Efﬁciency of conservation areas to protect orchid species in Benin,
, F.A. Azihou
, Y.S.C. Gogan
, M.D. Kouton
, B. Sinsin
Faculté des Sciences Agronomiques, Université d'Abomey-Calavi, 01 BP 526, Cotonou, Bénin
Faculté d'Agronomie, Université de Parakou, BP 123, Parakou, Bénin
Faculté des Lettres, Arts et Sciences Humaines, Département de Géographie, Université d'Abomey-Calavi, BP 248, Allada, Bénin
Faculté des Sciences et Techniques, Département de Biologie Végétale, Université d'Abomey-Calavi, 01 BP 4521, Cotonou, Bénin
Department of Plant and Soil Sciences, University of Pretoria, 1121 South Street, Pretoria 0002, South Africa
University of Rostock, Agricultural and Environmental Faculty, Grassland and Fodder Sciences, Justus-von-Liebig-Weg 6, 18059, Rostock, Germany
Received 21 November 2016
Received in revised form 13 February 2018
Accepted 23 February 2018
Available online xxxx
Edited by C Peter
The effectiveness of protected areas to guaranteefuture conservation of several plant species remains question-
able. This studywas carried out in the BiosphereReserve of Pendjari (BRP) and surrounding unprotected areas to
assess the efﬁciency of the reserve to conserve orchids. A total of 90 plots (52 in protected areas; 38 in unpro-
tected areas) were sampled. The recorded data include: orchid species, number of individuals per species, the
height and diameter at breast height of host trees. Diversity indices were used to assess the orchid diversity in
the protected and unprotected areas. Preferred habitat conditions of orchid species were investigated using
Constrained Correspondence Analysis. An independent t-test and two-way analysis of variance were performed
to assess an existing combined effect of vegetation type and the conservation status on the density of orchid spe-
cies. The Importance Value Index (IVI) was used to measure how dominant an orchid species is in a given zone
according to the conservation status of the zone. Only three epiphytic orchids (Calyptrochilum christyanum,
Cyrtorchis arcuata and Plectrelminthus caudatus) were recorded and all in gallery forest of unprotected areas. In-
deed, 67% and 58% of the orchid species were only recorded in unprotected areas and in gallery forest, respec-
tively. There was no signiﬁcant difference between the density of all recorded orchids in protected and
unprotected areas. The conservation status of the studied zone had a signiﬁcant effect on thedensities of Nervilia
kotschyi and Eulophia guineensis (p b0.0001). The highest IVI of N. kostchyi was observed in the protected area
and of E. guineensis was in the unprotected area.
This ﬁrst effort to compile a reference list of the orchid species ofthe BRP showed that some orchid species were
well represented within the protected area, but all of the epiphytic orchids were recorded from unprotected
areas. A representative gap can be assumed to existfor most epiphytic orchids only recorded in thegallery forests
of unprotected areas. Our results highlighted the need to redeﬁne protective management strategies for orchid
species in the BRP.
© 2018 SAAB. Published by Elsevier B.V. All rights reserved.
Biosphere Reserve of Pendjari
The role of protected areas in the preventionof extinction of species
has been much debated (Bruner et al., 2001). Several studies focused on
the effectiveness of the protected areas to ensure therepresentativeness
and persistence of biodiversity components (Defries et al., 2005;
Wittemyer et al., 2008; Houéhanou et al., 2011, 2012, 2013). Some of
the studies (Djossa et al., 2008; Gouwakinnou et al., 2009; Schumann
et al., 2010; Fandohan et al., 2011) have emphasized the positive effect
of protected areas to conserve some valuable species. However, a re-
view of conservation goals for different protected areas (Myers et al.,
2000; Diniz and Brito, 2015; Françoso et al., 2015)indicatedacleardif-
ference between expectations of conservation and the effectiveness in
species conservation. In addition, a gap of some priority areas for orchid
protection still needs to be ﬁlled by the existing protected areas net-
work (Wan et al., 2014). Furthermore, the number of speciesthreatened
with extinction far exceeds the projections of scientists (Myers et al.,
2000). This is the case of orchid species (CITES, 2017).
Orchids are distributed throughout the world from tropical to high
alpine areas (Delforge, 2001). The Orchid family is with the Asteraceae
the two largest families (Doyle and Luckow, 2003).
South African Journal of Botany 116 (2018) 230–237
⁎Corresponding author at: Faculté d'Agronomie, Université de Parakou, BP 123,
E-mail address: email@example.com (E.S.P. Assédé).
0254-6299/© 2018 SAAB. Published by Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
South African Journal of Botany
journal homepage: www.elsevier.com/locate/sajb
From the demographic explosion, correspondingly strong land mod-
iﬁcation was observed in West Africa (Wittig et al., 2007; Wittemyer
et al., 2008). Habitat alteration, including total loss, modiﬁcation, and
fragmentation was by far the main threat to most orchids in the tropics
(Dressier, 1981). As a result, signiﬁcant modiﬁcation of light intensity,
humidity, and other microclimatic factors affecting the survival of the
epiphytic orchids, were observed. Many orchid species in West Africa
are now considered to be at risk of extinction as a result of selective log-
ging of valuable timber species and clear-felling for agricultural devel-
opment (IUCN/SSC Orchid Specialist Group, 1996; Pillon et al., 2007;
Pant, 2013). Wild orchids have been overharvested at large scale to sup-
ply the medicinal, edible, and horticultural trades (Kasulo et al., 2009;
Pant, 2013; Ghorbani et al., 2014; Liu et al., 2014; Vermeulen et al.,
2014; Hinsley et al., 2015). Reliable statistics on the extent of the trade
in orchids in West Africa are scarce. However, several representatives
of the orchid family are under threat of extinction due to indiscriminate
collection (Cribb et al., 2005; Dunkan et al., 2005). In West Africa, ex-
traction of wild orchids for trade affects mostly those few orchid taxa
that either produce very showy ﬂowers or provide certain edible prod-
ucts (IUCN/SSC Orchid Specialist Group, 1996). As a result of these mul-
tiple threats, orchids featureprominently in theRed Data Book prepared
by International Union for Conservation of Nature (IUCN) (Pant, 2013).
