precision of Khaya species
density. The performance of SS in relation to that of StRS or SRS is greatly dependent on the
properties of the population (e.g. observations/items in random order, with linear trend,
stratification effects, periodic variation, auto correlated, or unpredictable) (Cochran 1977).
Following this author we might say that it is difficult to give general advice about the
situations in which SS is to be recommended in the study area, because the specific properties
of the Khaya species are not yet well known. From the data available, the vegetation map and
our own knowledge of the study area, we could not yet detect any particular property of
Khaya populations. Further research should document this issue.
Once a particular distribution pattern or population property is suspected or has been
detected, appropriate surveys designated to getting more precise and cost-efficient estimates
are available. It is now acknowledged that if a species is known or can be expected to cluster
125
Chapter 10: Assessing the density of Khaya species through SRS, StRS, and SS in Central Benin
(i.e. gregarious distribution of individuals), more efficient adaptive sampling designs are
available (Thompson 1992, Thompson & Seber 1996). Also, distance sampling provides
biologists with a powerful yet practical method for estimating density of various populations,
including plant populations (Buckland et al. 1993).
The approach followed here for Khaya species can be tested with other valuable
timber trees, such as Milicia excelsa, Antiaris toxicaria, Afzelia africana, Vitex doniana,
Diospyros mespiliformis, Albizia spp., Anogeissus leiocarpus, Pterocarpus erinaceus,
Holoptelea grandis, etc. Therefore we could get consistent results for most of the timber tree
in the study area. Starting from 1998, the Pénéssoulou protected forest management plan has
set as objective to harvest 30% of the increment of wood on the basis of a 10 year rotation.
Knowledge on the reliability of species density and precision of sampling designs can
eventually be useful for the management of tree populations in the study area, if consistent
results are obtained for other timber species. Although the Pénéssoulou protected forest
management plan has specified the attributions of each stakeholder, there is still room to
improve current and future management activities (e.g. number of trees to be harvested).
10.5. CONCLUSION
The present study facilitates the choice between several sampling designs for the estimation
of the density of Khaya species, which are endangered in Benin. In the study area, StRS with
proportional allocation of sampling units was the most precise sampling design in comparison
to SRS and SS to estimate Khaya stem density. The mean densities/ha were 5.4 ± 1.9 and 2.3
± 1.1 for Khaya senegalensis and K. grandifoliola trees, respectively (p = 0.05). Our results
also have confirmed empirical knowledge about the ecology of the two Khaya species. If
consistent results are also obtained for most timber trees in the study area, this may be helpful
to manage (e.g. for sustainable harvest) the remaining natural stands of these valuable trees.
Ongoing research should clarify several issues, such as the detection of population
characteristics and the assessment of non-conventional sampling designs. This could bring
out issues not covered by the three conventional sampling designs compared here.
126
Chapter 11
GENERAL DISCUSSION
A.K. Natta
Chapter 11: General discussion
Chapter 11
GENERAL DISCUSSION
11.1. CHARACTERISTICS OF RIPARIAN FORESTS
Riverine landscapes are heterogeneous, dynamic, and biologically and spatially complex
(Ward et al. 2002). Within riverine landscapes, riparian forests (RFs) are transitional between
terrestrial and aquatic ecosystems and are distinguished by gradients in biophysical
conditions, ecological process, and biota. They are portions of terrestrial ecosystems that
influence exchanges of energy and matter between aquatic and uplands ecosystems (NRC
2002), and contribute to the diversity and function of both terrestrial and aquatic ecosystems
(Acker et al. 2003). At all latitudes, riparian or streamside forests are recognized as distinct
components of the landscape because of their unique ecotonal nature, linking aquatic and
terrestrial ecosystems (Naiman & Décamps 1990, Gregory et al. 1991, Malanson 1993,
Hancock et al. 1996, Cordes et al. 1997). They occupy areas of intense land-water interaction
and are known to support concentrated and diverse assemblages of wildlife and plant species
(Nilsson 1992, Naiman et al. 1993). This enhanced biodiversity is believed to be related to
landscape mosaic, spatial heterogeneity, complex environmental gradients and unique natural
disturbance regimes (Robinson et al. 2002), which instigate a wide variety and abundance of
resources and substrates (Manuwal 1983, Knopf 1985, Szaro & Jakle 1985, Acker et al.
2003). As pointed out by Brinson (1990), RFs own their dynamics, structure, and
composition to river processes of inundation, transport of sediments, hence the erosive forces
of water.
The characteristic plant species, plant communities and associated aquatic or semiaquatic animal species in RFs are intrinsically linked to the role of water as both agent of
natural disturbance and as critical requirement of biota survival. Alternating environmental
stress (Hancock et al. 1996), such as periodic flooding, is a form of disturbance to which
many of the taxa occurring in riparian communities appear well adapted (Jongkind 1996), by
sprouting from remaining root or trunk (e.g. Pterocarpus santalinoides, Cola laurifolia,
Syzygium guineense, Cynometra megalophylla, personal observations). According to Acker et
al. (2003), flooding apparently promotes complexity at the life form, species composition,
and stand structural levels, thus the significance of channel constraint for severity of floods
may be different in mountainous versus relatively flat terrain. Frequent floods are known to
maintain the native riparian vegetation in a mosaic of different successional stages (Salo et al.
1986, Kalliola & Puhakka 1988, Naiman et al. 1993). Sediments deposited by floods can
create initial seedbeds for germination, but also injure seedlings (Levine & Stromberg 2001).
Research on the effects of flooding have shown that it increases wood and leaf decomposition
rates, and sites that are flooded generally have a lower accumulation of organic matter (Bell
& Sipp 1975, Bell et al. 1978, Peterson & Rolfe 1982, Shure et al. 1986, Ellis et al. 1999). As
a result RFs tend to be very diverse in species composition and physical structure (Gregory et
al. 1991).
Also, RFs at either waterway side are often related to gradients of environmental
factors perpendicular to the stream, all linked with water availability. Changes in
environmental factors across RFs are shown by: (a) decreasing humidity, soil moisture,
nutrient level, productivity (Hancock et al. 1996, Ettema et al. 1999); (b) decreasing flood
intensity, frequency and duration (Stromberg et al. 1993a, Hupp & Osterkamp 1996, Piegay
& Bravard 1997, Swanson et al. 1998); and (c) increasing light availability (Dignan & Bren
2003).
129
Chapter 11: General discussion
In the West-African savanna region, riparian forests are the natural wooded vegetation
types, of varying successional states, bordering (i.e. lining) fresh water bodies such as rivers,
streams, brooks and lakes. Riparian forests occupy 2.37 % of Benin’s the total area of the
country, and dense deciduous and semi-deciduous forests 1.05 %. Therefore dense forests in
Benin occupy some 3.42 % of total area of Benin (CENATEL 2001). Compared to the figure
given by FAO in 1980 (i.e. 0.55 % for deciduous, semi-deciduous forests, and important
riparian forests), the updated value seems more reliable because of computerised data
collection, rectification and analysis of remotely sensed data. Yet, it shows how big the
difference can be when various sources at different periods interpret and map tropical
vegetations.
11.2. ECOLOGICAL IMPORTANCE
RFs are under the influence of fluvial and geomorphic processes (Baker 1990, Fetherston et
al. 1995), and because of short- and long-term in waterways basin hydrological cycles, they
are in constant state of adjustment, with different abiotic and biotic components adjusting
with different lag-times, producing a complex-response situation (Cordes et al. 1997). Many
authors have documented on dynamic processes within riparian ecosystems, and interactions
of RFs with waterborne particles and chemicals, microbial assemblages, flood, flood plain,
water table, climate and micro-climate, soils, local geology, channel morphology, topography
and stream banks (Guillaumet 1967, Hall & Swaine 1976, 1981, Devineau 1975, 1984, van
Rompaey 1993, Leinard et al. 1999, NRC 2002). The contribution of RFs to the functioning
of rivers and biodiversity protection is recognised worldwide (Adjanohoun 1965, Baker 1990,
Tabacchi et al. 1990, Polansky 1994, Lykke 1996). More than other vegetation types, they
are well-known for their roles in controlling water runoff and quality, mineral nutrient flows,
bank erosion and stabilisation, sedimentation, flood, air and soil pollution, water shading and
cooling, sediment filtering, aquifer recharge, carbon storage, erosion control, etc. (Devineau
1984, Forman & Godron 1986, Gregory et al. 1991, Haycock et al. 1993, Harper et al. 1994,
Large & Petts 1994, Forniss 1996, Mander et al. 1997, Aide & Rivera 1998).
The processes involved in water quality improvement are: sediment retention,
preservation of the floodplain channels from excessive erosion, filtering and decomposition
of nutrients and water pollutants (e.g. NO3-, see Willems et al. 1997, Groffman et al. 2001),
improved soil water infiltration rates, etc. (Gilliam 1994, Lowrance et al. 1997, Trimble
1999). Riparian zones with various types of vegetation are reported to be effective in
reducing, removing, or assimilating nutrients, coming from upslope agricultural field via
shallow lateral flow (Peterjohn & Correl 1984, Pinay & Décamps 1988, Weller et al. 1994,
Hefting & Klein 1998, Ettema et al. 1999). Here, the processes involved include reduced
velocity of runoff flow, increased sedimentation of coarse particles, retention of suspended
particles by leaf and the soil, plant uptake, microbial immobilisation, nitrification and
denitrification (Lowrance et al. 1984, Ambus & Lowrance 1991, Groffman et al. 1991,
Ambus & Christensen 1993). Contrasting results were found concerning denitrification in
temperate countries: rates of denitrification in RFs were higher than in grassland zones
(Hefting & Klein 1998, Haycock & Pinay 1993), and lower than in grassland (Groffman et al.
1991, Schnabel et al. 1996).
Maintaining biodiversity is one of the most important functions of RFs, which are,
directly or indirectly, home to an abundance of animal life (e.g. fisheries, birds, invertebrates
and micro-organisms, amphibian and reptiles, mammal species, etc.). For Johnson et al.
(1977), riparian zones perform an essential and unique function in areas where water
availability is limited. According to Thomas (1996), typical riparian forest tree species are
dependent on river flows, a shallow aquifer, and the community and population structure of
130
Chapter 11: General discussion
riparian forests are related to spatial and temporal patterns of flooding. In most cases, they do
not occupy large areas (Varty 1990, Monnier 1990, Medley 1992), nevertheless they are
important in the conservation of a large range of plants, animals, water sources, soils and
watersheds. In the Amazon Basin, numerous streams and rivers provide huge potential for
increasing the conservation value of deforested and fragmented landscape through the
protection of linear remnants along waterways (de Lima & Gascon 1999). Not only do they
constitute a natural habitat or the last refuge for many species, but also they lodge many
endemic species and extinction-menaced species (Roggeri 1995), and usually act as routes for
movement of terrestrial plants and animals across the landscape (Forman & Godron 1986).
Mature riparian forests are known to store large quantities of biomass (above ground,
roots, litterfall, woody debris, soil carbon, submerged aquatic vegetation) than younger or
degraded stages (Burke et al. 1999, Giese et al. 2003). RFs regulate inputs to streams,
including solar radiation, dissolved nutrients, foliage and other food resources (Gregory et al.
1991), and large woods, which add to structural and biological diversity of streams by
creating pools and other hydraulic features needed by many fishes, algae, and invertebrates
(Sedell et al. 1988, Fetherston et al. 1995, Knutson & Nae 1997, Bragg & Kershner 1999).
All these fundamental and critical ecological functions that RFs perform better than
uplands and aquatic ecosystems, fall into three major categories: (a) hydrology and sediments
dynamics, (b) biogeochemistry and nutrient cycling, (c) species habitat diversity and food
web maintenance.
11.3. PLANT COMMUNITIES DIVERSITY IN RIPARIAN FORESTS
In addition to being characterized by a variety of plant species, and unique assemblages of
plants, compared to uplands and wetlands, RFs in Benin harbour a newly described rare and
endemic ornamental Acanthaceae (Thunbergia atacoriensis), which is found along certain
streams at hills feet. Compared to other tropical forests, such occurence is not negligible and
cannot be underestimated within the Dahomey Gap.
The structural complexity and species richness of tropical forests make difficult their
study by the traditional phytosociological procedures, but numerical methods have proved
successful, especially in demonstrating correlations between vegetation and environmental
factors in fairly small areas (Hall & Swaine 1976). Also studies utilising direct and indirect
gradient analysis have shown that each plant species is distributed in a unique pattern in the
landscape and that this pattern results from the interaction of each species with its physical
and biological environment. As result, the composition of plant communities changes more or
less continuously along environmental gradients as changing conditions cause species to
enter or leave (Whittaker 1975).
The direct ordination technique based on environmental variables, does not
recommend itself in the present study, which main interests are botany, phytosociology and
ecology. Meanwhile, ongoing research should clarify this issue. In the meantime, indirect
gradient analysis (here DCA) was applied and environmental gradients were inferred from
species composition data. Also it is worthwhile noting that, sometimes classification of
vegetation into communities or associations may obscure the true situation in the terrain that
is characterised by gradual changes of the floristic composition of plant assemblages over
environmental gradients.
Traditional methods of plant community ecology usually rely on the dominance of
overstorey trees to define species associations and establish plant community boundaries,
however, as shown by Sagers & Lyon (1997) species associations in the overstorey are not
necessarily good predictors of understorey associations. To overcome this, we have taken all
species of the least degraded relevés (presence/absence data file), and no species was down131
Chapter 11: General discussion
weighted in the classification analysis. Ordination and classification analyses provided
evidence that RFs are structured along gradients relating to moisture, flood conditions,
regional variations in climate and relief. These factors have created and maintained several
riparian forests plant communities, some of which have not yet been described. Also, microtopographic variations across waterways, created by diverse fluvial process, support floristic
richness that would not otherwise occur. The repetitive plant assemblages in riparian forests
are made of species that can tolerate greater water table and flow variation, physical
disturbance, and hydric stress either due to drought or soil saturation.
Elsewhere, e.g. in riparian ecosystems of the South Western USA, salinity,
disturbance from fire and community maturity were the most important gradients (Busch &
Smith 1995). Also, the distribution of vegetation, in general and forests in particular, within
riparian zones has been linked to numerous abiotic factors such as hydroperiod, floodplain
landform and sediment types as well as to competition and life history factors (Medley 1992,
Dunn & Stearns 1987, Nilsson et al. 1989, Hughes 1990). The type and severity of
disturbance also have a major impact on the diversity of RFs plant communities. In East
Gippsland in Victoria (Australia), Melick (1990) found that undisturbed riparian rain forests
were the most species poor plant community, and disturbed ones the most species rich. In a
riparian forest of the Buffalo National River (Arkansas, USA), species associations were
influenced by environmental gradients dominated by pH and elevation, but secondary
gradients differed among forest layers (Sagers & Lyon 1997).
Moreover dispersal (Skoglund 1990, Malanson 1993, Johansson et al. 1996,
Malanson & Armstrong 1996), floating (Danvind & Nilsson 1997, Jansson et al. 2000a), as
well as colonisation (Auble et al. 1997, Stromberg 1997) strategies of seeds in riparian
ecosystems, and the ability of riverbanks to trap waterborne diaspores (Andersson et al.
2000), are known to influence patterns of species richness and abundance of RF and predict
vegetation changes. However, diversity patterns of riparian plant community were found to
be complicated, with contrasting relationships.
11.4. RIPARIAN FORESTS AND WILDLIFE CONSERVATION
Waterways and their dynamic forested banks are known to serve as a focal point for animals
of many kinds. They host a diverse and abundant fauna of aquatic, terrestrial and amphibious
species (Robinson et al. 2002). Also, studies in very different areas have documented the
importance of riparian areas for birds, that these have typically different assemblages and
higher species richness than neighbouring areas (Skagen et al. 1998, Rottenborn 1999, Saab
1999, Woinarski et al. 2000). The importance of RFs as critical, seasonally or temporary
habitat for a variety of animals is reported by numerous authors (e.g. Knopf et al. 1988,
Hunter 1990) and confirmed by our own field observations in Benin. The reproduction,
growth and nutrition depend on the presence of more or less dense and safe stands of riparian
forests: e.g. nesting in trees along streams and rivers, food consisting of fish or other aquatic
animals from inundation plains near water courses.
In Benin, animals such as Cephalophus rufilatus, Tragelaphus scriptus, Kobus
defassa, Loxodonta africana africana, Python sebae, Crocodylus niloticus, Osteolamus
tetrapis, Heliosciurus gambianus, etc. very much like gallery forests and forests edges
(Sinsin et al. 1997). Likewise Cercopithecus mona, Cercopithecus nictitans, Cercopithecus
diana live in forest areas and large and dense galleries. Cercopithecus erythrogaster
erythrogaster, the endemic primate found so far in Benin (Grubb et al. 1999, Sinsin et al.
2002b) shares a part of only its time in riparian forests and forest patches in the Ouémé valley
region. Raynaud & Georgy (1969) reported that Syncerus cafer nanus that was usually seen
in woodlands and humid sites in the South and Central Benin follows gallery forests up to the
132
Chapter 11: General discussion
latitude of Djougou to Bembèrèkè. Tragelaphus scriptus and Cephalophus rufilatus main
habitats are forests, its edges and gallery forests. Limnotragus spekei lives in humid or
swampy areas, gallery forests and upland forests. Cephalophus sylvicultor lives in dense
forest or large and dense galleries. Phacochoerus aethiopicus lives in groups in savannas,
gallery and edges of forest. The Potamochoerus porcus was found in dense forest and
Guinean savanna and gallery forests as far as in the W National Park along the Mékrou
stream. Generally in regions where dense forest does not exist, riparian forests play the role
of residual forest and provide, when they are undisturbed, the best conditions for wildlife that
normally lives in forest zone.
In Central Africa, species such as squirrels (Helioscirus gambianus var. punctatus,
Protoxerus stangeri, Paraxerus poensis), anomalure (Anomalurops beecrofti, Anomalure
derbianus), and civet (Nandinia binotata) are mainly seen either in dense forest or RFs
(Depiere & Vivien 1992). Likewise certain bird species (e.g. Tigriornis leucolopha,
Butorides striatus atricapillus, Hagedashia hagedash brevirostris, Stephanoaetus coronatus,
Podica senegalensis, Tympanistria fraseri, Centropus monachus occidentalis, Musophaga
violacea) depend on RFs, dense forests or stream and river shores with periodic inundation
for feeding and nesting.
Many studies on wildlife ecology highlight the potential of linear remnant forests
along waterways to serve as habitat for small forest vertebrates and suggest they could
function as corridors for some species to increase landscape connectivity (Forman 1997,
Machtans et al. 1996). Examples are given by de Lima & Gascon (1999) in Central
Amazonia for small mammal and litter-frog communities; Burbrink et al. (1998) in Illinois
(USA) for frogs and lizards; Darveau et al. (2001) in the riparian zone of the Laurentian
mountains (Québec) for small mammals. In the USA, RFs are much more important to bird
communities than surrounding habitats because the vegetation is structurally complex and
floristically diverse within the vertical and horizontal dimensions (Anderson & Ohmart 1977,
Best et al. 1979, Hehnke & Stone 1979, Szaro 1980, Bull & Skovlin 1982, Szaro & Jakle
1985). It has also been shown that a mosaic of riparian woodlands, containing mixture of
native tree and shrub species of different classes, is necessary to maintain avian species
richness (Farley et al. 1994).
11.5. WATERWAYS ORDER AND SIZE RELATIONSHIP WITH RIPARIAN FOREST BIODIVERSITY
Outputs from the DCA and TWINSPAN suggest that waterways can be grouped into two
types: large (= rivers) and small (= streams). Field observations also support this
simplification of the reality. This analysis was based neither on measured environmental
variables nor on previous floristic data and hydrographic network maps. Indeed, physical
characteristics of streams vary from low to high order (Brierley & Hickin 1985, Forman &
Godron 1986, Billy & Ward 1989), and RF characteristics are known to change with
increasing stream order (Swanson & Lienkamper 1982, Swanson et al. 1982, Minshall et al.
1985, Nilsson et al. 1989). As pointed out by Gregory et al. (1991) and Naiman et al. (1992)
stream size or order, is an important factor that determines the nature of interactions between
geomorphology, fluvial and terrestrial disturbance, and riparian community characteristics.
Diversity of RFs and their distinctiveness from upslope forests tend to increase with
increasing stream size (Naiman et al. 1993). Therefore we expect concomitant changes in the
flora and plant community structure, and associated animals.
In Benin, RFs along streams are more species rich than those along rivers. In South
West Ghana, Jongkind (1996) found that the floristic difference between large rivers and
small streams is striking, and this was probably caused by the much smaller seasonal changes
in the water level and the absence of an open space above the water of the small streams (see
133
Chapter 11: General discussion
also Devineau 1984 in Côte d’Ivoire). Similar stream order patterns were identified in
numerous overseas studies. In South West of Western Australia, RFs along tributaries (low
order streams) had higher species richness (at p < 0.05) than those along the largest river
(higher order streams) (Hancock et al. 1996). This was related to the more intimate
association between streamside and upland communities along first order streams due to less
developed environmental gradient away from the usually ephemeral streams. In several RFs
in the USA, Naiman et al. (1987) and Lock & Naiman (1998) found that bird communities on
large rivers differ from those on small rivers and these differences occur predictably with
stream order. Also deciduous vegetation and patch types appeared to be linked to river size
through stream valley shape and disturbance history.
As waterways size and order have been shown to have a great impact on vegetation
types as well as wild- and bird life, ongoing research in Benin should integrate several
environmental (abiotic factors) variables (e.g. river order and size, sediment deposition,
debris flow, position in the watershed, patch size, RFs width and elevation within the
watershed, etc.) in multivariate analysis. Likewise, the concepts of ephemeral, perennial, and
quasi-perennial waterways should be re-investigated from field observations (e.g. hill
headwaters, stream related to hills, waterways on plateau with or no extended flood plains).
11.6. RIPARIAN FORESTS RELATIONSHIPS WITH UPLAND PLANT COMMUNITIES
RFs occur in various climatic, geomorphic, edaphic, hydroperiod and geographic settings,
and have similar as well as contrasting features with surrounding upland plant communities,
with regards to species composition, structure, and ecosystem diversity.
In most savanna regions, narrow bands of vegetation in the vicinity of the river
contrast strongly with open forests and savanna woodlands which otherwise dominate the
landscape (Woinarski et al. 2000). Numerous studies confirm not only the great richness of
RFs over upland forests (Brinson 1990, Nilsson 1992, Nilsson et al. 1994, Spackman &
Hughes 1995), but also that the structure of RFs is generally more complex than upland
forests (Hancock et al. 1996, Melick & Ashton 1991). It is now accepted that RFs harbour
more species of birds (Stauffer & Best 1980, Gates & Giffen 1991, McGarigal & McComb
1992, Larue et al. 1995, Whitaker & Montevecchi 1997), and mammals (Geier & Best 1980,
Doyle 1990) than uplands habitats. Roché (1993) found bird species richness in riverine flood
plains to be twice that in adjacent uplands, while Wakentin & Reed (1999) documented that
the majority of bird species in arid regions of Great Basin (USA) are associated with riparian
habitats. Also in the extensive natural landscape of Australia’s tropical savanna, species
richness and abundance of birds was significantly greater in riparian forests than in matched
non-riparian areas, and riparian vegetations allow many species to extend their distributions
into lower rainfall areas (Woinarski et al. 2000). The physical and vegetation characteristics
of the streamside area differ from those upslope because of frequent inundation, soil
saturation, and physical disturbance of streamside vegetation due to flood flows, mass soil
movement, etc. (Brinson 1990, Gregory et al. 1991, Fetherston et al. 1995).
On the contrary, along the Helena River and its tributaries, and in the Grampians
(Western Australia), RFs were generally less species rich than adjacent upland communities
(Enright et al. 1994, Hancock et al. 1996). According to these authors their results were
corroborated by other studies in other parts of Australia and South Africa (see Cowling &
Holmes 1992), but not in Europe. RFs in the Oregon Coast Range are said to have lower tree
density than upland stands (Gregory et al. 1990, Pabst & Spies 1999, Nierenberg & Hibbs
2000). Reasons of such richness pattern are not yet fully investigated.
Light penetration, which depends upon topographic position, slope and vegetation
characteristics, plays a major role in the ecology of the riparian forest, and protection of the
134
Chapter 11: General discussion
light environment is generally seen as an important function of buffers that aim to protect
riparian habitat values (Barton et al. 1985, MacDougall & Kellman 1992, Dignan & Bren
2003). The ecotone between the riparian and upland vegetation may be very narrow or
gradual depending on the fact that gradients are sharp or not. In the latter case it could be the
site of highest species richness because of the presence of species from both communities
(Tilman 1988, Hancock et al. 1996). In Benin, field observations show that RFs tree flora is
not only more diverse, but also has marked differences in structure, abundance and
composition with surrounding upland plant communities (e.g. in Yarpao, Pénéssoulou,
Idadjo, etc.). However, this issue was not analysed with environmental data and numerical
analysis of the whole floristic data. Therefore we could not yet draw any conclusion about the
entire RF flora (i.e. trees, shrubs and herbaceous).
11.7. DEGRADATION OF RIPARIAN FORESTS
It is generally agreed that species extinction is largely related to the reduction and
fragmentation of their habitats, therefore the protection of species can best be done through
protecting habitats (Swaminathan 1990). So, a wide range of plants and animals will benefit
from the protection of forest ecosystems along rivers and streams. Despite their importance to
biodiversity in many countries, native RFs have been severely degraded. This is attributed to
a variety of factors including dam construction, river channelisation, cattle ranching,
agricultural development, and recreational development (Décamps et al. 1988, Knopf et al.
1988, Roods & Mahoney 1990, Busch & Smith 1993, Busch 1995, Braatne et al. 1996,
Lonard et al. 2000). Dams and water diversions are known to modify surface flow rates,
flood periodicity, and sediment and nutrient transport, often to the detriment of riparian plants
(Jansson et al. 2000b, Levine & Stromberg 2001). When channelling occurs within riparian
systems, removal of sediment and nutrients from surface runoff is less effective (Norris
1993).
Modification of the frequency, duration and intensity of floods, draining and lowering
of floodplain water tables, contribute to change in RF communities (Stromberg et al. 1996,
Crawford et al. 1996, Molles et al. 1998). As sites become drier, width of riparian corridor
declines and species composition shifts from obligate and facultative wetland plants to those
with xerophytic adaptations (Baker 1989, Szaro 1990), and changes in physiognomy and
biomass may also occur, sometimes independently with compositional shifts (Stromberg et
al. 1993b, Stromberg 2001). Many authors (e.g. Medina 1990, Stromberg & Patten 1990,
Smith et al. 1991, Rood et al. 1995) observed that water reduction impact on riparian tree
populations include lowered tree densities, loss of class diversity due to death of susceptible
age classes, reduced reproductive output, increased seed size. Changes in water regime
inhibit recruitment of native plants, and, thus create ‘functional gaps’ in riparian communities
(Levine & Stromberg 2001).
Urbanisation on lands adjacent to intact riparian woodland has substantial impacts on
riparian bird species richness, density, and community composition (Katibath 1984, Smith &
Schaefer 1992, Cubbedge & Nilon 1993, Ohmart 1994, Rottenborn 1999). Likewise invasion
of the naturalized shrubs potentially alters competitive hierarchies and disturbance regimes in
riparian systems (Busch & Smith 1995, Ellis 2001).
Riparian plants are not dependent upon fire for renewal, but fire can influence the
composition and structure of riparian ecosystems (Kellman & Meave 1997), in particular the
understorey (personal observations), in combination or not with other factors such as flooding
(Kellman & Tackaberry 1993, Bendix 1994, DeBano & Neary 1996, Ellis 2001, Everett et al.
2003). In Central New Mexico (USA), the suppression of flooding along RFs has also
increased forest floor litter and woody debris, which may have contributed to the increased
135
Chapter 11: General discussion
frequency and severity of fires (Ellis et al. 1998, 1999). In tropical savanna regions, when the
gallery forests degrade, forest species become sparse, leaving the vegetation open for the
savanna species to invade (Lykke 1996). The environmental impacts of such degradations are
generally tremendous. For example, the changes in plant physiology, population dynamics,
and community structure affect functional roles such as provision of habitat for ripariandependent animals (Stromberg 2001).
Many studies have shown that in Benin riparian forests have waned during the last
decades. Interpretations of aerial photographs and satellite images show that degradation rates
varied from 0.14 to 8.6 during the period 1949 -1998 (Table 11.1). So far, no study has
shown an increase in riparian forests area during that period.
Table 11.1: Degradation rates of riparian forests in Benin
Study area
Latitude (°N) Degradation
rate (%)
Klouékanmè
6°55’ – 7°10 1.58
Dekpo-Lonkly
2
7° – 7°10
Kétou
7°20’ – 7°35 8.59
Dogo
7°30’ – 7°40 3.32
Ouémé-Boukou 7°45’ – 7°55 0.3
Bantè
8°05’ – 8°36 0.14
Wari-Maro
0.51
9°– 9°10
Djougou
3.4
9°45
Natitingou region 10°05’– 10°30 0.8
Péhunco
10°05’– 10°35 1.14
Alibori forest
10°40’– 10°50 3.5
Malanville
3.4
11°45
Period (number
of years)
1982-1994 (12)
1982-1994 (12)
1949-1994 (45)
1949-1995 (46)
1949-1998 (49)
1979-1987 (8)
1975-1997 (22)
1949-1975 (25)
1975-1994 (19)
1975-1997 (22)
1975-1998 (23)
1949-1975 (25)
Authors
Adambiokou-Akakpo (2001)
Gnele (1999)
Houndagba et al. (in press)
Houndagba et al. (in press)
Houndagba et al. (in press)
Akoègninou et al. 2001
Sounon (2001)
FAO (1980)
Tenté (2000)
Wotto (2001)
Arouna (2002)
FAO (1980)
Periodic removal of trees (either by selective tree cutting or farming) along waterways
has modified the structure, composition, hence the functions of riparian forests. Beside
farming and illegal tree cutting, livestock animals have a disproportionate effect on riparian
forests ecosystems, because they tend to concentrate in areas, which are close to rangelands
and water reservoirs. Long-term cumulative effects of domestic livestock grazing involve
changes in the structure, composition, and productivity of plants and animals at community,
ecosystem and landscape level (e.g. along the Ouémé river at Bétérou). Less often seen
disturbances in RFs in Benin are industrial, urbanization and recreational activities.
Nevertheless their impact is very destructive.
11.8. FLORISTIC RELATIONSHIPS OF RFS IN BENIN WITH OTHER TROPICAL DENSE FORESTS
There are marked floristic affinities as well as differences between RFs in the Dahomey Gap
and those in the Guinean and Congo forest blocks. Certain RF plants in the Dahomey Gap are
also found in forests along waterways in the Guinean and Congo rain forest blocks (Table
11.2). Among the most important we have Cola laurifolia, Pterocarpus santalinoides, Ficus
asperifolia, Entada manii, Leptoderris brachyptera, Psychotria calva, Manilkara multinervis,
Synsepalum brevipes, Parinari congensis and Uapaca heudelotii. These species can be
termed typical stream and riverside forest plants of West and Central Africa. They also can be
seen as specialists of a fixed habitat of limited area, because they are tied to special
topographic or edaphic site factors (see Hubbell & Foster 1986). However, riparian habitat
specialists are often common where RFs are extensive.
136
Chapter 11: General discussion
Among fresh water (swamp) forest species found in RFs in the Dahomey Gap,
Syzygium guineense var. guineense, Cleistopholis patens, Cynometra megalophylla, and
Mitragyna inermis are common. Floristic data indicate that large numbers of upland forest
species in drier parts of the rain forest zones (e.g. Dialium guineense, Diospyros
mespiliformis, Elaeis guineensis, Xylopia parviflora, Dennettia tripetala, Hexalobus
crispiflorus, Isolona thonneri, Drypetes floribunda, etc.), coexist locally within RF patches in
the Dahomey Gap. Also, large specimens of Cola gigantea, Ceiba pentandra, Milicia
excelsa, and Antiaris toxicaria are frequent at RF edges. Most of these species are typical of
the three drought resistant or dry semi-deciduous forest types of the Guineo-Congolian forest
belt (Ern 1988, Hall & Swaine 1981). Generally at RF edges we have a host of open forest,
forest outliers or woodland species, such as Albizia spp., Combretum spp., Terminalia spp.,
which augment RF phytodiversity.
Table 11.2: Typical riparian forest species of West and Central African regions
Species
Cola laurifolia
Cynometra megalophylla
Entada mannii
Ficus asperifolia
Gardenia imperialis
Irvingia smithii
Leptoderris brachyptera
Manilkara multinervis
Mimosa pigra
Morelia senegalensis
Napoleonaea vogelii
Ouratea glaberrima
Parinari congensis
Pterocarpus santalinoides
Psychotria calva
Synsepalum brevipes
Syzygium spp.
Uapaca heudelotii
Xylopia aethiopica
Life
Geographic
Habitat
forms* affinity**
mPh
GC
River banks tree in forest and savanna zone
mPh
GC
River banks and fresh water swamp forest species
mph
GC
A climbing forest shrub in streamside forest
nph
GC
Shrub near water
mph
GC
River banks and fresh water swamp forest species
mPh
PRA
Riparian forest tree
LmPh
GC
Climbing shrub in riparian forest
mPh
TA
Riparian forest tree
nph
Pan
Shrub often on sandy bars along rivers
mph
SZ
Shrub of gallery and swamp forests
mph
G
Riparian forest tree or near to sea shore
nph
GC
Erected shrub or small tree in forest by streams
MPH
GC
Evergreen tree on river banks and fringing forest
mPh
PRA
Forest tree on river banks
nph
GC
Erected shrub beside streams in forest
mph
GC
Riparian forest tree
mPh
River banks and fresh water swamp forest species
mPh
GC
River banks and fresh water swamp forest species
mPh
GC
Riverine forest tree, also in dry forest
Source
(1),(2),(7),(8)
(1),(2),(3),(6),(8)
(2)
(1),(2),(4),(9)
(1),(2),(4)
(4),(9)
(2)
(1),(4)
(9)
(1),(4),(9)
(1),(2)
(1)
(1),(2),(4),(6),(8),(9)
(1),(2),(3),(6),(7),(8),(9)
(2),(4)
(4)
(9)
(1),(4),(6),(9)
(5),(6)
* Life forms follow Raunkaier (1934), Schnell (1971), and Keay & Hepper (1954-1972).
