Tree diversity and conservation value of
Ngovayang’s lowland forests, Cameroon
Christelle F. Gonmadje, Charles
Doumenge, Doyle McKey, Gildas
P. M. Tchouto, Terry C. H. Sunderland,
Michael P. B. Balinga & Bonaventure
Sonké
Biodiversity and Conservation
ISSN 0960-3115
Volume 20
Number 12
Biodivers Conserv (2011) 20:2627-2648
DOI 10.1007/s10531-011-0095-z
1 23
Your article is protected by copyright and
all rights are held exclusively by Springer
Science+Business Media B.V.. This e-offprint
is for personal use only and shall not be selfarchived in electronic repositories. If you
wish to self-archive your work, please use the
accepted author’s version for posting to your
own website or your institution’s repository.
You may further deposit the accepted author’s
version on a funder’s repository at a funder’s
request, provided it is not made publicly
available until 12 months after publication.
1 23
Author's personal copy
Biodivers Conserv (2011) 20:2627–2648
DOI 10.1007/s10531-011-0095-z
ORIGINAL PAPER
Tree diversity and conservation value of Ngovayang’s
lowland forests, Cameroon
Christelle F. Gonmadje • Charles Doumenge • Doyle McKey •
Gildas P. M. Tchouto • Terry C. H. Sunderland • Michael P. B. Balinga
Bonaventure Sonké
•
Received: 23 December 2010 / Accepted: 15 June 2011 / Published online: 28 June 2011
Springer Science+Business Media B.V. 2011
Abstract The Ngovayang Massif of southern Cameroon is a range of small hills near the
Atlantic coast, in the Lower Guinea floristic region. This region is known to harbor forests
with high levels of biodiversity and endemism, but this Massif is botanically poorly known.
We assessed tree species diversity, floristic composition and level of endemism of the
Ngovayang forest, comparing it with other sites in Central Africa. Five 1-ha permanent
plots within old-growth lowland forests of the Ngovayang Massif were censused. A total of
2,658 individuals with dbh C 10 cm were recorded, belonging to 293 species, 170 genera
and 60 families. The mean number of stems was 532 ± 75 stems ha-1. Taking into
account other data available, the list of vascular plants known in the Massif reaches a total
of 450 species. We found 47 species of high conservation value, including Cameroon
endemics and other rare and threatened species. Species richness and endemism are
comparable to those of the richest known sites in Central African forests. The forests of
Electronic supplementary material The online version of this article (doi:10.1007/s10531-011-0095-z)
contains supplementary material, which is available to authorized users.
C. F. Gonmadje (&) B. Sonké
Plant Systematics and Ecology Laboratory, Higher Teacher’s Training College, University of Yaounde
I, P.O. Box 047, Yaounde, Cameroon
e-mail: cgonmadje@yahoo.fr
C. Doumenge
CIRAD, Campus International de Baillarguet, TA C-105/D, 34398 Montpellier Cedex 5, France
C. F. Gonmadje D. McKey
Centre d’Ecologie Fonctionnelle et Evolutive, UMR, CNRS 5175, 1919 route de Mende,
34293 Montpellier Cedex 5, France
G. P. M. Tchouto
National Forestry School, P.O. Box 69, Mbalmayo, Cameroon
T. C. H. Sunderland
Centre for International Forestry Research (CIFOR), P.O. Box 0113, BOBCD, Bogor 16000, Indonesia
M. P. B. Balinga
Centre for International Forestry Research (CIFOR), BP 9478 Ouagadougou, Burkina Faso
123
Author's personal copy
2628
Biodivers Conserv (2011) 20:2627–2648
Ngovayang were found to be particularly rich in Fabaceae-Caesalpinioideae. Topographic
heterogeneity, high precipitation and atmospheric humidity owing to the proximity of the
ocean, and permanence of a forest cover during past geological times probably all contribute to explaining the Massif’s high tree diversity and endemism. This study highlights
the botanical importance of the poorly studied Ngovayang forest within the Lower Guinea
region, justifying efforts for improved assessment of this value and for the development of
suitable national conservation strategies.
Keywords 1-ha permanent plots Cameroon Floristic patterns Endemic species
Lower Guinea floristic region Tree alpha diversity Tropical rainforest
Introduction
The Atlantic Central African rain forests, spanning parallel to the coast of the Gulf of
Guinea from Nigeria to Gabon, are among the most important biodiversity hotspots of the
world (Myers et al. 2000; Küper et al. 2004). This region is suspected to have harbored
several forest refugia during past drier geological periods (White 1983; Anhuf et al. 2006).
The Lower Guinea forests are well-known for their high biodiversity and endemism (White
1983; Sosef 1994). In fact, the peak of species richness and endemism of woody plants in
Africa is in these forests, especially in southwestern Cameroon (Lovett et al. 2000; Linder
2001). The Lower Guinea coastal region is transected in its northern part by the high
mountains of the Cameroonian volcanic line, and a discontinuous chain of small mountains
that extends from southwestern Cameroon southward through Equatorial Guinea and
Gabon, to near the mouth of the Congo River. The northernmost part of this coastal range
of hills is constituted by the Ngovayang Massif, in southwestern Cameroon, which still
maintains a largely intact continuous forest cover.
Despite its high potential for biodiversity conservation, the Ngovayang Massif has
attracted little attention from scientists. In mapping the vegetation of Cameroon, Letouzey
(1968, 1985) included this upland area in the ‘‘District atlantique biafre´en’’, characterized
by a high number of gregarious caesalpinioid legumes. However, Letouzey’s phytogeographic maps are largely based on prospection and collection of botanical specimens, and
very few quantitative inventories were performed. Since then, very little has been published on the Ngovayang Massif and it has been repeatedly overlooked in nationwide
censuses of critical forest sites for biodiversity conservation (Gartlan 1989; Doumenge
et al. 2003). However, some recent taxonomic studies, which focused on a small number of
families—i.e. Rubiaceae, Begoniaceae and Orchidaceae—have confirmed the high levels
of endemism of this area for these taxa (Sosef 1996; Sonké et al. 2006a, b, c, 2008a, b;
Droissart 2009). As strategies for conserving species and communities focuses as much on
species richness and endemism (Myers et al. 2000, Lovett et al. 2000, Kier and Barthlott
2001), more detailed information about the coastal forests of Cameroon, especially on the
poorly known Ngovayang Massif, is necessary. Better data from such unexplored areas
would greatly increase the reliability of modeling patterns of tree alpha diversity in African
rain forests (Parmentier et al. 2011).
Given the ever-increasing threats to local biodiversity in Cameroon, the need to study
the Ngovayang forest has never been greater. In addition to global threats such as climate
change, forest destruction for agriculture by local populations and timber exploitation have
both heavily impacted the natural vegetation for decades, mainly in the flat lowlands
(Letouzey 1985). Other human pressures, such as hunting, have reached alarming rates in
123
Author's personal copy
Biodivers Conserv (2011) 20:2627–2648
2629
the area. Owing to its rugged topography and low population density, the Ngovayang
Massif has so far been relatively preserved from these threats.
This paper is the first attempt to describe the tree diversity of the southern part of the
Ngovayang Massif using quantitative analysis. Because many other sites in Central Africa
now have comparable inventories, the Ngovayang Massif can be repositioned in this
context, so that we can determine whether our study site is important within the framework
of an overall conservation strategy for the forests of the Congo Basin. Specifically, we
aimed to answer the following questions:
1. What is the level of diversity of the Ngovayang Massif compared to other sites in the
Congo basin and, specifically, in the Lower Guinea floristic region?
2. Does the Ngovayang Massif exhibit levels of endemism similar to those of other sites
already identified as conservation priority areas in Central Africa?
Study sites and methods
Research area
The Ngovayang Massif is a range of hills located in southern Cameroon, 80 km east
of the Atlantic coast. The Massif covers an area of 102,000 ha and extends between
3120 –3250 N and 10300 –10450 E (Fig. 1).
The altitude varies from 50 m above sea level in the western part, to more than 1,000 m
in the eastern part, on the summits of the main hills. The area is characterized by highly
dissected terrain, with alternating ridges and valley bottoms. The main ridges are oriented
north-east to south-west. The southeastern slopes are steep, while a piedmont area of
varying altitude occupies the northwestern part of the area.
Soils in the Massif are derived from the Precambrian shield, which consists of metamorphic rocks (mainly gneiss, migmatite, quartzite and schist) and old volcanic intrusions.