The entire family is now included in Appendix II of the Convention on
International Trade in Endangered Species of Wild Fauna and Flora
(CITES, 2017). However, in West Africa, large populations of orchids
are still assumed to be present in their natural habitats in protected
areas. It is therefore paramount to assess how effective the protected
areas are in conserving the orchids in West Africa.
The Biosphere Reserve of Pendjari (BRP) is part of a well-managed
protected area network in West Africa. It conserves 28% of the total
ﬂora of Benin Republic (Assédé et al., 2012). Previous studies have
highlighted the importance of this reserve in plant conservation
(Gouwakinnou et al., 2009; Fandohan et al., 2011; Houéhanou et al.,
2011, 2013). Although the BRP is assumed to be the best way to con-
serve biodiversity of this area, its effectiveness in future conservation
of several plant species is not always guaranteed (Houéhanou et al.,
2013). Substantial representative gaps remain in its coverage of some
plant taxa. Terrestrial orchids are known to colonize both savanna and
forest areas (Delforge, 2001), while epiphytic orchids need appropriate
host plants on which to grow.
In the Sudanian zone of Benin Republic, the two major networks of
protected areas are focused on large mammal conservation. The
targeted zones to create the conservation areas are then savanna eco-
systems (the main habitat of those animals), covering up to 80% of the
total protected areas. Based on theecology of orchids, one might expect
a gap of representation in the network of the protected areas in the
study area. In Benin Republic, very few scientiﬁc studies have focused
on orchid species (Akoègninou et al., 2006).
The purpose of this paper is to assess the suitability of existing con-
servation areas to conserve orchid taxa in West Africa. This study there-
fore compares protected and unprotected areas to test the assumption
that the protected areas will have a higher conservation status of the or-
chid taxa than the unprotected areas. Speciﬁcally, we addressed two re-
search questions: (1) What are the habitat requirements of orchid
species occurring within the area? (2) How are orchid populations af-
fected by the conservation status of the land, i.e. in protected versus un-
2. Materials and methods
2.1. Study area
This research was conducted in the Biosphere Reserve of Pendjari
(BRP), located in the Sudanian zone of Benin Republic (West Africa)
and in its surrounding areas (Fig. 1). The BRP covers about
. It is composed of the National Park of Pendjari or core
zone (2660.4 km
) representing the protected area in this study, and
the hunting zones (Pendjari: 1750 km
and Konkombri: 251 km
the protected area, anthropogenic activities are strictly prohibited. The
surrounding areas representing unprotected areas in this study are
dominated by farmlands, fallows (disturbed savannas), and gallery for-
ests. The vegetation types in the unprotected areas were all subjected to
selective cutting of valuable tree species for timber and poles, livestock
grazing and harvesting of non-timber forest products. Gallery forest oc-
curs along the Pendjari river both inside and outside the reserve. The
dominant vegetation type in the protected area is savanna (wooded
grassland), intermingled with patches of woodland and grassland. The
climate is tropical with a ﬁve-month dry period (November–March).
The mean annual rainfall is 1000 mm with 60% rain between July and
September (Delvingt et al., 1989). Temperature varies between 21 °C
during the night, and up to 40 °C during the day (CENAGREF, 2016). Fer-
ruginous, indurate and swampy tropical soils occur in many areas of the
protected and unprotected areas. The BRP is surrounded by 20 villages
with subsistence agriculture as the main activity followed by livestock
breeding and natural resources harvesting. Logging and clearing of
land for agriculture remain the main sources of income for the local
population. Cultivated crops include rice, yams, maize, sorghum, millet,
and cotton; the latter being a cash crop and requires intense use of pes-
ticides (Delvingt et al., 1989).
2.2. Data collection
Data were collected between December 2014 and August 2015, i.e.
covering the two main seasons of the region: dry and rainy seasons.
The vegetation map of the reserve (König, 2005) was used to identify
the three main vegetation types (savannas, woodlands and gallery for-
ests). Both protected and unprotected areas were included in the
study. In each vegetation type, points were randomly selected to serve
as the starting point of a transect. In total, 65 transects of at least 3 km
long in each vegetation type were surveyed for the presence of orchid
Three-person teams were used to intensively survey the trees and
the vegetation on the ground for the presence of epiphytic and terres-
trial orchids. One team member monitored the compass bearing of the
transect, and the other two members scanned the vegetation for or-
chids. The presence of orchid species was conﬁrmed by two team mem-
bers. This method was adapted from Bergstrom and Carter (2008) and
Yulia et al. (2011).
Sample plots (Fig. 1) were selected alongthe transects, based on the
presence of orchid species. Rectangular plots (10 m × 50 m) were sam-
pled in gallery forests (due to the linear shape of the gallery forests) and
Square plots (30 m × 30 m) in savannas, woodlands and fallows. At least
25 plots were sampled per vegetation type, and 90 plots in total
(Table 1; with 52 plots in protected areas and 38 in unprotected areas).
The species and number of individuals of all orchid species (terres-
trial and epiphytic) and the identity of host trees of epiphytic orchids
were recorded within each plot. The dominant plant species of tree
and shrub layers were recorded. Simultaneously, ﬁeld data on environ-
mental variables were collected. These included the vegetation type,
vegetation cover, soil texture, and the presence of rocks or stones.
Signs of human disturbance, including agriculture, grazing, tree cutting
and pruning, were also collected within the plots of unprotected areas.