** Geographic affinity of each species (i.e. phytogeographic types) follows White (1986), Keay & Hepper
(1954-1972), Keay et al. (1964).
(1) Keay et al. (1964); (2) Keay & Hepper (1954-1972); (3) Aubreville (1950); (4) Pauwels (1993); (5) de
Koning (1983); (6) Hawthorne (1996); (7) Onochie (1979); (8) Adjanohoun (1968); (9) Schmitz (1988).
On the other hand, there are marked floristic differences between RFs in the rain
forest zone and those in the Dahomey Gap. Numerous plants bound to waterways in the rain
forest zones are absent in the Dahomey Gap. Among the most common riverine plants, we
have Brachystegia spp., Cathormion spp., Coelocaryon spp., Cynometra spp.,
Gilbertiodendron spp., Hymenocardia heudelotii, Millettia spp., Psychotria spp., Sacoglottis
gabonensis, etc. (see Keay & Hepper 1954-1972). So far, apart from a few evergreen forest
trees adapted to swampy situations (e.g. Cynometra megalophylla, Pentadesma butyracea,
Xylopia aethiopica, etc.), no true rain forest species of the Guinean-Congo regions are present
in riparian forests in the Dahomey Gap.
137
Chapter 11: General discussion
11.9. STRUCTURAL
RELATIONSHIPS OF
RFS
IN
BENIN
WITH OTHER TROPICAL DENSE
FORESTS
In contrast of savanna regions, the physiognomy of RFs in rain forest regions is similar to
that of adjacent rain forests. The distinction between the two forest types can therefore only
be made through their specific floristic composition and topographic position. In the Zaire
basin (Central Africa), RFs that fringe waterways are narrow where the banks are high but
expanding into great swamp forests in the sump-lands (Kingdon 1990). The difference
between RF in the Dahomey Gap and dense forest in the rain forest zone is well illustrated in
terms of tree density, basal area and plant richness. Although our own data indicate that high
plant species numbers can be sustained in isolated RFs habitats, tree richness and diversity in
RFs are lower in the Dahomey Gap than in the continuous West and Central Africa rain
forests (see Table 11.3).
Table 11.3: Characteristics of some riparian and rain forest sites in Africa and tropical
America
Locality
Shannon Tree density Basal area TR/ha SR/ha Rainfall Source
2
(mm)
index (H’) (stem/ha)*
**
***
(m /ha)
RIPARIAN FORESTS IN THE DAHOMEY GAP (BENIN & TOGO)
South Benin (Samiondji 7°20N)
2.4
726
Central Benin (Ouèssè 8°30N)
3.9
748-785
Benin: various RFs
3.7-4.9
312-665
South Togo: various RFs
1.89
-
41.7±17.4
45.6±15.8
21.8-41
-
27
34
-
249
129-195
-
1200
1150
1100
1000
(1)
(1)
(2)
(3)
RIPARIAN FORESTS IN SOUTH AND CENTRAL AMERICA
Belize: Río Bravo
Brazil: Mobi-Guaçu
476
Belize: Mountain Pine Ridge
766±241
21.9±8.8
56
52
-
1517
1280
-
(4)
(5)
(6)
360-523
523
520
375
461
-
30.5
-
34-70
42
109
138
-
2400
2080
2400
3250
3900
3900
(7),(8)
(8)
(9)
(9)
(10)
(11)
605-649
376-555
539-565
-
35-43
25.3-36.3
25.5
28.4-32.2
-
69-131
77
86-122
120
< 100
-
3000
2100
3000
2000
1640
-
(12),(13)
(14)
(15)
(16)
(17)
(11)
WEST AND CENTRAL AFRICAN RAIN FORESTS
Nigeria: Okumu forest reserve
Nigeria: Shasha forest reserve
Nigeria: Omo forest reserve
Cameroon: Bakundu reserve
Cameroon: Dja
5.4
Cameroon: Korup 0.65 ha plot
(African pleistocene refugia)
Gabon: various rain forests
Côte d’Ivoire: Yapo
Côte d’Ivoire: Taï National Park
Ghana: dense forests
Ghana: Kade research station
African non-refugial sites
-
* (dbh ≥ 10 cm); ** TR = Tree Species Richness per ha; *** SR = Species Richness per ha.
Sources: (1) Natta et al. (in prep.); (2) Sokpon et al. (2001); (3) Kokou et al. (2002); (4) Kellman et al. 1994; (5)
Gibbs & Leitão (1978); (6) Meave & Kellman (1994); (7) Jones (1955); (8) Richards (1939); (9) Richards
(1963); (10) Sonké & Lejoly (1998); (11) Whitmore (1990); (12) Reitsma (1988); (13) Hladik (1982), (14)
Corthay (1996); (15) Dengueadhe (1999); (16) Hall & Swaine (1981); (17) Swaine et al. (1987).
Also, most RFs of the Dahomey Gap are characterised by many small-stemmed
communities of low height (12-18 m) compared to tropical rain forest canopy height. In rain
forest regions, the uniformity in the physiognomy at riversides makes it difficult to
differentiate between RFs and adjacent dense evergreen forests. Swamps can be distinguished
easily, but lower slopes and riverbank forests, which are fed by groundwater as well, are
138
Chapter 11: General discussion
more difficult to distinguish from upland forest (van Rompaey 1993). Therefore from the
viewpoint of forest physiognomy, we prefer to use the term ‘riverine forest species’ instead
of ‘riparian forest’ for rain forest regions.
11.10. CONSERVATION OF DENSE FOREST SPECIES IN RIPARIAN FORESTS IN THE DAHOMEY
GAP: implications for periods of climate change
As many as 350 out of 410 (i.e. 85.4 %) species recorded in RFs in South and Central Benin
are not specialised to riparian systems, but are rather a subset of the general flora without
endemic species. One would expect many rain forest taxa in RFs of the Dahomey Gap,
because it is the wettest vegetation formation in this relatively dry savanna corridor.
However, present days distribution pattern of plants show that the greater water availability
owing to higher water tables in RFs seems not to favour many rain forest taxa. In Northern
Australia, riparian vegetation has many elements of monsoonal rain forests, and has probably
been an important refugium area for rain forest species in period of increased aridity during
the Pleistocene (Russell-Smith 1991), as it was postulated for RFs in neotropical systems (see
Meave & Kellman 1994). In Belize and Venezuela, RF fragments often contain a nonspecialised tree flora with species characteristic of continuous forests. Provided that the local
diversity of riparian systems is complemented by high regional diversity (this is lacking in the
Dahomey Gap), RFs may have provided plausible refugia for tropical forest taxa in drier
periods (Meave et al. 1991). Also, studies on the geographic patterns of genetic diversity of
certain lowland species in Central America showed that these species might have persisted in
riparian zones during the late Pleistocene (Aide & Rivera 1998). Likewise in the Neotropical
savanna region, it has been suggested that tropical rain forest taxa may have persisted in
small mesic, fire-protected pockets within Pleistocene non-forest communities (Leyden 1985,
Colinvaux 1987).
For Whitmore (1990) parts of the world’s rain forests that are most rich in species are
those that the evidence shows have been the most stable, where species have evolved and
continued to accumulate with the passage of time without episodes of extinction caused by
unfavourable climatic periods. Therefore the floristic dissimilarity between RFs in the
Dahomey Gap and those in the African rain forest zones as well as the presence of only a few
rain forest taxa in the former, suggest three hypotheses:
1 - The most recent regression of dense forests in West Africa, and subsequent
reopening of the Dahomey Gap, which occurred from the end of the middle Holocene to the
beginning of the late Holocene (see Sowunmi 1986, Maley 1991, 1997, 2001, 2002, Tossou
2002) has resulted in a localised extinction of African rain forest taxa in RFs of the Dahomey
Gap. Support to this hypothesis is given by e.g. Léonard (1965) and White (1979), who found
that the Dahomey Gap constitutes a barrier to the distribution of very few typical forest plant
species in West Africa (see also Kingdon 1990).
2 - Even during favourable periods, the ‘Gap’ may have been too much of an obstacle
for some species (Kingdon 1990), probably for rain forest taxa. We now known that the
shaping of the Dahomey Gap vegetation is linked to historical retreats and invasions of
forests, which occurred several times during the Pleistocene and Holocene (Guillaumet 1967,
Maley 1989, Dupond & Weinelt 1996, Dupond et al. 2000, Tossou 2002). Therefore the
absence of many rain forests taxa in RFs might be the result of cumulative effects of these
historical events at the actual location of the Dahomey Gap.
3 - Despite the higher water balance of RFs over upland forests, the environmental
stress (i.e. recurrent flooding but also periodic drought) is unfavourable for the establishment
and survival of many rain forest taxa in RFs of the Dahomey Gap.
139
Chapter 11: General discussion
These hypotheses need to be documented from palynologic records and
comprehensive floristic data from little disturbed sites. As RFs of the Dahomey Gap possess
a floristic richness lower than those of continuous tropical rain forests (see Table 11.3), and
sustain few typical rain forest species, they can not effectively act as refugia for many rain
forest taxa in the Dahomey Gap. Nevertheless, a significant number of species from
deciduous and semi-deciduous forest types can survive in RFs; therefore they could act as
refuge ecosystems for less luxuriant forest types taxa in the Dahomey Gap.
In addition to their importance in conserving biological diversity, RFs also provided
important ecosystems functions in geological times (Naiman et al. 1993) and may have
helped to maintain and create biodiversity in evolutionary time (Aide & Rivera 1998). RFs
are typically narrow, but can have high species diversity, and palaeoecological data suggest
that this diversity can be maintained for thousands of years, even during dry periods
(Kellman et al. 1996). This viewpoint is shared by Kingdon (1990) when he points out that
during the most arid periods, the riverine forest trailed over a much larger area than the
vestigial jungles of Biafra and Central African uplands. At such times, riverine galleries
became by far the most extensive microrefugia of all forest habitats in Africa. For Meave et
al. (1991), patches of RFs differ from neighbouring non forests plant communities in that
they are protected by a river, have high levels of humidity, and some occur within steep
ravines (see also RFs at hills feet in North Benin). These factors contribute to making RFs an
important habitat during periods of climate change. RFs are known to have played a crucial
role in maintaining biodiversity for more than 10,000 years in the past, so we expect that
these forests may be able to do the same in the future (Kellman et al. 1996). The mentioned
properties show the potential of riparian forests to act as refuge ecosystem during dry periods
in savanna as they are in rain forest regions.
140
Chapter 12
GENERAL CONCLUSION
TOWARDS A BETTER PRESERVATION AND REHABILITATION OF RIPARIAN
FORESTS IN BENIN
A.K. Natta
Chapter 12: General conclusion
Chapter 12
GENERAL CONCLUSION
TOWARDS A BETTER PRESERVATION AND REHABILITATION OF RIPARIAN
FORESTS IN BENIN
12.1. RIPARIAN FORESTS AS BIODIVERSITY VEGETATION HOTSPOTS IN BENIN
Hotspots of biodiversity are areas particularly rich in species, rare, endemic, threatened
species or some combination of these attributes (Reid 1998). In Benin, more than other
vegetation types, RFs can be considered as hotspots of biodiversity for several reasons:
- they are among the most vulnerable forest formations at all latitudes, yet of high ecological
importance.
- their high species richness on relatively small areas (patches and narrow corridors). We
collected 1/3 of the estimated number of species in the Benin flora in just 20 ha of RFs.
At this stage of the investigation on RFs diversity, we might say that 2.37 % of Benin’s
total area harbours at least 33 % of the estimated number of plant species in the country.
- their numerous species either rare, threatened, or with superior adaptability to a specific
habitat.
- they are habitat for an endemic plant species in a fire-prone environment. So far just a few
endemic plant species have been found in Benin.
- they are a vital ecosystem for numerous wild and bird life species.
- they protect many water resources all over the country.
Riparian ecosystems are essential multifunctional elements of the worldwide
ecological network (Mander et al. 1997), and the maintenance of riparian forests is
recognised as being vital to the integrity of waterways (Hancock et al. 1996). Given that RFs
can support high levels of biodiversity, and have played an important role during periods of
climate change over geological time (Kingdon 1990, Meave et al. 1991, Naiman et al. 1993,
Kellman & Tackaberry 1993, Meave & Kellman 1994, Kellman et al. 1994, Kellman et al.
1996, Aide & Rivera 1998), their conservation in Benin is of utmost importance with the
threat of a rapid increase in human occupation of land and possible change in climate.
12.2. LEGAL PROTECTION OF RIPARIAN FORESTS IN BENIN
The uniqueness, vulnerability and diversity of riparian forests are the most important criteria
to protect them. In Côte d’Ivoire, 29 out of 46 biological reserves are protected to preserve
genuine riparian forests, inter alia, because of their fragile soils in relationship with slope and
hydromorphy (Kadio 1999). The uniqueness of individual populations suggests that
conservation efforts directed only at species preservation can result in the loss of genetic
diversity (Aide & Rivera 1998). Therefore the preservation of endemic animal species (e.g.
Cercopithecus erythrogaster erythrogaster, a primate sub-species still found in Benin (Oates
1996, Grubb et al. 1999, Sinsin et al. 2002b), and the ornamental Thunbergia atacoriensis
(Acanthaceae), should be coupled with the protection of their habitats.
The distribution pattern of the endemic primate sub-species of Benin, suggest the
protection of all RFs of the Ouémé valley, which are among its most important habitat.
Likewise, RFs at Yarpao (Natitingou district) and many others in the Atacora hills should
have higher legal protection priority for their great plant diversity, and because they are
remnant habitats for Thunbergia atacoriensis. Likewise the boundaries of certain protected
areas (East of Kouffé Mountains; West of Kétou, Dogo and Ouémé-Boukou; and North East
143
Chapter 12: General conclusion
of the Pendjari Biosphere Reserve) that coincide with rivers should be extended to include
RFs at both riversides.
Legislation and policies for the protection or rehabilitation of RFs in Benin, like in
many developing countries, must take into account the economic and social costs as well as
the environmental benefits at various scales. Before improving the actual forest law, there is
an urgent requirement to increase enforcement of existing legislation, which is often
unknown or ignored by many stakeholders. As suggested by de Lima & Gascon (1999),
proactive educational initiatives are direly needed and should be coupled with legislation
enforcement to reduce deforestation of riverbanks and maintain corridors between
ecosystems in populated regions.
12.3. STRATEGIES FOR SUSTAINABLE MANAGEMENT OF RIPARIAN FORESTS
Cost-effective conservation of biodiversity requires that maximum biodiversity should be
protected in a minimum area (Williams et al. 1991). The current degraded status of many
riparian forests in Benin represents the cumulative and persistent impacts of unplanned and
uncontrolled land uses and poor management of RFs themselves but also of adjacent uplands.
Strategies that reflect a spectrum of goals linked to improving the ecological functions and
sustainability of riparian ecosystems (i.e. natural integrity and diversity of the system) are
needed. For example, strategies that focus on returning the hydrologic regime to a more
natural state have the greatest potential for restoring riparian vegetation, and forests in
particular that have evolved with and adapted to the patterns of changing flows in waterway
environments. As pointed out by Cordes et al. (1997), to maintain existing and restore
riparian woodland and to ensure their long-term regeneration, it has become important to
understand in some detail both the contemporary and historical relationships between riparian
vegetation and hydrological and geomorphological factors.
Also, management practices to maintain RF biodiversity need to be tailored to the
conditions of each particular area (Hancock et al. 1996), in particular one must take into
consideration the types of disturbances that typically affect these forests (Agee 1988).
Therefore, activities aiming at protecting and rehabilitating RFs in their watershed should be
implemented at the national, province, district and local levels, while taking into account
ecological benefits (i.e. physical variables) as well as social costs represented by the loss of
arable land (Cooper et al. 1987, Flanagan et al. 1989, Xiang 1993).
Because riparian forests perform a more than proportionate number of critical
biological and physical functions on a unit area basis, their restoration when degraded, and
protection of undisturbed stands can have a major influence on the local and regional
environment. Meanwhile, deciding on priorities between rehabilitation and preservation may
be difficult. Following Geraghty & Vollebergh (1993), we might say that it is more cost
effective initially to maintain systems that are in reasonable condition and build from them,
rather than at first attempting difficult rehabilitation tasks of very degraded river sections.
The preservation of intact riparian forest stands is vital because they represent
valuable reference sites for understanding the goals and efficacy of restoration efforts, and are
important sources of genetic material for the reintroduction of native species into degraded
areas. The restoration of RFs should refer firstly to the process of removing the degrading
factor or reducing its negative effects, and secondly repairing the condition and functioning
of degraded riparian vegetation. Ecological restoration of degraded RFs, in which
biodiversity is greatly decreased, has the stated goal to enhance the value of the ecosystem,
moderate degrading influences and regain pre-disturbance characteristics (NRC 2000). This
approach may include planting native trees, facilitating development of certain species,
preserving and rehabilitating threatened and endemic species. The problem of isolated
144
Chapter 12: General conclusion
populations could be partially solved by reforesting corridors between riparian forests patches
and belts along watercourses. Qualitative and quantitative knowledge of hydrology and
ecology (including the range of natural variability, disturbance regimes, soils and landforms,
and vegetation) are therefore required for successful rehabilitation of RFs.
12.4. CHALLENGES TO RIPARIAN FORESTS ASSESSMENT IN BENIN
As significant as climate changes are likely to be, land- and water-use changes have had and
will continue to have the greatest effects on riparian vegetation. Therefore, more qualitative
and quantitative assessments of the effects of land and water management on RFs have to be
undertaken. The main variables to be considered should be related to physical, biological and
anthropogenic factors. Research on RFs in Benin should document or clarify:
- ecological conditions (e.g. seed bank, germination, gap dynamics, edge effects, patterns in
the succession, and long-term survival of plant communities), geomorphic, as well as
hydrologic issues related to riparian systems;
- relationships between land forms, and RFs composition and structure dynamics;
- the effects of hydrologic (i.e. ground water table, surface water) changes on the
composition, physiognomy and productivity of RFs;
- relationships between RFs diversity and structure, and waterways order and size;
- the relationship between flooding, accumulation of organic debris, and severity of fire;
- the effects of dissolved nutrients, foliage, or woody debris from RFs, as food resources for
fishes, algae, and invertebrates;
- the influence of interactions of abiotic factors on relative growth and flood survivorship
rates of native RF species;
- the rate and pattern of tree mortality in RFs, and the role of subsequent delivery of wood to
waterways;
- the influence of fire on regeneration, recruitment patterns of plant species and RFs diversity;
- the potential role of such forest fragments as animal habitat or refuge, and movement
corridors throughout the landscape;
- the usefulness of direct gradient analysis based on environmental variables (e.g. soils
profile, structure, texture, fertility and nutrient content, water and flood related variables,
types and levels of disturbance) to explain RFs diversity and dynamics;
- the genetic diversity within and among little known, valuable or rare plant populations;
From the list of RFs species, a database of suitable riparian species needs to be
established to provide information on the type of plants and their ecological requirements at
various latitudes and river systems. Research should set up tolerances and thresholds at least
for the dominant RF trees regarding ground and surface waters, as a basis for predicting and
preventing adverse ecological effects from stream draining or severe drought. Likewise
dispersal and colonisation strategies of seeds, as well as ability of riverbank to trap
waterborne diaspores on riparian vegetation diversity should be investigated. Remote sensing
and geographical information systems are known to be powerful tools for assessing spatial
and temporal changes of natural resources in general, and riparian habitats in particular.
Effective management strategies must be devised to preserve and restore riparian
corridors (Brinson et al. 1981, Albernathy & Turner 1987, Gosselink & Lee 1989, Taylor et
al. 1990), while relying on accurate understanding of the structure and dynamics of RF
communities (Sagers & Lyon 1997). However, simply protecting RFs in a buffer zone may
not be adequate to ensure their existence in the long term (Hibbs & Bower 2001). Instead the
management of riparian forests must be a component (i.e. subset) of good watershed or
landscape management. So awareness has to be raised to various stakeholders on riparian
forests as unique physical and natural systems in their own right, and warranting special
145
Chapter 12: General conclusion
management and protection. Integrated management of riparian forests that optimise their
values as habitat for native plants and animals requires planning and acting, with all
stakeholders, at both site-specific and watershed levels. Through improved and long-term
scientific, educational and recreational programs, the ecological importance and intrinsic
human values associated with riparian forests may be better balanced against the competing
wants of unplanned urbanisation, short-term local inhabitants’ demands, and unknown
motivations of other stakeholders.
146
SUMMARY
A.K. Natta
Summary
SUMMARY
The present research deals with the flora, phytosociology and ecology of riparian forests in
Benin.
In chapter 1 (General introduction), the research background, objectives and approach for
riparian forests biodiversity assessment, and the organisation of the thesis are presented.
Chapter 2 introduces the study area which covered about 70 % of Benin, from 7° 10’ to 12°
20’ N.
Chapter 3 presents an overview of riparian forests biodiversity, their importance and the
threats they face making them endangered ecosystems. A definition of riparian forests (or
gallery forests) is given in the Benin context. The floristic characteristics of riparian forests in
each phytogeographic district are presented. Issues related to legal protection and
rehabilitation of the function and resources of riparian forests are documented: specifications
and weaknesses of the forest law regarding riparian forests are presented; challenges for
various stakeholders are discussed, and some improvements of the current forest law are
proposed.
Chapter 4 assesses plant species diversity, as well as species abundance models that best fit
representative collections of plant species of riparian forests throughout the country. This
study shows the richness and diversity of riparian forests in Benin, in comparison to other
vegetation types in this country. They harbour about 1/3 of the estimated total number of
plant species of the whole country in sample plots totalling 19 ha. This flora shares many
features with riparian forests and dense forests worldwide: e.g. most abundant families,
species richness/ha, trees species richness/ha, Shannon index, equitability index of Pielou,
and species abundance models. Endemism is very low compared to that in rain forests, what
is not surprising in the Dahomey Gap. The main conclusion is that relatively large numbers of
species are still maintained in small forest fragments along waterways. These remnants with
their specific plant species composition can be used for the restoration of degraded forest
stands.
Chapter 5 assesses the structure and ecological spectra of 19 ha of riparian forests through
selected parameters (e.g. life form, geographic affinity, diameter class distribution, basal area,
stem density, species dominance) that give a general picture of different vegetation types
present. Figures obtained for these parameters show that riparian forests in Benin are on the
one hand similar to many riparian forests in West Africa as well as in South and Central
America, and on the other hand to many tropical upland forests. A brief description of the
process of riparian forests degradation is also presented.
Chapter 6 deals with the phytosociological assessment of representative relevés of riparian
forests of Benin. Floristic ordination (DCA analysis) and classification (TWINSPAN) were
derived from a comprehensive floristic inventory of a data set of 818 plant species and 180
relevés. This yielded 12 plant communities or associations, most of which had not yet been
formally described:
1 - Community of Isolona thonneri and Callichilia barteri (10 relevés) along streams. This
community occurs at the lowest parts of the gallery forest with frequent inundation in the
centre of Pénéssoulou protected forest.
2 - Community of Motandra guineensis and Pararistolochia goldieana (24 relevés) along
streams at the East and West parts of Pénéssoulou reserve forest. This community is mainly
present on drained sites (i.e. seldom inundated).
3 - Community of Chrysobalanus icaco subsp. atacoriensis and Pentadesma butyracea (22
relevés) along streams at hill feet in the Atacora mountain chain.
4 - Community of Alchornea cordifolia and Ficus trichopoda (9 relevés) along streams on
regularly inundated plateaus all over the country.
149
Summary
5 - Community of Berlinia grandiflora and Khaya senegalensis (8 relevés) along streams on
drained plateaus (i.e. seldom inundated), mainly in the Sudanian region of the country.
6 - Community of Raphia sudanica and Oxytenanthera abyssinica (8 relevés) along streams
on drained plateaus, mainly in the Sudanian region.
7 - Community of Cynometra megalophylla and Parinari congensis (31 relevés) along the
Ouémé river in the Guinean region of Southern Benin.
8 - Community of Capparis thonningii and Crateva adansonii (30 relevés) along the Ouémé
river in the Sudano-Guinean region of Central Benin.
9 - Community of Lepisanthes senegalensis and Drypetes floribunda (17 relevés) along the
Ouémé river in the Sudano-Guinean region of Central Benin.
10 - Community of Uapaca heudelotii and Irvingia smithii (8 relevés) along the Sota river in
the Sudanian region of North East Benin.
11 - Community of Garcinia livingstonei and Combretum acutum (12 relevés) along the
Pendjari river in the Sudanian region of North West Benin.
12 - Community of Mimosa pigra and Ficus asperifolia (20 relevés) widely distributed on
sandy banks along rivers.
Ordination proved invaluable in the exploration of environmental characteristics of the
phytosociological groups. The environmental factors (waterways, relief, topography, latitude
and longitude) helped in the grouping of floristic relevés in the above mentioned 12 plant
communities. The distinguished plant communities were compared with syntaxonomic data
in literature. Riparian forests in Benin belong to the Mitragynetea Schmitz 1963, which is the
phytosociological class of hygrophile fresh water forests of tropical Africa. Based on
similarities of ecological conditions and floristic composition, the 12 plant communities can
be classified into 3 orders that are Alchornetalia cordifoliae Lebrun 1947, LanneoPseudospondietalia Lebrun & Gilbert 1954 and Pterygotetalia Lebrun & Gilbert 1954.
Chapter 7 presents the spatial distribution and ecological factors determining the occurrence
of Pentadesma butyracea (Clusiaceae), a rain forest and multipurpose species found in Benin
only along certain streams. Among the 224 tree species found along waterways, Pentadesma
is one of the least known, yet of great ecological and economic importance. Field survey
reveals the presence of this rain forest species in four non-contiguous remnant riparian areas,
some located far from its optimal ecological range. If urgent actions are not taken to protect
the remaining fragmented and dispersed riparian habitats, current human-induced disturbance
could result in the disappearance of this species in Benin.
Chapter 8 deals with the variation of the floristic composition, structural parameters (e.g.
abundance, average height, basal area, tree richness) and spatial distribution of tree species at
river edges across riparian forests. Horizontal and vertical structures of tree species exhibit
complex patterns at riverside. On the one hand, tree stems are characterised by an uneven
distribution across riparian forests, on the other hand height and basal area variations at
riverside do not show any easily interpretable patterns. The numerical analysis confirms a
gradual variation in the floristic composition across riparian forests and neighbouring plant
communities. These results suggest a partitioning of riparian forests in three habitats (i.e.
river front, middle and riparian forest edge). An implication for diversity assessment is that
plot size, shape and layout in the terrain should take into account the river front, the middle
and the edge of riparian forest. Due to the non-coverage of the whole riparian forest width
and unequal chance of species and stems to be sampled, circular and square plots are not
suitable for structural parameters and phytodiversity assessment in riparian forests. Instead
rectangular plots with varying length and width, and covering the whole cross section of
riparian forest are the most suitable sampling units under the study area conditions, and
probably for savanna regions too. The present study also provides scientific guidelines for an
150
Summary
improvement of the forest law regarding the distance to be protected at riverside, and
suggests 100 m instead of 25 m.
In chapter 9 the floristic composition, species richness and structure of two riparian
ecosystems in West Africa (the Comoé in Ivory Coast and the Ouémé in Benin), are
compared. The overall physiognomy of the two gallery forest sites seems similar and they
share the most prominent families. However, there are marked differences in terms of canopy
density and height, herb layer density, number of individuals, tree richness and diversity (H’),
and species composition. The phenomenon of single species dominance at both sites is
documented from Cynometra megalophylla, an evergreen tree species, which was time and
again the most frequent and dominant tree at both riversides and in the middle of the gallery
forests. Only detailed comparison shows the difference and complexity of ecological
processes between and within gallery forests sites.
The research carried out in chapter 10 facilitates the choice between several sampling
designs for the estimation of a population parameter for endangered species. This study was
carried out in the Pénéssoulou forest, in Central Benin. Stratified random sampling provided
the lowest variance, coefficient of variation, standard error and sampling error. This method
was taken as the most precise and reliable design over simple random and systematic
samplings for the density estimation of Khaya senegalensis and K. grandifoliola trees.
Results have confirmed empirical knowledge about the ecology of Khaya species and shown
that the selection of the most precise sampling design, with regards to estimating a given
parameter, can eventually be useful for the sustainable management of forest trees in the
study area. A reliable density estimate for Khaya species within the given vegetation types
facilitates the selection of areas to be protected and sustainably exploited.
Chapter 11 presents a general discussion on issues discussed in this thesis.
Sustainable rehabilitation and restoration of riparian forests biodiversity in Benin are
discussed in the general conclusion (chapter 12).
This study has provided detailed site-specific data on plant species that can serve for further
scientific research, as well as for conservation management and planning. It fills a gap of
knowledge on the flora of Benin, and can contribute to better land-use planning and
conservation of riparian forests.
151
RESUME
A.K. Natta
Résumés
RESUME
TITRE DE LA THESE
ETUDE ECOLOGIQUE DES FORETS GALERIES DU BENIN: PHYTODIVERSITE,
PHYTOSOCIOLOGIE ET DISTRIBUTION DES ESPECES LIGNEUSES.
Le chapitre 1 (Introduction générale) présente les fondements, objectifs et l’approche
suivie pour l’étude de la biodiversité des forêts galeries (i.e. riparian forests en Anglais) du
Bénin.
Le chapitre 2 (Milieu d’étude) introduit la zone d’étude qui couvre environ 70 % de la
surface totale du Bénin de 7° 10’ à 12° 20’ N. Tous les districts phytogéographiques de Bénin
reconnus par Adjanohoun et al. (1989), et Houinato et al. (2000) y ont été sillonnés et des
sites représentatifs installés.
Le chapitre 3 (Forêts galeries du Bénin: un écosystème unique mais en voie de
disparition) fait le point des connaissances actuelles sur la diversité de la flore, l’importance
écologique, économique et socioculturelle des forêts galeries du Bénin. Une définition des
forêts galeries dans le contexte du Bénin est proposée. Les causes de la dégradation actuelle
des forêts galeries sont présentées. Un accent particulier est mis sur la protection légale de ces
bandes boisées longeant les cours d’eau et les défis actuels pour la protection et
l’aménagement durable de leur biodiversité.
Le chapitre 4 (Diversité floristique des forêts galeries du Bénin) évalue la flore et la
phytodiversité de l’ensemble des 19 ha de forêts galeries étudiées de 1999 à 2002 au Bénin.
Au total quelque 1003 espèces ont été recensées, ce qui représente environ le 1/3 du nombre
total d’espèces estimé de la flore du Bénin. Ces espèces se regroupent en 120 familles et 513
genres. 224 espèces ligneuses ont été recensées. Les (sous) familles les plus représentées sont
les Papilionioideae, Poaceae, Rubiaceae et Euphobiaceae. Les forêts galeries sont
caractérisées par une flore typique adaptée aux inondations récurrentes et par une multitude
d’espèces typiques de forêt et savane de plateau. Selon les sites, l’indice de Shannon varie de
2,4 à 5,8 bits et l’Equitabilité de Pielou de 0,51 à 0,86. La richesse spécifique varie de 120 à
358 plantes par ha et celle des espèces ligneuses (dbh ≥ 10 cm) de 27 à 99 par ha. Cette étude
montre que dans la majorité des paysages savanicoles, les forêts galeries du Bénin constituent
des îlots de grande biodiversité qu’il faudrait restaurer et conserver durablement.
Le chapitre 5 (Structure et spectres écologiques des forêts galeries du Bénin) traite des
principales caractéristiques structurales de 19 ha de forêts galeries du Bénin, à travers : les
types biologiques, phytogéographiques, la distribution par classe de diamètre, la surface
terrière et la densité des ligneux. La physionomie des forêts galeries est très variable en terme
de stratification verticale et hauteur de la canopée, cependant le sous-bois est généralement
dense. Les forêts galeries du Bénin sont similaires à de nombreuses forêts galeries et forêts
semi-décidues d’Afrique de l’Ouest, concernant: les types phytogéographiques (forte
présence de espèces Guinéo-Congolaises), les types biologiques (forte abondance des
thérophytes et des micro et méso phanérophytes, faible abondance des méga-phanérophytes
et lianes ligneuses), la distribution par classe de diamètre (J renversé), la surface terrière (23
à 59 m3/ha) et la densité des ligneux (253 à 785 tiges/ha). La dominance d’une ou de
plusieurs espèces en terme d’abondance et de surface terrière est une caractéristique
155
Résumés
essentielle des forêts galeries du Bénin. Malheureusement les sites les plus denses et ayant
une structure assez développée sont assez rares de nos jours.
Le chapitre 6 (Phytosociologie des forêts galeries du Bénin) traite en large de la
classification et de l’ordination de relevés phytosociologiques les moins dégradés et les plus
représentatifs de l’ensemble du milieu d’étude. L’analyse multivariée de 180 relevés et 818
espèces a permis d’individualiser 12 associations qui s’ordonnent suivant cinq facteurs ou
gradients: l’importance du cours d’eau (rivière ou fleuve), le relief, la topographie, la latitude
et la longitude. Il s’agit des associations végétales à :
1 - Isolona thonneri et Callichilia barteri (10 relevés), le long de rivières dans le centre de la
forêt classée de Pénéssoulou, sur sol très inondable.