Following the FAO classification system, the dominant soils are xanthic ferralsols
(Franqueville 1973). Climate is sub-equatorial to tropical (Suchel 1972). Average annual
rainfall decreases eastward from the coast, from 2836 mm ± 393 at Kribi, west of the
Ngovayang Massif, to 2,096 mm ± 286 at Lolodorf (Olivry 1986), with about 2,000 mm
annual rainfall in the area between Bipindi and Lolodorf (Waterloo et al. 2000; Fig. 1).
Two distinct precipitation minima and maxima can be distinguished over the year. The wet
season extends from March to November, with a relatively drier period from July to
August; the main dry season occurs between December and February. Mean annual
temperature is 25C, decreasing from west (26.4C at Kribi) to east (24.6C at Lolodorf),
and with increasing altitude and distance from the ocean (Olivry 1986). The hydrography
of the area is dense, with many rivers, small river basins, fast-flowing creeks and rivers
flowing on rocky beds, and small waterfalls.
The main vegetation type is a dense evergreen rain forest described as rich in gregarious
caesalpinioid legumes (Letouzey 1968, 1985). Regrowth and secondary forests are present
in the lowlands, and very small areas of submontane forest are found on various summits of
the main ridge (Letouzey 1985). A few degraded forests occur locally in some of the most
accessible areas.
Most of the human population lives at the periphery of the Massif (Fig. 1). However,
temporary camps of Bagyeli pygmy hunter-gatherers are found throughout the forest.
Local people are very poor and rely solely on agriculture and forest resources to meet their
123
Author's personal copy
2630
Biodivers Conserv (2011) 20:2627–2648
Fig. 1 Location of the Ngovayang Massif in southern Cameroon, and locations of the five permanent plots.
Lambi, Bidjouka, Ngovayang are the successive main villages allowing access to the southern part of the
Ngovayang forest Massif. The first three letters of these villages were used for the plots’ code names
(LAM1, LAM2, BID1, BID2 and NGO1)
basic needs. The flat lowlands and some slopes (up to 300 m a.s.l.) have been converted to
slash-and-burn agriculture, and hunting is widespread. Timber exploitation has been
extensive in some lowland areas, mainly in the northern part of the Massif, but is becoming
123
Author's personal copy
Biodivers Conserv (2011) 20:2627–2648
2631
increasingly controlled, as timber companies are now engaged in developing sustainable
management plans in their concessions.
Data collection
Representative and homogeneous vegetation types were selected on the basis of physical
(altitude) and human (land use) factors. Five 1-ha (100 m 9 100 m) permanent plots were
established within the southern part of the Ngovayang Massif (Fig. 1) at altitudinal
intervals of about 200 m in three localities: Bidjouka, Lambi and Ngovayang I. The plot
code names were based on the three first letters of the localities’ names, followed by a
number going from the lower to the upper altitudes. Thus, we have BID1 and LAM1 at
400 m, and BID2, LAM2 and NGO1 at 600 m. The plots were chosen to cover mature
terra firme rain forest on flat or almost flat areas. We avoided secondary vegetation, rivers
and swampy vegetation types. Plots were all below 700 m altitude, as the objective of this
first study was to examine lowland vegetation, facilitating comparison with previous
studies, which have been mainly located in lowland and mid-elevation forests. The composition of the area’s submontane forests will be addressed in subsequent publications.
The establishment of the plots followed the field protocols of RAINFOR (Red
Amazonica Inventarios Forestales); details on measurement methods are given by Phillips
et al. (2004). This measurement protocol, modified from Dallmeier (1992), has been used
in several studies of tropical forests (Lieberman et al. 1996; Small et al. 2004), including
some in the Congo Basin (Reitsma 1988; van Gemerden et al. 2003). Within each plot, all
trees with a diameter at breast height (dbh) C 10 cm were recorded and characterized by
the following attributes: species or morpho-species name and dbh. The dbh was measured
with a diameter tape at 1.3 m above ground level, avoiding any protrusion on the trunk or
lianas growing around it. For buttressed trees, the dbh was taken 30 cm above the buttresses. The dbh C 10 cm size class of trees was chosen because it is a major component of
the rain forest structure, and because data are available for a large number of plots in
Central Africa, thus allowing comparisons among sites. All trees were permanently marked
and labeled with numbered aluminum tags. Identification of the most common species was
done directly in the field whenever possible. Herbarium specimens (200 samples) of most
species and morpho-species were collected, identified and preserved at the National
Herbarium of Cameroon (YA).
To obtain more accurate estimation of species richness and endemism, a plant species
checklist was generated from the tree inventory data, from the plant collections made
during the study, and from specimens previously collected in the area by other scientists
and stored in the Cameroon National Herbarium.
The African datasets for trees with dbh C 10 cm used for comparison with our data
were compiled from literature (Parmentier et al. 2011) and from unpublished data from the
Monitoring and Assessment of Biodiversity Program of the Smithsonian Institution. The
dataset of different sites comprises 70 1-ha plots and 264 0.1-ha transect sections (Fig. 2).
In common with our plots at Ngovayang, those selected for comparative purposes in this
study are all located in mature terra firme rain forest, avoiding secondary vegetation and
swampy vegetation types, and are all below 700 m altitude.
In the Angiosperm Phylogeny Group (APG 2003) classification system, several families
have been merged or split. However, to facilitate comparison with previous studies we
retain the old classification system used by Lebrun and Stork (1991–1997) for plant species
names and families.
123
Author's personal copy
2632
Biodivers Conserv (2011) 20:2627–2648
Fig. 2 Precipitation gradient and location of Central African 1-ha permanent plots mentioned in the text
Data analysis
We used a diversity indices to describe diversity patterns across the study plots. Alpha
diversity is the number of species in a chosen area or community (Kent and Coker 1992); it
is a diversity of a set of samples (Magurran 2004). Diversity indices take into account not
only the number of species but also whether species are more or less equally abundant, or
whether in contrast one or a few species dominate. We assessed diversity with Shannon
and Fisher’s a diversity indices, which are the most widely used and thus facilitate
comparisons. The Shannon diversity indices between the plots were compared using t tests,
as described by Hutcheson (1970).
The Shannon diversity index (H0 ) is:
H0 ¼
s
X
ni =N lnðni =NÞ
i¼1
where ni is the number of individuals of a given species i and N the total number of
individuals.
However, the entropy value (H0 ) of the Shannon index is not itself a measure of
diversity. Conversion of this value to effective number of species, or true Diversity (D), is
the key to a unified and intuitive interpretation of diversity (Jost 2006). Diversity is
calculated as:
D ¼ expðH 0 Þ
where D is Diversity, exp is an exponential function and H0 the Shannon diversity index
(Jost 2006).
123
Author's personal copy
Biodivers Conserv (2011) 20:2627–2648
2633
Tree alpha diversity was also estimated with Fisher’s a as follows:
S ¼ a lnð1 þ N=aÞ
where S is the number of species, N is the number of individuals, ln is the natural logarithm
and a is Fisher’s a (Magurran 2004).
These indices are complemented by another measure of a diversity, the S(50) index,
as data in the literature now allow wider comparison with various African forest sites
(Parmentier et al. 2011). S(50) is a non-parametric sample-size unbiased estimator of a
diversity: the expected number of species found in a subsample of 50 individuals. It was
computed following Hurlbert (1971).
We have calculated values of these diversity indices for our plots and selected plots
from other forests in Lower Guinea and elsewhere in Central Africa, and compared these
values using Kruskal–Wallis tests.
To describe the ecological importance of species, morpho-species and families within
each plot as well as for the total flora, the species Importance Value Index (IVI; Curtis and
McIntosh 1951) and the Family Importance Value index (FIV; Mori et al. 1983) were also
calculated:
Relative abundance ¼ ðnumber of trees of the species or family=total number of treesÞ
100:
Relative frequency ¼ ðfrequency of a species=sum of all frequenciesÞ 100:
Family relative diversity ¼ ðnumber of species in a family=total number of speciesÞ
100:
IVI ¼ relative density þ relative frequency þ relative dominance:
FIV ¼ family relative diversity þ relative density þ relative dominance:
The method used to construct the species accumulation curve is based on the calculation
of the average number of species for 1, 2,…, all individual plants.
A taxonomic search of potential taxa of high conservation value, such as those
endemic to Cameroon and other rare and threatened species according to Tchouto et al.
(2006), was conducted using the literature (Keay and Hepper 1954–1972; Aubréville
and Leroy (1961–1992; 1963–1978), supplemented by the CJB (2010) data base
(Conservatoire du Jardin Botanique de Genève: http://www.villege.ch/musinfo/bd/cjb/
africa/recherche.php) and the IUCN (2010) red list. In this study, the notion of rarity
was defined following three criteria which describe different ways in which a plant
species can be rare: (1) abundant somewhere or scarce everywhere (2) restricted to a
single habitat type or not, and (3) geographically widespread or restricted (Pitman et al.