Herbarium specimens of all recorded orchid species were prepared
and conﬁrmed with the National ﬂora (Akoègninou et al., 2006)and
at the National Herbarium of Benin Republic.
2.3. Data analyses
2.3.1. Orchid diversity and habitat requirements
Diversity indices were used to assess the orchid diversity in the two
zones (protected and unprotected areas). The taxonomic diversity con-
siders the number of species, genera, and families. The Shannon-
231E.S.P. Assédé et al. / South African Journal of Botany 116 (2018) 230–237
Weaver index (H′), the most important diversity index (Magurran,
2004), was calculated with the following formula:
where, pi = ni/N and ni = Number of individuals of an orchid species.
N = Total number of individuals of all orchid species. H′is low (if
between 0 and 2.5); moderate (if between 2.6 and 3.9); or high (if be-
tween 4 and 5).
The Shannon measure of evenness (J′) was calculated using the fol-
where Hmax is computed as H
(SR); SR representing the spe-
The Shannon measure of evenness (J') has a theoretical minimum
value of zero when all orchids belong to the same species, and the
index value increases to 1 when the number of species increases.
Preferred habitat conditions of orchid species were investigated
using Constrained Correspondence Analysis (CCA). The aim of CCA
was to display the plots of protected and unprotected areas as groups
Fig. 1. Location of the study area with the plot sites.
Distribution of the sampled plots in protected and unprotected areas.
Ecosystems Protected area Unprotected areas Total
Tree and shrub savannas 20 8 28
Woodlands 16 15 31
Gallery forests 16 15 31
Total 52 38 90
232 E.S.P. Assédé et al. / South African Journal of Botany 116 (2018) 230–237
in ordination space, based on habitat characteristics. The environmental
variables tested were tree cover, herb cover, topography, altitude and
soil texture. A weighted method was used and environmental data are
reweighted at each permutation step using permutated weights. All
qualitative environmental variables were coded. The CCA model and
the signiﬁcance of the ﬁtted environmental variables were evaluated
by the Monte Carlo permutation test with 499 permutations. Monte-
Carlo permutation tests were also used to test the signiﬁcance of the or-
dination axes (499 permutations under reduced model). A P-value of
≤0.05 for the ﬁrst ordination axis was accepted as an indicator for a sig-
niﬁcant relationship between the items. Data were computed using the
Vegan package of R software, version R-3.2.4.
2.3.2. Effect of conservation status of sites on orchids
The independent Student t-test was ﬁrst performed to examine
whether the conservation status of a study area inﬂuenced the density
of the overall orchid species. The mean density of all orchid individuals
was compared statistically between the protected and unprotected
areas. Data were log-transformed in order to normalize the distribution.
Two-way Analysis of Variance (Two-way ANOVA) was then
performed to assess an existing combined effect of vegetation type
(woodland, savanna and gallery forest) and the conservation status of
the two zones (protected and unprotected) on the density of orchid
species. Zones and vegetation types were used as categorical indepen-
dent variables. The continuous dependant variable was the density of
orchids. The density of orchids was therefore determined for each stud-
ied zone and vegetation type. Two-way ANOVA was also computed on
the two most common orchid species (Nervilia kotschyi (Rchb.f.) Schltr.
and Eulophia guineensis Lindl.) to assess the importance of the protected
area in their conservation. The average density per plot of orchids
was log-transformed in order to normalize the distribution. Data
were tested to check the homogeneity of variance with Levene's
test. Two-way ANOVA and Tukey's post-hoc tests were used in the
case of homoscedasticity. In the absence of homoscedasticity, samples
were compared using the Kruskal-Wallis and Tukey's post-hoc tests
In addition, the Importance Value Index (IVI) was used to measure
how dominant an orchid species is in a given zone according to the con-
servation status (protected and unprotected) of the area (Houéhanou
et al., 2012). The calculation of IVI for orchid species was based on Rel-
ative Density (RD) and Relative Frequency (RF).
IVI ¼RD þRF
RD ¼Number of individual of a species
Number of individual of all species X100
RF ¼Number of occurence of a species
Number of occurence of all species X100
A species with high IVI value is considered to be well represented
and thus, ecologically healthy in the given zone. The signiﬁcance level
of all analyses was set at 0.05. Data were computed using the Stats pack-
age of R software, version R-3.2.4.
3.1. Orchid diversity in protected versus unprotected areas
A total of 12 orchid species distributed over 7 genera were recorded
in the studied areas (Table 2). Three epiphytic orchid species, i.e.
Calyptrochilum christyanum (Rchb.f.), Cyrtorchis arcuata (Lindl.) Schltr.
and Plectrelminthus caudatus (Lindl.) Summerh., were recorded on
eight host trees (Pentadesma butyracea Sabine, Breonadia salicina
(Vahl) Hepper & J.R.I.Wood, Syzygium guineense (Willd.) DC., Berlinia
grandiﬂora (Vahl) Hutch. & Dalziel, Diospyros mespiliformis Hochst. ex
A.DC., Isoberlinia tomentosa (Harms) Craib & Stapf, Tamarindus indica L.
and Ficus spp.) in the gallery forest of the unprotected area. The two
most diversiﬁed genera were Eulophia and Habenaria.Nervilia kotschyi
(Rchb.f.) Schltr. and Eulophia guineensis Lindl. were the most common
and abundant terrestrial orchid species respectively in the protected
and unprotected areas (Table 2). C. christyanum (Fig. 2A, B) was the
most common epiphytic orchid with a high occurrence on S. guineense
(35.4%), B. salicina (24.6%) and B. grandiﬂora (17%). The rarest orchid
species recorded with less than three individuals each, were the terres-
trial Platycoryne paludosa (Lindl.) (Fig. 2C) Rolfe and epiphytic
Plectrelminthus caudatus (Fig. 2D). Gallery forests contained the most
important proportion of recorded orchid species (58%). The majority
of the orchid species (67%) were only recorded in unprotected areas.