2 - Motandra guineensis et Pararistolochia goldieana (24 relevés), le long des rivières à l’Est
et à l’Ouest de la forêt classée de Pénéssoulou, sur sol rarement inondé (i.e. bon ressuyage du
sol).
3 - Chrysobalanus atacoriensis et Pentadesma butyracea (22 relevés), le long des rivières de
bas de collines.
4 - Alchornea cordifolia et Ficus trichopoda (9 relevés), le long des rivières de plateaux sur
substrats engorgés d’eau pendant la saison de pluie (mauvais ressuyage du sol).
5 - Berlinia grandiflora et Khaya senegalensis (8 relevés), le long des rivières de plateaux à
bon ressuyage du sol en milieu Soudanien.
6 - Raphia sudanica et Oxytenanthera abyssinica (8 relevés), le long des rivières de plateaux
à bon ressuyage du sol en milieu Soudanien.
7 - Cynometra megalophylla et Parinari congensis (31 relevés), le long du fleuve Ouémé en
région Guinéenne.
8 - Capparis thonningii et Crateva adansonii (30 relevés), le long du fleuve Ouémé en région
Soudano-Guinéenne.
9 - Lepisanthes senegalensis et Drypetes floribunda (17 relevés), le long du fleuve Ouémé en
région Soudano-Guinéenne.
10 - Uapaca heudelotii et Irvingia smithii (8 relevés), le long de la Sota dans le Nord Est du
Bénin.
11 - Garcinia livingstonei et Combretum acutum (12 relevés), le long de la Pendjari dans le
Nord Ouest du Bénin.
12 - Mimosa pigra et Ficus asperifolia (20 relevés), sur les bancs de sable proche du lit des
cours d’eaux.
Une classification syntaxonomique est aussi présentée. Les forêts galeries du Bénin sont
rangées dans la classe des Mitragynetea Schmitz 1963 qui regroupe toutes les forêts
édaphiques hygrophiles d’eau douce d’Afrique tropicale. Les associations individualisées se
regroupent en trois ordres (Alchornetalia cordifoliae Lebrun 1947, LanneoPseudospondietalia Lebrun & Gilbert 1954, et Pterygotetalia Lebrun & Gilbert 1954). Trois
associations sur les douze ont pu être classifiées dans des Alliances préexistantes. L’approche
phytosociologique de Braun-Blanquet couplée aux analyses multivariées se sont avérées
concluantes quant à l’individualisation des groupements végétaux de forêts galeries et la
détection de gradients / facteurs majeurs de cette typologie.
Le chapitre 7 (Distribution spatiale et facteurs écologiques de la présence de Pentadesma
butyracea (Clusiaceae) au Bénin), traite de la répartition spatiale d’une espèce de forêt
sempervirente qui se retrouvent curieusement au Bénin uniquement dans les forêts galeries en
milieux Soudano-Guinéen (région de Bassila et près du village Agbassa, commune de
Tchaourou) et Soudanien (région de Perma à Tandafa et près du village de Gbèssè, commune
de Ségbana). Pentadesma est une espèce relativement peu connue du monde scientifique
156
Résumés
béninois, pourtant elle est à but multiple, et pourrait jouer un rôle de premier choix dans la
diversification des filières agro-forestières. La présence de cette espèce de forêt dense
sempervirente dans les forêts galeries du Bénin dans la zone Soudano-Guinéenne, et aussi
loin qu’à 11° N, nous a permis de valider quelques hypothèses concernant l’importance
écologique de ces formations édaphiques d’eau douce.
Le chapitre 8 (Variation de la structure et de la composition floristique dans la direction
perpendiculaire aux fleuves) fait usage de l’analyse multivariée (DCA et TWINSPAN) et
des courbes de distribution des ligneux pour tester l’existence d’espèces typiques du front
d’eau, du milieu et de lisière des galeries. L’abondance des ligneux décroît constamment du
lit du cours d’eau vers la lisière de la galerie. Cependant la hauteur des arbres et la surface
terrière varient très peu lorsqu’on s’éloigne du cours d’eau. Quant à la composition
floristique, elle varie graduellement dans la galerie perpendiculairement au cours d’eau vers
les groupements végétaux adjacents aux galeries. Ces résultats nous ont permis de définir des
espèces du front des fleuves, du milieu des galeries, de lisière de galeries, et des espèces
typiques des groupements végétaux contigus aux galeries.
Les espèces typiques de galeries constamment au contact de l’eau des fleuves sont: Syzygium
guineense, Pterocarpus santalinoides, Parinari congensis, Cola laurifolia et Napoleonaea
vogelii;
Les espèces constamment au centre des galeries sont: Cynometra megalophylla, Drypetes
floribunda et Manilkara multinervis;
Les espèces de bordure de galerie ou dans l’écotone avec les groupements végétaux adjacents
aux galeries sont: Dialium guineense, Diospyros mespiliformis, Elaeis guineensis, Cola
gigantea et Ceiba pentandra.
Cette recherche confirme le savoir empirique sur l’écologie des arbres en bordure des cours
d’eau, mais montre que, bien que linéaires et fragmentées, les galeries ne peuvent être
considérées comme uniformes sur les plans floristique et structural dans la direction
perpendiculaire aux cours d’eaux. Une conséquence de ces résultats est que les placeaux
rectangulaires avec des longueurs et largueurs variables et qui couvrent toute la section
perpendiculaire des galeries sont les mieux adaptés en milieux Soudanien et SoudanoGuinéen. Aussi la distance légale minimale à être protégée de part et d’autre des cours d’eau
(i.e. 25 m) apparaît-elle trop faible. Les résultats obtenus suggèrent de porter cette distance à
100 m, cela permettra aux galeries forestières de mieux assurer leurs importantes fonctions.
Le chapitre 9 (Comparaison de l’interface forêt galerie/ savanes le long de l’Ouémé
(Bénin) et de la Comoé (Côte d’Ivoire)), traite des relations forêt galerie / savane dans deux
systèmes riverains en milieu Soudano-Guinéen. Les deux sites comparés sont
approximativement à la même latitude, mais distants d’au moins 600 km à vol d’oiseau au
Bénin (Idadjo-Bétérou) et dans le Sud Ouest du Parc National de la Comoé en Côte d’Ivoire.
On constate une similarité concernant les familles ayant une grande richesse spécifique (i.e.
Leguminosae, Rubiaceae et Sapotaceae) mais aussi des différences notables (e.g. densité des
espèces de la canopée, hauteur moyenne, densité du sous-bois, richesse spécifique et
composition floristique). Cynometra megalophylla est dans les deux sites l’espèce la plus
fréquente et dominante. C’est un exemple typique de dominance mono-spécifique dans les
forêts galeries en Afrique de l’Ouest. Une analyse plus fine des interactions entre les trois
portions des galeries (front du cours d’eau, milieu de la galerie et l’écotone) montre la
variabilité et la complexité des phénomènes écologiques entre sites et au sein d’un même site.
Le chapitre 10 (Comparaison de trois méthodes d’échantillonnage pour l’estimation de
la densité de Khaya senegalensis et K. grandifoliola dans la forêt classée de Pénéssoulou)
157
Résumés
évalue les performances relatives des systèmes d’échantillonnage aléatoire, systématique et
stratifié pour l’obtention des valeurs crédibles et précises (en terme statistique) des densités
des deux espèces de Khaya présentent au Bénin. Il est généralement admis que l’obtention
des valeurs précises pour des paramètres d’espèces en voie de disparition est une nécessité
pour tout aménagement durable. Ainsi nous avons testé dans la partie la plus dense et la
moins perturbée de la forêt classée de Pénéssoulou, les trois méthodes d’échantillonnage
conventionnelles. L’échantillonnage stratifié, avec allocation proportionnelle de placeaux,
est le plus fiable et précis, puisqu’il donne les valeurs les plus faibles de la variance, du ratio
de variance, de l’erreur standard, et du coefficient de variation. Cette approche pourrait être
testée pour d’autres espèces de valeur dans la même zone d’étude. Cela permettrait d’avoir
des résultats consistants pour plusieurs espèces dans la zone d’étude.
Une discussion générale de tous les aspects abordés dans la présente thèse est présentée au
chapitre 11. La conclusion générale (Chapitre 12) traite des stratégies les plus adaptées et
les nouveaux défis pour une meilleure conservation de la biodiversité des forêts galeries au
Bénin.
158
SAMENVATTING
A.K. Natta
Samenvatting
SAMENVATTING
Dit onderzoek behandelt de flora, de vegetatiekundige aspecten en de ecologie van
oeverbossen in Benin.
In Hoofdstuk 1 (Algemene Inleiding) worden de achtergronden van het onderzoek, de
doelstellingen en benaderingen om de biodiversiteit van oeverbossen te beoordelen, evenals
de indeling van dit proefschrift beschreven.
Hoofdstuk 2 beschrijft het gebied in Benin waar het onderzoek plaatsvond. Het studiegebied
beslaat ca. 70% van het land en is gelegen tussen 7010 en 120 20’ Noorderbreedte.
Hoofdstuk 3 geeft een overzicht van de biodiversiteit in oeverbossen, het belang daarvan en
de aantastingen waaraan ze blootstaan. Dit ecosysteem moet in Benin als bedreigd worden
beschouwd. Een definitie van oever- (of galerij-)bossen is gegeven in de Beninese context.
De floristische eigenschappen van oeverbossen worden vermeld in elk fytogeografisch
district. Vraagstukken i.v.m. de wettige bescherming en herstel van de functie en bronnen van
het oeverbos komen aan de orde; uitdagingen voor diverse betrokkenen worden besproken,
en verbeteringen voor de huidige Boswet worden voorgesteld.
Hoofdstuk 4 beschrijft de soortenrijkdom, en de op de representatieve verzamelingen uit
oeverbossen uit de meeste delen van het land best passende abundantie-modellen. Dit
onderzoek laat zien hoe rijk en divers oeverbossen zijn in Benin, in vergelijking met
andersoortige bossen in het land. Oeverbossen herbergen ongeveer 1/3 van het aantal soorten
van de gehele flora van Benin, zoals aangetoond in vele proefvlakken die bij elkaar 19 ha
besloegen. De oeverbosflora heeft veel gemeen met andere oeverbossen en dichte
laaglandregenbossen in de wereld: bijv. de soortenrijkste plantenfamilies, soortenrijkdom per
ha, boomsoortenrijkdom per ha, Shannon index, de Equitability Index van Pielou, en
abundantie modellen. In de oeverbossen is het endemisme veel lager dan in dichtere
regenwouden, dat is echter niet verrassend in de Dahomey Gap. De belangrijkste conclusie is
dat relatief veel soorten zich nog steeds handhaven in kleine bosfragmenten langs
waterwegen. Deze bosfragmenten en bepaalde soorten kunnen worden gebruikt als
uitgangsmateriaal voor het herstel van gedegenereerde vegetaties.
Hoofdstuk 5 bekijkt de structuur en ecologische spectra van 19 ha oeverbos door middel van
geselecteerde parameters (levensvorm, geografische verbanden, verdeling van diameter
klassen, basis oppervlak, dichtheid van stammen, dominantie van soorten) welke een
overzicht leveren van vegetatietypes. Cijfers verkregen voor deze parameters bewijzen dat
oeverbossen in Benin vergelijkbaar zijn met vele oeverbossen in West Afrika en Zuid- en
Centraal Amerika, maar ook met vele tropische bossen op drogere gronden. Het
degradatieproces van oeverbossen wordt kort beschreven.
Hoofdstuk 6 behandelt de vegetatiekundige opnamen in representatieve proefvlakken in
Beninese oeverbossen. Ordinatie (DCA analyse) en classificatie (TWINSPAN) werden
toegepast op 180 proefvlakken en 818 plantensoorten. Dit leverde liefst 12
plantengemeenschappen op, waarvan de meeste nog niet eerder formeel zijn beschreven.
1 - Isolona thonneri en Callichilia barteri gemeenschap (10 proefvlakken) langs stroompjes
in het midden van het beschermde bos van Pénéssoulou. Deze associatie vindt men in de
laagstgelegen delen van het bos welke vaak overstroomd worden.
2 - Motandra guineensis en Pararistolochia goldieana gemeenschap (24 proefvlakken) langs
riviertjes in het oosten en het westen van het Pénéssoulou bos. Deze associatie komt voor op
goed ontwaterde plekken welke zelden overstromen.
3 - Chrysobalanus icaco subsp. atacoriensis en Pentadesma butyracea gemeenschap (22
proefvlakken) langs stroompjes aan de voet van de Atacora heuvels.
4 - Alchornea cordifolia en Ficus trichopoda gemeenschap (9 proefvlakken) langs riviertjes
op regelmatig overstroomde plateaus door het hele land.
161
Samenvatting
5 - Gemeenschap van Berlinia grandiflora en Khaya senegalensis (8 proefvlakken) langs
stroompjes op goed ontwaterde plateaus, vooral in de Sudan zone van het land.
6 - Gemeenschap van Raphia sudanica en Oxytenanthera abyssinica (8 proefvlakken), ook
langs stroompjes op goed gedraineerde plateaus in de Sudan zone.
7 - Gemeenschap van Cynometra megalophylla en Parinari congensis (31 proefvlakken)
langs de Ouémé rivier in de Guinee-zone van Zuid Benin.
8 - Gemeenschap van Capparis thonningii en Crateva adansonii (30 proefvlakken) langs de
Ouémé rivier in de Sudano-Guinee zone van midden-Benin.
9 - Gemeenschap van Lepisanthes senegalensis en Drypetes floribunda (17 proefvlakken)
langs de Ouémé rivier in de Sudano-Guinee zone van midden-Benin.
10 - Gemeenschap van Uapaca heudelotii en Irvingia smithii (8 proefvlakken) langs de Sota
rivier in het noordoosten van het land.
11 - Gemeenschap van Garcinia livingstonei en Combretum acutum (12 proefvlakken) langs
de Pendjari rivier in het noordwesten van Benin.
12 - Mimosa pigra en Ficus asperifolia gemeenschap (20 proefvlakken) zeer wijdverspreid
op zandbanken langs rivieren.
Ordinatie blijkt waardevol om de eigenschappen van het milieu van de fytosociologische
groepen te onderzoeken. De gevonden factoren (aard van watergangen, reliëf, topografie,
lengte- en breedtegraad) hielpen bij het groeperen van de floristische opnamen in de
bovengenoemde 12 gemeenschappen. Een syntaxonomische indeling van de onderscheiden
plantengemeenschappen wordt gegeven. Oeverbossen in Benin behoren tot het Mitragynetea
Schmitz 1963, de fytosociologische klasse van vochtminnende zoetwaterbossen van tropisch
Afrika. Gebaseerd op overeenkomsten van ecologische gesteldheid en floristische
samenstelling deelden wij de 12 plantengemeenschappen in 3 ordes in: Alchornetalia
cordifoliae Lebrun 1947, Lanneo-Pseudospondietalia Lebrun & Gilbert 1954 en
Pterygotetalia Lebrun & Gilbert 1954.
Hoofdstuk 7 onderzoekt de ruimtelijke verspreding en de ecologische factoren die het
voorkomen van Pentadesma butyracea (Clusiaceae) bepalen. Deze op diverse manieren
bruikbare soort uit het regenbos komt in Benin alleen langs sommige riviertjes voor. Onder
de 224 boomsoorten gevonden langs grote en kleine stromen, is Pentadesma een van de
minst bekende, ofschoon deze van groot ecologisch en economisch belang is. Veldstudies
toonden aan dat deze regenwoudsoort in vier disjuncte relictgebiedjes voorkomt, sommige
ver verwijderd van de optimale ecologische gebieden. De huidige verstoring door de mens
kan leiden tot de verdwijning van de soort in Benin tenzij onmiddellijk actie wordt
ondernomen de laatste versnipperde habitats, de oeverbossen, te beschermen.
Hoofdstuk 8 behandelt de variatie in de floristische samenstelling, structurele parameters
(bijv. abundantie, gemiddelde hoogte, basisoppervlakte, rijkdom aan bomen) en ruimtelijke
verdeling van boomsoorten in oeverbossen. Horizontale en verticale structuren laten
complexe patronen zien. Enerzijds wordt het aantal stammen gekarakteriseerd door een
ongelijke verdeling in het oeverbos dwars op de richting van de stroom, anderzijds zijn de
variaties van boomhoogte en basisoppervlakten langs de stromen niet eenvoudig te
interpreteren. De numerieke analyse bevestigt een geleidelijke variatie van de floristische
samenstelling vanaf de rivier, in het midden van het oeverbos, tot bij de naburige vegetatie.
De resultaten wijzen op een onderverdeling van het oeverbos in deze drie genoemde habitats
langs de horizontale gradiënt. Bij onderzoek aan diversiteit van oeverbossen, de omvang,
vorm en plaatsing van proefvlakken in het terrein moeten steeds de oevers, het middenstuk en
de buitenrand in beschouwing dienen genomen te worden. Ronde en vierkantige
proefvlakken zijn ongeschikt om structurele parameters en biodiversiteit te meten, omdat
hiermee niet de gehele breedte van het oeverbos wordt bemonsterd. Rechthoekige
proefvlakken van variabele lengte en breedte lijken beter geschikt te zijn voor het
162
Samenvatting
studiegebied, en vermoedelijk ook voor savannegebieden. Deze studie levert ook
wetenschappelijke onderbouwing ter verbetering van de Boswet wat betreft de te beschermen
breedte langs de rivieren, en stelt 100 m voor in plaats van 25 m.
In Hoofdstuk 9 worden de floristische samenstelling, soortenrijkdom en structuur van twee
oeverbossystemen in West Afrika (de Comoé in Ivoorkust en de Ouémé in Benin)
vergeleken. Hoewel de fysiognomie van de twee bossen overeen lijken te komen en dezelfde
plantenfamilies er prominent in voorkomen, zijn er duidelijke verschillen in dichtheid en
hoogte van het kronendak, dichtheid van de kruidlaag, aantal exemplaren, rijkdom aan bomen
en boomsoorten (H’) en soortensamenstelling. Het verschijnsel dat één soort dominant
voorkomt is gedocumenteerd aan de hand van Cynometra megalophylla, een altijdgroene
boomsoort, welke steeds de meest algemene de dominante soort is, zowel langs de rivier als
in het midden van het oeverbos, in beide landen. Alleen gedetailleerde vergelijkingen maken
de verschillen en complexiteit van de ecologische processen tussen en binnen oeverbossen
duidelijk.
Het onderzoek zoals gerapporteerd in hoofdstuk 10 vergemakkelijkt een keuze tussen
verschillende bemonsteringsmethoden om populatieparameters van bedreigde soorten in te
schatten. In het bos van Pénéssoulou bleek stratified random sampling de laagste variantie,
variatiecoëfficient, standaardafwijking en monsterfouten op te leveren. Deze methode werd
dus gekozen als de meest nauwkeurige in vergelijking met simple random en systematic
sampling, om de dichtheid te meten van Khaya senegalensis en K. grandifoliola bomen. De
resultaten bevestigen de empirische kennis van de ecologie van Khaya soorten en lieten zien
dat de keuze van de meest nauwkeurige monstermethode, om bepaalde parameters te
schatten, uiteindelijk nuttig kan zijn bosbomen in het studiegebied duurzaam te beheren. Een
betrouwbare schatting van Khaya in de betreffende vegetatietypes vergemakkelijkt de keuze
van te beschermen of duurzaam te exploiteren arealen.
Hoofdstuk 11 bevat een algemene bespreking over de problemen zoals gegeven in het
proefschrift als geheel. Duurzaam herstel van oeverbosdiversiteit in Benin wordt
gepresenteerd in de algemene conclusie (hoofdstuk 12).
Dit onderzoek levert nauwkeurige terrein-specifieke gegevens die kunnen dienen als
uitgangspunt voor verder wetenschappelijke studie, en ook voor beheer en planning van
behoudsmaatregelen. Het draagt bij aan de leemten in de kennis van de flora van Benin, en
wij verwachten dat het bijdraagt aan verbetering van het behoud van oeverbossen en een
verhoging van de prioriteit daarvan in landgebruiksplanning.
163
REFERENCES
A.K. Natta
References
REFERENCES
Abbiw D.K. 1990. Useful plants of Ghana: West African uses of wild and cultivated plants.
Intermediate Technology Publications and the Royal Botanic Gardens, Kew, London. 337p.
Acharya B. 1999. Forest biodiversity assessment: A spatial analysis of tree species diversity in Nepal.
PhD Thesis. ITC Forest Division, ITC Publication 72, Enschede, The Netherlands. 199 p.
Acker A.S., Lienkaemper G., McKee W.A., Swanson F.J. & Miller S.D. 2003. Composition,
complexity, and tree mortality in riparian forests in the central Western Cascades of Oregon.
Forest Ecology and Management 173: 293-308.
Adambiokou-Akakpo J. 2001. Gestion des terroirs et planification de l’utilisation des ressources
naturelles dans la commune de Klouékanmè. Mémoire DEA, Option Gestion de
l’environnement. FLASH-UAC, Abomey-Calavi, Bénin. 96p.
Adjakidjè V. 1984. Contribution à l’étude botanique des savanes guinéennes en République Populaire
du Bénin. Thèse de 3ème cycle, Université de Bordeaux III, Bordeaux, France. 245p.
Adjanohoun E. 1965. Végétation des savanes et des rochers découverts en Côte d’Ivoire Centrale.
Mémoire n° 7. ORSTOM, Paris. 178p.
Adjanohoun E. 1966. Existence au Dahomey de savanes côtières analogues à celles de Côte d’Ivoire.
Etudes dahoméennes 6-7: 48-71.
Adjanohoun, E. & Aké Assi L. 1967. Inventaire floristique des forêts claires soudaniennes et subsoudaniennes en Côte d’Ivoire septentrionale. Annales de l’Université d’Abidjan 3: 89-141.
Adjanohoun E. 1968. Le Dahomey. Acta Phytogeographica Suecica 54: 86-91.
Adjanohoun E. J., Adjakidjè V., Ahyi M. R. A., Aké Assi L., Akoègninou A., Akpovo F., d’Almeida
J., Gbaguidi N., Boukef K., Chadaré F., Cusset G., Dramane K., Eyme J., Gassita J-N.,
Goudoté E., Guinko S., Houngnon P., Issa Lo, Keita A., Kiniffo H. V., Koné Bamba D.,
Musampa Nseyya A., Saadou N., Sodogandji Th., de Souza S., Tchabi A., Zinsou Dossa C.
and Zohoun Th. 1989. Contribution aux études ethnobotaniques et floristiques en République
Populaire du Bénin. Médecine traditionnelle et pharmacopée. ACCT, Paris. 895p.
Adomako D. 1977. Fatty acid composition and characteristics of Pentadesma butyracea fat extracted
from Ghana seeds. Journal of the Science of Food and Agriculture 28 (4): 384 - 386.
Agbani P.O. 2002. Etudes phytosociologiques des groupements forestiers par bandes longitudinales à
grandes échelles: Cas du noyau central de la forêt dense semi-décidue de la Lama au Bénin.
Mémoire DEA/FLASH/UAC, Abomey-Calavi, Bénin. 73 p + annexes.
Agee J.K. 1988. Successional dynamics in forest riparian zones. In: Raedeke K. (ed.), Streamside
management: Riparian, Wildlife and Forestry Interactions. Institute of Forest resources,
Contribution 59, University of Washington, Seattle, WA. pp 31-44.
Aide M.T. & Rivera E. 1998. Geographic patterns of genetic diversity in Poulsenia armata
(Moraceae): implications for the theory of Pleistocene refugia and the importance of riparian
forest. Journal of Biogeography 25: 695-705.
Akoègninou A. 1984. Contribution à l’étude botanique des îlots de forêts denses humides semidécidues en République Populaire du Bénin. Thèse de 3ème cycle, Université de Bordeaux
III, Bordeaux, France. 181p.
Akoègninou A., Houndagba J.C., Tossou M.G., Essou J.P. & Akpagana K. 2001. La végétation d’une
zone de transition entre la forêt dense humide semi-décidue et les savanes: la région de Bantè
(Bénin, Afrique de l’Ouest). Journal Botanique de la Société Botanique de France 15: 99108.
Albernathy Y. & Turner R.E. 1987. United States Forested wetlands: 1940-1980. BioScience 37: 721727.
167
References
Ambus P. & Lowrance R. 1991. Comparison of denitrification in two riparian soils. Soil Science
Society of America Journal 55: 994-997.
Ambus P. & Christensen S. 1993. Denitrification variability and control in a riparian fen irrigated
with agricultural dainage water. Soil Biology and Biochemistry 25: 915-923.
Amétépé A.M. 1997. Forêts sacrées et conservation de la biodiversité au Bénin: Cas du département
du Mono. Mémoire d’Ingénieur Agronome. FSA/UAC, Abomey-Calavi, Bénin. 172p.
Anderson B.W. & Ohmart R.D. 1977. Vegetation structure and birduse in the lower Colorado River
valley. Importance, preservation, and management of riparian habitat: a symposium, USDA
Forest Service General Technical Report RM – 43, Rocky Mountain Forest and Range
Experiment Station, Fort Collins, Colorado. pp 23-34.
Andersson E., Nilsson C. & Johansson E. 2000. Plant dispersal in boreal rivers and its relation to the
diversity of riparian flora. Journal of Biogeography 27: 1095-1106.
Arouna O. 2002. L’exploitation des ressources biologiques et la dynamique de la forêt classée de
l’Alibori supérieur (secteur de Bagou). Mémoire de Maîtrise en Géographie. FLASH-UNB,
Abomey-Calavi, Bénin. 114p.
Auble G.T., Scott M.L., Friedman J.M., Back J. & Lee V.J. 1997. Contraints on establishement of
plains cottonwood in an urban riparian preserve. Wetlands 17: 138-148.
Aubreville A. 1950. Flore forestière soudano-guinéenne AOF – Cameroun – AEF. Office de la
Recherche Scientifique Outre-Mer. Société d’Editions Géographiques Maritimes et
Coloniales, Paris. 523p.
Aubreville A. 1959. La flore forestière de la Côte d’Ivoire. Tome 2, 2ème edition. Publication CTFT,
Paris. 332 p.
Baker W.L. 1989. Macro- and micro-scale influences on riparian vegetation in Western Colorado.
Annals of the Association of American Geographers 79: 65-78.
Baker W.L. 1990. Species richness of Colorado riparian vegetation. Journal of Vegetation Science 1:
119-124.
Bamps P. 1971. Répartition de Pentadesma butyracea en Afrique. Bulletin du Jardin Botanique
National de Belgique 41: 430.
Barton D.R., Taylor W.D. & Biette R.M. 1985. Dimensions of riparian buffer strips required to
maintain trout habitat in southern Ontario streams. North American Journal of Fisheries
Management 5: 364-378.
Baumer M. 1995. Arbres, arbustres et arbrisseaux nourriciers en Afrique Occidentale. ACCT, Paris.
260 p.
Begon M., Harper J.L. & Townsend C.R. 1986. Ecology, populations and communities. Blackwell
Scientific Publication, Oxford. 876p.
Bélem O.M. & Guinko S. 1998. La biodiversité des galeries forestières du Burkina Faso: cas de la
réserve de la biosphère de la mare aux hippopotames. In: Bâ A.T., Madsen J.E. & Sambou B.
(eds.), Atelier sur flore, végétation et Biodiversité au Sahel. Aarhus University Press, Aarhus
Denmark, AAU Reports 39: 75-86.
Bell D.T. & Sipp S.K. 1975. The litter stratum in a streamside forest ecosystem. Oikos 26: 391-397.
Bell D.T., Johnson F.L. & Gilmore A.R. 1978. Dynamics of litter fall, decomposition, and
incorporation in the streamside forest ecosystems. Oikos 30: 76-82.
Bendix J. 1994. Among-site variation in riparian vegetation in the Southern California Transverse
Ranges. American Midland Naturalist 132: 136-151.
Bennett A.F. 1999. Linkages in the landscape: the role of corridors and connectivity in wildlife
conservation. IUCN, Gland, Switzerland and Cambridge, UK. 254p.
Berhaut J. 1967. Flore du Sénégal, 2ème édition. Editions Clairafrique, Dakar, Sénégal. 485p.
168
References
Berhaut J. 1979. Flore illustrée du Sénégal. Tome IV. Edition Clairafrique, Dakar, Sénégal. 329p.
Bersier L-F. & Meyer D. 1994. Bird assemblages in mosaic forests: the relative importance of
vegetation structure and floristic composition along the successional gradient. Acta
Oecologica 15 (5): 561-576.
Best L.B., Stauffer D.F. & Geier A.R. 1979. Evaluating the effects of habitat alterations on birds and
small mammalls occupying riparian communities. In: Johnson R.R. & McCormick J.R. (eds.),
Strategies for protection and management of floodplain wetlands and other riparian
ecosystems. USDA Forest Service General Technical Report WO-12. Washington. D.C. pp
117-124.
Billy R.E. & Ward J.W. 1989. Changes in characteristics and functions of woody debris with
increasing size of streams in western Washington. Trans. Am. Fish. Soc. 118: 368-378.
Bocard D. 1992. Partialling out the spatial component of ecological variation. Ecology 73 (3): 10451055.
Bongers F., Popma J., Meave del Castillo J. & Carabias J. 1988. Structure and floristic composition of
the lowland rain forest of Los Tuxtlas, Mexico. Vegetatio 74:55-80.
Bonkoungou E.G. 1984. Inventaire et analyse biogégraphique de la flore des forêts galeries de la
Volta Noire en Haute Volta. Not. Doc.Volt. 15: 64-73.
Bossou B. 2001. Contribution à l’aménagement de la forêt classée de Dogo-Kétou : Structure et
dynamique des différents groupements végétaux et périodicité de coupe. Mémoire de 3ème
cycle DESS/FSA/UAC, Abomey-Calavi, Bénin. 141p + annexes.
Bowden W.B., McDowell W.H., Asbury C.E. & Finley A.M. 1992. Riparian nitrogen dynamics in
two geomorphological distinct tropical rain forest watersheds: nitrous oxide fluxes.
Biogeochemistry 18: 77-99.
Braatne J.H., Rood S.B. & Heilman P.E. 1996. Life history, ecology, and conservation of riparian
cottonwoods in North America. In: Settler R.F., Bradshaw Jr. H.D., Heilman P.E. & Hinckley
T.M. (eds.), Biology of Populus and its implications for Management and Conservation. NRC
Press, Ottawa, Ontario. pp 57-85.
Bragg D.C. & Kershner J.L. 1999. Coarse woody debris in riparian zones. Journal of Forestry 97(4):
30-35.
Braun-Blanquet J. 1972. Plant sociology: The study of plant communities (Facsimile of the edition of
1932, translated by Fuller G.D. & Conard H.S.). Hafner Publishing Company, New York. 439
p.
Brierley G.J. & Hickin E.J. 1985. The downstream gradation of particle sizes in the Squamish River,
Bristish Columbia. Earth Surface Processes and Landforms 10: 597-606.
Brinson M.M. 1990. Riparian forests. In: Lugo A.E., Brinson M. & Brown S. (eds.), Forested
Wetlands. Ecosystems of the world 15. Elsevier, Amsterdam. pp 87-141.
Brinson M.M., Swift B.L., Plantico R.C. & Barclay J.S. 1981. Riparian ecosystems: their ecology and
status. U.S. Fish and Wildlife Service, Biological Services Program, FWS/OBS - 81-17,
Washington D.C.
Brinson M.M. & Verhoeven J. 1999. Riparian forests. In: Hunter M.L.Jr. (ed.), Maintaining
Biodiversity in Forest Ecosystems. Cambridge University Press, Cambridge. pp 265-299.
Brunel J. F., Hiepko P. & Scholz H. (eds.) 1984. Flore analytique du Togo - Phanerogames. Englera
4. GTZ, Eschborn, Berlin. 751p.
Buckland S.T., Laake J.L., Anderson D.R. & Burnham K.P. 1993. Distance sampling. Estimating
abundance of biological populations. First edition. Chapman and Hall, London. 441p.
Bull E. & Skovlin J.M. 1982. Relationships between avifauna and streamside vegetation. 47th North
American Wildlife Conservation 47: 496-505.
169
References
Burbrink F.T., Phillips C.A. & Heske E.J. 1998. A riparian zone in Southern Illinois as a potential
dispersal corridor for reptiles and amphibians. Biological Conservation 86: 107-115.
Burke M.K., Lockaby B.G. & Conner W.H. 1999. Above ground production and nutrient circulation
along a flooding gradient in a South Carolina Coastal Plain forest. Canadian Journal of
Forest Resources 29: 1402-1418.
Busch D.E. 1995. Effects of fire on south-western riparian plant community structure. Southwestern
Naturalist 40: 259-267.
Busch D.E. & Smith S.D. 1993. Effects of fire on water and salinity relations of riparian forests
woody taxa. Oecologia 94: 186-194.
Busch D.E. & Smith S.D. 1995. Mechanisms associated with decline of woody species in riparian
ecosystems of the southwestern US. Ecological Monographs 65(3): 347-370.
Carrol R., Augspurger C., Dobson A., Franklin J., Orians G., Reid W., Tracy R., Wilcove D. &
Wilson D. 1996. Strengthening the use of science in achieving the goals of the endangered
species act: An assessment by the ecological society of America. Ecological Applications 6
(1): 1-11.
CENATEL (Centre National de Télédétection et de Surveillance du couvert forestier) 1995. Carte de
végétation du Bénin. DFRN/MDR. Cotonou, Bénin. 15p.
CENATEL 2000. The study on cartography, inventory and management of classified forest in
Northern area in Benin. Final Report, Main text. Japan International Cooperation Agency
(JICA), DFRN-MDR. Cotonou, Benin. 245p.
CENATEL 2001. Résultats statistiques de la carte de végétation du Bénin. DFRN/MAEP. Cotonou,
Bénin.