1999). Comparison of degree of rarity and threatened species was done only between
the Ngovayang Massif and Campo-Ma’an, for which data are available. Distribution
patterns of these species were tallied relative to the major African phytochoria (White
1979, 1983). We assigned each species to one of six categories: (1) Widespread (Ws)
including pan-African and paleotropical, (2) Guineo-Congolian (Gc), (3) Upper and
Lower Guinea (Gu), (4) Lower Guinea (Lg), (5) Cameroon (Cam) and (6) southwestern
Cameroon (Sw-Cam).
TM
The Vegan package of R software for Windows was used for statistical analysis
(Oksanen et al. 2010).
123
Author's personal copy
2634
Biodivers Conserv (2011) 20:2627–2648
Results
Diversity, species richness and endemism in the Ngovayang Massif
A total of 2,673 stems with dbh C 10 cm were recorded within the five plots, representing
293 species and morphospecies, 170 genera and 60 families. About 70% of morphospecies
were identified to species level, 27% identified to generic level, and 3% remain unidentified. The number of species per plot varied from 99 to 121, with a mean of 110 ± 9
species ha-1. The individual-based species accumulation curve displayed an increasing
trend that did not level off (Fig. 3). S(50) varied from 30 to 32 species. The Shannon
diversity index (H0 ) varied from 3.9 to 4.12. Values of the Shannon diversity index were
highest in LAM2 and NGO1 plots (H0 = 4.11 and 4.12) and significantly lower in BID2,
LAM1 and BID1 plots (t = 2.8; P \ 0.05 for H0 ). The same trend of diversity was found
with true Diversity (D) and Fisher’s a, which varied from 49 to 62 species, and from 39 to
46, respectively, indicating higher species diversity in LAM2 and NGO1 plots than in the
three other plots (Table 1).
The mean number of stems was 532 ± 75 stems ha-1 (range 451–634 stems ha-1). The
total basal area varied from 28.8 to 42.1 m2 ha-1 with a mean value of 34.6 ±
5.4 m2 ha-1.
The average true diversity (D) differed significantly among sites (Kruskal–Wallis test,
P = 0.0129; Table 4). Mean S(50) diversity measures based on 0.1-ha plots were significantly higher in most Atlantic Central African forests—including Bipindi-Akom
II–Lolodorf, Massif du Chaillu, Ngovayang Massif, Waka, Monte Mitra, Monts de Cristal
and Takamanda—than in continental plots such as Dja Fauna Reserve, Nouabale-Ndoki
National Park, Mbam-Djerem National Park, and some Atlantic forests plots such as
Ejagham Reserve and Korup National Park (Kruskal–Wallis test, P \ 0.0001). Of all sites,
Bipindi-Akom II–Lolodorf forest was the most diverse, whereas the Odzala forest of
Fig. 3 Species accumulation curves of the five 1-ha plots of Ngovayang Massif. Error bars represent
standard deviation. Line indicates species richness values by given number of censused individuals
123
Author's personal copy
Biodivers Conserv (2011) 20:2627–2648
2635
Table 1 Number of taxa, diversity indices and structural characteristics of five 1-ha plots in the Ngovayang
Massif, Cameroon
BID1
BID2
LAM1
LAM2
NGO1
Density (individuals ha-1)
451
478
513
634
582
Basal area (m2 ha-1)
37.7
30.3
42.1
28.8
34.1
Mean dbh (cm)
25.1 ± 20.7
23.3 ± 16.2
24.8 ± 20.7
20.8 ± 1
22.6 ± 15.2
Number of genera
73
73
83
76
80
Number of species
99
105
106
118
121
Shannon index (H0 )
3.90
3.93
3.91
4.11
4.12
Diversity (D)
49
51
50
61
62
Fisher’s a
39.2
41.6
40.6
42.7
46.5
S(50)
30.2
30.6
30.0
32.3
32.4
Elevation (m a.s.l.)
392
605
396
627
650
Latitude
3.14225
3.15948
3.10240
3.11833
3.23053
Longitude
10.47853
10.46520
10.42450
10.42436
10.58303
Congo was the least diverse (Kruskal–Wallis test, P \ 0.0001). Similar results were
obtained for the mean Fisher’s a diversity based on 1-ha plots where Mbam-Djerem forest
was the least diverse for this index (Kruskal–Wallis test, P \ 0.0001).
In addition to these 293 tree species and morphospecies encountered in the five 1-ha
plots, a provisional plant checklist comprising 450 species of vascular plants (trees, treelets, shrubs and herbaceous) was generated with the addition of other data sources. These
species belonged to 238 genera and 76 families, with 361 taxa identified to species level.
Among these 361 species whose ranges were known, a small proportion is comprised of
widespread species found in various African phytochoria (12.5%). Most species in the list
are Guineo-Congolian wide or restricted to a sub-phytochorium (Table 5). However, 153
species (43% of the total) are known only from the Lower Guinean domain. Of these, 36
species (10% of the total) have even narrower ranges, being found only in Cameroon, and
of these Cameroon endemics, 14 (4%) are restricted to southwestern Cameroon (e.g.
Aulotandra kamerunensis, Begonia mbangaensis, Gilbertiodendron pachyanthum, Pavetta
camerounensis,…). We produced a list of 47 plant species of high conservation priority,
including rare species, threatened species and Cameroon endemics, with information on
their growth forms and chorology. These include 12 threatened species recorded in the
IUCN red list for Cameroon (2010) and 22 rare species (Online Appendix 1).
Floristic composition
The 10 most important families (those with the highest values of the FIV index) represented less than 15% of all families, but accounted for almost 63% of the total FIV of all
plots considered together. They contributed more than 70% of the total number of individuals and of the total basal area of all plots. The three families or subfamilies with the
highest FIV and number of individuals were Fabaceae-Caesalpinioideae (347 individuals),
Euphorbiaceae (361) and Clusiaceae (248), followed by Myristicaceae, Olacaceae and
Ebenaceae (with 529 individuals for these three families together; Table 2). The most
species-rich families were Euphorbiaceae (36 species), Fabaceae-Caesalpinioideae (23),
Ebenaceae (19), Annonaceae (14), Rubiaceae (13), Olacaceae (11) and Clusiaceae (10).
123
Author's personal copy
2636
Biodivers Conserv (2011) 20:2627–2648
Table 2 Family Importance Value of the 10 most important families (in bold) of each plot, by decreasing
global FIV for all five 1-ha plots censused in the Ngovayang Massif
Family
Global FIV
BID1
BID2
LAM1
LAM2
NGO1
34.35
Fabaceae-Caesalpinioideae
39.21
42.40
22.97
36.41
60.67
Euphorbiaceae
34.64
38.38
32.67
28.01
38.03
32.32
Clusiaceae
23.01
5.85
30.00
33.14
16.31
39.86
Myristicaceae
17.33
27.04
16.72
20.19
7.19
25.18
Olacaceae
16.77
29.07
18.56
8.42
14.63
21.86
Ebenaceae
16.43
21.18
20.23
16.68
17.43
16.13
Annonaceae
11.91
9.07
11.80
15.17
13.78
11.57
Burseraceae
11.51
11.18
8.89
12.49
16.91
12.06
Rubiaceae
9.20
11.59
11.92
4.35
10.23
8.18
Scytopetalaceae
9.01
2.17
26.43
1.39
1.64
11.43
Anisophylleaceae
8.33
9.70
10.78
10.93
4.72
10.09
Sterculiaceae
8.11
4.37
18.31
4.97
4.77
5.13
Sapotaceae
7.13
1.52
10.04
9.01
8.76
3.76
Moraceae
6.98
13.89
10.82
4.54
6.08
Anacardiaceae
5.33
5.16
5.55
9.43
4.90
–
3.46
Species of these most important families differed in their distribution between plots
(Table 2). Fabaceae-Caesalpinioideae, Clusiaceae, Myristicaceae and Olacaceae were
most dominant, contributing primarily to the floristic composition of the canopy: many of
their species were emergent or large canopy trees that accounted for much of the basal area
recorded in the plots. Euphorbiaceae and Ebenaceae owed their high FIV values to their
abundance, contributing mostly to the understory with numerous individuals but exhibiting
a rather low basal area. Other families, such as Burseraceae, contributed in nearly equal
proportions to both canopy and understory strata (Fig. 4).