The taxonomic diversity (SR) was moderate in unprotected areas
and low in protected areas (Table 3). The Shannon-Weaver diversity
index (H′) was low in both the protected and in unprotected areas,
but close to zero in the protected areas. The Evenness index of Shannon
(J') was low in protected areas but relatively high in unprotected areas
3.2. Orchid distribution and habitat condition requirements
The CCA explained 26.8% of the total variation (6.8). Table 4 showed
the correlation of environmental variables with the ﬁrst two canonical
axes. The major ﬂoristic groups correlated with the tree cover gradient
(axis 1). Topography, tree cover, herb cover and soil texture correlated
best with the ﬁrst axis (CCA1) whereas altitude correlates with the
Distribution of the orchid species in protected and unprotected areas.
RA = Relative abundance of orchid species.
Protected area Unprotected area
Number of individuals RA (%) Number of individuals RA (%)
Species Type of orchid Woodland Savanna Gallery Total Woodland Savanna Gallery Total
Calyptrochilum christyanum (Rchb.f.) Summerh. Epiphytic 0 0 0 0 0.00 0 328 284 612 13.18
Cyrtorchis arcuata (Lindl.) Schltr. Epiphytic 0 0 0 0 0.00 0 26 0 26 0.56
Eulophia spp Terrestrial 0 33 21 54 1.14 0 0 21 21 0.45
Eulophia guineensis Lindl. Terrestrial 9 27 139 175 3.69 0 0 2183 2183 47.02
Eulophia horsfallii (Bateman) Summerh. Terrestrial 0 0 0 0 0.00 0 0 116 116 2.50
Habenaria cirrhata (Lindl.) Rchb. f. Terrestrial 0 0 41 41 0.87 0 0 22 22 0.47
Habenaria ﬁlicornis Lindl. Terrestrial 0 0 0 0 0.00 0 79 0 79 1.70
Habenaria schimperiana Hochst. ex A.Rich. Terrestrial 0 0 0 0 0.00 0 171 0 171 3.68
Nervilia bicarinata (Blume) Schltr. Terrestrial 0 0 0 0 0.00 0 0 942 942 20.29
Nervilia kotschyi (Rchb.f.) Schltr. Terrestrial 1847 2269 352 4468 94.30 0 0 417 417 8.98
Platycoryne paludosa (Lindl.) Rolfe Terrestrial 0 0 0 0 0.00 0 53 0 53 1.14
Plectrelminthus caudatus (Lindl.) Summerh. Epiphytic 0 0 0 0 0.00 0 0 1 1 0.02
Total 1856 2329 553 4738 100.00 0 657 3986 4643 100.00
233E.S.P. Assédé et al. / South African Journal of Botany 116 (2018) 230–237
second axis (CCA2). The environmental variables are projected in the
ﬁrst two axes as well as the discriminated habitat groups (Fig. 3). The
pattern of these variables conﬁrmed effectively that the ﬁrst axis
showed a decreasing tree cover gradient.
A differentiation appeared between groups of gallery forests of
protected and unprotected areas (G1), woodland of protected areas
(G2) and grassland of unprotected areas (G3) (Fig. 3). The two sub-
groups on the top and bottom of G1 (Fig. 3) were respectively domi-
nated by the genus Eulophia and the epiphytic orchid C. christyanum
on respectively relatively low (mean of 220 m) and high altitudes
(mean of 363 m). Both subgroups were characterized by a high cover
of the woody layer (mean of 85%), a low cover of the herb layer
(mean of 15%), clayey soil with presence of rocks and boulders, and a
steep slope. The dominant tree species were S. guineense and
B. salicina. The third subgroup of G1 dominated by Nervilia bicarinata
and E. guineensis was a mixed stand of terrestrial and epiphytic orchids,
recorded on clay-sandy soil at relatively low altitude (mean of 267 m)
and moderate slope in unprotected areas. In this subgroup (G1), the
vegetation was tree and shrub savanna dominated by Khaya
senegalensis. The tree layer covered 45%–50% and the herb layer 25%–
30%. The second group G2, constituted by N. kotschyi, is a mixed stand
of woodland and gallery forest of the protected and unprotected areas
on clayey soil, relatively low altitude (mean of 246 m), and slope. The
average cover of the tree and herb layers was respectively 70% and
30%. The dominant tree species was Anogeissus leiocarpa (DC.) Guill. &
Perr. The Group G3, was clearly differentiated from the other groups,
being represented by plots from unprotected areas and dominated by
H. schimperiana,H. ﬁlicornis and P. paludosa. This group was associated
with the plains at 245 m, silty soil, and characterized by a high cover
of the herbaceous layer (95%) dominated by Cyperus spp. and
Andropogon spp. and a relative absence of the tree layer.
3.3. Inﬂuence of conservation status and vegetation types on orchid density
Results from the Student t-test (p = 0.3) and that from the two-way
ANOVA inside zones (p = 0.2) did not show any signiﬁcant difference
between the density of all recorded orchids in protected and unpro-
tected areas. However, there was a signiﬁcant difference in the orchid
density when considering the vegetation types (p = 0.004) as well as
the interaction between zones and vegetation types. Based on Tukey's
post-hoc tests, the gallery forests presented a higher orchid density
than woodlands (p = 0.03). The protected area presented the highest
density of N. kotschyi, and signiﬁcantly more than in the unprotected
areas (p b0.0001), but the vegetation type, and the interaction between
zones and vegetation type, had no effect. Only vegetation type effect
was signiﬁcant on E. guineensis density (p b0.0001). The result from
Tukey's post-hoc test showed signiﬁcant differences between gallery
forest and woodland (p b0.0001), and between gallery forest and sa-
vanna (p b0.0001). E. guineensis was more abundant in gallery forest
than in woodland and savanna (p = 0.03).