Chestnut T.J. & McDowell W.H. 2000. C and N dynamics in the riparian and hyporheic zones of a
tropical stream, Luquillo Mountains, Puerto Rico. Journal of the North American
Benthological Society 19: 199-214.
Cochran W.G. 1977. Sampling techniques. 3rd edition. John Wiley, New York. 428p.
Colinvaux P. 1987. Amazon diversity in light of the paleoecological record. Quaternary Science
Reviews 6: 93-114.
Cooper J.R., Gilliam J.W., Daniels R.B., Ponarge W.P. 1987. Riparian areas as filters for agricultural
sediments. Soil Science Society of America Journal 51: 416-420.
Cordes L.D., Hughes F.M.R. & Getty M. 1997. Factors affecting the regeneration and distribution of
riparian woodlands along a northern prairie river: the red Deer River Alberta, Canada. Journal
of Biogeography 24: 675-695.
Corthay R. 1996. Analyse floristique de la forêt sempervirente de Yapo (Côte d’Ivoire). Mémoire de
l’Université de Genève, Suisse. 152p.
Cowling R.M. & Holmes P.M. 1992. Flora and vegetation. In: Cowling R.M. (ed.), The Ecology of
Fynbos: Nutrients, Fire and Diversity. Oxford University Press, Cape Town.
Crawford C.S., Ellis L.M. & Molles Jr.M.C. 1996. The middle Rio Grande bosque: an endangered
ecosystem. New Mexico Journal of Science 36: 276-299.
Cubbedge A.W. & Nilon C.H. 1993. Adjacent land use effects on the flood plain forest bird
community of Minnesota Valley National Wildlife Refuge. Natural Areas Journal 13: 220.
Danvind M. & Nilsson C. 1997. Seed floating ability and distribution of alpine plants along a northern
Swedish river. Journal of Vegetation Science 8: 271-276.
Darveau M., Beauchesne P., Bélanger L., Huot J. & Larue P. 1995. Riparian forests strips as habitat
for breedings birds in boreal forest. Journal of Wildlife Management 59: 67-78.
Darveau M., Huot J. & Bélanger L. 1998. Riparian forests strips as habitat for snowshoe hare in
boreal balsam fir forest. Canadian Journal of Forest Resources 28: 1494-1500.
170
References
Darveau M., Labbé P., Beauchesne P., Bélanger L. & Huot J. 2001. The use of riparian forest strips
by small mammals in a boreal balsam fir forest. Forest Ecology and Management 143: 95104.
DeBano L.F. & Neary D.G. 1996. Effects of fire on riparian systems. In: Folliott P.F., DeBano L.F. &
Baker M.B. (eds.), Effects of fire on Madrean Province ecosystems. A symposium
proceedings. General Technical Report RMGTR – 289. US Department of Agriculture, Forest
Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, CO. pp 69-76.
Décamps H., Fortuné M., Gazelle F. & Pautou G. 1988. Historical influence of man on the riparian
dynamics of a fluvial landscape. Landscape Ecology 1: 163-173.
de Koning J. 1983. La forêt du Banco. Thèse de Doctorat. Section de Taxinomie et de Géographie
botaniques, Université Agronomique de Wageningen, Pays-Bas. 921 p.
de Lima M.G. & Gascon C. 1999. The conservation value of linear forest remnants in central
Amazonia. Biological Conservation 91: 241-247.
Dengueadhe K.Th.S. 1999. Etude de la végétation du Sud du Parc National de Taï: Identification des
milieux. Résultats du Projet ‘Flore’, Deuxième partie. Tropenbos, Côte d’Ivoire. 62p.
Depiere D. & Vivien J. 1992. Mammifères sauvages du Cameroun. Ministère du Développement
Rural. Office National des Forêts, Fontainebleau, France. 249p.
de Souza S. 1979. Contribution à l’étude biologique et écologique de quelques Chrysobalanacées des
genres Chrysobalanus, Parinari et Maranthes au Bénin. Thèse de Doctorat de l’Université de
Bordeaux III, Bordeaux, France. 317p.
Devineau J.L. 1975. Etude quantitative des forêts galeries de Lamto (Moyenne Côte d’Ivoire). Thèse
3ème cycle. Université Pierre et Marie Curie, Paris VI, France. 190p.
Devineau J.L. 1976. Principales caractéristiques physiognomiques et floristiques des formations
forestières de Lamto (Moyenne Côte d’Ivoire). Annales de l’Université d’Abidjan série E 9:
274-303.
Devineau J.L. 1984. Travaux des chercheurs de la station de Lamto: Structure et dynamique de
quelques forêts tropophiles de l’Ouest Africain (Côte d’Ivoire). Programme MAB savanes no.
5. Université d’Abidjan, Station d’Ecologie tropicale de Lamto, Côte d’Ivoire. 294p.
Dignan P. & Bren L. 2003. A study of logging on the understorey light environment in riparian buffer
strips in a south-east Australian forest. Forest Ecology and Management 172: 161-172.
Djego J.G.M. 2000. Impacts écologiques des plantations forestières sur la diversité biologique des
écosystèmes côtiers du Bénin. Cas du périmètre de reboisement de Sèmè. Mémoire
DEA/FLASH/UAC. Abomey-Calavi, Bénin. 129p.
DNRE 1996. Code of Practice: Code of Forest practices for Timber Production. Department of
Natural Resources and Environment, Victoria, Australia. 68p.
Doyle A.T. 1990. Use of riparian and upland habitats by small mammals. Journal of mammalogy 71:
14-23.
Dunn C.P. & Stearns F. 1987. Relationship of vegetation layers to soils in Wisconsin forested
wetlands. American Midland Naturalist 118: 366-374.
Dupond L.M. & Weinelt M. 1996. Vegetation history of the savanna corridor between the Guinean
and the Congolian rain forest during the last 150,000 years. Vegetation History
Archaeobotany 5: 273-292.
Dupond L.M., Jahn S., Marret F. & Ning S. 2000. Vegetation change in equatorial West Africa:
times-slices for the last 150 ka. Paleogeography, Paleoclimatology, Paleoecology 155: 95122.
Eldin M. 1971. Le climat. In: Le milieu naturel de la Côte d’Ivoire. Mémoire ORSTOM 50: 73-108.
171
References
Ellis L.M., Crawford C.S. & Molles Jr.M.C. 1998. Comparison of litter dynamics in native and exotic
riparian vegetation along the Middle Rio Grande of Central New Mexico, USA. Journal of
Arid Environments 38: 283-296.
Ellis L.M., Molles Jr.M.C. & Crawford C.S. 1999. Influence of experimental flooding on litter
dynamics in a Rio Grande riparian forest of Central New Mexico. Restoration Ecology 7: 113.
Ellis L.M. 2001. Short-term response of woody plants to fire in a Rio Grande riparian forest of
Central New Mexico, USA. Biological Conservation 97:159-170.
Enright N.J., Miller B.P. & Crawford D. 1994. Environmental correlates of vegetation patterns and
species richness in the northern Grampians, Victoria. Australian Journal of Ecology 19: 159168.
Ericsson K.A. & Schimpf D.J. 1986. Woody riparian vegetation of a Lake superior tributary.
Canadian Journal of Botany 64: 769-773.
Ern H. 1979. Die Vegetation Togos. Gliederung, Gefährdung, Erhaltung. Willdenowia 9: 295-312.
Ern H. 1988. Flora and vegetation of the Dahomey Gap. A contribution to the plant geography of west
tropical Africa. Monographs Systematic Botany from the Missouri Botanical Garden 25: 517520.
Essou J-P. 1991. Contribution à l’étude des pâturages et systèmes agro-pastoraux du Sud-Bénin.
Thèse de Doctorat Université de Bordeaux III, Bordeaux, France. 370p.
Ettema C.H., Lowrance R. & Coleman D.C. 1999. Riparian soil response to surface nitrogen input:
the indicator potential of free-living soil nematode populations. Soil Biology and Biochemistry
31: 1625-1638.
Everett R., Schellhaas R., Ohlson P., Spurbeck D. & Keenum D. 2003. Continuity in fire disturbance
between riparian and adjacent sideslope Douglas-fir forests. Forest Ecology and Management
173: 293-308
Eyog-Matig O., Adjanohoun E., de Souza E. & Sinsin B. (eds.) 2001. Réseau “espèces ligneuses
Médicinales”. Compte rendu de la première réunion du Réseau, 15-17 Décembre 1999.
Station IITA Cotonou, Bénin. 131p.
Eyog-Matig O., Gaoué O.G. & Obel-Lawson E. (eds.) 2002. Development of appropriate
conservation strategies for African forest trees identified as priority species by SAFORGEN
member countries. IPGRI Nairobi, Kenya. 83p.
Faber-Langendoen D. & Gentry A.H. 1991. The structure and diversity of rain forests at Bajo Calima,
Choco region, western Colombia. Biotropica 23(1): 2-11.
FAO 1980. Cartographie du couvert végétal et étude de ses modifications. Projet pilote sur la
surveillance de la couverture forestière tropicale. UN 32/6. 1102-75-005. Rome, Italie. 75p +
2 cartes.
FAO 2000. Africa Forests Shrinking. Environment News Service. Rome, Italia.
Farley G.H., Ellis L.M., Stuart J.N. & Scott Jr. 1994. Avian species richness in different-aged stands
of riparian forest along the Middle Rio Grande, New Mexico. Conservation Biology 8(4):
1098-1108.
Fetherston K.L., Naiman R.J. & Billy R.E. 1995. Large woody debris, physical process, and riparian
forest development in montane river networks of the Pacific Northwest. Geomorphology 13:
133-144.
FGU Kronberg 1979. Gegenwärter Status der Comoé – und Tai – Nationalparks sowie des AzagnyReservats und Vorschläge zu deren Erhaltung und Entwicklung zur Förderung des Tourismus.
II. Comoé-Nationalpark. 1. Bestandsaufnahme der ökologischen Verhältnisse. GTZ,
Eschborn. Germany.
172
References
Flanagan D.C., Foster G.R., Neibling W.H. & Burt J.P. 1989. Simplified equations for filter strip
design. Transactions of the ASAE 32(1): 2001-2007.
Folke C., Holling C.S. & Perrings C. 1996. Biological diversity, ecosystems, and the human scale.
Ecological Applications 6(4): 1018-1024.
Forman R.T.T. 1997. Land Mosaics. Cambridge University Press. Cambridge, UK.
Forman R.T.T. & Godron M. 1986. Landscape ecology. John Wiley & Sons. New York. 619p.
Forniss M.J. 1996. Index for riparian bibliography. Interagency Watershed Analysis Centre, USDA –
Forest Service.
Freese F. 1990. Statistics for land managers. Paeony Press, Edinburg, Scotland. 176p.
Ganglo J.C. 2000. Phytosociologie de la végétation naturelle de sous-bois, écologie et productivité des
plantations de teck (Tectona grandis L.f.) du sud et du centre Bénin. Thèse Doctorat
Université Libre de Bruxelles. Section Interfacultaire d’Agronomie, Belgique. 366p.
Gartlan J.S., Newbery D.M., Thomas D.W. & Waterman P.G. 1986. The influence of topography and
soil phosphorus on the vegetation of Korup forest Reserve, Cameroon. Vegetatio 65: 131-148.
Gates J.E. & Giffen N.R. 1991. Neotropical migrant birds and edge effects at a forest-stream ecotone.
Wilson Bulletin 103: 204-217.
Geier A.R. & Best L.B. 1980. Habitat selection by small mammals of riparian communities:
evaluating effects of habitat alterations. Journal of Wildlife Management 44: 16-24.
Geraghty P. & Vollebergh P. 1993. Riparian zone management in Victoria: Policy and practice. In:
Bunn S.E., Pusey B.J. & Price P. (eds.), Ecology and Management of Riparian Zones in
Australia. LWRRDC Occasional Paper Series No: 05/93.
Gentry A.H. 1982. Patterns of neotropical plant species diversity. In: Hecht M.K., Wallace B. &
Prance G.T. (eds.), Evolutionary Biology. Plenum Press, New York. pp 123-135.
Gentry A.H. 1988. Changes in plant community diversity and floristic composition on environmental
and geographical gradients. Annals of the Missouri Botanical Garden 75:1-34.
Gentry A.H. 1992. Tropical forest biodiversity: distributional patterns and their conservational
significance. Oikos 63: 19-28.
Gibbs P.E. & Leitão F.H.F. 1978. Floristic composition of an area of gallery forest near Mogi Guaçu,
State of São Paulo, S.E. Brazil. Revista Brasileira de Botanica 1: 151-156.
Giese L.A.B., Aust W.N., Kolka R.K. & Trettin C.C. 2003. Biomass and carbon pools of disturbed
riparian forests. Forest Ecology and Management 6198: 1-16.
Gilliam J.W. 1994. Riparian wetlands and water quality. Journal of Environment Quality 25: 896-900.
Gnele J.E. 1999. Dynamique de l’environnement et utilisation des bas-fonds dans le Mono-Nord: Cas
du secteur Dekpo-Lonkly. Mémoire de Maîtrise en Géographie FLASH-UNB. AbomeyCalavi, Bénin. 89p.
Gosselink J.P. & Lee L.C. 1989. Cumulative impact assessment in bottomland hardwood forests.
Wetlands 9: 89-174.
Goudiaby A. 1998. Diversité floristique et conservation de la forêt galerie de la cascade de Dindelello,
sud-est du Sénégal. In: Bâ A.T., Madsen J.E. & Sambou B. (eds.), Atelier sur flore, végétation
et Biodiversité au Sahel. Aarhus University Press, Aarhus, Denmark. AAU Reports 39: 67-74.
Green A.A. 1979a. Développement des Parcs Nationaux: La rivière Pendjari (Parc National de la
Pendjari). PNUD/FAO/ 77/011 Document de travail No. 6. Cotonou, Bénin. 42p.
Green A.A. 1979b. Développement des Parcs Nationaux: La végétation du Parc National de la
Pendjari et des régions avoisinantes. PNUD/FAO/ 77/011 Document de travail No. 8.
Cotonou, Bénin. 86p.
173
References
Gregory S.V., Beschta R, Swanson F.J., Sedell J., Reeves G. & Everest F. 1990. Abundance of
conifers in Oregon’s riparian forests. COPE Report, Vol 3, No. 2. College of Forestry, Oregon
State University, Corvallis, OR. pp 5-6.
Gregory S.V., Swanson F.J., MsKee W.A. & Cummins K.W. 1991. An ecosystem perspective of
riparian zones: focus on links between land and water. BioScience 41: 540-551.
Groffman P.M., Axelrod E.A., Lemunyon J.L. & Sullivan W.M. 1991. Denitrification in grass and
forest vegetated filter strips. Journal of Environmental Quality 20: 671-674.
Groffman P.M., McDowell W.H., Myers J.C. & Merriam J.L. 2001. Soil microbial biomass and
activity in tropical riparian forests. Soil Biology & Biochemistry 33: 1339-1348.
Grubb P., Lernould J.M. & Oates J.F. 1999. Validation of Cercopithecus erythrogaster pococki as the
name for Nigerian white-throated guenon. Mammalia 63(3): 389-392.
Guelly K.A. 1994. Les savanes des plateaux de la zone subhumide du Togo. Thèse de Doctorat de
l’Université Paris 6, Spécialité Botanique tropicale. Paris, France. 163 p.
Guillaumet J.L. 1967. Recherche sur la végétation et la flore de la région du Bas-Cavally (Côted’Ivoire). Mémoire ORSTOM 20: 247p.
Guillaumet J.L. & Adjanohoun E. 1971. La végétation de la Côte d’Ivoire. In: Le milieu naturel de la
Côte d’Ivoire. Mémoire ORSTOM 50:157-263.
Guinko S. 1974. Contribution à l’étude écologique des savanes marécageuses du Bas-Dahomey.
Thèse 3ème cycle. Université d’Abidjan, Côte d’Ivoire. 142p.
Gunasekera S.P., Sivapalan K., Sultanbawa M. & Ollis W.D. 1977. Chemical investigation of
ceylonese plants. Extractives of Pentadesma butyracea Guttiferae. J.C.S. Perkin 1: 15051511.
Hall J.B. & Swaine M.D. 1976. Classification and ecology of closed-canopy forest in Ghana. Journal
of Ecology 64: 913-951.
Hall J.B. & Swaine M.D. 1981. Distribution and Ecology of vascular plants in a tropical rain forest,
Forest Vegetation in Ghana. Geobotany 1, Dr. W. Junk, The Hague. 383 p.
Hancock C.N., Ladd P.G. & Froend R.H. 1996. Biodiversity and management of riparian vegetation
in Western Australia. Forest Ecology and Management 85: 239-250.
Hanson J.S., Malanson G.P. & Armstrong P.M. 1990. Landscape fragmentation and dispersal in a
model of riparian dynamics. Ecological Modelling 49 (3): 277-296.
Harper D., Smith C., Barham P. & Howell R. 1994. The ecological basis for the management of the
natural river environment. In: Harper D. & Ferguson A.J.D. (eds.), The ecological basis for
river management. Wileym Chichester. pp 219-238.
Hart T.B. 1990. Monospecific dominance in tropical rain forests. Tropical Ecology and Evolution 5:
6-12.
Hartshorn G.S. 1988. Tropical and subtropical vegetation of meso-America. In: Barbour M.A. &
Billings W.D. (eds.), North American terrestrial vegetation. Cambridge University Press,
Cambridge. pp 366-390.
Hawthorne W.D. 1996. Guide de terrain pour les arbres des forêts denses de la Côte d’Ivoire et pays
limitrophes. Avec clés végétatifs sur plus que 650 espèces d’arbres, à partir de 5 cm de
diamètre. Projet ECOSYN, Université de Wageningen, Pays-Bas. 276 p.
Haycock N.E. & Pinay G. 1993. Groundwater nitrate dynamics in grass and poplar vegetated riparian
buffer strips during the winter. Journal of Environmental Quality 22: 273-278.
Haycock N.E., Pinay G. & Walter C. 1993. Nitrogen retention in river corridors: European
perspective. Ambio 22: 340-346.
Hefting M.M. & de Klein J.J.M. 1998. Nitrogen removal in buffer strips along a lowland stream in the
Netherlands: a pilot study. Environmental Pollution 102(S1): 521-526.
174
References
Hegarty E.E. & Caballé G. 1991. Distribution and abundance of vines in forests communities. In: Putz
F.E. & Mooney H.A. (eds.), The biology of vines. Cambridge University Press, Cambridge.
pp 238-250.
Hehnke M. & Stone C. 1979. Value of riparian vegetation to avian populations along the Sacramento
River system. In: Johnson R.R. & McCormick J.R. (eds.), Strategies for protection and
management of floodplain wetlands and other riparian ecosystems. USDA Forest Service
General Technical Report WO-12, Washington D.C. pp 228-235.
Hermans M. 2002. Primary Producers at War with Parasites: A survey of medicinal plants used to
treat malaria in West Africa with emphasis on Benin. MSc thesis, Laboratory for Plant
Taxonomy. Wageningen University, The Netherlands. 60p.
Hibbs D.E. & Bower A.L. 2001. Riparian forests in the Oregon Coast Range. Forest Ecology and
Management 154: 201-213.
Hill M.O. & Gauch Jr.H.G. 1980. Detrended Correspondance Analysis: An improved ordination
technique. Vegetatio 42: 47-58.
Hill M.O. 1979. TWINSPAN- a FORTAN Program for arranging Multivariate Data in an Ordered
Two Way Table by Classification of the individuals and the Attributes. Cornell University,
Department of Ecology and Systematics, Ithaca. New York.
Hladik C. 1982. Dynamique d’une forêt équatoriale africaine: mesures en temps réel et comparaison
du potentiel de croissance des différentes espèces. Acta Oecologica, Oecologia Generalis 3:
373-392.
Hommel P.W.F.M. 1990. A phytosociological study of a forest area in the humid tropics (Ujung
Kulon, West Java, Indonesia). Vegetatio 89: 39-54.
Houinato M., Delvaux C. & Pauwels L. 2000. Les Eragrostis (Poaceae) du Bénin. Belgian Journal of
Botany 133(1-2): 21-35.
Houinato M.R.B. 2001. Phytosociologie, écologie, production et capacité de charge des formations
végétales pâturées dans la région des Monts Kouffè, Bénin. Thèse de doctorat de l’Université
Libre de Bruxelles, Belgique. 255p.
Houndagba C.J., Tenté A.B.H. & Guédou G.R. (in press). Dynamique des forêts classées dans le
cours moyen de l’Ouémé (Kétou, Dogo et Ouémé-Boukou) au Bénin. In : Actes du Premier
séminaire régional sur l’Aménagement et la Gestion des Aires Protégées d’Afrique de
l’Ouest. Parakou, Bénin, 14-19 Avril 2003.
Houngbedji A. 1997. Etude phytotechnique, écologique et des technologies endogènes de
transformation du Pentadesma butyracea, espèce des galeries forestières de la région de
Bassila. Mémoire du DEAT – LTAM, Sékou, Bénin. 59 p.
Hountondji Y.C.H. 1998. Contribution à l’étude des possibilités d’aménagement de la forêt de Niaouli
au Sud-Bénin. Mémoire de fin d’étude, CPU/UAC, Abomey-Calavi, Bénin. 114p.
Hubbell S.P. & Foster R.B. 1986. Commonness and rarity in a neotropical forests: implications for
tropical tree conservation. In: Soulé M.J. (ed.), Conservation Biology: the Science of Scarcity
and Diversity. Sinauer, Massachusetts. pp 205-231.
Hughes F.M.R. 1990. The influence of flooding regimes on forest distribution and composition in the
Tana River floodplain, Kenya. Journal of Applied Ecology 27: 475-491.
Hunter Jr.M.L. 1990. Wildlife, Forest and Forestry: Principles of Managing Forests for biological
diversity. Prentice-Hall, Englewood Cliffs, NJ. pp 148-149.
Hupp C.R. & Osterkamp W.R. 1996. Riparian vegetation and fluvial geomorphic processes.
Geomorphology 14: 277-295.
Huston M.A. 1994. Biological diversity: The coexistence of species on changing landscapes.
Cambridge University Press, Cambridge. 681p.
175
References
Hutcheson K. 1970. A test for comparing based on the Shannon formula. Journal of Theoretical
Biology 29: 151-4.
INSAE (Institut National de la Statistique et de l’Analyse Economique) 1979. Recensement Général
de la Population et de l’Habitat (RGPH). Tableaux statistiques, Volume National (Tome 1:
mars 1979). Ministère du Plan, Cotonou, Bénin.
INSAE 2002. Recensement Général de la Population et de l’Habitat (RGPH). Résultats provisoires.
Ministère du Plan, Cotonou, Bénin. 4 p.
Jansson R., Nilsson C., Dynesius M, & Andersson E. 2000a. Effects of river regulation on rivermargin vegetation: a comparison of eight boreal rivers. Ecological Applications 10: 203-224.
Jansson R., Nilsson C. & Renöfält B. 2000b. Fragmentation of riparian floras in rivers with multiple
dams. Ecology 81: 899-903.
Johansson M.E., Nilsson C. & Nilsson E. 1996. Do rivers function as corridors for plant dispersal?
Journal of Vegetation Science 7: 593-598.
Jenik J. 1994. The Dahomey-Gap: an important issue in African phytogeography. Mémoire de la
Société Biogéographique (3ème série) IV: 125-133.
Johnson R.R., Haight L.T. & Simpson J.M. 1977. Endangered species versus endangered habitats: a
concept. Importance, preservation, and management of riparian habitat: a symposium. USDA
Forest Service General Technical Report RM – 43, Fort Collins, Colorado. pp 68-79.
Jones A.P.D. 1945. Notes on terms for use in vegetation description in southern Nigeria. Farm and
Forest 6: 130-136.
Jones E.W. 1955. Ecological studies on the rain forest of southern Nigeria. IV. The plateau forest of
the Okumu Forest Reserve. Part I. The environment, the vegetation types of the forest, and the
distribution of species. Journal of Ecology 43: 564-594.
Jongkind C.C.H. 1996. Environmental Impact Assessment of Black Volta River in south west Ghana.
Report for the ODA - Ghana. (Unpublished).
Kadio A. 1999. Constitution d’un réseau de “série de protection” dans les forêts classées gérées par la
Société de Développement des Forêts (SODEFOR): une contribution à la sauvegarde de la
diversité biologique. In: ECOSYN (ed.), Guide, Atlas et Ecologie des arbres et lianes de
l’Afrique de l’Ouest, de la Guinée au Ghana, pour soutenir la gestion écologique des forêts
denses tropicales humides. Séminaire d’Abidjan 13-14 décembre 1999, Côte d’Ivoire. pp 3235.
Kalanda K. 1981. La végétation du parc de la N’sele (Zaire). Bulletin du Jardin Botanique National
de Belgique 51 (3-4): 347-357.
Kalliola R. & Puhakka M. 1988. River dynamics and vegetation mosaicism: a case study of the River
Kamajohka, northernmost Finland. Journal of Biogeography 15: 703-719.
Katibath E.F. 1984. A brief history of riparian forests in the Central Valley of California. In: Warner
R.E. & Hendrix K.M. (eds.), California Riparian Systems: Ecology, Conservation, and
Productive management. University of California Press, Berkeley. pp 23-29.
Keay R.W.J. 1949. An example of Sudan zone vegetation in Nigeria. Journal of Ecology 37: 335-64.
Keay R.W.J. 1959. Derived savanna - derived from what? Bulletin IFAN série A 21: 427-38.
Keay R.W.J. 1960. An example of Northern Guinean Zone vegetation in Nigeria. Nigerian Forestry
Information Bulletin 1: 1-46.
Keay R.W.J & Hepper F.N. (eds.). 1954-1972. Flora of West Tropical Africa, 2nd edition. Millbank,
London.Vol. I, (828p), Vol. II (544p), Vol. III (574p).
Keay R.W.J., Onochie C.F.A. & Standfield D.P. 1964. Nigerian trees. Federal Department of Forest
Research, Ibadan, Nigeria. Vol 1 (334p), Vol 2 (495p).
176
References
Keita S.M., Arnason J.T., Baum B.R., Marles R., Camara F. & Traoré A.K. 1999. Etude
ethnopharmacologique traditionnelle de quelques plantes médecinales anthelminthiques de la
Haute-Guinée (République de Guinée). Revue de Médecines et Pharmacopées Africaines 13:
49-64.
Kellman M. & Tackaberry R. 1993. Disturbance and tree species coexistence in tropical riparian
forest fragments. Global Ecology and Biogeography Letters 3: 1-9.
Kellman M., Tackaberry R., Brokaw N. & Meave J. 1994. Tropical gallery forests. National
Geographical Research and Exploration 10: 92-103.
Kellman M, Tackaberry R. & Meave J. 1996. The consequences of prolongued fragmentation: lessons
from tropical gallery forests. In: Schelhas J. & Greenberg R. (eds.), Forest patches in tropical
landscapes. Island Press, Washington D.C. pp 37-58.
Kellman M. & Meave J. 1997. Fire in the tropical gallery forests of Belize. Journal of Biogeography
24 (1): 23-34.
Kent M. & Coker P. 1992. Vegetation description and analysis: A practical approach. John Wiley &
Sons, England. 363 p.
Kerharo J. & Adam J.G. 1974. La pharmacopée sénégalaise traditionnelle : Plantes médecinales et
toxiques. Edition Vigot Frères, Paris, France. 1011p.
Kingdon J. 1990. Island Africa: The evolution of Africa’s rare animals and plants. William Collins
Sons & Co., London. 287p.
Kish L. 1965. Survey sampling. John Wiley & Sons, New York.
Knopf F.L. 1985. Significance of riparian vegetation to breeding birds across an altitudinal cline.
USDA Forest Service, General Technical Report RM – 120, Rocky Mountain Forest and
Range Experiment Station, Fort Collins, Colorado. pp 105-111.
Knopf F.L., Johnson R.R., Rich T., Samson F.B. & Szaro R.C. 1988. Conservation of riparian
ecosystems in the United States. Wilson Bulletin 100: 272-284.
Knutson K.L. & Nae V.L. 1997. Management recommendations for Washington priority habitats:
riparian. Washington Departement of Fish and Wildlife, Olympia, WA.
Kokou K. 1998. Les mosaïques forestières au sud du Togo: Biodiversité, dynamique et activités
humaines. Thèse Doctorat de l’Université de Montpellier II. Montpellier, France. 140p +
Annexes.
Kokou K. & Caballé G. 2000. Les îlots forestiers de la plaîne côtière togolaise. Bois et Forêts
Tropicaux 263 (1): 39-50.
Kokou K., Caballé G., Akpagana K. & Batawila K. 1999. Les îlots forestiers au sud du Togo:
Dynamique et relations avec les végétations périphériques. Revue d’Ecologie (Terre et Vie)
54: 301-314.
Kokou K., Couteron P., Martin A. & Caballé G. 2002. Taxonomic diversity of lianas and wines in
forests fragments of southern Togo. Revue d’Ecologie (Terre Vie) 57: 3-17.
Kouamé F.N. & Traoré D. 2001. Exploitation forestière et les lianes de forêt dense tropicale humide:
cas de la forêt classée du Haut sassandra. In: Rapport d’activité 1999-2000 du Centre Suisse
de Recherche Scientifique en Côte d’Ivoire (CRRS). Abidjan, Côte d’Ivoire. pp 20-21.
Large A.R.G. & Petts G.E. 1994. Rehabilitation of rivers margins. In: Calow P. & Petts G.E. (eds.),
The rivers handbook, hydrological and ecological principles. Blackwell Scientific Oxford. pp
401-418.
Larue P., Bélanger L. & Huot J. 1995. Riparian edge effects on boreal balsam fir bird communities.
Canadian Journal of Forest Resources 25: 555-566.
Lawson G.W., Armstrong-Mensah K.O. & Hall J.B. 1970. A catena in tropical moist semi-deciduous
forest near Kade, Ghana. Journal of Ecology 58: 371-398.
177
References
Lebrun J. 1947. La végétation de la plaine alluviale au sud du lac Edouard, Exploration du Parc
National Albert. Mission J. Lebrun. Institut des Parcs Nationaux du Congo Belge. 800p.
Lebrun J. & Gilbert G. 1954. Une classification écologique des forêts du Congo. Publication
I.N.E.A.C. Série Science 63: 89p.
Lebrun J-P. & Stork L. 1991-1997. Enumération des plantes à fleur d’Afrique Tropicale. CIRATEMVT- Montpellier. Editions des Conservatoire et Jardin botaniques de Genève. Vol. I
Publication hors-série no.7; 249p.; Vol. II Publication hors-série no.7a; 257p.; Vol. III
Publication hors-série no.7b; 341p.; Vol. IV Publication hors-série no.7c; 712p.
Leinard R.E., Wyman S.K. Sue Noggles R. & Volk W.P. 1999. Classification of riparian plant
communities using the ecological site concept. VIth International Rangeland Congress
Proceedings 1: 249-251.
Lejoly J. 1994. Mise en place des travaux en vue des inventaires de biodiversité dans la forêt de
Ngotto (République Centrafricaine). Projet Ecofac, Agreco-CTFT, Bruxelles. 109p.
Lejoly J. & Mandango A. 1982. L’association arbustive ripicole à Alchornea cordifolia dans le hautZaire. Studies on Aquatic Vascular Plants, Royal Botanical Society of Belgium, Brussels. pp
257-265.
Lejoly J., Sonké B. & van Essche K. 1996. Utilisation de la méthode du transect en vue de l’étude de
la biodiversité dans la réserve de Faune du Dja (Cameroun). In: van der Maesen L.J.G., van
der Burgt X.M. & van Medenback de Rooy J.M. (eds.), The biodiversity of African plants.
Proceedings of the XIVth AETFAT congress. Wageningen, The Netherlands. pp 150-154.
Léonard J. 1947. Contribution à l’étude des formations ripicoles arbustives et arborescentes de la
région d’Eala. C.R. semaine agricole de Yangambi (du 26 février au 5 mai 1947). Publication
I.N.E.A.C. Buxelles, 2ème partie 67: 863-877.
Léonard J. 1952. Aperçu préliminaire des groupements végétaux pionniers dans la région de
Yangambi (Congo Belge). Vegetatio 3: 279-297.
Léonard J. 1965. Contribution à la subdivision phytogéographique de la région guinéo-congolaise
d’après la répartition géographique d’Euphorbiacées d’Afrique tropicale. Webbia 19: 627649.
Levaux P. 1990. Etude des relations faune-végétation sur transects au Parc National de la Pendjari au
Bénin. Ecole Nationale du Génie Rural des Eaux et Forêts-Centre de Montpellier, France. 37p
+ annexes.
Levine C.M. & Stromberg J.C. 2001. Effects of flooding on native and exotic plant seedlings:
implications for restoring south-western riparian forests by manipulating water and sediment
flows. Journal of Arid Environments 49: 111-131.
Leyden B. 1985. Late quaternary aridity and Holocene moisture fluctuations in the Lake Valencia,
Venezuela. Ecology 66: 1279-1295.
Lock P.A. & Naiman R.J. 1998. Effects of stream size on bird community structure in coastal
temperate forests of the Pacific Northwest, USA. Journal of Biogeography 25: 773-782.
Loizeau P.A. 1992. La diversité: exemple des forêts denses humides amazoniennes. Saussurea 23: 4982.
Lonard R.I., Judd F.W., Everitt J.H., Escobar D.E., Davis M.R., Crawford M.M. & Desai M.D. 2000.
Evaluation of color-infrared photography for distinguishing annual changes in riparian forests
vegetation of the lower Rio Grande in Texas. Forest Ecology and Management 128: 75-81.
Lowrance R.R., Todd R.L. & Asmussen L.E. 1984. Nutrient cycling in an agricultural watershed-I
phreatic movement. Journal of Environmental Quality 13: 22-27.
Lowrance R.R., Altier L.S., Newbold J.D., Schnabel R.R., Groffman P.M., Denver J.M., Correll D.L.,
Gilliam J.W., Robinson J.L., Brinsfield R.B., Staver K.W., Lucas W. & Todd A.H. 1997.