The four species with the highest IVI values were Coelocaryon preussii (Myristicaceae), Guibourtia tessmannii, Tetraberlinia bifoliolata (both Fabaceae-Caesalpinioideae)
and Allanblackia floribunda (Clusiaceae; Table 3). The families containing the ten species
with the highest IVI values included Fabaceae-Caesalpinioideae (three species), Myristicaceae and Clusiaceae (two each), and Moraceae, Olacaceae and Euphorbiaceae (one
each). A large group of species was represented by only one or two individuals in the
overall sample of the five plots together (147 species, representing 49% of the total).
Although commonly found in all plots, the Burseraceae, Ebenaceae, Euphorbiaceae and
Fabaceae-Caesalpinioideae exhibited remarkable variation in species composition among
sites. Fabaceae-Caesalpinioideae was widely represented in plots BID1 and LAM1 by
Guibourtia tessmannii and Tetraberlinia bifoliolata, whereas in plot LAM2, the family was
represented mostly by Didelotia letouzeyi and Tetraberlinia bifoliolata. These species were
absent in plot BID2, where the family was mainly represented by Hymenostegia afzelii.
Euphorbiaceae was the most important family in plot BID2 but was represented there by
many species which were not among the ten most important species, whereas in plot
NGO1, this family was mainly represented by Uapaca heudelotii (Table 3).
Some species were found exclusively in one or two plots but had high IVI values in
these (Oubanguia alata and Cola rostrata in plot BID2; Garcinia lucida in plot NGO1 and
Guibourtia tessmannii in plots BID1 and LAM1). In contrast, Coula edulis was present in
123
Author's personal copy
Biodivers Conserv (2011) 20:2627–2648
2637
Fig. 4 Relation between relative density and relative dominance for all families in five 1-ha plots of the
Ngovayang Massif
all the plots but its IVI value was high only in plot BID1. Only 13 species (4.3%) occurred
in all five plots. Of these, only three ranked among the ten most important species:
Coelocaryon preussii, Coula edulis and Staudtia kamerunensis. Other ubiquitous species
had lower IVI values, being everywhere relatively rare. These included Strombosiopsis
tetrandra, Santiria trimera and Dacryodes edulis. In strong contrast, the majority of
species (171 species or 57% of the total number) were sampled in only one plot, and 18,
12.7 and 8% were found in two, three and four plots, respectively.
Discussion
The diversity of Ngovayang forests
Forest communities considered rich are characterized by a Shannon diversity value of
about 3.5 or higher (Kent and Coker 1992). The plots of Ngovayang, which all have high
values of Shannon diversity (H0 [ 3.5; Table 1) and of Fisher’s a, can accordingly be
considered very diverse. The lowland forests of the southwestern part of the Ngovayang
Massif harbor a rich and diverse flora. As shown by similar results in previous studies
(Pitman et al. 2002), such high diversity seems also to derive from a great abundance of
rare species: almost half of the species found in the Ngovayang plots (49% of the total)
were represented by three or fewer individuals. All tropical forests have many rare species,
which generally present high risk of at least local extinction (Kenfack et al. 2006). In the
present study, 35–41 families were inventoried in each plot, which falls in the medium of
the range of 16–58 families found for tropical forests as a whole (Gentry 1988; Campbell
123
Author's personal copy
2638
Biodivers Conserv (2011) 20:2627–2648
Table 3 Importance Value Index of the 10 most important species (in bold) of each plot, by decreasing
global IVI for all five 1-ha plots censused in the Ngovayang Massif
Species
Global IVI
BID1
BID2
LAM1
LAM2
NGO1
Coelocaryon preussii
8.96
14.93
12.40
Guibourtia tessmannii
8.44
28.63
–
20.49
9.83
3.09
17.39
Tetraberlinia bifoliolata
7.33
9.21
–
12.43
–
10.08
–
13.43
Allanblackia floribunda
7.16
0.97
30.34
2.59
7.09
Treculia obovoidea
6.17
18.59
–
12.86
2.20
7.96
Coula edulis
5.80
17.63
4.84
4.45
Didelotia letouzeyi
5.79
–
–
Staudtia kamerunensis
5.42
10.87
4.76
Dichostemma glaucescens
5.07
25.06
0.55
Pentadesma grandifolia
5.07
2.40
10.80
Dacryodes klaineana
4.96
4.61
Oubanguia alata
4.34
–
–
–
–
24.29
11.05
–
–
18.98
–
Strombosiopsis tetrandra
3.88
7.15
7.77
3.13
Poga oleosa
3.77
7.14
1.63
10.81
2.09
1.54
29.87
3.19
2.36
2.39
–
2.34
5.73
14.63
1.67
5.76
–
–
2.14
4.99
–
–
Tabernaemontana brachyantha
3.75
1.14
9.61
6.77
0.86
5.20
Santiria trimera
3.37
3.41
7.14
0.50
1.59
4.28
Diospyros iturensis
3.32
1.62
1.32
2.53
10.70
4.36
Dacryodes edulis
3.21
2.09
5.79
0.50
7.38
0.42
Anthonotha fragrans
3.07
–
3.46
–
7.49
6.49
Cola rostrata
2.93
–
19.31
–
–
–
Scytopetalum klaineanum
2.71
1.78
5.50
–
–
8.18
Uapaca staudtii
2.55
–
4.66
2.00
7.46
–
Diospyros sp. 1
2.52
2.54
7.87
Pentadesma butyracea
2.49
0.58
10.09
1.28
0.41
1.80
–
–
0.45
0.53
Hymenostegia afzelii
2.41
–
12.40
–
0.67
Trichoscypha sp. 1
2.36
0.85
1.51
–
10.42
Garcinia lucida
2.35
–
–
–
–
–
14.92
Mareyopsis longifolia
2.28
4.67
0.55
–
7.67
0.85
Heisteria zimmereri
2.09
1.33
–
–
1.82
9.16
Uapaca heudelotii
2.06
1.08
–
–
0.60
7.54
Xylopia aethiopica
2.05
–
8.96
0.84
0.94
Uapaca cf. guineensis
1.19
7.16
–
1.04
–
–
Baphia capparidifolia
1.17
–
–
–
7.36
–
Oubanguia laurifolia
0.95
–
–
–
–
6.56
Drypetes simulans
0.94
–
–
5.75
––
12.55
–
et al. 1992). Moreover, the individual-based species accumulation curve showed that the
rate of species increase with sampling effort had not yet reached an asymptote, indicating
that the diversity of the area had not yet been fully captured (Fig. 3).
The number of 450 species assessed in this study is higher than the number previously
recorded by Letouzey (1985) during itinerant surveys in the Ngovayang Massif. He
recorded only 86 species (including 13 Caesalpinioideae), which were represented by a
123
Author's personal copy
Biodivers Conserv (2011) 20:2627–2648
2639
total of 250 individuals in our study site. Furthermore, non-tree taxa are also highly
diverse. Droissart (2009) mentioned that out of ten taxa of Orchidaceae recently recorded
for the first time from Cameroon, two were collected in the Ngovayang Massif. This area
has also yielded new species of Rubiaceae (Sonké et al. 2006a, b, 2008a, b), mostly
understory treelets. Collectively, these studies show that the Ngovayang flora is particularly rich, and still remains incompletely known.
Diversity and endemism in a regional context
The Southern Ngovayang is among the most species-rich sites of the range of small
mountains bordering the gulf of Guinea, with a mean of 110 tree species ha-1 (Table 4).
Based on the S(50) diversity measure of the dataset from 264 0.1-ha transect sections in
Central African forests, the Ngovayang lowland forest (31.1 ± 1.2) stands fourth in terms
of tree biodiversity after the neighboring Bipindi-Akom II–Lolodorf area (33.8 ± 1.2),
Massif du Chaillu 1 (32.2 ± 3.1) and Monts de Cristal 2 (32.0 ± 5). S(50) diversity for
Ngovayang was slightly but not significantly lower than that for Bipindi-Akom II–Lolodorf
with the highest value (Kruskal–Wallis test, P = 0.188). Similar results were obtained for
Fisher’s a values for seventy 1-ha plots, with 42 ± 2.5 for Ngovayang lowland forest, and
54.2 ± 7 for Bipindi-Akom II–Lolodorf area, (Kruskal–Wallis test, P = 0.116). Thus,
values of diversity indices for the Ngovayang Massif are close to those previously found in
other Atlantic Central African forests (Monts de Cristal, Monte Mitra, Campo-Ma’an,
Bipindi-Akom II–Lolodorf) considered to be species-rich and with high levels of endemism (White 1983; Davis et al., 1994; van Gemerden et al. 2003; Tchouto et al. 2006,
2009). Furthermore, the Fisher’s a value for Ngovayang is significantly higher (Kruskal–
Wallis test, P = 0.039) than that for forests of Korup National Park (mean of 30.5), widely
known for its high diversity and endemism.