3.4. Conservation status of orchid species based on IVI
The IVI of the recorded orchid species varied between the protected
and unprotected areas (Table 5). Nervilia kostchyi had the highest IVI
Fig. 2. Orchidspecies in the Biosphere Reserve of Pendjari: the most common orchid, Calyptrochilum christyanum individual (A) andﬂower (B); the rarest terrestrial orchid,Platycoryne
paludosa (C); the rarest epiphyte orchid Plectrelminthus caudatus (D).
Diversity indices of protected and unprotected areas.
Index Protected areas Unprotected areas
Species richness: SR 4 12
Shannon-Weaver diversity: H′0.3 2.3
Shannon evenness: J' 0.1 0.6
Correlation of environmental variables with ordination axes of CCA. Only values N│0.5│
contribute substantially to the axis.
Variables CCA1 CCA2
Soil texture 0.5284 −0.0464
Topography −0.7751 −0.5304
Altitude −0.6688 −0.7227
Herb cover 0.7463 0.0992
Tree cover −0.7472 −0.1027
234 E.S.P. Assédé et al. / South African Journal of Botany 116 (2018) 230–237
(169.3) in the protected areas. In unprotected areas, the species with
the highest IVI values were Eulophia guineensis and Calyptrochilum
christyanum (respectively 69.52 and 45.68).
4.1. Orchid diversity and habitat characteristics
The Biosphere Reserve of Pendjari (BRP) is known to be a corner
stone for strategic conservation of plant species in the Sudanian zone,
with 35.6% of all species and 33.3% of all genera listed for Benin
(Akoègninou et al., 2006). However, only 23% of recordedorchid species
was found in the core zone of the reserve with no epiphytic species. It
has been assumed that the protection status of an area would normally
conserve more animal and plant species (Dudley and Bean, 2012). This
assumption may be true only when the vegetation structure is suitable
to the ecology required by the species. For example, gallery forests had
the highest orchid density and a great diversity of terrestrial orchids in
the protected area of the BRP (a mosaic of savannas and woodlands)
compared to unprotected areas (Table 3). Several factors inﬂuence the
distribution pattern of orchid species. Epiphytic orchid diversity in-
creases along moisture and latitudinal gradients (Gentry and Dodson,
1987;Phillips et al., 2011). At larger scale, orchid richness is highest in
the high rainfall zones with closed vegetation cover. In addition, closed
formations with high tree density were found to be providing ideal de-
velopment conditions for epiphytic orchids. Therefore, with the low
rainfall (an average of 1000 mm per year) and long dry season (ﬁve
month dry period) observed in the BRP compared to the national scale
(up to 1400 mm per year), several orchid species may experience less
drought stress in gallery forests. However, the conditions required by
some terrestrial orchids are an open area with thin ground litter. The
vegetation found across both protected and unprotected areas and the
difference in orchid species might also be a consequence of different
land uses. The vegetation in unprotected areas was more disturbed by
anthropogenic activities, especially logging and farming system. In the
protected areas there was no human activity. But, if gallery forests of un-
protected areas still conserve more orchids than gallery forests in
protected areas, it is probably because some speciﬁc ecological condi-
tions were also required by orchids even if the vegetation was not
The small size of orchid seeds and the fact that they lack endo-
sperm make these plants dependent on mycorrhizal symbioses to
provide energy and nutrients during their early development stages
(Otero and Flanagan, 2006; Barthlott et al., 2014). Mycorrhizal spec-
iﬁcity and habitat specialization (Gravendeel et al., 2004; Otero and
Flanagan, 2006)havebothbeenimplicatedinthediversiﬁcation of
the orchid family. The diversity of mycorrhiza was demonstrated as
afactortoinﬂuence seed germination and greater growth for all or-
chid species (Phillips et al., 2011; McCormick et al., 2016). However,
the real beneﬁt conferred by mycorrhizal associations to plants may
also depend on soil conditions, especially fertility. Therefore, identi-
fying the mycorrhiza associated with the recorded terrestrial orchids
and their role in orchid distribution in protected vs unprotected zone
of the BRP should be the next goal in the study of the ecology of
The habitat conditions required by the orchid species were shown in
the CCA (Fig. 3). There is a differentiation between habitats of protected
and unprotected areas in terms of tree cover, altitude, soil type and to-
pography that inﬂuence the pattern of orchid distribution in both
zones, andthe importance of woodland in the protected area. Therefore,
orchids associatedwith woodland, the dominant vegetation type in the
protected area, characterized by an average tree cover, were assumed
more conserved in the protected area as conﬁrmed by the conservation
status of N. kotschyi. However, a gap of conservation can be assumed to
exist for most of the epiphytic orchids conﬁned to the gallery forests of
Setyawan (2000) highlighted the importance of the height of the
host trees in the distribution, diversity and density of epiphytic
plants. Even though data were not collected and tested with refer-
ence to this assumption, it was observed during this study that the
taller host tree species were the most colonized by C. christyanum
in gallery forest. Similarly, Yulia and Budiharta (2011) also showed
that the characteristics of the host tree (height and bark type) impact
on the establishment and development of epiphytic orchids. How-
ever, several other environmental factors not quantiﬁed by this
study have been highlighted as determinants of orchid distribution.