178
References
Water quality functions of riparian forest buffers in Chesapeak Bay watersheds. Environment
Management 21: 687-712.
Lykke A.M. & Goudiaby A. 1999. Structure, floristic composition, and conservation potentials of a
remnant gallery forest at ‘Mare du Dragon’, Senegal. Nordic Journal of Botany 19: 561-574.
Lykke A.M. 1996. How gallery forest turms into savanna: an example from Senegal. In: van der
Maesen L.J.G., van der Burgt X.M. & van Medenbach de Rooy J.M. (eds.), The biodiversity
of African plants. Proceedings of the XIVth AETFAT congress. Wageningen, The
Netherlands. pp 150-154.
MacDougall A. & Kellman M. 1992. The understorey light regime and patterns of tree seedlings in
tropical riparian forest patches. Journal of Biogeography 19: 667-675.
Machtans C.S., Villard M.A. & Hannon S.J. 1996. Use of riparian buffer strips as movement corridors
by forest birds. Conservation Biology 10: 1366-1379.
Magurran A.E. 1988. Ecological diversity and its measurement. Cambridge University Press,
Cambridge. 179 p.
Magurran A.E. & Henderson P.A. 2003. Explaining the excess of rare species in natural species
abundance distributions. Nature 422: 714-716.
Malanson G.P. 1993. Riparian lansdcapes. Cambridge University Press, Cambridge.
Malanson G.P. & Armstrong M.P. 1996. Dispersal probability and forest diversity in a fragmented
landscape. Ecological Modelling 87: 91-102.
Maley J. 1989. 20.000 ans d’évolution des climats du Sahara Central aux savanes tropicales humides.
In: Centre Culturel Français d’Abidjan (ed.), Les climats, Sud-Sahara, Sahel Nord. pp 34-39
et 42-50.
Maley J. 1991. The African rain forest vegetation and paleoenvironments during late quaternary.
Climatic Change 19: 79-98.
Maley J. 1996. The African rain forest – main characteristics of changes in vegetation and climate
from the Upper Cretaceous to the Quaternary. In: Alexander I.J., Swaine M.D. & Watling R.
(eds.), Essays on the Ecology of the Guinea-Congo Rain Forest. Proceedings of the Royal
Society of Edinburgh 104B: 31-73.
Maley J. 1997. Middle to late Holocene changes in tropical Africa and the other continents: paleomonsoon and sea surface temperature variations. In: Dalfer H.N., Kukla G. & Weis H. (eds.),
Third Millenium BC Climate Change and World Collapse. Nato ASI Series I 49: 611-640.
Maley J. 2001. Si la forêt tropicale m'était contée. La destruction catastrophique des forêts d'Afrique
centrale survenue il y a environ 2500 ans exerce encore une influence majeure sur la
répartition actuelle des formations végétales. Canopée (Journal du Programme Européen
ECOFAC) Libreville 19: 6-9.
Maley J. 2002. A catastrophic destruction of African forests about 2,500 years still exerts a major
influence on the present vegetation formations. IDS Bulletin 33 (1): 13-30.
Maley J. & Chepstow-Lusty A. 2001. Elaeis guineensis Jacq. (oil palm) fluctuations in central Africa
during the late holocene: climate or human driving forces for this pioneering species.
Vegetation History and Archaeobotany 10: 117-120.
Mandango A. 1982. Flore et végétation des îles du fleuve Zaïre dans la sous-région de la Tshopo
(Haut-Zaire). Thèse Doctorat de l’Université de Kisangani. Zaire, 2 tomes. 425p.
Mandango A. & Ndjele M. 1986. Végétation aquatique et semi-aquatique de l’île Kongolo à
Kisangani (Zaire). Bulletin de la Société Royale Botanique de Belgique 119(1): 63-73
Mander Ü., Kuusemtes V., Lõhmus K. & Mauring T. 1997. Efficiency and dimensioning of riparian
buffer zones in agricultural catchments. Ecological Engineering 8: 299-324.
179
References
Mangenot G. 1955. Etude sur les forêts des plaines et plateaux de la Côte d’Ivoire. Etudes
Eburnéennes IFAN 4 : 5-61.
Manuwal D.A. 1983. Spring bird communities in the southern Washington Cascade range. In:
Ruggiero L.F., Aubry K.B., Carey A.B. & Huff M.H. (eds.), Wildlife and vegetation of
unmanaged Douglas-fir forests. USDA Forest Service General Technical Report PNW-GTR285. Pacific Northwest Research Station, Portland, Oregon.
Martijena N.E. 1998. Soil properties and seedling establishment in soils from monodominant and
high-diversity stands of the tropical deciduous forests of Mexico. Journal of Biogeography
25: 707-719.
Martijena N.E. & Bullock S.H. 1994. Monospecific dominance of a tropical deciduous forests in
Mexico. Journal of Biogeography 21: 63-74.
Martin C. 1991. The rainforests of West Africa. Birkhauser Verlag Publishers. Basel. Switzerland.
235p.
McClain M.E., Richey J.E. & Pimentel T.P. 1994. Groundwater nitrogen dynamics at the terrestriallotic interface of a small catchment in the Central Amazon Basin. Biogeochemistry 27:113127.
McDowell W.H., Bowden W.B. & Asbury C.E. 1992. Riparian nitrogen dynamics in two
geomorphological distinct tropical rain forest watersheds: subsurface solute patterns.
Biogeochemistry 18: 77-99.
McGarigal K. & McComb W.C. 1992. Streamside versus upslope breeding bird communities in the
central Oregon Coast Range. Journal of Wildlife Management 56: 10-23.
Meave J., Kellman M., MacDougall A. & Rosales J. 1991. Riparian habitats as tropical forest refugia.
Global Ecology and Biogeography Letters 1: 69-76.
Meave J. & Kellman M. 1994. Maintenance of rain forest diversity in riparian forests of tropical
savannas: implications for species conservation during Pleistocene drought. Journal of
Biogeography 21: 121-135.
Medina A.L. 1990. Possible effects of residual development on streamflow, riparian plant
communities, and fisheries on small mountain streams in Central Arizona. Forest Ecology
and Management 33-34: 351-361.
Medley M.E. 1992. Patterns of forest diversity along the Tana River, Kenya. Journal of Tropical
Ecology 8: 358-371.
Melick D.R. 1990. Ecology of rainforest and sclerophyllous communites in the Mitchell River
National Park, Gippsland, Victoria. Proccedings of the Royal Society of Victoria 102: 71-87.
Melick D.R. & Ashton D.H. 1991. The effects of natural disturbances on warm temperate rainforests
in south-eastern Autralia. Austalian Journal of Ecology 39:1-30.
Meurer M. 1994. Etudes sur le potentiel d’herbage dans les savanes du Nord Ouest du Bénin.
Agriculture et Développement Rural 1(1): 37- 41.
Minshall G.W., Petersen C.R.Jr. & Curtis N.F. 1985. Species richness in streams of different size
from the same drainage basin. American Naturalist 125(1): 16-38.
Molles Jr.M.C., Crawford C.S., Ellis L.M., Valett H.M. & Dahm C.N. 1998. Managed floods:
restoration of riparian forest ecosystem structure and function along the Middle Rio Grande.
BioScience 48: 749-756.
Mondjannagni A. 1969. Contribution à l’étude des paysages végétaux du Bas-Bénin. Annales de
l’Université d’Abidjan Série C 1(2): 191p.
Monnier Y. 1990. La poussière et la cendre: Paysages, dynamique des formations et stratégies des
sociétés en Afrique de l’Ouest. 2ème édition. ACCT, Paris, France. 264 p.
Montgomery D.R. 1997. What’s best on the banks? Nature 388: 328-329.
180
References
Mori S.A. & Boom B.M. 1987. The Lecythidaceae of a lowland Neotropical Forest. La fûmée
Mountain, French Guiana. Memoire New York Botanical Garden 44: 9-29.
Murray N.L. & Stauffer D.F. 1995. Nongame bird use of habitat in central Appalachian riparian
forests. Journal of Wildlife Management 59: 78-88.
Naiman R.J., Beechie T.J., Benda L.E., Berg D.R., Bison P.A., MacDonald L.H., O’Connor M.D.,
Olson P.L. & Stell E.A. 1992. Fundamental elements of ecologically healthy watershed. In:
Naiman R.J. (ed.), Watershed management: balancing sustainability and environmental
change. Springer-verlag, New York. pp 127-188.
Naiman R.J. & Décamps H. 1997. The ecology of interface: riparian zones. Annual Review of Ecology
and Systematics 28: 621-658.
Naiman R.J. & Décamps H. (eds.) 1990. The ecology and management of aquatic-terrestrial ecotones.
Man and the Biosphere series Vol. 4. UNESCO, Paris and Parthenon Publishing group,
Carnforth. 316p.
Naiman R.J., Décamps H. & Polleck M. 1993. The role of riparian corridors in maintaining regional
biodiversity. Ecological Applications 3: 209-212.
Naiman R.J., Melillo J.M., Lock M.A., Ford T.E. & Reice S.R. 1987. Longitudinal patterns of
ecosystem process and community structure in a subarctic river continuum. Ecology 68: 11391156.
Nakashizuka T., Zulkifli Yusop & Abdoul Rahim Nik 1992. Altitudinal zonation of forest
communities in Selangor, Peninsular Malaysia. Journal of Tropical Forest Science 4 (3): 233244.
National Research Council (NRC) 2002. Riparian areas: Functions and strategies for management.
Summary. WSTB, CRZFSM. National Academy Press, Washington. 22p.
Natta A.K. 1997. Contribution à l’étude et aménagement des pâturages naturels dans le Nord-Ouest de
l’Atacora. Mémoire d’Ingénieur Agronome. DAGE/FSA/UNB, Abomey-Calavi, Bénin. 199p.
Natta A.K. 2000. Tree species diversity assessment in riparian forests using remote sensing,
geographical information systems and geostatistics (case study of Toui-Kilibo protected forest
in Benin). MSc thesis, ITC Forest Science Division, Enschede, The Netherlands. 91p.
Natta A.K. & Porembski S. (in press). Ouémé and Comoé: forest-savanna border relationships in two
riparian ecosystems in West-Africa. Botanische JahrbĦcher 00: 00-00.
Natta A.K., Sinsin B. & van der Maesen L.J.G. 2002. Riparian forests, a unique but endangered
ecosystem in Benin. Notulae Florae Beninensis 4. Botanische JahrbĦcher 124: 55-69.
Natta A.K., Sinsin B. & van der Maesen L.J.G. (in press). A phytosociological study of Riparian
forests in Benin (West Africa). Belgian Journal of Botany 00:00-00.
Natta A.K., Sinsin B. & van der Maesen L.J.G. A phytosociological study of the Garcinia livingstonei
and Combretum acutum gallery forest plant community of North West Benin (West Africa)
(Submitted to the Acta Botanica Gallica).
Natta A.K. & van der Maesen L.J.G. Assessment of riparian forest fragments plant diversity in West
African savanna region: an overview from Benin Republic. (Submitted to the Journal of
Biogeography).
Nierenberg T. & Hibbs D.E. 2000. A characterisation of unmanaged riparian areas in the central Coast
range of western Oregon. Forest Ecology and Management 129: 195-206.
Nilsson C.1992. Conservation management of riparian communities. In: Hansson L. (ed.), Ecological
principles of nature conservation. Elsevier Applied Science, London. pp 352-372.
Nilsson C., Ekbald A., Dynesius M.A., Backe S., Gardfjell M., Carlberg B., Hellqvist S & Jansson R.
1994. A comparison of species richness and traits of riparian plants between a main river
channel and its tributaries. Journal of Ecology 82: 281-298.
181
References
Nilsson C., Grelsson G., Johansson M. & Sperens U. 1989. Patterns of plant species richness along
riverbanks. Ecology 70(1): 77-84.
Nilsson C., Jansson R. & Zinko U. 1997. Long-term responses of river-margin vegetation to waterlevel regulation. Science 276: 798-800.
Norris V.O.L. 1993. The use of buffer zones to protect water quality: a review. Water Resources
Management 7: 257-272.
Nye H. 1954. Some soil processes in humid tropics. 1. A field study of a catena in the West African
Forest. Journal of Soil Science 5: 7-21.
Oates J.F. 1996. Survey of Cercopithecus erythrogaster populations in the Dahomey Gap. African
Primates 2(1): 9-11.
Ohmart R.D. 1994. The effects of human-induced changes on the avifauna of western riparian
habitats. Studies in Avian Biology 15: 273-285.
Oldfield S., Lusty C. & Mackinven A. 1998. The World List of Threatened Trees. World
Conservation Press, Cambridge, UK. 650p.
Onochie C.F.A. 1979. The Nigerian rainforest ecosystem. In: Okali D.U.U (ed.), The Nigerian
rainforest ecosystem. Proceedings of the Man and Biosphere Workshop on the Nigerian
rainforest Ecosystem. University of Ibadan 24-26 January 1979. Ibadan, Nigeria. pp 1-37.
Ott L.R. & Longnecker M. 2001. An introduction to Statistical methods and data analysis. Fifth
edition. Thomson Learning Academic resource Center. Duxbury, USA. 1152p.
Otto T.S. 1993. Biodiversity: A means for coping with environmental changes, and a basic property of
life. Wildlife and Nature 9(3): 5-9.
Ouattara N. 1999. Evolution du taux de germination de semences oléagineuses en fonction du mode et
de la durée de conservation, Cas de Pentadesma butyracea Sabine (Lami). In: Ouédraogo
A.S., Boffa J.-M. (eds.), Vers une approche régionale des ressources génétiques forestières en
Afrique sub-saharienne. Actes du premier atelier régional de formation sur la conservation et
l’utilisation durable des ressources génétiques forestières en Afrique de l’Ouest, Centrale et
Madagascar, Centre National de Semences Forestières, Ouagadougou, Burkina-Faso 1627/03/98. pp 170-174.
Pabst R.J. & Spies T.A. 1999. Structure and composition of unmanaged riparian forests in the coastal
mountains of Oregon. Canadian Journal of Forest Resources 29: 1557-1573.
Paradis G. 1975a. Observations sur les forêts marécageuses du Bas-Dahomey: localisation, principaux
types et évolution au cours de quaternaire récent. Annales de l’Université d’Abidjan E 8(1):
281-315.
Paradis G. 1975b. Physionomie, composition floristique et dynamisme des formations végétales d’une
partie de la basse Vallée de l’Ouémé (Dahomey). Annales de l’Université d’Abidjan C 11: 55101.
Paradis G. 1976. Contribution à l’étude de la flore et de la végétation littorale du Dahomey. Bulletin
du Musée National d’Histoire Naturelle Paris 383: 33-68.
Paradis G. 1979. Observation sur Laguncularia racemosa Gaertn. et Dalbergia ecasstaphyllum (L.)
Taub. dans les successions secondaires de la mangrove du Bénin. Bulletin de l’Institut
Francais d’Afrique Noire A 41(1): 92-102.
Paradis G. 1980. Un cas particulier de zones dénudées dans les mangroves d’Afrique de l’Ouest:
celles dues à l’extraction de sel. Bulletin du Musée National d’Histoire Naturelle Paris, 4,
serie 2, section B 3: 227-261.
Paradis G. 1981. Ecologie et géomorphologie littorale en climat sub-équatorial sec: la végétation
côtière du Bas-Bénin occidental. Annales de l’Université d’Abidjan E 14: 7-63.
Paradis G. 1983. A phytogeographic survey of Southern Benin. Bothalia 14 (3-4): 579-585.
182
References
Paradis G. 1988. Etude comparative des végétations littorales du Bénin et de la Côte-d’Ivoire. Thèse
de Doctorat es Sciences. Université de Bordeaux III, Bordeaux, France. 295p.
Paradis G. & Adjanohoun E. 1974. L’impact de la fabrication du sel sur la végétation de mangrove et
la géomorphologie dans le Bas-Dahomey. Annales de l’Université d’Abidjan E 7: 599-612.
Paradis G., de Souza S. & Houngnon P. 1978. Les stations à Lophira lanceolata dans la mosaïque
forêt-savane du Sud-Bénin (ex Sud-Dahomey). Bulletin du Musée National d’Histoire
Naturelle Paris, Botanique 521: 39-58.
Paradis G. & Houngnon P. 1977. La végétation de l’aire classée de la Lama dans la mosaïque forêtsavane du Sud-Bénin (ex Sud-Dahomey). Bulletin du Musée National d’Histoire Naturelle
Paris 503: 169-198.
Paradis G. & Rabier J. 1979. La végétation de la lagune de Porto-Novo (Bénin) avant la fermeture du
chenal de Cotonou en 1978. Annales de l’Université d’Abidjan E 12: 7-28.
Parren M.P.E. 2003. Lianas and logging in West Africa. Tropenbos-Cameroon Series 6. PhD thesis,
Wageningen University, The Netherlands.168p.
Pauwels L. 1993. Nzayilu N’Ti. Guide des arbres et arbustres de la région de Kinshasa - Brazaville.
Jardin Botanique National de Belgique, Meise, Belgique. 495p.
Peterjohn W.T. & Correl D.L. 1984. Nutrient dynamics in an agricultural watershed: observation on
the role of a riparian forest. Ecology 65(5): 1466-1475.
Peterson D.L. & Rolfe G.L. 1982. Nutrient dynamics and decomposition of litterfall in floodplain and
upland forest of Central Illinois. Forest Science 28: 667-681.
PGRN/IUCN. 1994. Preparation of a strategy for conservation and management of protected areas:
Ecology and biodiversity components. MDR, Cotonou, Benin. 87p + annexes.
Piegay H. 1997. Riparian forests, a structural compartment of the intra-alpine and piedmont river
hydrosystem. Houille-Blanche 52(1-2): 13-18.
Piegay H. & Bravard J-P. 1997. Response of a Mediterranean riparian forests to a 1 in 400 year flood,
Ouveze River, Drome-Vaucluse, France. Earth Surface Processes and Landforms 22(1): 3143.
Pinay G. & Décamps H. 1988. The role of riparian woods in regulating nitrogen fluxes between
alluvial aquifer and surface water: a conceptual model. Regulated Rivers 2: 507-516.
Poilecot P., Bonfou K., Dosso H., Lauginie F., N’Dri K., Michel N. & Sangaré Y. 1991. Un
écosystème de savane soudanienne: Le parc de la Comoé, Côte d’Ivoire. Projet
UNESCO/PNUD, No. IVC/87/007. Note technique 2 MAB. Paris, France. pp 1-163.
Polansky C. 1994. Sommaire de l’inventaire quantitatif de la forêt classée de Toui-Kilibo.
PGRN/VAF. DFRN/MDR, Cotonou, Bénin. 68 p.
Popma J., Bongers F. & Meave J. 1988. Patterns in the vertical structure of the tropical lowland
rainforest at Los Tuxtlas, Mexico. Vegetatio 74: 81-91.
Porembski S. 2001. Phytodiversity and structure of the Comoé River Gallery forest (NE Ivory-Coast).
Life forms and dynamics in Tropical Forests. Dissertationes Botanicae 346:1-10.
PRRF 1998. Plan d’aménagement participatif de la forêt classée de Pénéssoulou. Projet de
Restauration des Ressources Forestières dans la région de Bassila, PRRF/DFRN/MAEP –
GTZ/Luso - Consult. Bassila, Bénin. 86 p + cartes.
Projet Flore du Bénin (PFB). 1997. Projet d’Aménagement d’un Herbier National et de Rédaction de
la flore Analytique du Bénin. Wageningen et Cotonou. 25p.
Pulliam H.R., Dunning J.B. & Lui J. 1992. Populations dynamics in complex landscapes: a case
study. Ecological Applications 2: 165-177.
Purba K. & Sumarua E. 1987. Chemical analysis of twenty seven wood species from West-Java.
Jurnal Penelitian Hasil Hutan 4: 23-29.
183
References
Rachman O., Balfas J. & Sifat P.K. 1987. Machining properties of wood species from West-Java.
Jurnal Penelitian Hasil Hutan 4: 54-64.
Raunkiaer C. 1934. The life forms of plants and statistical plant geography. Clarendron Press,
London. 632p.
Raynaud J. & Georgy G. 1969. Nature et chasse au Dahomey: Manuel pratique. Secrétariat d’Etat aux
Affaires Etrangères pour la Coopération. Paris, France. 320p.
Reid W.V. 1998. Biodiversity hotspots. Tree 13(7): 275-280.
Reitsma J.M. 1988. Végétation forestière du Gabon. Technical series 1. Tropenbos foundation, Ede,
The Netherlands. 142p.
République du Bénin (RB). 1993. Loi Nº 93-009 du 2 juillet 1993 portant régime des forêts en
République du Bénin. Présidence de la République, Cotonou, Bénin. 26p.
Richards P.W. 1939. Ecological studies on the rain forest of southern Nigeria. I. The structure and
floristic composition of the primary forest. Journal of Ecology 27: 1-61.
Richards P.W. 1952. The tropical rain forest, an ecological study. Cambridge University Press,
Cambridge. 450p.
Richards P.W. 1963. Ecological notes on West African vegetation. II. Lowland forest of the Southern
Bakundu Forest Reserve. Journal of Ecology 51: 123-149.
Robinson C.T., Tockner K & Ward J.V. 2002. The fauna of dynamic riverine landscape. Freshwater
Biology 47: 661-677.
Roché J. 1993. The use of historical data in the ecological zonation of rivers: the case of the ‘tern
zone’. Vie et Milieu 43: 27-41.
Roggeri H. 1995. Zones humides tropicales d'eau douce. Guide des connaissances actuelles et de la
gestion durable. WIW, Union Européenne, Université de Leiden. Kluwer Academic
Publishers, Pays-Bas. 385p.
Roods S.B. & Mahoney J.M. 1990. Collapse of riparian poplar forest downstream from dams in
western prairies: probable causes and prospects for mitigation. Environmental Management
14: 451-464.
Roods S.B., Mahoney J.M., Reid D.E. & Zilm L. 1995. Instream flows and the decline of riparian
cottonwoods along the St. Mary River, Alberta. Canadian Journal of Botany 73:1250-1260.
Rottenborn S.C. 1999. Predicting the impacts of urbanisation on riparian bird communities. Biological
conservation 88: 289-299.
Russell-Smith J. 1991. Classification, species richness, and environmental relations of monsoon rain
forests in northern Australia. Journal of Vegetation Science 2: 259-278.
Saab V. 1999. Importance of spatial scale to habitat use by breeding birds in riparian forests: a
hierarchical analysis. Ecological Applications 9: 135-151.
Sagers C.L. & Lyon J. 1997. Gradient analysis in a riparian landscape: contrasts among forests layers.
Forest Ecology and Management 96: 13-26.
Sala O.E., Chapin I.F.S., Armesto J.J., Berlow E., Bloomfield J., Dirzo R., Huber Sanwald E.,
Huenneke L.F., Jackon R.B., Kinzig A., Leemans R., Lodge D.M., Mooney H.A., Oesterheld
M., Poff N.L., Sykes M.T., Walker B.H., Walker M. & Wall D.H. 2000. Global biodiversity
scenarios for the year 2100. Science 287: 1770-1774.
Salley M.N. & Nieng F.S. 1997. Ouvrage sur l'aménagement durable des forêts. Etude FAO des forêts
n° 122. FAO Rome, Italie. 305p.
Salo J., Kalliola R., Häkkinen I., Mäkinen Y., Niemelä P., Puhakka M. & Coley P.D. 1986. River
dynamics and the diversity of Amazon lowland forest. Nature 322: 254-258.
184
References
Sayer J.A. & Wegge P. 1992. Biological conservation issues in forest management. In: Blockhus
J.M., Dillenbeck M., Sayer J.A. & Wegge P. (eds.), Conserving biological diversity in
managed tropical forests. Workshop held at the IUCN General Assembly at Perth, Australia,
30 november – 01 december 1990. pp 1-4.
Schmitz A. 1963. Aperçu sur les groupements végétaux du Katanga. Bulletin Société Royale
Botanique de Belgique 96(2): 233-347.
Schmitz A. 1971. La végétation de la plaine de Lubumbashi (Haut-Katanga). Publications INEAC,
R.D.C (Zaire). Série Scientifique no. 113. 388p.
Schmitz A. 1988. Révision des groupements végétaux décrits du Zaire, du Rwanda et du Burundi.
Musée Royal de l’Afrique Centrale, Tervuren, Belgique. Annales Sciences Economiques 17:
315p.
Schnabel R.R., Cornish L.F., Stout W.L. & Shaffer J.A. 1996. Denitrification in grassed and a
wooded, valley and ridge, riparian ecotone. Journal of Environmental Quality 25: 1230-1235.
Schnell R. 1952. Contribution à l’étude phytosociologique et phytogéographique de l’Afrique
occidentale. Les groupements et les unités géobotaniques de la région Guinéenne. Mémoire
IFAN, Dakar 18:145-234.
Schnell R. 1971. Intoduction à la phytogéographie des pays tropicaux. Volume II: Milieux, les
groupements végétaux. Edition Gauthier-Villars, Paris, France. pp 503-951.
Schnell R. 1976. La flore et la végétation de l’Afrique tropicale. Volume III, 1ère partie. Edition
Gauthier-Villars, Paris, France. 460 p.
Schreckenberg K. 1996. Forests, field and markets: A study of indigenous Tree Products in the
Woody savannas of the Bassila region. PhD thesis. Department of Geography, School of
African and Oriental studies. University of London, UK. 326p.
Sedell J.R., Bisson P.A., Swanson F.J. & Gregory S.V. 1988. What we know about trees that fall into
streams and rivers. In: Maser C., Tarrant R.F. & Trappe J.M. (eds.), From the Forest to Sea: a
story of fallen trees. General Technical Report PNW-GTR-229. USDA Forest Service Pacific
Northwest Resource Station, Portland, OR. pp 47-81.
Shure D.J., Gottschalk M.R. & Parsons K.A. 1986. Litter decomposition processes in a floodplain
forest. American Midland Naturalist 115: 314-327.
Simberloff D. & Gotelli N. 1984. Effects of insularisation on plant species richness in the prairieforest ecotone. Biological Conservation 29: 27-46.
Sinadouwirou Th.A. 1997. Forêts rémanentes et conservation de la biodiversité: Etude écologique de
quelques forêts dans le département de l’Atlantique au Sud-Bénin. Mémoire d’Ingénieur
Agronome. DAGE/FSA-UNB, Abomey-Calavi, Bénin. 132p.
Sinadouwirou Th.A. 2000. Produits forestiers non ligneux et développement durable: structure des
peuplements naturels et importance socio-économique du Pentadesma butyracea dans la
région de Bassila au Bénin. Mémoire de Master FASA/Université de Dschang, CRESA ForêtBois, Yaoundé, Cameroon. 78 p.
Sinsin B. 1991. Exploitation des pâturages dans un système traditionnel d’élevage bovin dans le
périmètre Nikki-Kalalé. IVth International Rangeland Congress. Montpellier, France 2: 660662.
Sinsin B., Daouda I., Ahokpè E., Téhou A., Toko I., Coubéou P. & Fandohan S. 1997. Faune sauvage
des forêts classées de Goungoun de la Sota et des environs. MDR/DFRN. Projet UNSO.
Laboratoire d’Ecologie Appliquée (LEA/FSA/UNB). Cotonou, Bénin. 158p.
Sinsin B., Eyog-Matig O., Sinadouwirou Th.A. & Assogbadjo A. 2002a. Caractérisation écologique
des essences fourragères Khaya senegalensis (Desr.) et Afzelia africana Sm. suivant les
gradients de latitude et de station au Bénin. In: Eyog-Matig O., Gaoué O.G. & Obel-Lawson
E. (eds.), Development of appropriate conservation strategies for African forest trees
185
References
identified as priority species by SAFORGEN member countries. IPGRI Nairobi, Kenya. pp
15-50.
Sinsin B., Nobimé G., Téhou A., Bekhuis P. & Tchibozo S. 2002b. Past and present distribution of
red-bellied monkey Cercopithecus erythrogaster erythrogaster in Benin. Folia
primatologia 73: 116-123.
Skagen S.K., Melcher C.P., Howe W.H. & Knopf F.L. 1998. Comparative use of riparian corridors
and oases by migrating birds in southeast Arizona. Conservation Biology 12: 896-909.
Skoglund J. 1990. Seed dispersing agents in two regularly flooded river sites. Canadian Journal of
Botany 68: 754-760.
Smith S.D., Willington A.B., Nachlinger J.L. & Fox C.A. 1991. Functional responses of riparian
vegetation to streamflow diversion in the eastern Sierra Nevada. Ecological Applications 1:
89-97.
Smith R.J. & Schaefer J.M. 1992. Avian characteristics of urban riparian strip corridor. Wilson
Bulletin 104: 732-738.
Snedecor G.W. & Cochran W.G. 1989. Statistical methods. 8th edition. Iowa State University Press,
Iowa Ames. 503p.
Sokpon N. 1995. Recherches écologiques sur la forêt dense semi-décidue de Pobè au Sud-Est du
Bénin. Groupements végétaux, structure, régénération naturelle et chute de litière. Thèse
Doctorat en Sciences Agronomiques. Université Libre de Bruxelles, Belgique. 350 p.
Sokpon N. & Adjakidjè V. 1999. Inventaire et caractérisation des espèces et formations végétales des
zones humides du Sud-Bénin (Rapport provisoire). Projet d’Aménagement des Zones
Humides (PAZH). Cotonou, Bénin. 35p.
Sokpon N. & Ouinsavi C. 2002. Utilisations du Khaya senegalensis en médecine traditionnelle au
Benin. Revue de Médecines et Pharmacopées Africaines 16: 9-19.
Sokpon N., Sinadouwirou Th., Gbaguidi F. & Biaou H. 2001. Aperçu sur les forêts édaphiques
hygrophiles du Bénin. Belgian Journal of Botany 134(1): 79-93.
Sonké B. & Lejoly J. 1998. Biodiversity study in Dja fauna reserve (Cameroon): using the transect
method. In: Huxley C.R., Lock J.M. & Cutler D.F. (eds.), Chorology, Taxonomy and Ecology
of the Floras of Africa and Madagascar. Royal Botanic Gardens, Kew. pp 171-179.
Sounon B.B. 2001. Facteurs pédologiques, biologiques et socio-économiques de mise en jachère des
terres dans le secteur de Wari-Maro - Igbomakoro. Mémoire DEA, Option Gestion de
l’environnement. FLASH-UAC, Abomey-Calavi, Bénin. 89p.
Sowunmi M.A. 1986. Change of vegetation with time. In: Lawson G.W. (ed.), Plant Ecology in West
Africa. pp 273-307.
Sowunmi M.A. 1999. The significance of oil palm (Elaeis guineensis Jacq.) in the late Holocene
environments of west and west central Africa: a further consideration. Vegetation History and
Archeobotany 8: 199-210.
Spackman S.C. & Hughes J.W. 1995. Assessment of stream corridor width for biological
conservation: species richness and distribution along mid-order streams in Vermont, USA.
Biological Conservation 71:325-332.
Sparovek G., Ranieri S.B.L., Gassner A., de Maria I.C., Schnug E.E., dos Santos R.F. & Joubert A.
2002. A conceptual framework for the definition of the optimal width of riparian forests.
Agriculture, Ecosystems and Environment 90: 169-175.
Stauffer D.F. & Best L.B. 1980. Habitat selection by birds of riparian communities: evaluating effects
of habitat alterations. Journal of Wildlife Management 44: 1-15.
STATISTICA® 5.1. 1998. Logiciel Statisitca. Editions 1998. Maisons Alfort, Paris Statsoft, France.
186
References
Sterck F.J. & Bongers F. 2001. Crown development in tropical rain forests trees: patterns with tree
height and light availability. Journal of Ecology 89: 1-13.
Stromberg J.C. & Patten D.T. 1990. Variation in seed size of a Southwestern riparian tree, Arizona
walnut (Juglans major). American Midland Naturalist 124: 269-277.
Stromberg J.C., Richter B.D., Patten D.T. & Wolden L.G. 1993a. Response of a Sonoran riparian
forest to a 10-year return flood. Great Basin Naturalist 53: 118-130.
Stromberg J.C., Wilkins S.D. & Tress J.A. 1993b. Vegetation-hydrology models as management tools
for velvet mesquite (Prosopis velutina) riparian ecosystems. Ecological Applications 3: 307314.
Stromberg J.C., Tiller R. & Richter B. 1996. Effects of groundwater decline on riparian vegetation of
semiarid regions: The San Pedro, Arizona. Ecological Applications 6(1): 113-131.
Stromberg J.C. 1997. Growth and survivorship of Fremont cottonwood, Gooding willow, and salt
cedar seedlings after large floods in central Arizona. Great Basin Naturalist 57: 198-208.
Stromberg J.C. 2001. Biotic integrity of Platanus wrightii riparian forests in Arizona: first
approximation. Forest Ecology and Management 142: 251-266.
Stutz De Ortega L.C. 1987. Etudes floristiques de divers stades secondaires des formations forestières
du Haut Parana (Paraguay Oriental). Structure, composition floristique et régénération
naturelle: comparaison entre la forêt primaire et la forêt sélectivement exploitée. Candollea
42(1): 205-262.
Swaine M.D. & Hall J.B. 1976. An application of ordination to the identification of forest types.
Vegetatio 32(2): 83-86.
Swaine M.D., Hall J.B. & Alexander I.J. 1987. Tree population dynamics at Kade, Ghana (19681982). Journal of Tropical Ecology 3: 331-345.
Swaminathan S.M. 1990. Foreword. In: McNeely J.A., Miller K.M., Reid W.V., Mittermeier R.A. &
Werner T. (eds.), Conserving the world’s biological diversity. IUCN, Gland Switzerland;
WRI, CI, WWF- US and the Word Bank, Washington D.C. 155p.