These results confirm those of Parmentier et al. (2011), which predict that Atlantic
Central African forests are among the most species-rich in Africa and that the highdiversity areas identified in Atlantic Central African forests correspond to hilly regions at
medium elevation. These authors mentioned the area stretching from Campo-Ma’an to
Bipindi-Akom II–Lolodorf as the most diverse area (Fisher’s a [ 40) in southern Cameroon and confirm the importance of this area in terms of species richness. The new data
we report show that Ngovayang should be included in a broad Campo-Ma’an–Ngovayang
area of high species richness.
Most studies in the sites mentioned above focus on a diversity; very few studies have
examined endemism based on quantitative data. Furthermore, the use of different inventory
methods makes it difficult to compare endemism based on published data. Species
inventories using standardized methods are required for rigorous comparison of endemism
among sites. However, the comparisons done in this study already give a first idea of the
level of endemism in different Central African forest sites.
Within this region of high biodiversity, the Ngovayang Massif includes a large proportion of species restricted to Lower Guinea (43%), comparable to those for Korup (44%)
and Monte Alen National Parks (45%) but higher than those for the Dja Fauna Reserve,
central Gabon forests and Campo-Ma’an National Park (Table 5). The proportion of
Lower-Guinea species in different sites is highest near the coast (23% for Dja and central
Gabon forests to 43% for Monte Alen, Ngovayang and Korup). In Campo-Ma’an National
Park, identified as a biodiversity hotspot due to its high levels of species richness and
endemism (Tchouto et al. 2006, 2009), the proportion of Lower Guinea endemics
(including Cameroonian endemics) is lower (29%), but this apparent difference may
123
Diversity
Rainfall
(mm year-1)
Bipindi-Akom
2-Lolodorf
(Cameroon)
19
2,000
Massif du Chaillu
1 (Gabon)
47
Monts de Cristal 2
(Gabon)
28
2,700
Ngovayang
(Cameroon)
5
2,000
532 – 75
34.6 – 5.4
110 – 9
4.00 – 0.1ab
54 – 6ab
Monte Mitra
(Equatorial
Guinea)
3
3,250
548 ± 108
31.2 ± 2.8
108 ± 16
3.92 ± 0.3ab
Monts de Cristal 1
(Gabon)
5
2,700
562 ± 17
39.5 ± 6.5
100 ± 10
Waka (Gabon)
2
1,700
589 ± 50
46.3 ± 2.0
106 ± 4
Mean
density
Mean basal
area
(m2 ha-1)
Mean
species ha-1
Mean
Shannon
diversity
S(50)
Source
54.2 ± 7.0b
33.8 ± 1.2c
Parmentier
et al. (2011)
–
–
–
–
–
–
–
–
–
–
–
32.2 ± 3.1bc
Parmentier
et al. (2011)
–
–
–
–
–
–
32.0 ± 5bc
Parmentier
et al. (2011)
42.0 – 2.5ab
31.1 – 1.2bc
This study
52 ± 17ab
41.4 ± 13.6ab
30.3 ± 3.4bc
Balinga et al.
(2005)
3.92 ± 0.14ab
51 ± 8ab
35.7 ± 5.9ab
30.2 ± 1.9bc
Sunderland
et al. (2004)
3.90 ± 0.11ab
49 ± 5ab
37.0 ± 0.4ab
30.3 ± 1bc
Balinga (2006)
ab
ab
ab
3.86 ± 0.13
48 ± 6
37.6 ± 6.8
29.8 ± 1.3bc
Sunderland
et al. (2003)
3.72 ± 0.02ab
41 ± 1ab
33.2 ± 3.3a
28.7 ± 0.9ab
Sunderland and
Balinga
(2005)
–
28.3 ± 5.7b
Parmentier
et al. (2011)
–
–
27.6 ± 3.4ab
Parmentier
et al. (2011)
–
–
27.4 ± 4.8ab
Parmentier
et al. (2011)
Takamanda
(Cameroon)
6
4,500
446 ± 40
33.5 ± 7.6
95.5 ± 11
Nouabale-Ndoki
(Congo)
2
1,700
300 ± 11
19.9 ± 3.7
77 ± 4
Campo-Ma’an
(Cameroon)
67
2,300
–
–
–
–
–
Massif du Chaillu
2 (Gabon)
14
2,800
–
–
–
–
6
2,350
–
–
–
–
Rabi-Mayombe
(Gabon)
Mean
Fisher’s a
Author's personal copy
Number
of plots
Biodivers Conserv (2011) 20:2627–2648
Locality
2640
123
Table 4 Comparison of Ngovayang tree diversity with that of other forest sites in Central Africa and the Neotropics, based on 1-ha (100 m 9 100 m) plots, by decreasing
values of Fisher’s a indices
Locality
Central Gabon
forests (Gabon)
Number
of plots
Rainfall
(mm year-1)
Mean
density
Mean basal
area
(m2 ha-1)
2,350
–
14
5,272
492 ± 5
Dja (Cameroon)
9
1,500
–
–
Ejagham
(Cameroon)
2
–
–
Mbam-Djerem
(Cameroon)
3
1,650
Odzala (Congo)
19
1,500
Korup (Cameroon)
–
Diversity
Mean
Fisher’s a
Source
27.1 ± 6.3ab
Parmentier
et al. (2011)
–
–
–
30.4 ± 6.5a
25.6 ± 3ab
Parmentier
et al. (2011)
–
–
–
30.9 ± 20.6a
23.5 ± 11.9ab
Parmentier
et al. (2011)
–
–
–
–
28.1 ± 4.2a
22.2 ± 2.8ab
Parmentier
et al. (2011)
–
–
–
–
–
14.0 ± 5.7a
19.5 ± 3.9ab
Parmentier
et al. (2011)
–
–
–
–
–
17.8 ± 2.9a
Parmentier
et al. (2011)
86 ± 12
–
S(50)
–
26 ± 4.0
–
Mean
Shannon
diversity
–
The average true diversity (D) and Shannon diversity index (H0 ) differed significantly among Central African sites (Kruskal–Wallis test, P = 0.0129). Values for Fisher’s a
and S(50) showed highly significant differences among Central African sites (Kruskal–Wallis test, P \ 0.0001). All plots included in the comparison were located below
700 m altitude. Means ± standard deviation are presented. a, b, c: means followed by different letters in superscript are significantly different (P \ 0.05). Values in the same
column that are followed by the same letter in superscript are not significantly different
Data selected from Parmentier et al. (2011) were used for mean comparisons of S(50) and alpha Fisher indices
Author's personal copy
9
Mean
species ha-1
Biodivers Conserv (2011) 20:2627–2648
Table 4 continued
2641
123
Author's personal copy
2642
Biodivers Conserv (2011) 20:2627–2648
Table 5 Percentages of chorological categories of tree species in the Ngovayang Massif and other forest
sites in Central Africa
Site
Percentage of species in chorological categories
Lower
Guinea
Cameroon
Source
Gabon
Total
number of
speciesa
Widespread
GuineoCongolian
CampoMa’an
(Cameroon)
17
49
5
23
6
0
2,297
Tchouto
et al.
(2006)
Central
Gabon
forests
(Gabon)
10
55
6
22
0
7
585
Doucet
(2003)
Dja
(Cameroon)
12
49
16
23
0
0
267
Senterre
(2005)
Korup
(Cameroon)
6
29
21
33
11
0
365
Kenfack
et al.
(2006)
Monte Alen
(Equatorial
Guinea)
8
32
15
45
0
0
718
Senterre
(2005)
Ngovayang
(Cameroon)
12
38
7
32
11
0
361
This
study
a
Upper and
Lower
Guinea
The total number of species mentioned here are those which were identified to species level in each site
simply reflect the inclusion in the inventory for this site of secondary-forest and swampy
environments, which harbor numerous species with large geographic ranges.
Additional indications of high endemism in the Massif come from studies focusing on
the distribution of particular taxonomic groups. For Rubiaceae, Sonké et al. (2006c) and
Nguembou (2006) showed that the two Massifs of Ngovayang (or the ‘‘Bipindi Massif’’ as
they called it) and Akom 2 exhibit higher values of species richness and endemism for this
plant family than do Mount Cameroon and Korup, usually cited as the areas with the
highest levels of species richness and endemism in Central Africa. These authors also
showed that these areas of high endemism for Rubiaceae coincide with those previously
identified for Begonia (Sosef 1994, 1996) and Orchidaceae (Droissart et al. 2006). More
recently, phytogeographical studies focusing on the distribution of Orchidaceae endemic to
Atlantic Central Africa revealed the concordance between centres of endemism and the
Ngovayang-Mayombe line of hills and small mountains (Droissart 2009).