Indeed, the constraint axis of the CCA explained only 26.8% of the
total variance between plots. The micro-climatic factors (sunshine
intensity, humidity and air temperature) and the potential for soil
resources (moisture and soil pH) have also been reported to inﬂu-
ence the orchid distribution (McCormick et al., 2012). Furthermore
the difference in host tree composition and pollination strategy
(Jersàkovà et al., 2006) are key aspects to examine in the future be-
cause this may help explain the observed distribution patterns of or-
Fig. 3. Canonical Correspondence Analysis (CCA) diagram representing the ﬁrst two axes
that explained 72.8% (CCA1: 39.2% and CCA 2: 33.6%) of all varianc e explained by the
CCA. Empty circle designate the plots scores in each discriminated habitat group (np:
unprotected area; p: protected area) by orchid species composition; G1: gallery forest
plots (dominate d by those of unprotected areas), G2: protected woodland and tree
savanna plots, G3: unprotected grassland plots. Environmental variables are represented
by blue vectors (Alt: Altitude, Topo: To pography, Tex_sol: Soil texture, S_Arb: Tree
cover, S_Herb: Herb cover) that determine additional arrowed axes in the diagram.
Importance Value Index (IVI) of orchid species within the studied zones.
Species Protected areas Unprotected areas
RD RF IVI RD RF IVI
Calyptrochilum christyanum 0 0 0 13.18 32.50 45.68
Cyrtorchis arcuata 0 0 0 0.56 2.50 3.06
Eulophia spp 1.14 4.20 5.34 0.45 2.50 2.95
Eulophia guineensis 3.69 16.60 20.29 47.02 22.50 69.52
Eulophia horsfallii 0 0 0 2.50 5.00 7.50
Habenaria ﬁlicornis 0 0 0 1.70 2.50 4.20
Habenaria schimperiana 0 0 0 3.68 2.50 6.18
Habenaria cirrhata 0.86 4.20 5.06 0.47 7.50 7.97
Nervilia bicarinata 0 0 0 20.29 5.00 25.29
Nervilia kotschyi 94.30 75.00 169.30 8.98 12.50 21.48
Platycoryne paludosa 0 0 0 1.14 2.50 3.64
Plectrelminthus caudatus 0 0 0 0.02 2.50 2.52
235E.S.P. Assédé et al. / South African Journal of Botany 116 (2018) 230–237
4.2. Conservation status and orchid density
Although there was low species richness in protected areas com-
pared to unprotected areas, the absence of conservation effect on orchid
density (when all species are considered), was observed mainly because
the species recorded in the protected area were characterized by a
high density. Thus, this effect became signiﬁcant when considering the
orchid species separately. N. kotschyi seems more common in protected
than in unprotected areas, probably because of its habitat preference. As
a terrestrial orchid, the species grew more in woodland (Akoègninou
et al., 2006), the dominant vegetation type of the protected area. The
habitat suitability, a factor that determines where a species is found,
can explain the afﬁnity of some recorded orchids to particular tree spe-
cies associated with a speciﬁc community of gallery forest occurring
within unprotected areas. E. guineensis is an example of this with the
highest density and IVI in unprotected areas. E. guineensis was always
recorded under shade in dense vegetation of gallery forest with a
steep slope (Akoègninou et al., 2006). This species could be assumed
to be ecologically suited to speciﬁc gallery forest communities of unpro-
tected areas. The gallery forest community in unprotected areas would
also be one of the suitable habitats for Calyptrochilum christyanum and
Plectrelminthus caudatus where these orchids were the most abundant.
Hence, the prevalence of speciﬁc trees (e.g., Berlinia grandiﬂora,
Breonadia salicina,Pentadesma butyracea) and high tree cover in the gal-
lery forests of unprotected areas should be one of the factors explaining
both E. guineensis abundance and the occurrence of three epiphytic or-
chid species in unprotected areas. Therefore, the establishment of
C. christyanum, the most common epiphytic orchid, should not be af-
fected by human activities. Even if recorded under closed canopy,
C. christyanum can establish on a host tree under severe pruning. In ad-
dition, one of the factors which determined the epiphytic orchid estab-
lishment may be the presence of rock slabs or sandstone boulders.
Indeed, contrary to the situation in the protected area, the edaphic sup-
port of more than 90% of investigated gallery forests in unprotected
areas, wasrepresented by therock slabs and large deposits of sandstone
boulders of varying size. A high density of epiphytic orchids, in particu-
lar C. christyanum, was observed. However, these assumptions need to
be tested in future studies. The environmental conditions required at
ﬁner scale for epiphytic orchids in gallery forests in unprotected areas
need to be ascertained. Are those speciﬁc site conditions requirements
provided by tree species only present in gallery forest in unprotected
areas, or by speciﬁc physical habitats provided by rock slabs and/or
sandstone boulders, or both?
5. Conclusion and suggestions for management
The Biosphere Reserve of Pendjari was found to be a habitat for or-
chid species of the Benin Republic. Three epiphytic orchid species
were recorded from eight host tree species. The two most diverse orchid
genera were Eulophia and Habenaria. Some orchid species (N. kotschyi)
were well represented within the protected area, while all the recorded
epiphytic orchids and E. guineensis (67% of all species recorded in the
study) were only recorded in unprotected areas. Gallery forests had a
higher orchid density than woodlands. The most common orchid
species in unprotected areas were C. christyanum (epiphyte) and
E. guineensis (terrestrial), while in protected area it was N. kotschyi
Based on our extensive ﬁndings, we propose a new approach for the
joint protection and management of orchid species. A special manage-
ment plan should be developed for gallery forest in unprotected areas.
Control measures should be reinforced in those areas left to the local
population for their livelihoods, especially in the gallery forests, to facil-
itate the sustainable management of orchid populations. The current
deforestation and anthropogenic activities observed in unprotected
areas should be mitigated to maintain the habitats of the orchids.