Swanson F.J. & Lienkamper G.W. 1982. Interaction among fluvial process, forest vegetation and
aquatic ecosystems, South Fork Hoh River, Olympic National Park. In: Starkey E.E., Franklin
J.F. & Mattews J.W. (eds.), Ecological research in the national parks of the Pacific Northwest.
Oregon State University, Corvalis, Oregon. pp 30-35.
Swanson F.J., Gregory S.V., Sedell J.R. & Campbell H.G. 1982. Land-water interactions: the riparian
zone. In: Edmonds R.L. (ed.), Analysis of coniferous forest ecosystems in the western United
States. Hutchinson Ross Publishing Company. Stroudsburg, Pennsylvania. pp 267-291.
Swanson F.J., Johnson S.L., Gregory S.V. & Acker A.S. 1998. Flood disturbance in a forested
mountain landscape. BioScience 48: 681-689.
Szaro R.C. 1980. Factors influencing bird populations in southwestern riparian forests. Management
of western forests and grassland for nongame birds. In: USDA Forest Service General
Technical Report INT-86. Intermountain Forest and range Experiment Station, Ogden, Utah.
pp 403-418.
Szaro R.C. & Jakle M.D. 1985. Avian use of a desert riparian island and its adjacent shrub habitat.
Condor 87: 511-519.
Szaro R.C. 1990. Southwestern riparian forests plant communities: site characteristics, tree species
distributions, and size-class structures. Forest Ecology and Management 33-34: 315-334.
Tabacchi E., Planty-Tabacchi A-M. & Décamps O. 1990. Continuity and discontinuity of the riparian
vegetation along a fluvial corridor. Landscape Ecology 5(1): 9-20.
Taylor J.R., Cardonne M.A. & Mitsch W.J. 1990. Bottomland hardwood forests: their functions and
values. In: Gosselink J.P., Lee L.C. & Muir T.A. (eds.), Ecological Processes and Cumulative
187
References
impacts: Illustrated by Bottomland Hardwood Wetlands Ecosystems. Lewis Publishers,
Chelsea, MI. pp 13-86.
Tenté A.B.H. 2000. Dynamique actuelle de l'occupation du sol dans le massif de l'Atacora: Secteur
Perma-Toukountouna. Mémoire DEA, Spécialité Gestion des Ressources Naturelles,
Aménagement du Territoire et Politique Environnementale. FLASH/UAC, Abomey-Calavi,
Bénin. 83p.
Ter Braak C.J.F. 1987. The analysis of vegetation-environment relationships by canonical
correspondence analysis. Vegetatio 69: 69-77.
Ter Braak C.J.F. 1995. Ordination, Chapter 5. In: Jongman R.H.G., Ter Braak C.J.F. & Van Tongeren
O.F.R. (eds), Data Analysis in Community and Landscape Ecology. Cambridge University
Press, Cambridge, UK. pp 91-173.
Ter Braak C.J.F. & Smilauer P. 1998. CANOCO reference manual and User’s guide to Canoco for
Windows. Software for Canonical Community Ordination (version 4). Centre for Biometry
Wageningen, CPRO-DLO. Wageningen, the Netherlands. 352p.
Texier H., Colleuil B., Profizi J-P. & Dossou C. 1980. Le Lac Nokoué, environnement du domaine
margino-littoral sud-béninois. Bulletin de l’Institut Géologique du Bassin d’Aquitaine
Bordeaux 28: 115-142.
Thomas D.H.L. 1996. Dam construction and ecological change in the riparian forest of the HadejiaJama'are floodplain, Nigeria. Land Degradation and Development 7(4): 279-295.
Thompson S.K. 1992. Sampling. John Wiley, New York. 343p.
Thompson S.K. & Seber G.A.F. 1996. Adaptive sampling. John Wiley, New York.
Tilman D. 1988. Plant strategies and the dynamics and structure of plant communities. Princeton
University Press, Princeton. 360p.
Tongeren O. & Prentice C. 1986. A spatial simulation for vegetation dynamics. Vegetatio 65: 163173.
Touré S.F. 1991. Arbres et arbustes fourragers dans l’alimentation des ruminants en zones sahéliennes
et soudaniennes. Valeur nutritive espèces appétées (Sénégal). IVth International Rangeland
Congress. Montpellier, France 2: 670-673.
Tossou G.M. 2002. Recherche palynologique sur la végétation holocène du sud-Bénin (Afrique de
l’Ouest). Thèse de Doctorat de l’Université de Lomé, Faculté des Sciences, Togo. 136p.
Trimble S.W. 1999. Decreased rates of alluvial sediment storage in the Coon creek basin, Wisconsin,
1975-1993. Science 285: 1244-1246.
Trochain J.L. 1957. Accord interafricain sur la définition des types de végétation de l’Afrique
Tropicale. Bulletin de l’Institut d’Etude Centrafricaines, nouvelle série no. 13-14: 55-93.
Tuani G.K., Cobbinah J.R. & Agbodaze P.K. 1994. Bioactivity and phytotechemical studies on
extractives from some Ghanaian plants. Ghana Journal of Forestry 1: 44 - 48.
UNCED. 1992. Biological Diversity Convention. United Nations Conference on Environment and
Development. Rio de Janeiro June 1992, Brazil.
van Groenewoud H. 1975. Theoretical considerations on the covariation of plant species along
ecological gradients with regard to multivariate analysis. Journal of Ecology 63: 837-47.
van der Maarel E. 1990. Ecotones and ecoclines are different. Journal of Vegetation Science 1: 135138.
van Etten E.J.B. 1999. Vegetation maps and geographic information systems as tools for the
management and planning of arid lands. VIth International Rangeland Congress Proceedings
1: 251-253.
van Rompaey R.S.A.R. 1993. Forest gradients in West Africa: a spatial gradient analysis. PhD Thesis.
Wageningen University, The Netherlands. 138p.
188
References
Varty N. 1990. Ecology of small mammals in the Riverine forests of the Jubba Valley, Southern
Somalia. Journal of Tropical Ecology 6: 179-189.
Vivien J. & Faure J.J. 1985. Arbres des forêts d’Afrique Centrale. Ministère des Relations Extérieures
de la Coopération et du Développement. Paris, France. 526 p.
Vought L.B.M., Dahl J., Pedersen C.L. & Lacoursiére J.O. 1994. Nutrient retention in riparian
ecotones. Ambio 23: 324-348.
Wakentin I.G. & Reed J.M. 1999. Effects of habitat type and degradation on avian species richness in
great Basin riparian habitats. Great Basin Naturalist 59: 205-212.
Ward J.V., Tockner K., Arscott D.B. & Claret C. 2002. Riverine landscape diversity. Freshwater
Biology 47: 517-539.
Warner R.E. 1979. California riparian study program. Background information and proposed study
design. California Resources Agency, Department of Fish and Game, Sacramento, California
(USA). 177p.
Warner R.E. & Katibath E.F. 1980. Measurement techniques for inventorying riparian systems. An
International Workshop, November 30-December 6, 1980. La Paz, Mexico. pp 487-495.
Weller D.E., Correll D.L. & Jordan T.E. 1994. Denitrification in riparian forests receiving agricultural
discharges. In: Mitsch W.J. (ed.), Global wetlands: Old Worlds and New. Elsevier,
Amsterdam. pp 117-132.
Whitaker D.M. & Montevecchi W.A. 1997. Breeding bird assemblages associated with riparian,
interior forest, and non riparian edge habitats in a balsam fir ecosystems. Canadian Journal of
Forest Resources 27: 1159-1167.
White F. 1979. The Guineo-Congolian region and its relationships to other phytocoria. Bulletin du
Jardin Botanique National de Belgique 49: 11-55.
White F. 1986. La végétation de l’Afrique. Mémoire accompagnant la carte de végétation de
l’Afrique. UNESCO/AETFAT/UNSO. 384p.
Whitmore T.C. 1990. An introduction to tropical rain forests. Clarendon Press, Oxford. 225p.
Whittaker R.H. 1972. Evolution and measurement of species diversity. Taxon 21: 213-251.
Whittaker R.H. 1975. Communities and ecosystems. 2nd edition. Macmillan Comp, New York. 385p.
Willems H.P.L., Rotelli M. D., Berry D.F., Smith E.P., Reneau R.B. & Mostaghimi S. 1997. Nitrate
removal in riparian wetlands soils: effects of flow rate, temperature, nitrate concentration and
soil depth. Water Resources 31(4): 841-849.
Williams P.H., Humphries C.J. & Vane-Wright R.I. 1991. Measuring biodiversity: Taxonomic
relatedness for conservation priorities. Australian Systematic Botany 4: 665-79.
Williamson M. 1975. The design of wildlife reserves. Nature 256: 519.
Woinarski J.C.Z., Brock C., Armstrong M., Hempel C., Cheal D. & Brennan K. 2000. Bird
distribution in riparian vegetation in the extensive natural landscape of Australia’s tropical
savanna: a broad-scale survey and analysis of a distributional data base. Journal of
Biogeography 27: 843-868.
Wotto J. 2001. Dynamique des pâturages naturels dans le système agraire de la sous-préfecture de
Péhunco. Mémoire de Maîtrise en Géographie. FLASH-UNB, Abomey-Calavi, Bénin. 116p.
Xiang W.N. 1993. Application of a GIS-based stream buffer generation model to environmental
policy evaluation. Environment Management 17: 817-827.
Yédomonhan H. 2002. Contribution à l’étude des formations saxicoles au centre méridional du Bénin.
Mémoire de Maîtrise en Géographie. FLASH/UAC, Abomey-Calavi, Bénin. 115p.
Zepernick B. & Timler F.K. 1984. Guttiferae. In: Brunel J.F., Hiepko P., Scholz H. (eds), Flore
analytique du Togo, Phanerogames, Englera 4. GTZ, Berlin. pp 216-217.
189
ANNEX
LIST OF RIPARIAN FORESTS PLANT SPECIES OF BENIN
A.K. Natta
Riparian forests plant species of Benin
List of riparian forests plant species of Benin
no.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
LF
PT
Lnph
Lnph
Lnph
Ch
Lmph
mPh
mph
mPh
mph
mph
Th
Th
Lmph
He
He
Lnph
Lmph
Lnph
Lnph
Ch
Lmph
Th
Th
Ch
Ep
Ch
Th
Ge
Ge
mPh
Th
LmPh
LmPh
LmPh
mPh
mPh
mph
mPh
Lmph
mph
mph
Th
Th
Th
Th
Th
mPh
Ge
Ge
Ge
Ge
Lmph
Lmph
Lmph
mph
Ge
Ge
TA
TA
Pan
TA
TA
S
SZ
SZ
SZ
_
Pan
Cosmo
GC
SG
Pan
GC
GC
GC
G
SZ
TA
TA
GC
Pan
GC
PRA
Pan
GC
_
SZ
Pan
GC
GC
GC
PRA
GC
GC
GC
GC
GC
GC
TA
Pan
Cosmo
Pan
Pan
S
G
G
G
G
GC
GC
PRA
Pan
G
SZ
Families
Papilionaceae
Papilionaceae
Papilionaceae
Malvaceae
Mimosaceae
Mimosaceae
Mimosaceae
Mimosaceae
Mimosaceae
Mimosaceae
Euphorbiaceae
Amaranthaceae
Malpighiaceae
Poaceae
Poaceae
Passifloraceae
Passifloraceae
Passifloraceae
Passifloraceae
Papilionaceae
Cucurbitaceae
Asteraceae
Asteraceae
Adiantaceae
Orchidaceae
Papilionnaceae
Papilionnaceae
Zingiberaceae
Zingiberaceae
Caesalpiniaceae
Asteraceae
Apocynaceae
Apocynaceae
Apocynaceae
Mimosaceae
Mimosaceae
Mimosaceae
Mimosaceae
Euphorbiaceae
Sapindaceae
Sapindaceae
Poaceae
Amaranthaceae
Amaranthaceae
Amaranthaceae
Amaranthaceae
Papilionnaceae
Araceae
Araceae
Araceae
Araceae
Vitaceae
Vitaceae
Vitaceae
Anacardiaceae
Araceae
Araceae
Plant list
Abrus canescens Welw. ex Wight & Arn.
Abrus fruticulosus Wall. ex Wight & Arn syn. A. pulchellus Wall. ex Thwaites
Abrus precatorius L.
Abutilon mauritianum (Jacq.) Medic.
Acacia erythrocalyx Brenan syn. Acacia pennata (L.) Willd.
Acacia gourmaensis A. Chev.
Acacia macrostachya Reichenb. ex DC.
Acacia polyacantha Willd. subsp. campylacantha (Hochst. ex A. Rich.) Brenan
Acacia sieberiana DC. var. sieberiana
Acacia sp.
Acalypha ciliata L.
Achyranthes aspera L.
Acridocarpus alternifolius (Schum. & Thonn.)
Acroceras amplectens Stapf.
Acroceras zizanoides (Kunth.) Dandy
Adenia cissampeloides (Planch. ex Hook) Harms
Adenia lobata (Jacq.) Engl.
Adenia rumicifolia Engl. & Harms var. miegei
Adenia tennuispira (Stapf.) Engl.
Adenodolichos paniculatus (Hua) Hutch. & Dalz. syn. Dolichos paniculatus Hua
Adenopus brevifolius Benth.
Adenostemma caffrum DC. var. caffrum
Adenostemma perrottetii DC.
Adiantum philippensis L.
Aerangis biloba (Lindl.) Schltr. excl.
Aeschynomene afraspera J. Leonard
Aeschynomene indica L.
Aframomum sceptrum (Oliv. & Hanb.) K. Schum.
Aframomum sp.
Afzelia africana Smith ex Pers.
Ageratum conyzoides L. subsp. conyzoides
Alafia barteri Oliv.
Alafia benthamii (Baill. ex Stapf) Stapf var. benthamii
Alafia scandens (Thonning) De Wild. syn A. landolphioides
Albizia coriaria Welw. ex Oliv.
Albizia ferruginea (Guill. & Perr.) Benth.
Albizia glaberrima (Schum. & Thonn.) Benth.
Albizia zygia (DC.) J. F. Macbr.
Alchornea cordifolia (Schum. & Thonn.) Müll. Arg.
Allophyllus spicatus (Poir.) Radlk.
Allophyllus africanus P. Beauv. syn A. cobbe (L.) Raeusch.
Alloteropsis paniculata (Benth.) Stapf. syn. Urochloa paniculata Benth.
Alternanthera sessilis (L.) DC.
Amaranthus hybridus L.
Amaranthus spinosus L.
Amaranthus viridis L.
Amblygonocarpus andongensis (Welw. ex Oliv.) Exell & Torre syn. Tetrapleura andongensis
Amorphophallus aphyllus (Hook.) Hutch.
Amorphophallus dracontioides (Engl.) N. E. Br.
Amorphophallus flavovirens N. E. Br.
Amorphophallus johnsonii N. E. Br.
Ampelocissus bombycina Planch.
Ampelocissus leonensis (Hook. f.) Planch.
Ampelocissus multistriata (Baker) Planch. syn. A. pentaphylla (Guill. & Perr.) Gilg. & Brandt
Anacardium occidentale L.
Anchomanes difformis (Blume) Engl.
Anchomanes welwitschii Rendle
193
Riparian forests plant species of Benin
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
Lmph
mph
He
He
He
He
He
He
He
He
He
Lnph
Lmph
nph
mPh
Ge
Th
mPh
mPh
mPh
MPh
mph
mph
mph
LGe
LGe
Lnph
Ge(Fern)
Lnph
Th
Th
Th
Th
Ch
Ge(Fern)
Th
Th
He
mPh
mph
He
Lnph
Th
mPh
Th
Th
Ch
mPh
mph
Th
Th
Th
Th
Ge(Fern)
mPh
mPh
He
Th
He
Th
GC
SZ
SG
SG
SG
GC
_
SG
GC
TA
GC
Pan
G
PRA
PRA
_
PRA
TA
G
GC
GC
GC
TA
GC
SG
AA
GC
TA
SZ
TA
GC
G
G
SG
GC
G
Pan
Pan
Paleo
SZ
Pan
SG
TA
TA
Pan
Pan
TA
Pan
GC
SG
SG
Pan
Cosmo
PRA
SZ
SZ
TA
Paleo
SZ
Paleo
Apocynaceae
Ancylobotrys scandens (Schumach. & Thonn.) Pichon
Papilionaceae
Andira inermis (Wright) DC. subsp. rooseveltii (De Wild.) Gillett ex Polhill
Poaceae
Andropogon gayanus Kunth. var. bisquamulatus (Hochst.) Hack.
Poaceae
Andropogon gayanus Kunth. var. gayanus
Poaceae
Andropogon gayanus Kunth. var. polycladus (H.) W.D. Clayton syn. A. gayanus var. squamulatus
Poaceae
Andropogon macrophyllus Stapf.
Poaceae
Andropogon sp.
Poaceae
Andropogon tectorum Schum. & Thonn.
Commelinaceae Aneilema beninense (P. Beauv.) Kunth
Commelinaceae Aneilema dispermum Brenan
Commelinaceae Aneilema umbrosum (Vahl) Kunth
Convolvulaceae Aniseia martinicensis (Jacq.) Choisy
Asclepiadaceae Anisopus manii N.E.Br.
Annonaceae
Annona senegalensis Pers.
Combretaceae
Anogeissus leiocarpa (DC.) Guill. & Perr. syn. A. leiocarpus (DC.) Guill. & Perr.
Liliaceae
Anthericum sp.
Melastomataceae Antherotoma naudini Hook. F.
Loganiaceae
Anthocleista djalonensis A. Chev.
Loganiaceae
Anthocleista nobilis G. Don
Loganiaceae
Anthocleista vogelii Planch.
Moraceae
Antiaris toxicaria Lesch. var africana A. Chev. syn. A. africana Engl.
Euphorbiaceae
Antidesma membranaceum Müll. Arg.
Euphorbiaceae
Antidesma venosum Tul.
Sapindaceae
Aphania senegalensis (Juss. ex Poir.) Radlk. syn. Lepisanthes senegalensis (Juss. ex Poir.) Leenh
Aristolochiaceae Aristolochia albida Duchartre
Aristolochiaceae Aristolochia ringens Vahl.
Annonaceae
Artabotrys velutinus Sc. Elliot
Davalliaceae
Artopteris orientalis (Gmel.) Posth.
Liliaceae
Asparagus africanus Lams
Asteraceae
Aspilia africana (Pers.) C.D. Adams
Asteraceae
Aspilia angustifolia Oliv. &Hiern.
Asteraceae
Aspilia bussei (Schum. & Thonn.) Oliv. & Hiern
Asteraceae
Aspilia paludosa Berhaut
Asteraceae
Aspilia rudis C. D. Adams subsp. fontinaloides Adams
Aspleniaceae
Asplenium diplazorum Hieron.
Acanthaceae
Asystasia calycina Benth.
Acanthaceae
Asystasia gangetica (L.) T. Anders
Poaceae
Axonopus flexuosus (Peter) C.E. Hubbard ex Troupin
Meliaceae
Azadirachta indica A. Juss.
Balanitaceae
Balanites aegyptiaca (L.) Del.
Poaceae
Bambusa vulgaris Schrader ex Wendel.
Acanthaceae
Barleria oenotheroides Dum.
Lamiaceae
Basilicum ploystachion (L.) Moench.
Caesalpiniaceae Berlinia grandiflora (Vahl) Hutch. & Dalz.
Asteraceae
Bidens pilosa L.
Oxalidaceae
Biophytum umbraculum Welw. syn. B. petersianum Klotszch
Acanthaceae
Blepharis maderaspatensis (L.) Heyne ex Roth.
Sapindaceae
Blighia sapida Koenig
Sapindaceae
Blighia unijugata Baker
Asteraceae
Blumea viscosa (Mill.) Badillo syn. B. aurita DC.
Nyctaginaceae
Boerhavia coccinea Mill.
Nyctaginaceae
Boerhavia diffusa L.
Nyctaginaceae
Boerhavia erecta L.
Lomariopsidaceae Bolbitis heudelotii (Bory ex Fee) Aston
Bombacaceae
Bombax costatum Pellegr. & Vuill.
Arecaceae
Borassus aethiopum Mart.
Poaceae
Brachiaria brizantha (Hochst. ex A. Rich.) Stapf
Poaceae
Brachiaria deflexa (Schumach.) Robyns
Poaceae
Brachiaria jubata (Figari & De Notaris) Stapf.
Poaceae
Brachiaria lata (Schumach.) C.E. Hubbard
194
Riparian forests plant species of Benin
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
He
Th
mph
mph
mPh
mph
mPh
Th
Ge
mph
nph
nph
nph
Ep(fern)
Ep(fern)
nph
nph
mph
Lmph
Lmph
Th
Th
Lnph
Lnph
Lnph
Lnph
mph
Th
nph
Lnph
Lmph
Lnph
Th
Lnph
Lnph
mph
mph
Ch
mPh
Par
MPh
Th
LTh
LTh
mPh
Th
Lnph
Th
Ge
Ge
Ge
nph
Lnph
Th
nph
mph
He
He
He
He
_
SZ
GC
SZ
TA
SC
SZ
Pan
_
SZ
Paleo
Paleo
GC
GC
GC
GC
SG
GC
GC
Pan
Pan
Pan
SG
GC
SG
GC
TA
GC
S
SZ
G
G
Pan
AA
Pan
Pan
SZ
_
GC
Pan
Pan
Paleo
GC
Paleo
SZ
G
AA
TA
G
_
G
Pan
GC
SZ
GC
SZ
PRA
GC
GC
PRA
Poaceae
Poaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Rubiaceae
Cyperaceae
Cyperaceae
Caesalpiniaceae
Capparidaceae
Papilionaceae
Apocynaceae
Orchidaceae
Orchidaceae
Ochnaceae
Ochnaceae
Ochnaceae
Celastraceae
Papilionaceae
Gentinaceae
Gentinaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Euphorbiaceae
Capparidaceae
Capparidaceae
Capparidaceae
Capparidaceae
Solanaceae
Sapindaceae
Sapindaceae
Caricaceae
Caesalpiniaceae
Caesalpiniaceae
Rhizophoraceae
Lauraceae
Bombacaceae
Amaranthaceae
Amaranthaceae
Amaranthaceae
Ulmaceae
Poaceae
Papilionaceae
Adiantaceae
Asclepiadaceae
Asclepiadaceae
Asclepiadaceae
Caesalpiniaceae
Menispermaceae
Poaceae
Rubiaceae
Oleaceae
Poaceae
Liliaceae
Liliaceae
Liliaceae
Brachiaria sp.
Brachiaria villosa (Lam.) A. camus syn. B. distichophylla (Trin.) Stapf.
Bridelia atroviridis Müll. Arg.
Bridelia ferruginea Benth.
Bridelia micrantha (Hochst.) Baill.
Bridelia scleroneura Müll. Arg.
Brenadia salicina (Vahl) Hepper & Wood syn. Adina microcephala (Del.) Hiern
Bulbostylis barbata (Rottb.) C.B.Cl.
Bulbostylis sp.
Burkea africana Hook.
Cadaba farinosa Forsk.
Cajanus cajan (L.) Millsp.
Callichilia barteri (Hook. f.) Stapf syn. Hedranthera brateri (Hook. f.) pichon
Calyptrochilum christyanum (Rchb. f.) Summerh.
Calyptrochilum emarginatum (Sw.) Schltr.
Campylospermum flavum (Schumach.&Thonn. ex Stapf) Farron syn. Ouratea flava
Campylospermum glaberrimum (P. Beauv.) Farron syn. Ouratea glaberrima
Campylospermum reticulatum (P.Beauv) Farron var. reticulatus syn. Ouratea reticulata
Campylostemon warneckeanum Loes. ex Fritsch
Canavalia ensiformis (L.) DC.
Canscora decussata (Roxb.) Roem. & Schult.
Canscora diffusa (Vahl) R. Br. ex Roem. & Schult
Canthium cornelia Cham. & Schlecht.
Canthium henriquesianum (K. Schum.) G. Tayl.
Canthium horizontale (Schum. & Thonn.) Hiern
Canthium setosum Hiern
Canthium vulgare (K. Schum.) Bullock. F.R.
Caperonia senegalensis Müll. Arg.
Capparis decidua (Forsk.) Edgew.
Capparis sepiaria L. var. fischeri (Pax) De Wolf syn. C. corymbosa Lam.
Capparis thonningii Schum. syn. C. brassii DC.
Capparis viminea Hook. f. & Th. ex. Oliv. var. viminea
Capsicum frutescens L.
Cardiospermum grandiflorum Swartz
Cardiospermum halicacabum L.
Carica papaya L.
Cassia sieberiana DC.
Cassia sp.
Cassipourea congoensis R. Br. ex DC.
Cassytha filiformis Linn.
Ceiba pentandra (L.) Gaertn.
Celosia argentea L.
Celosia isertii C. Towns. syn. C. laxa Schum. & Thonn.
Celosia trigyna L.
Celtis toka (Forssk) Hepper & Wood syn. C. integrifolia
Centosteca latifolia (Osb.) Trin. syn. C. lappacea (L.) Desv.
Centrosema pubescens Benth.
Ceratopterix cornuta (P. Beauv.) Lepr.
Ceropegia fusiformis N. E. Br.
Ceropegia sp.
Ceropegia yorubana Schlechter
Chamaecrista mimosoides (L.) Greene syn. Cassia mimosoides L.
Chasmanthera dependens Hochst.
Chasmopodium caudatum (Hack.) Stapf
Chassalia kolly (Schumach.) Hepper
Chionanthus niloticus (Oliv.) Stearn syn. Linociera nilotica Oliv.
Chloris gayana Kunth
Chlorophytum alismifolium Baker
Chlorophytum andongense Baker
Chlorophytum blepharophyllum Schweinf. ex. Baker
195
Riparian forests plant species of Benin
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
He
Ge
Ge
Ch
mPh
Lnph
Lnph
Lnph
Lnph
Lnph
Lnph
Lnph
Lnph
Lnph
Lmph
LTh
LGe
LTh
mph
nph
mPh
Th
Th
Lnph
Lnph
Lnph
Lnph
nph
MPh
mPh
mph
Lmph
mph
mph
mph
mph
Lnph
mph
LmPh
mph
LmPh
Lmph
Lnph
He
He
He
He
Th
He
He
Lmph
Th
Th
Th
mph
mph
nph
Ge
mph
Lmph
Liliaceae
Chlorophytum macrophyllum (A. Rich.) Aschers.
GC
Liliaceae
Chlorophytum sp.
_
Liliaceae
Chlorophytum togoense Engl.
SG
Asteraceae
Chromolaena odorata (L.) R.M.King & H. Robinson
Pan
Chrysobalanaceae Chrysobalanus icaco L. subsp. atacorensis
SC
Menispermaceae Cissampelos mucronata A. Rich.
SZ
Menispermaceae Cissampelos owariensis P. Beauv. ex DC.
GC
Vitaceae
Cissus aralioides (Welw. ex Baker) Planch.
TA
Cissus cymosa Schum. & Thonn.
PRA Vitaceae
Vitaceae
Cissus glaucophylla Hook.
G
Vitaceae
Cissus gracilis Guill. & Perr.
TA
Vitaceae
Cissus kouandeensis A. Chev.
S
Vitaceae
Cissus palmatifida (Baker) Planch.
SG
Cissus petiolata Hook. f.
PRA Vitaceae
Cissus populnea Guill. & Perr. var. populnea
PRA Vitaceae
Vitaceae
Cissus rubiginosa (Welw. ex Baker) Planch
TA
Cissus rufescens Guill. & Perr.
PRA Vitaceae
Vitaceae
Cissus sp.
_
Aurantiaceae
Citrus limonum Risso
Pan
Rutaceae
Clausena anisata (Willd.) Benth.
TA
Annonaceae
Cleistopholis patens (Benth.) Engl. & Diels
GC
Cappariadaceae Cleome rutidosperma DC. syn. C. ciliata Schum. & Thonn.
S
Cappariadaceae Cleome viscosa L.
SG
Clerodendrum capitatum (Willd.) Schum. & Thonn.
PRA Verbenaceae
Verbenaceae
Clerodendrum dusenii Gurke
GC
Verbenaceae
Clerodendrum polycephalum Baker
GC
Coccinia grandis L.
Paleo Cucurbitaceae
Cochlospermaceae Cochlospermum planchonii Hook. f.
SG
Sterculiaceae
Cola gigantea A. Chev.
GC
Sterculiaceae
Cola laurifolia Mast.
GC
Sterculiaceae
Cola millenii K. Schum.
G
Combretaceae
Combretum acutum Laws.
S
Combretaceae
Combretum adenogonium Steud. ex A. Rich. syn. C.ghasalense Engl. & Diels
SG
Combretaceae
Combretum collinum Fresen subsp. collinum
SG
Combretaceae
Combretum collinum Fresen subsp. hypopilinum
S
Combretaceae
Combretum glutinosum Perr. ex DC.
SZ
Combretaceae
Combretum lecardii Engl. & Diels
G
Combretaceae
Combretum molle R. Br. ex G. Don
SZ
Combretaceae
Combretum mucronatum Schum. & Thonn. syn. C. smeathmannii G. Don
GC
Combretaceae
Combretum nigricans Lepr. ex Guill. & Perr.
SG
Combretaceae
Combretum paniculatum Vent.
TA
Combretaceae
Combretum racemosum P. Beauv.
GC
Combretaceae
Combretum tomentosum G. Don
G
Commelinaceae Commelina benghalensis L. var. bengalensis
GC
Commelinaceae Commelina diffusa Bum. f. subsp. diffusa
Pan
Cosmo Commelinaceae Commelina erecta L. subsp. erecta
Commelinaceae Commelina erecta subsp livingstonei (C.B. Cl.) J. K. Morton
TA
PRA Commelinaceae Commelina nigritana Benth.
Commelinaceae Commelina sp.
_
Commelinaceae Commelina thomasii Hutch.
G
Connaraceae
Connarus africanus Lam.
SG
Tiliaceae
Corchorus aestuans L.
Pan
Tiliaceae
Corchorus fascicularis Lam.
Pan
Tiliaceae
Corchorus olitorius L.
Pan
Boraginaceae
Cordia africana Lam.
TA
Cordia myxa L.
Paleo Boraginaceae
Boraginaceae
Cordia senegalensis Juss.
GC
Zingiberaceae
Costus afer Ker-Gawl.
TA
Crateva adansonii DC. subsp. adansonii syn. C. religiosa auct. Afr. Fl. non Forst. f.
Paleo Cappariadaceae
Rubiaceae
Cremaspora triflora (Thonn.) K. Schum.
TA
196
Riparian forests plant species of Benin
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
Ge
Ge
Ge
mph
Th
Ch
Th
Th
Th
Ch
Ch
Ch
Th
Th
Lmph
Th
Lnph
Lmph
Ge
mph
nph
Th
Ch
mPh
Ge
Ge
Ge
Ge
Th
LGe
Th
Lmph
Lmph
Lmph
Lmph
LmPh
mPh
mPh
mph
Ch
Ch
nph
Th
Th
Ch
Th
LTh
Lnph
Ch
mph
mPh
mPh
mph
nph
Th
Th
Th
Th
Th
Lnph
SG
GC
SG
SZ
Pan
TA
SC
S
PRA
PRA
Pan
Pan
_
Pan
SG
Pan
PRA
G
PRA
SG
GC
Pan
GC
GC
Pan
Pan
Pan
_
Pan
_
Pan
GC
PRA
G
GC
GC
SG
Pan
G
AA
AA
GC
TA
PRA
GC
_
_
Pan
Pan
SZ
GC
GC
G
TA
TA
TA
Pan
Pan
_
_
Amaryllidaceae
Amaryllidaceae
Amaryllidaceae
Rubiaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Euphorbiaceae
Asclepiadaceae
Poaceae
Cucurbitaceae
Araceae
Hypoxidaceae
Araliaceae
Amaranthaceae
Amaranthaceae
Thelypteridaceae
Caesalpiniaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Vitaceae
Poaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Caesalpiniaceae
Caesalpiniaceae
Annonaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Caesalpiniaceae
Caesalpiniaceae
Caesalpiniaceae
Dichapetalaceae
Mimosaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Dilleniaceae
Crinum glaucum A. Chev.
Crinum jagus (Thomps.) Dandy
Crinum zeylanicum (L.) L. syn. C. ornatum (Alt.) Bury
Crossopteryx febrifuga (G. Don) Benth.
Crotalaria calycina Schrank
Crotalaria cephalotes Steud. ex A. Rich.
Crotalaria comosa Baker
Crotalaria deightonii Hepper
Crotalaria hyssopifolia Klotszch
Crotalaria lachnosema Stapf
Crotalaria pallida var. mucronata Desv.
Crotalaria retusa L.
Crotalaria sp.
Croton lobatus L.
Cryptolepis sanguinolenta (Lind.) Schltr.
Ctenium newtonii Hack
Cucumis metuliferus E. Mey. ex Naudin
Culcasia scandens P. Beauv.
Curculigo pilosa (Schum. & Thonn.) Engl.
Cussonia arborea Hochst. ex A. Rich. syn. C. barteri Seem. & C. kirkii Seem.
Cyathula achyranthoides (H. B. et K.) Moq
Cyathula prostrata (L.) Blume
Cyclosorus striatus Cop.
Cynometra megalophylla Harms
Cyperus difformis L.
Cyperus distans L.f.
Cyperus haspan L.
Cyperus sp.
Cyperus sphacelatus Rottb.
Cyphostema sp.
Dactyloctenium aegyptium (L.) P.Beauv
Dalbergia dalzielii Baker f. ex. Hutch. & Dalz.
Dalbergia lactea Vatke
Dalbergia rufa G. Don
Dalbergia saxatilis Hook. f. var. saxatilis
Dalbergiella welwitschii (Baker) Baker f.
Daniellia oliveri (Rolfe) Hutch. & Dalz.
Delonix regia (Boj. ex Hook.) Raf.