One of the specificities of Ngovayang is the presence of certain rare (and rarely collected) species previously known only from one or two sites. For example, Hymenostegia
klainei (otherwise known only from Gabon; Doucet 2003) and Tricalysia achoundongiana
(known previously only from Korup and Monte Alen; Senterre 2005) were also found in
the Ngovayang Massif.
The Ngovayang Massif, located in a region influenced by the Atlantic monsoon, benefits
from moisture supplied by the Atlantic Ocean and receives high amounts of precipitation
(C2000 mm year-1), considerably more than the sites inland such as Nouabale-Ndoki
(Suchel 1972 and Fig. 2). The mountainous relief leads also to great variation in rainfall
and temperature (Suchel 1972), and certain climatic conditions (such as wind exposure,
persistent high cloudiness, mist and fog, or foehn effect) are found only in these
123
Author's personal copy
Biodivers Conserv (2011) 20:2627–2648
2643
geomorphologically heterogeneous areas (Sosef 1994), leading to a great diversity of
microclimates. In consequence, the Ngovayang Massif, although not very high in altitude,
might present over small distances a wide variety of ecological niches, from lowland to
submontane forests, providing habitats for a wide range of species, including endemic
species (Sosef 1994; Leal 2004).
Also, owing to its geographically intermediate position, the Ngovayang Massif may
have been able to recruit species from two distinct pools in phytogeographic regions
known for their high diversity and endemism; according to Linder 2001, the first one to the
south (from Equatorial Guinea to southern Congo) and the second one to the north (from
Nigeria to the Cameroon Volcanic Line and southern Cameroon). This position, at the
hinge of various phytogeographic influences, might also contribute to explaining the high
biodiversity levels recorded in this area.
Apart from niche partitioning and migration of flora, the history of the forests bordering
the Gulf of Guinea may have played a role in the high levels of biodiversity of Ngovayang
and other small hills. Maley (1987) postulated that, during arid phases, persistent stratiform
clouds along the Atlantic coast of Central Africa provided a source of fine precipitation and
humidity in the small hills and mountains, within a generally dryer climate. The combination of these climatically and geomorphologically favorable characteristics may have
favored the maintenance of a forest cover during past geological times in the Gulf of
Guinea coastal regions, including the Ngovayang Massif, which—in turn—favored the
maintenance of species and a rich ‘‘paleo-diversity’’ (Maley 1987, 1989; Sosef 1994; Plana
2004). Moreover, according to Linder (2001), there is a remarkable congruence between
centres of high diversity with areas of high endemism in Lower Guinea; he mentions the
Atlantic forests of Cameroon among these centres. He also showed that rainfall is the best
predictor of diversity, whereas climatic stability is an important determinant of high degree
of endemism. Thus, the lower annual rainfall (1,100–1,700 m; Table 2) and longer periods
of dry-season stress (1–4 months) observed in Nouabale-Ndoki, Dja, Mbam-Djerem and
Odzala, might contribute to explaining the lower diversity of these inland sites, compared
to most of the sites bordering the Gulf of Guinea.
Elements of the flora and their significance
In terms of FIV, Fabaceae-Caesalpinioideae was the most important (sub) family throughout
the area (Table 3). This result corroborates, with quantitative data, the findings of Letouzey
(1985), who characterized forests of this region as Atlantic Biafran evergreen forest rich in
Fabaceae-Caesalpinioideae. The importance of this legume subfamily in our study site is one
of the characteristics of the Guineo-Congolian forests (White 1983). This subfamily is also
considered characteristic of old forests and of forests suspected to have been rainforest refugia
(Rietkerk et al. 1996; Leal 2001; Maley 2002). The same trend was observed in the ‘‘Bipindi
area’’ (van Gemerden 2004) and in the Campo-Ma’an, both located to the south of the
Ngovayang Massif (Fig. 2), and considered to be part of a series of forest refugia bordering
the Atlantic coast, from Cameroon to the mouth of the Congo River (Tchouto et al. 2009).
Tchouto et al. (2009) found that Fabaceae-Caesalpinioideae forest had a high proportion of
bio-indicator species of rainforest refugia (i.e., those that reflect changes in environmental
conditions; Heywood and Watson 1995). This also appears to be the case in the Ngovayang
Massif, which is also rich in Fabaceae-Caesalpinioideae (Table 3).
The gregarious nature of numerous species of Fabaceae-Caesalpinioideae is also a
characteristic feature of this subfamily in the Atlantic Biafran evergreen forest (Letouzey
1985). This is confirmed by our results, which show a marked importance (high FIV
123
Author's personal copy
2644
Biodivers Conserv (2011) 20:2627–2648
values) and clumped distribution for some of the key species within our plots (Table 3),
Guibourtia tessmannii and Didelotia letouzeyi being typical examples. This subfamily
includes many typical canopy species, which were key species of the upper strata of the
forest structure in our plots (Fig. 4).
Other important families such as Olacaceae and Burseraceae are also known to be good
indicators of old evergreen Atlantic forests (Doumenge 1992; Senterre and Lejoly 2001).
Doucet (2003) also showed, in the case of Gabon’s rainforests, that high levels of endemism were associated with high dominance of Fabaceae-Caesalpinioideae, Burseraceae
and Olacaceae.
Another family, Rubiaceae, typically diverse in African rainforests, was one of the most
important families (FIV) inventoried in this study (Table 2). Many species of this family
were recognized as bio-indicators of rainforest refugia within the Campo-Ma’an area
(Tchouto et al. 2009). Because many narrow endemic species are also known to occur in
the Akom II-Bipindi–Lolodorf area, which lies between Ngovayang and Campo-Ma’an,
Tchouto et al. (2009) postulated that the Campo-Ma’an area refugium probably extended
along the range of hills right to the Bipindi area. Although the Ngovayang Massif was not
explicitly mentioned in that study, it is near Bipindi and lies within the ‘‘Bipindi area’’
defined by Tchouto et al. (2009). The proximity of the Ngovayang Massif to these other
small mountains may have facilitated dispersal between them, favoring the maintenance of
more species than if the Ngovayang Massif had been isolated from other hills and other
putative refugia.
Conclusion
The importance of the Ngovayang forest Massif for the conservation of tree diversity
within Cameroon, and within African forests more generally, has long been overlooked
(Gartlan 1989; Doumenge et al. 2003). We have shown here that, in terms of tree alpha
diversity, it is one of the richest sites within Central African rainforests. Its diversity and
level of endemism are similar to those previously found for other Atlantic Central African
forests reputed for their high species richness and endemism. These findings may also be
corroborated by further studies, both on flora and fauna. Although very few studies have
focused on fauna, some non-governmental organisations such as Bird Life International
have tentatively designated the Ngovayang Massif as a potential Important Bird Area
(IBA) on the basis of its suspected ornithological importance, in addition to the 33 IBAs
already identified in Cameroon (Fishpool and Evans 2001). At least one threatened bird
species, the grey-necked rockfowl (Picathartes oreas), is known to occur in the Ngovayang
Massif. Although data on the mammal fauna are lacking, the presence of Gorilla (Gorilla
gorilla gorilla) has also been confirmed by local hunters to one of us. Our results show
that, by the criteria of high diversity, high levels of endemism and the presence of a large
number of rare species (Myers et al. 2000; Kier and Barthlott 2001), the Ngovayang
forests, at least those of the southern quarter of the Massif, are of great conservation value.
Their protection and sustainable use should be of high priority for conservation decisionmakers in Cameroon.
Unless the area is protected, human-generated disturbances such as uncontrolled logging, hunting or transient farming may rapidly reduce the biodiversity value of the
Ngovayang Massif in the near future, by exerting severe negative impacts on its rich but
poorly known fauna and flora. There is an urgent need to consider this area within the
national strategy for conservation and sustainable development, fostering actions to
123
Author's personal copy
Biodivers Conserv (2011) 20:2627–2648
2645
preserve this Massif and its potential for providing ecosystem services both to local
populations and to the nation and the region. This study begins to fill in an important gap in
the floristic knowledge of southern Cameroon. Its results give a first idea of the diversity
and endemism of the southern part of the Massif. Further studies characterizing plant and
animal diversity, presence and distribution of endemic and rare species in the entire
Ngovayang Massif are required to confirm these first conclusions, and thereby gain the data
needed to support informed decisions on conservation and management of these forests.