Finally, our study did not consider the inﬂuence of several factors,
such as soil nutrients, humidity and light conditions, and host tree iden-
tity and characteristics that could explain more of the high occurrence
of orchids,both epiphytic and terrestrial, in unprotected areas. In future,
factors such as these should be included in the analyses to reﬁne our un-
derstanding of the effects of distribution patterns and conservation
status on the orchid species. In so doing, we will further improve con-
servation strategies that aim to protect these orchid species. Despite
this limitation, our results highlighted the need to redeﬁne protective
and management strategies for orchid species in the Biosphere Reserve
We are grateful to The Rufford Foundation for ﬁnancial support
through Rufford Small Grants for Nature Conservation to Eméline Sêssi
Pélagie ASSEDE (grant reference: 16855-1). Eméline Sêssi Pélagie
ASSEDE was further supported by the University of Pretoria Postdoc-
toral Fellowship Program. We thank Adandé Belarmain Fandohan and
Paxie W. Chirwa for their useful comments. We also thank the staff of
the Biosphere Reserve of Pendjari for logistic support during this study
and anonymous reviewers for their feedback. Most importantly, we
are grateful to the ﬁeld guides and farmers of the villages surrounding
the reserve, who collaborated during data collection.
Akoègninou, A., Van der Burg, W.J., Van der Maesen, L.J.G., 2006. Flore Analytique du
Bénin. Backhuys Publishers, Wageningen.
Assédé, P.S.E., Adomou, A.C., Sinsin, B., 2012. Magnoliophyta, biosphere Reserve of
Pendjari, Atacora, Benin. Check List 8, 642–661.
Barthlott, W., Große-Veldmann,B., Korotkova, N.,2014. Orchid Seed Diversity: A Scanning
Electron Microscopy Survey. Botanic Garden and Botanical Museum Berlin-Dahlem,
Berlin (245 pp.).
Bergstrom, B.J., Carter, R., 2008. Host-tree selection by an epiphytic orchid, Epidendrum
magnoliae Muhl. (green ﬂy orchid), in an inland hardwood hammock in Georgia.
Southeast. Nat. 7, 571–580.
Bruner, A.G., Gullison, R.E., Rice, R.E., da Fonseca, G.A.B., 2001. Effectiveness of parks in
protecting tropical biodiversity. Report 291, 125–128.
CENAGREF, 2016. Plan d'Aménagement et de Gestion Participative de la Réserve de
Biosphère de la Pendjari. Programme d'appui aux Parcs de l'Entente, Composante 2
CITES, 2017. Numbers of species listed in the CITES Appendices as of October 2017. The
Convention on the International Trade in Endangered Species of Flora and Fauna
https://www.cites.org/eng/app/appendices.php, Accessed date: November 2017.
Cribb, P., Roberts, D., Hermans, J., 2005. Distribution, ecology, and threat to selected
Madagascan orchids. Selbyana 26, 125–135.
Defries, R., Hansen, A., Newton, A.C., Hansen, M.C., 2005. Increasing isolation of protected
areas in tropical forests over the past twenty years. Ecol. Appl. 15, 19–26.
Delforge, P., 2001. Guide des Orchidées d'Europe, d'Afrique du Nord et du Proche-Orient.
2nd edition. Delachaux et Niestlé, Neuchâtel, Paris.
Delvingt, W., Heymans, J.C., Sinsin, B., 1989. Guide du Parc National de la Pendjari: Pro-
gramme d'aménagement des parcs nationaux et de protection de l'environnement.
Coordination AGRER, Bruxelles.
Diniz, M.F., Brito, D., 2015. Protected areas effectiveness in maintaining viable giant ant-
eater (Myrmecophaga tridactyla) populations in an agricultural frontier. Natureza &
Conservação 13, 145–151.
Djossa, B., Fahr, J., Wiegand, T., Ayihouénou,B.E., Kalko, E.K., Sinsin, B., 2008. Land use im-
pact on Vitellaria paradoxa C.F. Gaertn. stand structure and distribution patterns: a
comparison of Biosphere Reserve of Pendjari in Atacora district in Benin. Agrofor.
Syst. 72, 205–220.
Doyle, J.J., Luckow,M.A., 2003. The rest ofthe iceberg. Legume diversity and evolution in a
phylogenetic context. Plant Physiol. 131, 900–910.
Dressier, R.L., 1981. The Orchids: Natural History and Classiﬁcation. Harvard University
Press, Camb ridge, Massachusetts.
Dudley, T.L., Bean, D.W., 2012. Tamarisk biocontrol, endangered species risk and resolu-
tion of conﬂict through riparian restoration. BioControl 57, 331–347.
Dunkan, M., Pritchard, A., Coates, F., 2005. Major threats to endangered orchids of
Victoria, Australia. Selbyana 26, 189–195.
Fandohan, B., Assogbadjo, A.E., Glèlè Kakaï, R., Sinsin, B., 2011. Effectiveness of a
protected areas network in the conservation of Tamarindus indica (Leguminosae-
Caesalpinioideae) in Benin. Afr. J. Ecol. 49, 40–50.
Françoso,R., Brandão, R., deCampos, N.C., Salmona, Y., Machado, R., Colli, G., 2015.Habitat
loss and the effectiveness of protected areas in the Cerrado bi odiversity hotspot.
Natureza & Conservação 13, 35–40.
Gentry, A.H., Dodson, C.H., 1987. Diversity and biogeography of neotropical vascular epi-
phytes. Ann. Mo. Bot. Gard. 74, 205–233.
236 E.S.P. Assédé et al. / South African Journal of Botany 116 (2018) 230–237
Ghorbani, A., Gravendeel, B., Naghibi, F., de Boer, H., 2014. Wild orchid tuber collection in
Iran: a wake-up call for conservation. Biodivers. Conserv. 23, 2749–2760.
Gouwakinnou, G.N., Kindomihou, V., Assogbadjo, A.E., Sinsin, B., 2009. Population struc-
ture and abundance of Sclerocarya birrea (A. Rich) Hochst subsp. birrea in two con-
trasting land-use systems in Benin. Int. J. Biodivers. Conserv. 1, 194–201.