Dennettia tripetala Baker f.
Desmodium adscendens (Sw.) DC var. adscendens
Desmodium adscendens (Sw.) DC var. robustrum Schubert
Desmodium gangeticum (L.) DC.
Desmodium hirtum Guill. & Perr.
Desmodium ramosissimum G. Don
Desmodium salicifolium (Poir.) DC. var. salicifolium
Desmodium sp1.
Desmodium sp2.
Desmodium triflorum (L.) DC.
Desmodium velutinum (Willd.) DC.
Detarium microcarpum Harms
Detarium senegalense J. F. Gmelin
Dialium guineense Willd.
Dichapetalum madagascariense Poir var. madagascariense syn. D. guineense
Dichrostachys cinerea (L.) Wight & Arn
Digitaria argyrotricha (Anderss.) Chiov.
Digitaria ciliaris (Retz.) Koel.
Digitaria gayana (Kunth) Stapf ex A. Chev.
Digitaria horizontalis Willd. ssp. proximus (Hocht. ex A.Rich.) Maire & Weiler
Digitaria sp.
Dillenia sp.
197
Riparian forests plant species of Benin
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
Lmph
LTh
LGe
LGe
LGe
LGe
LGe
LGe
LGe
LGe
LGe
LGe
LGe
LGe
Lnph
mPh
mPh
mph
Ge(fern)
Ch
Ge
Ge
Ge
Ch
mPh
mph
mph
mph
Ch
Th
Th
Th
mPh
mPh
Ch
Th
Th
He
He
Th
Th
mph
mph
Lmph
He
Th
Th
nph
mPh
Ch
Ch
mPh
mph
mPh
mPh
mph
mph
Th
Ge
Ge
Pan
Pan
SG
Pan
SZ
SZ
SG
GC
SG
GC
SG
GC
_
GC
_
GC
SZ
GC
GC
GC
SG
S
S
_
GC
TA
G
GC
G
TA
Pan
Pan
SZ
GC
Pan
SG
Pan
SZ
GC
SG
PRA
TA
PRA
GC
Paleo
_
Paleo
GC
GC
PRA
Pan
GC
SG
GC
GC
TA
GC
Pan
PRA
PRA
Papilionaceae
Rubiaceae
Dioscoraceae
Dioscoraceae
Dioscoraceae
Dioscoraceae
Dioscoraceae
Dioscoraceae
Dioscoraceae
Dioscoraceae
Dioscoraceae
Dioscoraceae
Dioscoraceae
Dioscoraceae
Menispermaceae
Ebenaceae
Ebenaceae
Ebenaceae
Athyriaceae
Umbelliferae
Melastomataceae
Melastomataceae
Melastomataceae
Melastomataceae
Sterculiaceae
Sterculiaceae
Agavaceae
Euphorbiaceae
Acanthaceae
Acanthaceae
Poaceae
Asteraceae
Meliaceae
Arecaceae
Asteraceae
Asteraceae
Poaceae
Poaceae
Acanthaceae
Lamiaceae
Lamiaceae
Mimosaceae
Mimosaceae
Mimosaceae
Poaceae
Poaceae
Poaceae
Acanthaceae
Sapindaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Caesalpiniaceae
Caesalpiniaceae
Erytroxylaceae
Capparidaceae
Poaceae
Orchidaceae
Orchiadaceae
Dioclea reflexa Hook. f.
Diodia sarmentosa Sw. syn. D. scandens auctt
Dioscorea abyssinica Hochst. ex Kunth.
Dioscorea bulbifera L. var. bulbifera
Dioscorea dumetorum (Kunth) Pax
Dioscorea hirtiflora Benth. subsp. hirtiflora
Dioscorea lecardii De Wild.
Dioscorea multiflora Pax syn. D. minutiflora Engl.
Dioscorea odoratissima Pax syn. D. praehensilis Benth.
Dioscorea preussii Pax
Dioscorea sagittifolia Pax
Dioscorea sansibarensis Pax
Dioscorea sp.
Dioscorea togoensis Kunth syn. D. callei A. Chev. ex De Wild.
Dioscoreophyllum sp.
Diospyros abyssinica (Hiern) White
Diospyros mespiliformis Hochst. ex A. DC.
Diospyros monbuttensis Gürke
Diplazium sammatii (Kuhn) C. Chr.
Diplolophium africanum Turcz.
Dissotis anthennima (Sm.) Triana
Dissotis grandiflora (Sm.) Benth. var. lambii (Hutch.) Keay
Dissotis irvingiana Hook. syn. D. senegambiensis (Guill. & Rev.) triana
Dissotis sp.
Dombeya ledermanii Engl.
Dombeya quinqueseta (Del.) Exell
Dracaena arborea Baker
Drypetes floribunda (Müll. Arg.) Hutch.
Dyschoriste heudelotiana (Nees) O. Ktze.
Dyschoriste perrottetii (Nees) O. Kuntze
Echinochloa colona (L.) Link. syn. E. colonum (L.) Link.
Eclipta prostata (L.) L.
Ekebergia capensis Sparrm. syn. E. senegalensis A. Juss.
Elaeis guineensis Jacq.
Elephantopus mollis Kunth
Elephantopus senegalensis (Klatt) Oliv. & Hiern
Eleusine indica Gaertn.
Elymandra androphila (Stapf) Stapf
Elytraria marginata Vahl
Englerastrum gracillimum Th. C. E. Fries
Englerastrum schweinfurthii Briq.
Entada abyssinica Steud. ex A. Rich.
Entada africana Guill. & Perr.
Entada mannii (Oliv.) Tisserant
Eragrostis atrovirens (Desf.) Trin. ex Steud. (Hochst. ex A. Rich.) Maire & Weiler
Eragrostis sp.
Eragrostis tremula Horst. & Steud.
Eremomastax speciosa (Hochst.) Cuf.
Eriocoelum kerstingii Gilg. ex Engl. var. kerstingii
Eriosema glomeratum (Guill & Perr) Hook.f. syn. Rhynchosia glomerata Guill. & Perr.
Eriosema psoraleoides (Lam.) G. Don syn. Crotaliaria psoralioides Lam.
Erythrina excelsa Baker
Erythrina senegalensis A. DC.
Erythrophleum africanum (Wel. ex Benth.) Harms
Erythrophleum suaveolens (Guill. & Pierr.) Brenan syn. E. guineense G. Don
Erythroxylum emarginatum Thonn.
Euadenia trifoliolata (Schum. & Thonn.) Oliv.
Euclasta condylotricha (Steud.) Stapf (Hochst. ex A. Rich.) Maire & Weiler
Eulophia guineensis Lindl.
Eulophia quartiniana A. Rich.
198
Riparian forests plant species of Benin
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
Th
Th
Th
Th
Th
Th
Th
Th
Th
Ch
nph
nph
mph
mPh
mph
mph
mPh
mPh
mPh
mph
mPh
mph
MPh
mPh
mph
mph
mph
mPh
mph
mph
mPh
mph
He
He
Th
Lmph
mph
Lnph
He
Th
nph
Th
Ch
Th
Th
mPh
mPh
mph
mPh
mph
nph
mph
nph
nph
LTh
LTh
Th
Ch
Th
LGe
AA
Pan
Paleo
AA
Pan
Pan
AA
Pan
S
SZ
GC
SZ
SC
GC
GC
TA
SZ
SZ
S
GC
SC
SG
GC
_
SG
SZ
TA
G
TA
GC
GC
Pan
SG
_
GC
GC
GC
GC
SZ
Pan
TA
TA
_
PRA
GC
GC
GC
SZ
SZ
SG
GC
_
SG
GC
Pan
Paleo
_
Pan
GC
Asteraceae
Asteraceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Rubiaceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Moraceae
Cyperaceae
Cyperaceae
Cyperaceae
Malpighiaceae
Flacourtiaceae
Flagellariaceae
Commelinaceae
Commelinaceae
Euphorbiaceae
Urticaceae
Urticaceae
Cyperaceae
Cyperaceae
Apocynaceae
Apocynaceae
Rubiaceae
Clusiaceae
Clusiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Molluginaceae
Iridaceae
Molluginaceae
Liliaceae
Eupatorium microstemon Cass.
Eupatorium triplinerve Vahl.
Euphorbia aegyptiaca Boiss.
Euphorbia glomerifera (Millesp.) Wheeler
Euphorbia heterophylla L.
Euphorbia hirta L.
Euphorbia hyssopifolia L.
Euphorbia sp.
Euphorbia thymifolia L.
Excoecaria grahamii Stapf. syn. Sapium grahamii (Stapf) Prain
Feretia apodanthera Del.
Ficus asperifolia Miq.
Ficus capreifolia Del.
Ficus congensis Engl. syn. Ficus trichopoda Baker
Ficus cordata Thunb. Warb.
Ficus dicranostyla Mildbr.
Ficus exasperata Vahl
Ficus glumosa Del.
Ficus ingens (Miq.) Miq.
Ficus kerstingii Hutch. syn. Ficus abutilifolia (Miq.) Miq.
Ficus lyrata Warb.
Ficus ovata Vahl
Ficus platyphylla Del.
Ficus polita Vahl
Ficus sp.
Ficus sur Forssk. syn. Ficus capensis Thunb.
Ficus sycomorus L. subsp. gnaphalocarpa (Miq.) Berg
Ficus sycomorus L. var psychomorus
Ficus tesselata Warb.
Ficus thonningii Blume syn. F. basarensis Mildbr. & Burret
Ficus vogeliana (Miq.) Miq.
Ficus vogelii (Miq.) Miq.
Fimbristylis dichotoma (L.) Vahl subsp. dichotoma
Fimbristylis hispidula (Vahl.) Kunth
Fimbristylis sp.
Flabellaria paniculata Cav.
Flacourtia flavescens Willd.
Flagellaria guineensis Schumacher
Floscopa africana (P. Beauv.) C. B. Clarke
Floscopa flavida C. B. Cl.
Flueggea virosa (Roxb ex Willd) Voigt subsp.virosa syn. Securinega virosa (Roxb ex Willd)
Fluerya aestuans (L.) ex Miq.
Fluerya ovalifolia (Schum. & Thonn.) Dandy
Fuirena sp.
Fuirena stricta Steud.
Funtumia africana (Benth.) Stapf
Funtumia elastica (Preuss) Stapf
Gaertnera paniculata Benth.
Garcinia livingstonei T. Anders.
Garcinia ovalifolia Oliv.
Gardenia erubescens Stapf & Hutch.
Gardenia imperialis K. Schum.
Gardenia sp.
Gardenia ternifolia Schum. & Thonn. syn. G. triacantha DC.
Geophila obvallata (Schumach.) F. Didr.
Geophila reniformis D. Don syn. G. repens (L.) I.M. Johnston
Gisekia pharnacioides L.
Gladiolus sp.
Glinus oppositifolus (L.) Aug. DC.
Gloriosa superba L. syn. G. simplex L.
199
Riparian forests plant species of Benin
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
mph
mPh
Lmph
Ep
Lmph
nph
nph
nph
Lnph
Ge
Ge
MPh
Th
Ch
Th
mPh
mph
Th
Th
nph
Ch
Ch
Lmph
Ch
Ch
Ch
Ch
nph
Th
Lmph
mPh
nph
Ch
Th
Th
Th
mph
nph
Th
He
Th
Th
Th
Ge
Ge
Th
Th
Th
He
Th
Th
Ch
Th
Lnph
Ch
Th
Ch
Th
Th
Th
TA
Pan
TA
PRA
GC
Paleo
SZ
Paleo
Pan
PRA
TA
GC
Pan
Paleo
SG
GC
SZ
TA
TA
G
Paleo
Paleo
Pan
PRA
TA
Pan
PRA
_
Pan
GC
TA
PRA
Pan
PRA
Pan
_
SZ
PRA
S
Pan
_
GC
_
GC
GC
AA
Paleo
Pan
Pan
SZ
GC
SG
GC
GC
GC
PRA
TA
S
_
Pan
Tiliaceae
Verbenaceae
Asclepiadaceae
Orchidaceae
Tiliaceae
Tiliaceae
Tiliaceae
Tiliaceae
Asclepiadaceae
Orchidaceae
Amaryllidaceae
Simaroubaceae
Boraginaceae
Boraginaceae
Compositeae
Annonaceae
Annonaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Hippocrateaceaae
Apocynaceae
Lamiaceae
Violaceae
Hippocrateaceaae
Acanthaceae
Acanthaceae
Euphorbiaceae
Rubiaceae
Poaceae
Poaceae
Poaceae
Acanthaceae
Acanthaceae
Marantaceae
Marantaceae
Lamiaceae
Lamiaceae
Lamiaceae
Poaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Glyphaea brevis (Spreng.) Monachino
Gmelina arborea Roxb.
Gongronema latifolium Benth.
Graphorkis lurida (Sw.) O. Ktze.
Grewia carpinifolia Juss.
Grewia flavescens Juss.
Grewia mollis Juss. var. mollis syn. G. pubescens P. Beauv.
Grewia ternax (Forssk.)
Gymnema sylvestre (Retz.) Schultes
Habenaria laurantii De Wild.
Haementhus multiforus Martyn
Hannoa klaineana Pierre & Engl. syn. Quassia undulata (Guill. & Perr.) D. Dietr.
Heliotropium indicum L.
Heliotropium strigosum Willd.
Herderia truncata Cass.
Hexalobus crispiflorus A. Rich.
Hexalobus monopetalus (A. Rich.) Engl. & Diels var. monopetalus
Hibiscus asper Hook. F.
Hibiscus esculentus L.
Hibiscus grewioides Baker. F.
Hibiscus lunariifolis Willd.
Hibiscus micranthus L.
Hibiscus panduriformis Burm. f.
Hibiscus physaloides Guill. & Perr.
Hibiscus rostellatus Guill. & Perr.
Hibiscus sabdariffa L.
Hibiscus sidiformis Baill.
Hibiscus sp.
Hibiscus surratensis L.
Hippocratea welwitchii Oliv. syn. Simirestis welwitchii (Oliv.) Halle
Holarrhena floribunda (G. Don) Dur. & Schinz
Hoslundia opposita Vahl
Hybanthus enneaspermus (L.) F. Muell.
Hydrolea glabra Schum. & Thonn.
Hygrophila auriculata (Schumach.) Heine
Hygrophila sp.
Hymenocardia acida Tul. var. acida
Hymenodiction floribundum (Steud. & Hochst.) B.L.Rob.
Hyparrhenia involucrata Stapf
Hyparrhenia rufa (Nees) Stapf
Hyparrhenia sp.
Hypoestes cancellata Nees
Hypoestes sp.
Hypselodelphys violacea (Ridley) Milne. Redh.
Hypselodelphys poggeana (K. Schum.) Milne-Redh.
Hyptis lanceolata Poir
Hyptis spicigera Lam.
Hyptis suaveolens Poit.
Imperata cylindrica (L.) Raeuschel
Indigofera dendroides Jacq.
Indigofera geminata Baker
Indigofera hirsuta L. var hirsuta
Indigofera longicalyx Gillett
Indigofera macrophylla Schum.
Indigofera nigritana Hook. f.
Indigofera polysphaera Baker
Indigofera rhynchocarpa Baker
Indigofera secundiflora Poir.
Indigofera sp.
Indigofera subulata Vahl ex Poir.
200
Riparian forests plant species of Benin
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
Ch
Lnph
LGe
Lnph
Lnph
Lnph
Lnph
Lnph
LGe
Lnph
Lnph
mPh
mPh
mPh
mPh
Ch
mPh
mph
Lmph
Lmph
Lnph
Lnph
Lmph
Th
Ge
Lmph
Lnph
Lmph
MPh
mPh
mPh
Ch
Ge
Ge
Ge
Ge
Th
Th
LmPh
LMPh
Lmph
Lmph
mPh
mPh
mPh
mPh
mph
Ch
He
Lnph
Lmph
Lnph
Th
LmPh
Lmph
mph
mph
Th
Th
Ge
TA
S
Pan
SG
Pan
Pan
TA
Pan
Pan
PRA
_
GC
PRA
SZ
SZ
G
GC
PRA
PRA
SG
TA
G
GC
TA
TA
GC
PRA
GC
GC
SZ
GC
PRA
PRA
AA
Cosmo
_
G
Pan
GC
PRA
_
G
PRA
PRA
SZ
G
GC
TA
Pan
TA
Paleo
PRA
TA
GC
_
Pan
GC
GC
_
TA
Papilionaceae
Convolvulaceae
Convolvulaceae
Convolvulaceae
Convolvulaceae
Convolvulaceae
Convolvulaceae
Convolvulaceae
Convolvulaceae
Convolvulaceae
Convolvulaceae
Simaroubaceae
Simaroubaceae
Caesalpiniaceae
Caesalpiniaceae
Lamiaceae
Annonaceae
Rubiaceae
Oleaceae
Oleaceae
Oleaceae
Oleaceae
Connaraceae
Acanthaceae
Zingiberaceae
Rubiaceae
Rubiaceae
Rubiaceae
Meliaceae
Meliaceae
Bignoniaceae
Asteraceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Compositeae
Asteraceae
Apocynaceae
Apocynaceae
Apocynaceae
Apocynaceae
Anacardiaceae
Anacardiaceae
Anacardiaceae
Anacardiaceae
Sapindaceae
Leeaceae
Poaceae
Convolvulaceae
Asclepiadaceae
Asclepiadaceae
Poaceae
Papilionaceae
Papilionaceae
Mimosaceae
Gentianaceae
Scrophulariaceae
Scrophulariaceae
Cyperaceae
Indigofera trita L.f.
Ipomoea acanthocarpa (Hochst. ex Choisy) Aschers. & Schweinf.
Ipomoea alba L.
Ipomoea argentaurata Forssk.
Ipomoea asarifolia (Desr.) Roen. & Schult.
Ipomoea eriocarpa R. Br.
Ipomoea involucrata P. Beauv.
Ipomoea mauritiana Hall. f.
Ipomoea muricata (L.) Jacq.
Ipomoea rubens Choisy
Ipomoea sp.
Irvingia gabonensis (Aubry-Lecomte ex O'Rorke) Baill.
Irvingia smithii Hook. F.
Isoberlinia doka Craib & Stapf.
Isoberlinia tomentosa (Harms) Craib & Stapf.
Isodictyophorus reticularus (A. Chev.) J. K. Morton
Isolona thonneri (De wild. Th. Dur.) Engl. & Diels
Ixora brachypoda DC.
Jasminum dichotomum Vahl
Jasminum obtusifolium Baker
Jasminum pauciflorum Benth.
Jasminum preussii Engl. & Knobl.
Jaundea pinnata (P. Beauv.) Schellenb.
Justicia anselliana (Nees) T. Anderson
Kaempferia aethiopica (Solms-Laub.) Benth.
Keetia hispida (Benth.) Bridson
Keetia multiflora (Schum.& Thonn.) Bridson syn. Canthium multiflorum (S.&T.) Hiern
Keetia venosa (Oliv.) Bridson syn. Canthium venosum (Oliv.) Hiern
Khaya grandifoliola C. DC.
Khaya senegalensis (Desv.) A. Juss.
Kigelia africana (Lam.) Benth.
Kinghamia macrocephala (Oliv. & Hiern.) Jeffrey
Kyllinga erecta Schumach. var. erecta
Kyllinga odorata Vahl
Kyllinga pumila Michx.
Kyllinga sp.
Laggera alata (D.Don) Sch. Bip. ex Oliv.
Laggera aurita (L.f.) Benth ex C.B. Clarke syn. Blumea aurita (L.f.) DC.
Landolphia hirsuta (Hua) Pichon
Landolphia owariensis P. Beauv.
Landolphia sp.
Landolphia togolana (Hallier f.) Pichon
Lannea acida A. Rich.
Lannea kerstingii Engl. & K. Krause syn. Lannea barteri (Oliv.) Engl.
Lannea microcarpa Engl. & K. Krause
Lannea nigritana (Sc. Elliot) Keay var. nigritana
Lecaniodiscus cupanioides Planch.
Leea guineensis G. Don
Leersia hexandra Sw.
Lepistemon owariense (P. Beauv.) Hall. f.
Leptadenia arborea (Forsk.) Schweinf.
Leptadenia hastata (Pers.) Decne.
Leptochloa caerulescens Steud.
Leptoderris brachyptera (Benth.) Dunn.
Leptoderris sp.
Leucaena leucocephala Benth.
Lindackeria dentata (Oliv.) Gilg.
Lindernia diffusa (L.) Wettst.
Lindernia sp.
Lipocarpha albiceps Ridl.
201
Riparian forests plant species of Benin
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
Ge
Ge
nph
Th
LmPh
LmPh
LmPh
mph
mPh
mph
He
He
Th
He
He
Th
Th
Th
Th
Ch
Lnph
Th
He(Fern)
nph
mph
mph
mPh
Ge
mPh
mPh
mph
mph
mPh
Ge
Ge
mph
He
Th
Ge
Ge
He
He
mph
Th
mph
mph
He
Th
Ch
Ch
Th
mph
Lnph
Lnph
Ge
Lmph
mph
Ch
Lmph
MPh
TA
TA
SG
PRA
TA
_
GC
PRA
PRA
GC
TA
_
S
TA
_
PRA
TA
TA
_
TA
Pan
Cosmo
Pan
SG
TA
PRA
Pan
Pan
G
TA
SG
SG
G
G
_
TA
Pan
Paleo
AA
TA
GC
_
GC
PRA
SZ
PRA
GC
TA
G
Pan
_
G
AA
Pan
_
G
GC
Paleo
Pan
G
Cyperaceae
Cyperaceae
Verbenaceae
Campanulaceae
Hippocrateaceae
Hippocrateaceae
Papilionaceae
Papilionaceae
Papilionaceae
Ochnaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Onagraceae
Onagraceae
Onagraceae
Onagraceae
Onagraceae
Cucurbitaceae
Solanaceae
Lycopodiaceae
Rubiaceae
Capparidaceae
Euphorbiaceae
Anacardiaceae
Euphorbiaceae
Euphorbiaceae
Sapotaceae
Chrysobalanaceae
Chrysobalanaceae
Chrysobalanaceae
Maranthaceae
Maranthaceae
Euphorbiaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Bignoniaceae
Marsileaceae
Celastraceae
Celastraceae
Marantaceae
Asteraceae
Melastomataceae
Sterculiaceae
Sterculiaceae
Melastomataceae
Convolvulaceae
Convolvulaceae
Eriocaulaceae
Caesalpiniaceae
Euphorbiaceae
Asteraceae
Asteraceae
Moraceae
Lipocarpha atra Ridl.
Lipocarpha prieuriana Steud.
Lippia multiflora Moldenke
Lobelia sapinii De Wild.
Loeseneriella africana Willd. var. africana syn. Hippocratea africana (Willd.) Loes.
Loeseneriella sp.
Lonchocarpus cyanescens (Schum. & Thonn.) Benth.
Lonchocarpus laxiflorus Guill. & Perr. syn. Philenoptera laxiflora (Guill. & Perr.) G. Roberty
Lonchocarpus sericeus (Poir.) H. B. & K.
Lophira lanceolata Van Tiegh. ex Keay
Loudetia phragmitoides (Peter) C. E. Hubb.
Loudetia sp.
Loudetia togoensis (Pilg.) C. E. Hubb.
Loudetiopsis ambiens (K. Schum.) Conert
Loudetiopsis sp.
Ludwigia abyssinica A. Rich. syn. Jussiaea abyssinica (A. Rich.) Dandy & Brenan
Ludwigia hyssopifolia (G. Don) Exell syn. Jussiaea linifolia Vahl
Ludwigia octovalvis (Jacq.) Raven
Ludwigia sp.
Ludwigia stenorraphe (Brenan) Hara subsp. stenorraphe
Luffa cylindrica (L.) M. J. Roem syn. Luffa aegyptiaca Mill
Lycopersicon esculentum Mill syn. Solanum lycopersicon L.
Lycopodium cernuum L.
Macrosphyra longistyla (DC.) Hiern
Maerua angolensis DC.
Mallotus oppositifolius (Geisel.) Müell. Arg. var. oppositifolius
Mangifera indica L.
Manihot esculenta Crantz
Manihot glaziovii Müll. Arg.
Manilkara obovata (Sabine & G. Don) syn. M. multinervis (Baker) Dubard
Maranthes kerstingii (Engl.) Prance
Maranthes polyandra (Benth.) Prance
Maranthes robusta (Oliv.) Prance syn. Parinari robusta Oliv.
Maranthochloa purpurea (Ridley) Milne. Redh.
Maranthochloa sp.
Margaritaria discoidea (Baill.) Webster
Mariscus cylindristachyus Steud. syn. M. alternifolius auct.
Mariscus dubius (Rottb.) C.E.C. Fisch.
Mariscus flabelliformis Kunth
Mariscus longibracteatus Cherm.
Mariscus soyauxii (Boech.) C.B. Clarke
Mariscus sp.
Markhamia tomentosa (Benth.) K. Schum. ex Engl.
Marsilea crenulata Desv.
Maytenus heterophylla (Eckl. & Zeyh.) Robson
Maytenus undatus (Thunb.) Blakelock
Megaphrynium macrostachyum (Benth.) Milne-Redh.
Melanthera scandens (Schum. & Thonn.) Rob.
Melastomastrum segregatum (Benth) A.&R.Fern. syn. Dissotis segregata (Benth) H.f.
Melochia corchorifolia Linn.
Melochia sp.
Memecylon afzelii G. Don var. afzelii
Merremia cissoides (Lam.) Hallier f.
Merremia hederacea Burm. f.
Mesanthemum sp.
Mezoneuron benthanianum Baill.
Microdesmis puberula Hook. f. ex Planch.
Microglossa pyrifolia (Lam.) Kuntze
Mikania cordata (Burm. f.) B. L. Robins var. cordata syn. M. Chenopodiifolia Willd.
Milicia exelsa (Welw.) Berg syn. Chlorophora excelsa (Welw.) benth.
202
Riparian forests plant species of Benin
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
Lmph
mPh
Lmph
nph
mPh
mph
mPh
Th
Lnph
Lnph
Lnph
Lnph
Lnph
Ch
Ch
mph
mph
mph
mph
mph
LmPh
Lmph
LmPh
Lmph
LTh
He
He
Lmph
Lmph
Th
mph
Th
LTh
mph
Ge
Ge
Ge
Ge
mPh
mPh
Ge
mph
mph
mph
mph
mph
Ch
Lmph
Th
Th
Th
Th
Th
Lmph
Lmph
mph
LmPh
Th
Th
Ch
GC
G
G
Pan
GC
SZ
SZ
Paleo
Pan
G
GC
G
G
TA
SZ
G
SZ
SZ
SG
G
GC
TA
Pan
Pan
Paleo
TA
Pan
GC
GC
_
G
Pan
GC
G
Pan
_
PRA
PRA
GC
G
Pan
G
GC
SG
SZ
_
Paleo
GC
Pan
Paleo
AA
_
GC
G
G
SZ
SZ
SG
SG
TA
Papilionaceae
Papilionaceae
Papilionaceae
Mimosaceae
Sapotaceae
Sapotaceae
Rubiaceae
Molluginaceae
Cucurbitaceae
Cucurbitaceae
Cucurbitaceae
Annonaceae
Annonaceae
Acanthaceae
Acanthaceae
Annonaceae
Dipterocarpaceae
Rubiaceae
Rubiaceae
Rubiaceae
Apocynaceae
Papilionaceae
Papilionaceae
Papilionaceae
Cucurbitaceae
Commelinaceae
Musaceae
Rubiaceae
Rubiaceae
Najadaceae
Lecythidaceae
Acanthaceae
Papilionaceae
Annonaceae
Davalliaceae
Davalliaceae
Orchidaceae
Orchidaceae
Boraginaceae
Simaroubaceae
Nymphaeaceae
Ochnaceae
Ochnaceae
Ochnaceae
Ochnaceae
Ochnaceae
Lamiaceae
Olacaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Poaceae
Apocynaceae
Apocynaceae
Flacourtiaceae
Opiliaceae
Poaceae
Poaceae
Papilionaceae
Millettia barteri (Benth.) Dunn.
Millettia thonningii (schum. & Thonn.) Baker
Millettia warneckei Harms
Mimosa pigra L.
Mimusops andogensis Hiern
Mimusops kummel Bruce ex A. DC.
Mitragyna inermis (Willd.) O. Kuntze
Mollugo nudicaulis Lam.
Momordica balsamina L.
Momordica charantia L.
Momordica cissoides Benth.
Monanthotaxis parviflora (Oliv.) Verdc.
Monanthotaxis whytei (Stapf) Verdc.
Monechma ciliatum (T. Anders.) C. B. Cl.
Monechma depauperatum (Jacq.) Milne
Monodora tenuifolia Benth.
Monotes kerstingii Gilg
Morelia senegalensis A. Rich.
Morinda geminata DC.
Morinda lucida Benth.
Motandra guineensis (Thonning) A. DC.
Mucuna poggei (L.) DC. var. poggei
Mucuna pruriens (L.) DC. var. pruriens
Mucuna sloanei Fawc. & Rendle
Mukia maderaspatana (L.) M. J. Roem. syn. Melothria maderaspatana (L.) Cogn.
Murdannia simplex (Vahl) Brenan
Musa sapientum L.
Mussaenda elegans Schum. & Thonn.
Mussaenda isertiana DC.
Naja sp.
Napoleonaea vogelii Hook. & Planch. syn. N. leonensis Hutch. & Dalz.
Nelsonia canescens (Lam.) Spreng.
Neorautanenia mitis (A. Rich.) Verdc.
Neostenanthera myristicifolia (Oliv.) Exell
Nephrolepis biserrata (Sw.) Schott
Nephrolepis sp.
Nervilia petraea (Afzel. ex Pers.) Summerch.
Nervilia umbrosa (Rchb.f.) Schltr.
Newbouldia laevis (P. Beauv.) Seem. ex Bureau
Nothospondias staudtii Engl.
Nymphaea lotus L.
Ochna afzelii R. Br. ex Oliv.
Ochna membranacea Oliv.
Ochna rhizomatosa (van Tiegh.) Keay
Ochna schweinfurthiana F. Hoffm.
Ochna sp.
Ocimum gratissimum L.
Olax subscorpioidea Oliv. var. subscorpioidea
Oldenlandia corymbosa L.
Oldenlandia herbacea (L.) Roxb.
Oldenlandia lancifolia (Schumach.) DC.
Oldenlandia sp.
Olyra latifolia L.
Oncinotis glabrata (Baill.) Stapf ex Hiern
Oncinotis nitida Benth.
Oncoba spinosa Forssk.
Opilia amentacea Roxb. syn. O. celtidifolia (Guill. & Perr) Endl. ex Walp.
Oplismenus burmannii (Retz.) P. Beauv.
Oplismenus hirtellus (L.) P. Beauv.
Ormocarpum sennoides (Willd.) DC.
203
Riparian forests plant species of Benin
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
He
He
Th
mph
nph
nph
nph
mph
mph
Lnph
He
nph
Lnph
mph
Th
Ch
Th
He
Th
LGr
MPh
mph
mPh
LmPh
He
Lnph
Lmph
mph
nph
nph
Th
Th
Th
He
mPh
Th
Lmph
mph
Th
Ch
Th
Ch
Th
mph
He
He
Ch
Th
mph
Ch
LmPh
Th
Lnph
Th
nph
Th
mPh
mph
Th
Th
Pan
Pan
PRA
GC
GC
GC
_
TA
GC
GC
TA
PRA
Pan
SG
PRA
SZ
GC
G
_
GC
GC
SZ
SZ
SG
Pan
Pan
AA
SG
SZ
SG
Paleo
Pan
_
TA
GC
TA
TA
SG
Paleo
PRA
TA
GC
TA
TA
PRA
Paleo
G
Pan
Pan
Pan
TA
Paleo
TA
_
_
Pan
G
TA
Pan
Paleo
Poaceae
Osmundaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Asclepiadaceae
Poaceae
Anacardiaceae
Papilionaceae
Pandanaceae
Amaranthaceae
Amaranthaceae
Poaceae
Poaceae
Poaceae
Aristolochiaceae
Chrysobalanaceae
Chrysobalanaceae
Mimosaceae
Asclepiadaceae
Poaceae
Passifloraceae
Sapindaceae
Rubiaceae
Rubiaceae
Rubiaceae
Poaceae
Poaceae
Poaceae
Poaceae
Clusiaceae
Rubiaceae
Asclepiadaceae
Papilionaceae
Poaceae
Acanthaceae
Acanthaceae
Acanthaceae
Acanthaceae
Arecaceae
Poaceae
Poaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Euphorbiaceae
Solanaceae
Simaroubaceae
Caesalpiniaceae
Araceae
Lamiaceae
Oryza sativa L.
Osmunda regalis L.
Otomeria elatior (A. Rich. ex. DC.) Verdc.
Oxyanthus formosus Hook. f. ex Planch.
Oxyanthus pallidus Hiern
Oxyanthus racemosus (Schum. & Thonn.) Keay
Oxyanthus sp.
Oxyanthus speciosus DC.
Oxyanthus unilocularis Hiern
Oxystelma bornouense R. Br.
Oxytenanthera abyssinica (A. Rich.) Munro
Ozoroa insignis Del. syn. Heeria insignis (del.) O. Ktze. Rev.
Pachyrrhizus angulatus Rich.
Pandanus candelabrum P. Beauv.
Pandiaka angustifolia (Vahl) Hepper
Pandiaka involucrata (Moq.) B. D. Jackson
Panicum brevifolium L.
Panicum maximum Jacq.
Panicum sp.
Pararistolochia goldieana (Hook. f.) Hutch. & Dalz.
Parinari congensis F. Didr.
Parinari curatellifolia Planch. ex Benth.
Parkia biglobosa (Jacq.) R. Br. ex G. Don f.
Parquetina nigrescens (Afzel.) Bullock
Paspalum scrobiculatum L. syn. P. orbiculare G. Forest
Passiflora foetida L.