Acknowledgments We thank ‘‘Sud Expert Plantes, project#374’’ (SEP), a program funded by the French
Ministry of Foreign Affairs, and the Centre de Cooperation Internationale en Recherche Agronomique pour
le Developpement (CIRAD), for financial assistance through research grants which allowed this study. We
are grateful to N. Fauvet for his technical assistance in GIS and the preparation of the maps, and to
G. Walter, C. Geldenhuys, B. Sinsin and V. Freycon for valuable comments on earlier drafts of this article.
We also thank two anonymous reviewers whose constructive remarks improved the manuscript. Special
thanks are extended to M. Sainge, P. Mezili and J.-M. Onana for their technical expertise on specimen
identification in the field and at the National Herbarium, Cameroon. We thank the people of Ngovayang for
allowing us to work on their land and for assisting us in fieldwork. The authors would like to thank the Plant
Biology Department of the University of Yaoundé I for their co-operation in facilitating this research.
References
Anhuf D, Ledru M-P, Behling H, Da Cruz FW Jr, Cordeiro RC, Van der Hammen T, Karmann I, Marengo
JA, De Oliveira PE, Pessenda L, Siffedine A, Albuquerque AL, Da Silva Dias PL (2006) Paleoenvironmental change in Amazonian and African rainforest during the LGM. Palaeogeogr Palaeoclimatol Palaeoecol 239:510–527
APG (the Angiosperm Phylogeny Group) (2003) An update of the angiosperm phylogeny group classification for the orders and families of flowering plants: APG II. Bot J Linn Soc 141:399–436
Aubréville A, Leroy J-F (1961–1992) Flore du Gabon, Muséum National d’Histoire Naturelle, Paris
Aubréville A, Leroy J-F (1963–1978) Flore du Cameroun, vols 1–20. Muséum National d’Histoire
Naturelle, Paris
Balinga MPB (2006) A vegetation assessment of the Waka national park, Gabon. CARPE Report.
http://carpe.umd.edu/resources/Documents/. Cited 4 July 2009
Balinga MPB, Issembe YA, Sunderland TCH, Nzabi T, Obiang D, Nyangadouma R (2005) Quantitative
vegetation assessment of the Monte Mitra forest using 1 hectare biodiversity plots (BDP’s). In:
Sunderland TCH (ed) A biodiversity assessment of the Monte Mitra forest, Monte Alen National Park,
Equatorial Guinea. http://carpe.umd.edu/resources/Documents/. Cited 4 July 2009
Campbell DG, Stone JL, Rosas A Jr (1992) A comparison of the phytosociology and dynamics of three
floodplain (Varzea) forests of known ages, Rio Jurua, western Brazilian Amazon. Bot J Linn Soc
108:213–237
CJB (2010) African flowering plants database. http://www.ville-ge.ch/musinfo/bd/cjb/africa. Cited 11 May
2010
Curtis JT, McIntosh RP (1951) An upland forest continuum in the prairie-forest border region of Wisconsin.
Ecology 32:476–496
Dallmeier F (ed) (1992) Long-term monitoring of biological diversity in tropical forest areas. Methods for
establishment and inventory of permanent plots. MAB Digest Series, 11. UNESCO, Paris
Davis SD, Heywood VH, Hamilton AC (1994) Centres of plant diversity: a guide and strategy for their
conservation, vol 1: Europe South West Asia and the Middle East Africa. IUCN and WWF, Cambridge
Doucet JL (2003) L’alliance délicate de la gestion forestière et de la biodiversité dans les forêts du centre du
Gabon. Thèse de Doctorat, Faculté universitaire des sciences agronomiques de Gembloux, Belgique
Doumenge C (1992) La Réserve de Conkouati: Congo Le secteur sud-ouest. UICN, Gland, Suisse
Doumenge C, Ndinga A, Fomete Nembot T, Tchanou Z, Micha Ondo V, Ona Nze N, Bourobou Bourobou
H, Ngoye A (2003) Conservation de la biodiversité forestière en Afrique centrale atlantique. II.
Identification d’un réseau de sites critiques. Bois For Trop 276:43–58
Droissart V (2009) Etude taxonomique et biogéographique des plantes endémiques d’Afrique centrale
atlantique : le cas des Orchidaceae. Thèse de doctorat, Université Libre de Bruxelles, Belgique
123
Author's personal copy
2646
Biodivers Conserv (2011) 20:2627–2648
Droissart V, Sonké B, Stévart T (2006) Les Orchidaceae endémiques d’Afrique centrale atlantique présentes
au Cameroun. Syst Geogr Pl 76:3–84
Fishpool LDC, Evans MI (eds) (2001) Important Bird Areas in Africa and associated islands: priority sites
for conservation. BirdLife Conservation Series No. 11. Pisces Publications and BirdLife International,
Newbury
Franqueville A (1973) Atlas Regional Sud-Ouest 1. République du Cameroun. ORSTOM, Yaoundé
Gartlan JS (1989) La conservation des écosystèmes forestiers du Cameroun. IUCN, Suisse
Gentry AH (1988) Tree species richness of upper Amazonian forests. Proc Natl Acad Sci USA 85:156–159
Heywood VH, Watson RT (1995) Global biodiversity assessment. Cambridge University Press, Cambridge
Hurlbert SH (1971) The non concept of species diversity: a critique and alternative parameters. Ecology
52:577–586
Hutcheson K (1970) A test for comparing diversities based on the Shannon formula. J Theor Biol 29:
151–154
IUCN (2010) IUCN Red List of threatened species, version 2010.4. www.iucnredlist.org. Downloaded on 01
May 2011
Jost L (2006) Entropy and diversity. Oikos 113:363–375
Keay RWJ, Hepper FN (1954–1972) Flora of West Tropical Africa, 2nd edn, vols 1–3. Crown Agents,
London
Kenfack D, Thomas DW, Chuyong G, Condit R (2006) Rarity and abundance in a diverse African forest.
Biodivers Conserv. doi:10.1007/s10531-006-9065-2
Kent M, Coker P (1992) Vegetation description and analysis. Belhaven Press, London
Kier G, Barthlott W (2001) Measuring and mapping endemism and species richness: a new Methodological
approach and its application on the flora of Africa. Biodivers Conserv 10:1513–1529
Küper W, Kreft H, Nieder J, Koster N, Barthlott W (2004) Large-scale diversity patterns of vascular
epiphytes in neotropical montane rain forests. J Biogeogr 31:1477–1487
Leal ME (2001) Microrefugia, small scale ice age forest remnants. In: Robbrecht E, Degreef J, Friis I (eds)
Plant systematics and phytogeography for the understanding of African biodiversity. Proceedings
XVIth AETFAT congress. Systematics and Geography of Plants, vol 71. National Botanic Garden of
Belgium, Meise, pp 1073–1077
Leal ME (2004) The African rain forest during the last glacial maximum, an archipelago of forests in a sea
of grass. Ph.D Thesis, Wageningen University, Wageningen, The Netherlands
Lebrun JP, Stork AL (1991–1997) Enumération des plantes à fleur d’Afrique tropicale, vol 1 (1991), vol 2
(1992), vol 3 (1995), vol 4 (1997). Conservatoire et Jardin botaniques, Genève
Letouzey R (1968) Etude phytogéographique du Cameroun. Lechevallier P, Paris
Letouzey R (1985) Notice de la carte phytogéographique du Cameroun, vols 1–5. Institut de la carte
Internationale de la Végétation, Toulouse
Lieberman D, Lieberman M, Peralta R, Hartshorn GS (1996) Tropical forest structure and composition on a
large-scale altitudinal gradient in Costa Rica. J Ecol 84:137–152
Linder HP (2001) Plant diversity and endemism in sub-Saharan tropical Africa. J Biogeogr 28:169–182
Lovett JC, Rudd S, Taplin J, Frimodt-Moller C (2000) Patterns of plant diversity in Africa south of the
Sahara and their implications for conservation management. Biodivers Conserv 9:37–46
Magurran AE (2004) Measuring biological diversity. Blackwell, Oxford
Maley J (1987) Fragmentation de la forêt dense humide africaine et extension des biotopes montagnards du
quaternaire récent: nouvelles données polliniques et chronologiques. Implications paléoclimatiques et
biogéographiques. Paleoecol Afr 18:307–334
Maley J (1989) Late quaternary climatic changes in African rain forest: the question of forest refuges and the
major role of sea surface temperature variations. In: Lewen M, Sarthein M (eds) Paleoclimatology and
paleometeorology: modern and past pattern of global atmospheric transport. Kluwer, London,
pp 585–616
Maley J (2002) A catastrophic destruction of African forests about 2,500 years ago still exerts a major
influence on present vegetation formations. IDS Bull 33:13–30
Mori SA, Boom BM, Carvalino AM, Dos Santos TS (1983) The ecological importance of Myrtaceae in an
eastern Brazilian wet forest. Biotropica 15:68–70
Myers N, Mittermeier RA, Mittermeier CG, da Fonseca GAB, Kent J (2000) Biodiversity hot spots for
conservation priorities. Nature 403:853–858
Nguembou KC (2006) Distribution des Rubiaceae endémiques du Cameroun. Mémoire de DEA, Université
Libre de Bruxelles, Belgique
Oksanen J, Blanchet G, Kindt R, Legendre P, O’Hara RB, Simpson GL, Solymos P, Stevens MHH, Wagner
H (2010) Vegan: community ecology package, version 1.17-3. http://CRAN.R-project.org/package=
vegan
123
Author's personal copy
Biodivers Conserv (2011) 20:2627–2648
2647
Olivry JC (1986) Fleuves et rivières du Cameroun Collection Monographies Hydrologiques d’ORSTOM 9.