Gravendeel, B., Smithson, A., Slik, F.J.W., Schuiteman, A., 2004. Epiphytism and pollinator
specialization: drivers for orchid diversity? Philos. Trans. R. Soc. B 359, 1523–1535.
Hinsley, A., Verissimo, D., Roberts, D.L., 2015. Heterogeneity in consumer preferences for
orchids in international trade and the potential for the use of ma rket research
methods to study demand for wildlife. Biol. Conserv. 190, 80–86.
Houéhanou, T.D., Kindomihou, V., Sinsin, B., 2011. Effectiveness of conservation areas in
protecting Shea trees against hemiparasitic plants (Loranthaceae) in Bénin, West
Africa. Plant Ecol. Evol. 144, 267–274.
Houéhanou, T.D., Glèlè Kakaï, R.L., Assogbadjo, A.E., Kindomihou, V., Houinato, M., Wittig,
R., Sinsin, B.A., 2012. Change in the woody ﬂoristic composition, diversity and struc-
ture from protected to unprotected savannahs in PendjariBiosphere Reserve (Benin,
West Africa). Afr. J. Ecol. 51, 358–365.
Houéhanou, D.T., Assogbadjo, A.E., Glèlè Kakaï, R., Kyndt, T., Houinato, M., Sinsin, B., 2013.
How far a protected area contributes to conserve habitat species composition and
population structure of endangered African tree species. Ecol. Complex. 13, 60–68.
IUCN/SSC Orchid Specialist Group, 1996. Orchids - Status Survey and Conservation Action
Plan. IUCN, Gland Switzerland and Cambridge, UK.
Jersàkovà, J., Johnson, S.D., Kindlmann, P., 2006. Mechanisms and evolution of deceptive
pollination in orchids. Biol. Rev. Camb. Philos. Soc. 81, 219–235.
Kasulo, V., Mwabumba, L., Munthali, C., 2009. A review of edible orchids in Malawi.
J. Hortic. For. 1, 133–139.
König, K., 2005. Carte de Végétation du Parc National de la Pendjari et ses Zones
Cynégétiques. BIOTA W11. Institut de Geographie, Cotonou.
Liu, H., Luo, Y.B., Heinen, J., Bhat, M., Liu, Z.J., 2014. Eat your orchid and have it too: a po-
tentially new conservation formu la for Chinese epiphytic medicinal orchids.
Biodivers. Conserv. 23, 1215–1228.
Magurran, A.E., 2004. Measurement Biological Diversity. Blackwell Science Ltd., United
McCormick, M.K., Lee Taylor, D., Juhaszova, K., Brunett, R.K., Whigham, D.F., O'Neill, J.P.,
2012. Limitations on orchid recruitment: not a simple picture. Mol. Ecol. 21,
McCormick, M.K., Taylor,D.L., Whigham, D.F., Burnett Jr., R.K., 2016. Germination patterns
in three terrestrial orchids relate to abundance of mycorrhizal fungi. J. Ecol. 104,
Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, G.A.B., Kent, J., 2000. Biodiversity
hotspots for conservation priorities. Nature 403, 853–858.
Otero, J.T.,Flanagan, S.N., 2006.Orchid diversity –beyonddeception. Trends Ecol.Evol. 21,
Pant, B., 2013. Medicinal orchids and their uses: tissue culture a potential alternative for
conservation. African J. Plant Sci. 7, 448–467.
Phillips, R.D., Barrett, M.D., Dixon, K.W., Hopper, S.D., 2011. Do mycorrhizal symbioses
cause rarity in orchids? J. Ecol. 99, 858–869.
Pillon, Y., Qamaruz-Zaman, F., Michael, F.F., Hendoux,F., Piquot, Y., 2007.Genetic diversity
and ecological differentiation in the endangered fen orchid (Liparis loeselii). Conserv.
Genet. 8, 177–184.
Scherrer, B., 2007. Biostatistique. Gaëtan Morin, Paris.
Schumann, K., Wittig, R., Thiombiano, A., Beckerc, U., Hahn, K., 2010. Impact of land-use
type and bark- and leaf-harvesting on population structure and fruit production of
the baobab tree (Adansonia digitata L.) in a semi-arid savanna, West Africa. For.
Ecol. Manag. 260, 2035–2044.
Setyawan, A.D., 2000. Epiphytic plants on stand of Schima wallichii (D.C.) Korth. at Mount
Lawu. Biodiversitas 1, 14–20.
Vermeulen, J.J., Phelps, J., Thavipoke, P., 2014. Notes on Bulbophyllum (Dendrobiinae;
Epidendroideae; Orchidaceae): two new species and the dilemmas of species discov-
ery via illegal trade. Phytotaxa 184, 012–022.
Wan, J., Wang, C., Han, S., Yu, J., 2014. Planning the priority protected areas ofendangered
orchid species in northeastern China. Biodivers. Conserv. 23, 1395–1409.
Wittemyer, G., Elsen, P., Bean, W.T.,Burton, A.C., Brashares, J.S., 2008. Accelerated human
population growth at protected area edges. Science 320, 123–126.
Wittig, R., König, K., Schmidt, M., Szarzynski, J., 2007. A study of climate change and an-
thropogenic impacts in West Africa. Environ. Sci. Pollut. Res. 14, 182–189.
Yulia, N.D., Budiharta, S., 2011. The diversity of epiphytic orchid and its host tree along
Cemoro Sewu hiki ng pathway, Lawu mountain, district of Magetan, East Java,
Indonesia. J. Nat. Stud. 10, 26–31.
Yulia, N.D., Budiharta, S., Yulistyarini, T., 2011. Analysis of epiphytic orchid diversity and
its host tree at three gradient of altitudes in Mount Lawu, Java. Biodiversitas 12,
237E.S.P. Assédé et al. / South African Journal of Botany 116 (2018) 230–237