Paullinia pinnata L.
Pavetta corymbosa (DC.) F. N. Williams
Pavetta crassipes K. Schum.
Pavetta oblongifolia (Hiern) Bremek.
Pennisetum pedicellatum Trin.
Pennisetum polystachion (L.) Schult.
Pennisetum sp.
Pennisetum unisetum (Nees) Benth. syn. Beckeropsis uniseta (Nees & Thonn.)
Pentadesma butyracea Sab.
Pentodon pentandrus (Schum. & Thonn.) Vatke
Pergularia daemia (Forssk.) Chiov.
Pericopsis laxiflora (Benth.) van. Meeuwen
Perotis indica (L.) Kuntze
Phaulopsis barteri (T. Anders.) Lindau
Phaulopsis ciliata (Willd.) Hepper
Phaulopsis imbricata (Forssk.) Sweet. subsp imbricata
Phaulopsis sylvestre (Lindau) Lindau
Phoenix reclinata Jacq.
Phragmites australis subsp. altissimus (Benth.) W. D. Clayton
Phragmites karka (Retz.) Trin. ex Steud.
Phyllanthus alpestris Beille
Phyllanthus amarus Schum. & Thonn.
Phyllanthus kerstingii Brunel (ined.) syn. P. beillei auct.
Phyllanthus maderaspatensis L.
Phyllanthus muellerianus (O. Ktze.) Exell
Phyllanthus niruri L.
Phyllanthus reticulatus Poir. var. reticulatus
Phyllanthus sp1.
Phyllanthus sp2. (Yarpao stream)
Physalis angulata L.
Pierreodendron kerstingii (Engl.) Little
Piliostigma thonningii (Schumach.) milne-Redh.
Pistia stratrioides L.
Platostoma africanum P. Beauv.
204
Riparian forests plant species of Benin
718
719
Ep
Th
TA
TA
Polypodiaceae
Platycerium angolense Welw.
Caryophyllaceae Polycarpaea eriantha Hochst. ex A. Rich.
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
Th
He
mph
Ep
Th
nph
mph
Th
Lmph
Lmph
Lmph
Lnph
mPh
mph
mPh
Lnph
mph
Lnph
mph
nph
nph
nph
nph
nph
nph
nph
mph
Ch
mPh
mPh
Lnph
Th
Ge
He
He
Lmph
mPh
Lnph
mph
Lmph
mph
Th
LTh
LTh
LTh
LTh
Lnph
He
Th
nph
mph
Lmph
Lmph
Lnph
mph
nph
Th
Cosmo
TA
SZ
_
Pan
SG
GC
PRA
GC
G
G
_
SZ
SZ
PRA
TA
Pan
GC
SG
G
GC
GC
GC
GC
_
GC
SG
Pan
SZ
PRA
Paleo
Pan
Cosmo
Pan
_
Paleo
SZ
G
G
Paleo
TA
Pan
GC
Paleo
Pan
_
GC
Pan
Paleo
Pan
G
G
G
_
GC
_
Pan
Polygonaceae
Polygonaceae
Rubiaceae
Orchidaceae
Portulacaceae
Rubiaceae
Sapotaceae
Urticaceae
Verbenaceae
Verbenaceae
Verbenaceae
Verbenaceae
Mimosaceae
Meliaceae
Anacardiaceae
Papilionaceae
Myrtaceae
Papilionaceae
Hypericaceae
Hypericaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Combretaceae
Adianthaceae
Papilionaceae
Papilionaceae
Papilionaceae
Amaranthaceae
Cyperaceae
Cyperaceae
Cyperaceae
Combretaceae
Arecaceae
Icacinaceae
Apocynaceae
Celastraceae
Anacardiaceae
Poaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Cyperaceae
Poaceae
Euphorbiaceae
Violaceae
Capparidaceae
Capparidaceae
Capparidaceae
Rubiaceae
Rubiaceae
Poaceae
Polygonum salicifolium Brouss. ex Willd. syn. Persicaria salicifolia (Brouss. ex Willd)
Assenov (1966)
Polygonum senegalense Meisn. syn. Persicaria senegalensis (Meisn.) Sojak
Polysphaeria arbuscula K. Schum.
Polystachya sp.
Portulaca oleracea Linn.
Pouchetia africana A. Rich. ex DC.
Pouteria alnifolia (Baker) Roberty syn. Malacantha alnifolia (Baker) Pierre
Pouzolzia guineensis Benth
Premna angolensis Gürke
Premna hispida Benth.
Premna lucens A. Chev.
Premna sp.
Prosopis africana (Guill. & Perr.) Taub.
Pseudocedrela kotschyi (Schweinf.) Harms
Pseudospondias microcarpa (A. Rich) Engl.
Pseudovigna sp.
Psidium guajava L.
Psophocarpus tetragonolobus (L.) DC.
Psorospermum glaberrimum Hochr.
Psorospermum senegalense Spach
Psychotria calva Hiern
Psychotria latistipula Benth.
Psychotria linderi Hepper
Psychotria psychotrioides (DC.) Roberty
Psychotria sp.
Psychotria vogeliana Benth.
Pteleopsis suberosa Engl. & Diels
Pteris atrovirens Willd.
Pterocarpus erinaceus Poir subsp. lucens
Pterocarpus santalinoides DC.
Pueraria phaseoloides (Roxb.) benth. var javanica
Pupalia lappacea (L.) A. Juss.
Pycreus lanceolatus (Poir.) C.B. Clarke
Pycreus polystachyos (Rottb.) P.Beauv.
Pycreus sp.
Quisqualis indica L.
Raphia sudanica A. Chev.
Raphiostylis beninensis (Hook. f. ex Planch.) Planch. ex Benth.
Rauvolfia vomitoria Afzel.
Reissantia indica (Willd.) N. Hallé syn. Hippocratea indica Lam.
Rhus natalensis Bernh. ex Krause
Rhynchelytrum roseum (Nees) Stapf. & C.E. Hubbard ex Bews
Rhynchosia congensis Baker
Rhynchosia densiflora (Roth) DC.
Rhynchosia minima (L.) DC. var. minima
Rhynchosia sp.
Rhynchosia viscosa (Roth) DC.
Rhynchospora corymbosa (L.) Britt.
Rhytachne triaristata (Steud.) Stapf
Ricinus communis L.
Rinorea dentata (P. Beauv.) O. Ktze
Ritcheia capparioides (Andr.) Britten var. capparoides
Ritcheia longipedicellata Gilg.
Ritcheia sp.
Rothmannia urcelliformis (Hiern) Robyns
Rothmannia sp.
Rottboellia cochinchinensis (Lour.) syn. R. exaltata L. f.
205
Riparian forests plant species of Benin
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
nph
Ch
Ch
Ch
Ch
Lmph
mph
mph
nph
nph
LmPh
LmPh
Lnph
Lnph
Lnph
He
Lmph
Lmph
nph
nph
nph
Lmph
Ge
Ge
Ge
mph
Lmph
Th
Ge
Th
Th
He
He
Th
Lnph
MPh
He
He
He
He
He
He
Th
Lmph
mph
Th(fern)
nph
nph
nph
Th
Ch
Ch
Th
nph
nph
Ch
Th
Th
He
He
TA
GC
G
GC
G
_
GC
SG
_
GC
AA
GC
GC
SG
_
SZ
GC
GC
PRA
G
_
GC
GC
GC
G
TA
TA
AA
TA
Pan
PRA
PRA
Pan
_
Pan
GC
TA
GC
GC
AA
GC
_
TA
GC
SZ
_
Pan
AA
Pan
Paleo
G
SG
TA
GC
Pan
SG
Paleo
SG
TA
GC
Connaraceae
Acanthaceae
Acanthaceae
Acanthaceae
Acanthaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Apocynaceae
Apocynaceae
Rubiaceae
Rubiaceae
Rubiaceae
Poaceae
Celastraceae
Celastraceae
Celastraceae
Celastraceae
Celastraceae
Celastraceae
Agavaceae
Agavaceae
Agavaceae
Euphorbiaceae
Rubiaceae
Ochnaceae
Amaryllidaceae
Poaceae
Poaceae
Poaceae
Poaceae
Poaceae
Mimosaceae
Oleaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Cyperaceae
Scrophulariaceae
Asclepiadaceae
Polygalaceae
Selaginaceae
Caesalpiniaceae
Caesalpiniaceae
Caesalpiniaceae
Caesalpiniaceae
Caesalpiniaceae
Rubiaceae
Pedaliaceae
Papilionaceae
Papilionaceae
Papilionaceae
Poaceae
Poaceae
Poaceae
Poaceae
Rourea coccinea (Baker) Jongkind subsp. Coccinea syn. Byrsocarpus coccineus Thonn.
Ruellia praetermissa Schweinf. ex Lindau
Ruellia togoensis (Lindau) Heine
Rungia guineensis Heine
Ruspolia hypocrateriformis (Vahl) Milne-Redhead
Rutidea sp.
Rytigynia canthioides (Benth.) Robyns
Rytigynia senegalensis Blume
Rytigynia sp.
Rytigynia umbellata (Hiern) Robyns
Saba comorensis (Bojer) Pichon
Saba thompsonii (A. Chev.) Pichon
Sabicea brevipes Wernh.
Sabicea calycina Benth.
Sabicea sp.
Sacciolepis africana C E. Hubbard & Snowden
Salacia debilis (G. Don.) Walp.
Salacia hispida Blakelock
Salacia leptoclada Tul.
Salacia pallescens Oliv.
Salacia sp.
Salacia staudtiana Loes.
Sansevieria guineensis (L) Willd. syn. S. trifasciata prain
Sansevieria liberica Gérôme & Labory
Sansevieria longiflora Sims
Sapium ellipticum (Hochst. ex Krauss) Pax
Sarcocephalus latifolius (Smith) Bruce syn. Nauclea latifolia Smith
Sauvagesia erecta L.
Scadoxus multiflorus (Martyn) Raf. subsp multiflorus
Schizachyrium brevifolium Nees var. brevifolium
Schizachyrium exile (Hochst.) Pilger
Schizachyrium platyphyllum (Franch.) Stapf
Schizachyrium sanguineum (Retz.) Alston
Schizachyrium sp.
Schrankia leptocarpa DC.
Schrebera arborea A. Chev.
Scleria achtenii De Wild.
Scleria bulbifera A. Rich.
Scleria depressa (C. B. Cl.) Nelmes
Scleria lacustris Wright ex Sauvelle
Scleria naumanniana Boeck.
Scleria sp.
Scoparia dulcis Linn
Secamone afzelii (Schultes) K. Schum.
Securidaca longipedunculata Fres.
Selaginela sp.
Senna alata (L.) Roxb. syn. Cassia alata L.
Senna hirsuta (L.) HS Irwin & Barneby syn. Cassia hirsuta (L.) Irwin & Barneby
Senna obtusifolia (L.) HS Irwin & Barneby syn. Cassia tora L. var obtusifolia (L.) Haines
Senna occidentalis (L.) Link syn. Cassia occidentalis L.
Senna podocarpa (frill. & Perr.) Lock syn. Cassia podocarpa Guill. & Perr.
Sericanthe chevalieri (K. Krause) Robbrecht var chevalieri
Sesamum indicum Linn.
Sesbania leptocarpa DC.
Sesbania sesban (L.) Merrill
Sesbania sudanica Gillett subsp. occidentalis Gillett
Setaria barbata (Lam.) Kunth
Setaria gracilipes C. E. Hubbard
Setaria longisepa P. Beauv.
Setaria megaphylla (Steud.) Th. Dur. & Schinz syn. S. chevaleri Stapf.
206
Riparian forests plant species of Benin
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
Th
Th
Th
He
Th
Ch
Ch
Ch
Th
Ch
Lnph
Ch
Lmph
LGe
Ch
Ch
Ch
Th
Th
Th
Th
He
He
nph
Th
mPh
Th
Th
Th
Th
Ch
Th
mPh
He
Th
Lnph
Th
Ge
mph
mph
mPh
mph
He
Th
Lmph
Lmph
LmPh
Th
Lmph
LmPh
LMPh
LmPh
mph
LmPh
LmPh
LmPh
Ge
Ge
Ge
mph
Pan
TA
_
TA
Pan
Pan
AA
GC
_
Pan
GC
Pan
GC
TA
Cosmo
Pan
Cosmo
TA
TA
_
SZ
SG
AA
G
Pan
TA
SZ
PRA
_
SG
SG
AA
TA
SZ
_
G
AA
GC
TA
SZ
GC
SG
Paleo
PRA
GC
GC
GC
_
GC
G
GC
GC
SZ
GC
G
PRA
G
SZ
SZ
SG
Poaceae
Poaceae
Poaceae
Poaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Malvaceae
Hippocrateaceae
Smilacaceae
Solanaceae
Solanaceae
Solanaceae
Lamiaceae
Lamiaceae
Lamiaceae
Poaceae
Poaceae
Poaceae
Anacardiaceae
Asteraceae
Bignoniaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Menispermaceae
Loganiaceae
Anacardiaceae
Poaceae
Poaceae
Icacinaceae
Verbenaceae
Commelianaceae
Apiaceae
Sterculiaceae
Sterculiaceae
Bignoniaceae
Poaceae
Scrophlariaceae
Apocynaceae
Apocynaceae
Apocynaceae
Asteraceae
Loganiaceae
Loganiaceae
Loganiaceae
Loganiaceae
Loganiaceae
Loganiaceae
Loganiaceae
Loganiaceae
Araceae
Araceae
Araceae
Papilionaceae
Setaria pallidae-fusca (Schumach.) Stapf & C.E.Hubb. ex M.B. Moss
Setaria restioidea (Franch.) Stapf
Setaria sp.
Setaria sphacellata (Schumach.) Stapf & C.E.Hubb. ex M.B. Moss
Sida acuta Burm. f. subsp. acuta
Sida cordata (Burm. f.) Borss. Waalk.
Sida linifolia Juss. ex Cav.
Sida rhombifolia L.
Sida sp.
Sida spinosa L. syn. S. alba L.
Sida urens L. var. urens
Sida veronicifolia Lam.
Simicratea welwitschii (Oliv.) N. Hallé syn. Simirestis welwitschii (Oliv.) N. Hallé
Smilax anceps Willd. syn. S. kraussiana Meissner
Solanum nigrum L.
Solanum torvum Sw.
Solanum verbascifolium L.
Solenostemon latifolius (Hoechst. ex Benth.) J. K. Morton
Solenostemon monostachyus (P. Beauv.) subsp. monostachyus
Solenostemon sp.
Sorghastrum bipennatum (Hack.) Pilg.
Sorghum arundinaceum (Desv.) Stapf
Sorghum bicolor (L.) Moench
Sorindeia warneckei Engl.
Sparganophorus sparganophara (L.) Jeffrey syn. Struchium sparganophora (L) O. Ktze
Spathodea campanulata P. Beauv.
Spermacoce filifolia Perr.& Lepr. ex DC. syn. Borreria filifolia (S.&T) K. Schum.
Spermacoce ruelliae DC. syn. Borreria scabra (Schumach. & Thonn.) K. Schum.
Spermacoce sp.
Spermacoce stachydea DC. syn. Borreria stachydea (DC.) Hutch. & Dalz.
Sphenostylis schweinfurthii Harms
Spigelia anthelmia L.
Spondias mombin L.
Sporobolus pyramidalis P. Beauv.
Sporobolus sp.
Stachyanthus occidentalis (Keay & Miege) Boutique
Stachytarpheta indica (L.) Vahl syn. S. angustifolia (Mill.)
Stanfieldiella imperforata (C. B. Cl.) Breman
Steganotaenia araliacea Hochst. var. araliacea
Sterculia setigera Del.
Sterculia tragacantha Lindl.
Stereospermum kunthianum Cham. var. kunthianum
Streptogyna crinita P. Beauv.
Striga micrantha (Benth.) Benth.
Strophanthus hispidus DC.
Strophanthus preussii Engl. & Pax
Strophanthus sarmentosus DC.
Struchium sp.
Strychnos afzelii Gilg.
Strychnos barteri Solered.
Strychnos congolana Gilg.
Strychnos floribunda Gilg.
Strychnos innocua Del.
Strychnos nigritana Baker
Strychnos splendens Gilg.
Strychnos usambarensis Gilg.
Stylochaeton hostifolius Engl.
Stylochaeton hypogaeus Lepr.
Stylochaeton lancifolius Kotschy & peyr.
Swartzia madagascariensis Desv.
207
Riparian forests plant species of Benin
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
Th
mPh
mph
mPh
Lmph
Ge
Th
mPh
mPh
mPh
Th
Ch
mph
mPh
mph
mph
mPh
LmPh
LmPh
mPh
Ch(Fern)
Ge(Par)
Lnph
Lnph
Lnph
Lnph
Ge
mph
Th
Lnph
mph
mph
nph
mph
nph
mph
mph
mPh
mph
mph
Lnph
Th
mPh
Ch
Ch
Ch
Th
Th
Ch
Lnph
Lnph
Lnph
Lnph
Lnph
mPh
mPh
Ch
Ge
Lmph
Lmph
Pan
GC
TA
TA
TA
Pan
Pan
Pan
GC
Paleo
SG
TA
SG
SG
SZ
_
GC
GC
GC
G
_
G
Pan
S
GC
GC
Pan
GC
Pan
G
SZ
SZ
G
_
GC
SZ
GC
GC
G
_
G
Pan
GC
Pan
GC
S
Pan
_
TA
SG
GC
G
_
GC
GC
TA
Pan
SG
GC
PRA
Asteraceae
Sapotaceae
Sapotaceae
Myrtaceae
Asclepiadaceae
Taccaceae
Portulacaceae
Caesalpiniaceae
Rubiaceae
Verbenaceae
Papilionaceae
Papilionaceae
Combretaceae
Combretaceae
Combretaceae
Combretaceae
Combretaceae
Dilleniaceae
Dilleniaceae
Mimosaceae
Thelypteridaceae
Balanophoraceae
Acanthaceae
Acanthaceae
Acanthaceae
Acanthaceae
Cyperaceae
Ulmaceae
Aïzoaceae
Malpighiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Meliaceae
Rubiaceae
Meliaceae
Meliaceae
Meliaceae
Anacardiaceae
Anacardiaceae
Menispermaceae
Asteraceae
Moraceae
Tiliaceae
Tiliaceae
Tiliaceae
Tiliaceae
Tiliaceae
Tiliaceae
Cucurbitaceae
Asclepiadaceae
Asclepiadaceae
Asclepiadaceae
Asclepiadaceae
Euphorbiaceae
Euphorbiaceae
Malvaceae
Liliaceae
Loganiaceae
Annonaceae
Synedrella nodiflora Gaertn.
Synsepalum brevipes (Baker) Pennington syn. Pachystela brevipes (Baker) Engl.
Synsepalum passargei (Engl.) Pennington syn. Vincentella passargei Engl. (Aubréville)
Syzygium guineense (Willd.) DC. subsp. guineense
Tacazzea apiculata Oliv.
Tacca leontopetaloides (L.) O. Ktze syn. T. involucrata Schum
Talinum triangulare (Jacq.) Willd.
Tamarindus indica L.
Tarenna eketensis Wernham.
Tectona grandis L. f.
Tephrosia platycarpa Guill. & Perr. syn. T. humilis. & Perr.
Tephrosia vogelii Hook. f.
Terminalia avicennioides Guill. & Perr.
Terminalia glaucescens Planch. ex Benth.
Terminalia macroptera Guill. & Perr.
Terminalia sp.
Terminalia superba Engl. & Diels
Tetracera alnifolia Willd.
Tetracera potatoria Afzel. ex G. Don
Tetrapleura tetraptera (Schum. & Thonn.) Taub.
Thelypteris sp.
Thonningia sanguinea Vahl.
Thunbergia alata Boj. ex Sims.
Thunbergia atacoriensis Akoègninou & Lisowski
Thunbergia cynanchifolia Benth.
Thunbergia togoensis Lindau
Torulinium odoratum (L.) Hopper
Trema orientalis (L.) Blume syn. T. guineensis (Schum. & Thonn.) Ficalho
Trianthema portulacastrum L.
Triaspis odorata (Willd.) A. Juss.
Tricalysia okelensis Hiern var. okelensis
Tricalysia okelensis Hiern var. pubescens
Tricalysia reticulata (Benth.) Hiern
Tricalysia sp.
Tricalysia subquadrata A. Rich. ex DC.
Trichilia emetica Vahl subsp. emetica
Trichilia prieuriana A. Juss.
Trichilia retusa Oliv.
Trichoscypha smythei Hutch. & Dalz.
Trichoscypha sp.
Triclisia subcordata Oliv.
Tridax procumbens L.
Trilepiseum madagascariense DC. syn. Bosqueia angolensis Ficalho
Triumfetta cordifolia A. Rich
Triumfetta dubia De Wild.
Triumfetta lepidota K. Schum.
Triumfetta rhomboidea Jacq. var. rhomboidea
Triumfetta sp.
Triumfetta tomentosa Jacq.
Trochomeria macrocarpa (Sond.) syn. T. macroura Hook. F.
Tylophora camerooniana N. E. Br.
Tylophora dahomensis K. Schum.
Tylophora sp.
Tylophora sylvatica Decne.
Uapaca heudelotii Baill.
Uapaca togoensis Pax syn. U. somon Aubrév. & Léandri
Urena lobata L.
Urginea altissima (L.f.) Baker
Usteria guineensis Willd.
Uvaria chamae P. Beauv.
208
Riparian forests plant species of Benin
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
nph
nph
mph
nph
mph
nph
Ch
Ch
mph
Ch
Th
Ch
Ch
Ch
He
Th
LTh
Lnph
Lnph
Lnph
LTh
LGe
mPh
mph
mPh
nph
mph
Th
Ch
Th
mph
nph
mPh
mPh
mPh
Ge
mPh
mph
Th
Th
Lnph
Lnph
LTh
Lnph
mph
mph
mph
G
G
GC
_
SG
_
SG
G
SZ
SG
TA
G
G
_
SZ
TA
GC
PRA
PRA
_
Pan
AA
SG
SZ
SZ
SZ
TA
AA
Pan
SZ
SZ
Pan
GC
G
GC
PRA
SG
SG
Cos
Pan
TA
PRA
_
PRA
Paleo
PRA
Paleo
Annonaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Rubiaceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Asteraceae
Poaceae
Asteraceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Papilionaceae
Sapotaceae
Verbenaceae
Verbenaceae
Verbenaceae
Apocynaceae
Campalunaceae
Sterculiaceae
Malvaceae
Papilionaceae
Olacaceae
Annonaceae
Annonaceae
Annonaceae
Xyridaceae
Sapindaceae
Rutaceae
Mimosaceae
Poaceae
Cucurbitaceae
Cucurbitaceae
Cucurbitaceae
Cucurbitaceae
Rhamnaceae
Rhamnaceae
Rhamnaceae
Uvaria ovata (Dunal) A. DC.
Vangueria kerstingii Robyns
Vangueria madagascariensis Gmelin syn. V. venosa Robyns
Vangueria sp.
Vangueriella spinosa (Schum & Thonn) syn.Vangueriopsis spinosa (Schumach & Thonn)
Vangueriopsis sp.
Vernonia bambilorensis Berhaut
Vernonia camporum A. Chev.
Vernonia colorata (Willd.) Drake
Vernonia glaberrima Welw. ex O. Hoffm.
Vernonia pauciflora (willd.) Less.
Vernonia poskeana Vartke & Hildebrandt
Vernonia purpurea Sch. Bip. ex Walp.
Vernonia sp.
Vetiveria nigritana (Benth.) Stapf
Vicoa leptoclada (Webb) Dandy
Vigna gracilis (Guill. & Perr.) Hook. f. syn. Dolichos gracilis Guill. & Perr.
Vigna racemosa (G. Don) Hutch. & Dalz
Vigna reticulata Hook. f.
Vigna sp.
Vigna unguiculata (L.) Walp. var. unguiculata
Vigna vexillata (L.) A. Rich. var. vexillata
Vitellaria paradoxa C. F. Gaertn. subsp. paradoxa
Vitex chrysocarpa Planch. ex Benth.
Vitex doniana Sweet
Vitex madiensis Oliv.
Voacanga africana Stapf
Wahlenbergia perrottetii (A. DC.) Thulin syn. Cephalostigma perrottetii A. DC.
Waltheria indica L.
Wissadula amplissima (L.) R. E. Fries
Xeroderris stuhlmannii (Taub) Mendonça & E.P. Sousa
Ximenia americana L.
Xylopia aethiopica (Dunal) A. Rich.
Xylopia elliotii Engl. & Diels
Xylopia parviflora (A. Rich.) Benth.
Xyris capensis Thunb.
Zanha golungensis Hiern
Zanthoxylum zanthoxyloides (Lam.) Zepernick & Timber
Zapoteca portoricensis (Jacq.) H. M. Hern. syn. Calliandra portoricensis (Jacq.) benth.
Zea mays L.
Zehneria capillacea (Schumach.) C. Jeffrey syn. Melothria capillacea (Schumach.) Cogn.
Zehneria hallii C. Jeffrey syn. Melothria deltoidea Benth.
Zehneria sp.
Zehneria thwartesii (Schweinf.) C. Jeffrey syn. Melothria tridactyla Hook. f.
Ziziphus mauritiana Lam.
Ziziphus mucronata Willd. subsp. mucronata
Ziziphus spina-cristi (L.) Desf. var. microphylla Hochst. A. ex A. Rich.
Life forms (LF) followed Raunkaier (1934), Schnell (1971) and Keay & Hepper (1954-1972). They were:
- Phanerophytes (Ph): mega-phanerophyte (MPh > 30 m), meso-phanerophyte (mPh: 8 to 29 m), microphanerophyte (mph: 2 to 7 m) and nano-phanerophyte (nph < 2m);
- Therophytes (Th); - Hemicrytophytes (He); - Chamaephytes (Ch); - Lianas (L); - Geophytes (Ge); - Epiphytes
(Ep) and - Parasites (Par).
The phytogeographic types (PT) were named after White (1986) and Keay & Hepper (1954-1972). They were:
- Species widely distributed in the tropics (Cosmopolitan-Cosmo, Pantropical-Pan, Afro-American-AA and
Paleotropical-Paleo).
- Species widely distributed in Africa (Tropical Africa-TA, Pluri Regional in Africa-PRA)
- Regional species in Africa (Sudanian-S, Guinean-G, Sudano-Guinean-SG, Sudano-Zambesian-SZ, GuineoCongolian-GC).
209
Acknowledgements
ACKNOWLEDGEMENTS
This thesis could have not been completed without the help of many people and institutions. I
have received a great deal of moral support from my family, friends and colleagues. Several
people have read the various chapters and helped me very much in polishing the English.
In the first place, my great appreciation and deep gratitude are due to my supervisor in
Benin, Professor Dr. B. Sinsin, Faculty of Agronomic Sciences, Department of Environment
Management, University of Abomey-Calavi. He has proposed and supported my candidature
for the fellowship of the Project Flora of Benin. I wish to extend my sincere thanks for
proposing the topic of the work and for his excellent guidance and constructive criticisms.
I would like to thank my promotor Professor Dr. L.J.G. van der Maesen, for his open
mind, gentleness, and excellent guidance. He is gratefully acknowledged for his skilful
improvement of the English text. His critical reading and subtle comments were of great help
in improving all the drafts of my chapters. I learned from him that patience is always
rewarding. Thanks are due to Professor Dr. A. de Gier, Forest Science Division ITC,
Enschede, for his valuable contribution, especially in methodological aspects of my research
and chapter 10.
Special gratitude goes to the Dutch Government for financing the Project Flora of
Benin, which granted me financial support to carry out MSc and PhD studies. The staff of the
Project Flora of Benin in Abomey-Calavi and Wageningen helped me during all the steps of
this research. I take this opportunity to thank Dr. N. Sokpon, Dr. A. Akoègninou, Dr. J-P.
Essou, Dr. V. Adjakidjè, and all lecturers of the Department of Environment Management of
the Faculty of Agronomic Sciences, University of Abomey-Calavi, Benin. Special thanks are
due to the colleagues of the Project Flora of Benin, Paul H. Yédomonhan, Pierre O. Agbani,
Aristide C. Adomou, Ebenezer Ewedje, Raoul A. Adjobo and Nestor Gbéssi for their
valuable help during data collection, plants determination and data analysis. The convivial
atmosphere we had helped me very much overcome difficulties inherent to such studies.
I would like to thank Dr. R. N’tia, Dr. Kaki, Dr. S. Seibou Toléba (University of
Abomey-Calavi, Benin), G. Ossory, Dr. J.-P. Séro, who have constantly encouraged me.
I would like to express my gratitude to the staff of the Direction des Forêts et
Ressources Naturelles (DFRN), the Centre National de Gestion des Réserves de Faune
(CENAGREF) and the Projet de Restauration des Ressources Forestières (PRRF – Bassila)
for granting me special permission for working in the protected areas of Benin. Very kind
hospitality and cooperation have been provided by all the foresters, fauna guards and all those
who tirelessly worked with me all the days of my data collection at Samiondji, Toui-Kilibo,
Idadjo, Bétérou, Pénéssoulou, Affon, Yarpao, Boukoumbé, Toucountouna, Kandi, Gbèssè
and in the Pendjari Biosphere Reserve. It is practically impossible to name them all here.
Without their help I could not have been able to complete my fieldwork. I also appreciated
the hospitality of the people of the villages I visited throughout the country.
I appreciated the very helpful reading and useful comments of Professor S. Porembski
of Rostock University (Germany), especially for chapters 3 and 9. Also, I would like to
express my sincere regards to Professor F. Bongers, of Wageningen University for his critical
comments.
My gratitude goes to all the staff and members of the Biosystematics group,
Department of Plant Sciences, Wageningen University, The Netherlands, for the pleasant
working atmosphere and their sincere help throughout my stay at Wageningen, and
finalisation of this thesis. I also thank Kadry N.E. Abdel Khalik and other PhD students and
project workers for the literature research they did for me during the time I was in Benin.
They were crucial for the finalisation of my chapters.
211
Acknowledgements
Ebenezer Ewedje should be thanked for the maps he draw, and Grace Nangendo for
comments and helpful reading. I would like to acknowledge the companionship of Adi
Mama, Marcel Houinato, Madjidou Oumorou, Théophile Sinadouwirou, Richard Bawa,
Raoul Adjobo, Antonio Sinzogan, Roger Sodjinou, Florent Okry, Honoré Biaou, Brice Tenté,
Yacoubou Boni, Etotépé Sogbohossou, Bienvenu Gangboché, and Barthélemy Kassa. I also
appreciated the help of all the colleagues of the Laboratoire d’Ecologie Appliquée
(FSA/DAGE), and all those whose names are not mentioned here.
Finally my special thanks to my lovely mother for her never-ending support, and my
gratitude to my sister Sylvia, to her family and to Espérance Obonté who have constantly
encouraged me to achieve my aims.
212
Curriculum Vitae
CURRICULUM VITAE
The author of this dissertation, Armand Kuyéma NATTA, was born in Tanguiéta (North
Benin), on 3 May 1972. He joined the Faculty of Agronomic Sciences, University of
Abomey-Calavi in 1992. He obtained his BSc degree in General Agronomy in 1995. He
continued his post-graduate studies in Agronomy (option Natural Resources Management) at
the same University and obtained the diploma of Agronomic Engineer in 1997. He worked a
short time as technical director of a Company specialised in the distribution of fertilisers and
advice to cotton farmers. In 1998, he came back to the Faculty of Agronomic Sciences as
research assistant. The same year he was granted a fellowship by the Project Flora of Benin
to study Remote Sensing and Geographical Information Systems at the International Institute
for Aerospace Survey and Earth Sciences (ITC), Enschede, The Netherlands. In February
2000, he was awarded an MSc degree in the Forest Science Division by ITC, with an overall
qualification of very good. In March 2000, he began his PhD studies on riparian forests
diversity in Benin, within the Project Flora of Benin, and in close cooperation with the
Biosystematics group and National Herbarium of the Netherlands, Wageningen University
Branch, Department of Plant Sciences, Wageningen University, The Netherlands. He has
participated in several regional workshops in Benin, Burkina-Faso and Côte d’Ivoire, and in
the XII World Forestry Congress held in Québec, Canada, in September 2003.
His addresses in Benin are:
Office:
Faculty of Agronomic Sciences, Department of Environment Management
(FSA/DAGE), University of Abomey-Calavi. 01 BP 526 Cotonou, Benin
(West Africa).
Tel/Fax: 00 229 303084 or Tel 00 229 360126
Home:
02 BP 906 Cotonou, Benin (West Africa)
Tel: 00 229 055682 or 00 229 490087 or 00 229 822172
e-mail:
natta@bj.refer.org
aknatta@yahoo.com
aknatta@hotmail.com
213
List of publications
LIST OF PUBLICATIONS
Natta A.K., Sinsin B. & van der Maesen L.J.G. 2002. Riparian forests, a unique but
endangered ecosystem in Benin. Notulae Florae Beninensis 4. Botanische JahrbĦcher
124: 55-69.
Natta A.K., Sinsin B. and van der Maesen L.J.G. 2003. Riparian forests and biodiversity
conservation in Benin (West Africa). Paper presented at the XII World Forestry
Congress, organised by FAO in Québec City. In: Proceedings of the XII World
Forestry Congress Area B: Forest for the planet. B2a: Maintenance of Biodiversity.
Voluntary paper of major interest to the deliberations (level 2, No. 0356). Canada,
September 21 to 28, 2003. pp 126-127.
Natta A.K. & Porembski S. Ouémé and Comoé: forest-savanna border relationships in two
riparian ecosystems in West Africa. Notulae Florae Beninensis 8. Botanische
JahrbĦcher. (in press).
Natta A.K., Sinsin B. & van der Maesen L.J.G. A phytosociological study of Riparian forests
in Benin (West Africa). Belgian Journal of Botany. (in press).
215