MESRES-ORSTOM, Paris
Parmentier I, Harrigan RJ, Buermann W, Mitchard ETA, Saatchi S, Malhi Y, Bongers F, Hawthorne WD,
Leal ME, Lewis SL, Nusbaumer L, Sheil D, Sosef MSM, Affum-Baffoe K, Bakayoko A, Chuyong GB,
Chatelain C, Comiskey JA, Dauby G, Doucet J-L, Fauset S, Gautier L, Gillet J-F, Kenfack D, Kouame
FN, Kouassi EK, Kouka LA, Parren MPE, Peh KS-H, Reitsma JM, Senterre B, Sonke B, Sunderland
TCH, Swaine MD, Tchouto MGP, Thomas D, Van Valkenburg JLCH, Hardy OJ (2011) Predicting
alpha diversity of African rain forests: models based on climate and satellite-derived data do not
perform better than a purely spatial model. J Biogeogr. doi:10.1111/j.1365-2699.2010.02467.x
Phillips OL, Baker TR, Arroyo L, Higuchi N, Killeen TJ, Laurance WF, Lewis SL, Lloyd J, Malhi Y,
Monteagudo A, Neill DA, Vargas PN, Silva JNM, Terborgh J, Martinez RV, Alexiades M, Almeida S,
Brown S, Chave J, Comiskey JA, Czimczik CI, Di Fiore A, Erwin T, Kuebler C, Laurance SG,
Nascimento HEM, Olivier J, Palacios W, Patino S, Pitman NCA, Quesada CA, Salidas M, Lezama AT,
Vinceti B (2004) Pattern and process in Amazon tree turnover, 1976–2001. Philos Trans R Soc Lond B
359(1443):381–407
Pitman NCA, Terborgh JW, Silman MR, Muñez PV (1999) Tree species distributions in an upper Amazonian forest. Ecology 80:2651–2661
Pitman NCA, Terborgh JW, Silman MR, Nùñez PV, Neill DA, Cerón CE, Palacios WA, Aulestia M (2002)
A comparison of tree species diversity in two upper Amazonian forests. Ecology 83(11):3210–3224
Plana V (2004) Mechanisms and tempo of evolution in the African Guineo–Congolian rain forest. Philos
Trans R Soc Lond B 359(1450):1585–1594
Reitsma JM (1988) Forest vegetation of Gabon. Tropenbos Technical Series 1. The Tropenbos Foundation,
Wageningen
Rietkerk M, Ketner P, De Wilde JJFE (1996) Caesalpinioideae and the study of forest refuges in Central
Africa. In: van der Maesen LJG, van der Burgt XM, de Rooy JM (eds) The biodiversity of African
plants. Kluwer, Dordrecht, pp 619–623
Senterre B (2005) Recherches méthodologiques pour la typologie de la végétation et la phytogéographie des
forêts denses d’Afrique tropicale. Thèse de doctorat, Université Libre de Bruxelles, Belgique
Senterre B, Lejoly J (2001) Tree diversity in the Nsork rain forest (Rio Muni, Equatorial Guinea). Acta Bot
Gallica 148:227–235
Small A, Martin TG, Litehing RL, Wong KM (2004) Contribution of tree species to the biodiversity of a
1 ha Old World rainforest in Brunei, Borneo. Biodivers Conserv 13:2067–2088
Sonké B, Nguembou KC, Cheek M, Davis AP (2006a) A new species of Chassalia (Rubiaceae, Rubioideae)
from southern Cameroon: C. bipindensis. Kew Bull 61:585–589
Sonké B, Nguembou KC, Davis AP (2006b) A new dwarf Coffea (Rubiaceae) from southern Cameroon. Bot
J Linn Soc 151:425–430
Sonké B, Nguembou KC, Neuba D (2006c) Les Rubiaceae endémiques du Cameroun et leur statut de
conservation. In: Beetje H, Ghazanfar S (eds) Taxonomy and ecology of African plants, their conservation and uses. Royal Botanic Gardens, Kew, pp 97–109
Sonké B, Dessein D, Taedoumg H, Groeninckx I, Robbrecht E (2008a) A new species of Colletoecema
(Rubiaceae) from southern Cameroon with a discussion of relationships among basal Rubioideae.
Blumea 53:533–547
Sonké B, Djuikouo KM-N, Robbrecht E (2008b) A new species of Calycosiphonia (Afrotropical Rubiaceae)
from the ‘Lower Guinea’ area. Nord J Bot 25:275–280
Sosef MSM (1994) Refuge Begonias. Taxonomy, phylogeny and historical biogeography of Begonia sect.
Loasibegonia and sect. Scutobegonia in relation to glacial rain forest refuges in Africa. PhD Thesis,
Wageningen Agricultural University, The Netherlands
Sosef MSM (1996) Begonias and African rain forest refuges: General aspects and recent progress. In: Van
Der Maesen LJG, Van Der Burgt X M, Van Mendenbach De Rooy JM (eds) The Biodiversity of
African plant. Wageningen, 14:602-611
Suchel J-B (1972) La répartition des pluies et les régimes pluviométriques au Cameroun. Travaux et
Documents de Géographie Tropicale, CEGET-CNRS, Bordeaux, France et Université Fédérale du
Cameroun, Yaoundé, Cameroun
Sunderland TCH, Balinga MB (2005) Évaluation préliminaire de la végétation du parc national de Nouabalé-Ndoki et de sa zone tampon, Congo. CARPE Report. http://carpe.umd.edu/resources/Documents/.
Cited 4 July 2009
Sunderland TCH, Comiskey JA, Besong S, Mboh H, Fonwebon J, Dione MA (2003) Vegetation assessment
of Takamanda Forest Reserve, Cameroon. In: Comiskey JA, Sunderland TCH, Sunderland-Groves JL
(eds) Takamanda: the biodiversity of an African rainforest, SI/MAB series no. 8. Smithsonian
123
Author's personal copy
2648
Biodivers Conserv (2011) 20:2627–2648
Institution, Washington. http://nationalzoo.si.edu/ConservationAndScience/MAB/conservation/about/
publications.cfm. Cited 4 July 2009
Sunderland TCH, Walters G, Issembe Y (2004) A preliminary vegetation assessment of the Mbé National
Park, Monts de Cristal, Gabon. CARPE Report. http://carpe.umd.edu/resources/Documents/. Cited 4
July 2009
Tchouto MGP, Yemefack M, De Boer WF, De Wilde JJFE, Cleef AM (2006) Biodiversity hotspots and
conservation priorities in the Campo-Ma’an rainforests, Cameroon. Biodivers Conserv 15:1219–1252
Tchouto MGP, De Wilde JJFE, De Boer WF, Van der Maesen LJG, Cleef AM (2009) Bio- indicator species
and Central African rain forest refuges in the Campo-Ma’an area, Cameroon. Syst Biodivers 7:21–31
van Gemerden BS (2004) Disturbance, diversity and distributions in Central African rain forest. PhD Thesis,
Wageningen University, The Netherlands
van Gemerden BS, Olff H, Parren MPE, Bongers F (2003) The pristine rain forest? Remnants of historical
human impacts on current tree species composition and diversity. J Biogeogr 30:1381–1390
Waterloo MJ, Ntonga JC, Dolman AJ, Ayangma AB (2000) Impact of shifting cultivation and selective
logging on the hydrology and erosion of rain forest land in south Cameroon. Tropenbos-Cameroon
Documents 3. Tropenbos-Cameroon Programme, Wageningen
White F (1979) The Guineo-Congolian Region and its relationships to other phytochoria. Bull Jard Bot Nat
Belg 49:11–55
White F (1983) The vegetation of Africa. UNESCO, Paris
123
Michael Balinga