Forest Ecology and Management 569 (2024) 122196

Contents lists available at ScienceDirect

Forest Ecology and Management

journal homepage: [Link]/locate/foreco

Disturbance and climate affect species richness and aboveground biomass

relationship in a forest–savanna transition ecosystem
Melanie C. Nyako a, 1 , Moses B. Libalah a, b, *, 2 , Louis-Paul-Roger B. Kabelong a, 3 , Stéphane
T. Momo b, c, 4 , Marius G. Babonguen d, 5 , Cyntia K. Djouking a, 6 , Marius R.M. Ekué e, 7 ,
Laure B. Fabo b, e, 8 , Alex B.T. Fomekong f, 9 , Nicole M. Guedje g, 10 , Tagnang N. Madountsap h, 11 ,
Adrian Medou a, 12 , Marie C.S. Momo f, 13 , Géraldine D. Nguemo b, 14 , Narcisse E.N. Njila a, 15 ,
Le Bienfaiteur T. Sagang i, 16 , Lise Zemagho b, 17 , Bonaventure Sonké b, 18 , Louis Zapfack a, 19
a
Department of Plant Biology, Faculty of Science, P.O. Box 812 Yaoundé, University of Yaoundé I, Yaoundé, Cameroon
b
Plant Systematics and Ecology Laboratory (LaBosystE), P.O. Box 047 Yaoundé, Higher Teacher’s Training College, University of Yaoundé I, Yaoundé, Cameroon
c
Gembloux Agro-Bio Tech, TERRA Teaching and Research Centre, Forest is Life, University of Liège, Gembloux, Belgium
d
Department of Geography, Faculty of Arts Letters and Social Sciences, University of Yaoundé I, Yaoundé, Cameroon
e
Bioversity International, Yaoundé, Cameroon
f
Département de Biologie Végétale, Université de Dschang, Faculté des Sciences, B.P: 67, Dschang, Cameroon
g
Department of Pharmacy and African Pharmacotherapy, Faculty of Medicine and Biomedical Sciences, University of Yaoundé I, Yaoundé, Cameroon
h
Department of Plant Biology, University of Douala, Douala, Cameroon
i
Institute of the Environment and Sustainability, University of California, Los Angeles, Los Angeles, CA 90095, USA

A R T I C L E I N F O A B S T R A C T

Keywords: The relationship between species richness and aboveground biomass (richness–biomass relationship) is a major
Aboveground biomass facet of ecosystem functions and has stimulated debates over the past decades. However, we still lack basic
Climate change knowledge on whether this relationship has consistent patterns across natural and human-impacted vegetation
Congo Basin
cover types. More importantly, the effects of disturbance and climatic conditions on species richness and
Ecosystem functions
Tropical forests
aboveground biomass remain controversial. Using data from 197 0.5-ha plots established within a forest-savanna
Restoration transition ecosystem, we determined three ecological vegetation covers via non-metric multidimensional scaling,
obtained gridded climate data and calculated disturbance index from species succession guilds. We examined the

* Corresponding author at: Department of Plant Biology, Faculty of Science, P.O. Box 812 Yaoundé, University of Yaoundé I, Yaoundé, Cameroon.
E-mail address: libalah_moses@[Link] (M.B. Libalah).
1
[Link]
2
[Link]
3
[Link]
4
[Link]
5
[Link]
6
[Link]
7
[Link]
8
[Link]
9
[Link]
10
[Link]
11
[Link]
12
[Link]
13
[Link]
14
[Link]
15
[Link]
16
[Link]
17
In honour of deceased co-author
18
[Link]
19
[Link]

[Link]
Received 23 May 2024; Received in revised form 1 August 2024; Accepted 2 August 2024
Available online 8 August 2024
0378-1127/© 2024 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
M.C. Nyako et al. Forest Ecology and Management 569 (2024) 122196

consistency of the richness–biomass relationship across the vegetation covers using general linear models and
further analysed the effects of disturbance and climatic conditions on this relationship using non-mixed and
mixed linear models. We identified three vegetation cover types corresponding to Old-growth forests, Cocoa
agroforests and Woodland savannas, suggesting a succession gradient from late-, mid-, and early-successions,
respectively. Aboveground biomass consistently increased with species richness across the three vegetation
cover types (i.e. positive relationship). Species richness interacts with climate humidity to increase aboveground
biomass while interaction of species richness and disturbance rather decreases aboveground biomass consistently
across the vegetation cover types. Our results provide insights into the relationship between diversity and
ecosystem functions, having implications for ecosystem restoration and understanding ecological consequences
of disturbance and climate change in the forest-savanna ecosystem.

1. Introduction enhances biomass, niche complementary would likely be the underlying

mechanism (Cardinale et al., 2007; Loreau and Hector, 2001). However,
Tropical forests are extremely important to humanity. They contain dominance by species with particular traits, (e.g. monodominant forest
about 50 – 80 % of the Earth’s biodiversity (Rajpar, 2018; Raven, 1988), stands; Djuikouo et al., 2010) may impose a selection effect and result in
harbour over 95 % of the world’s estimated 40,000 tree species (Slik a negative relationship, with subsequent effects on ecosystem processes
et al., 2015), and store half of the world’s biomass carbon in their (Loreau and Hector, 2001). A hump-shaped relationship is evident when
terrestrial vegetation (Lewis et al., 2015). Despite the provision of these large heterogenous environments are studied. As such a gradient
important ecosystem services, tropical forests are experiencing unprec- involving the positive and negative conditions that make up the two
edented human disturbance by land conversion (Foley et al., 2005) sides of the of the hump-shaped curve with the possibility of a plateau
which is depleting biodiversity ecosystem functions (BEF) (Naeem et al., are studied (Guo and Berry, 1998). Studies therefore need to sample
2012). The BEF has attracted interests and debates in recent decades and wider gradient to cover wide environmental conditions influencing the
the relationship between species richness and biomass has been at the relationship between species richness and biomass.
core of this debates (Cardinale et al., 2007; Chisholm et al., 2013; Guo The forest-savanna transition (sensu Oliveras and Malhi, 2016) de-
and Berry, 1998; Mittelbach et al., 2001; Roswell et al., 2023; Schuldt fines the ecosystem where both forests and open woodland savannas
et al., 2023). Controversies range from the simple existent of the rela- exist as discrete patches under a similar climate zone (Veenendaal et al.,
tionship (Adler et al., 2011), to its variation over time (Cardinale et al., 2015). Elsewhere, forest-savanna transitions have often been referred to
2007) and space (Chisholm et al., 2013), the magnitude and direction as savanna sensu lato (Bouvet et al., 2018), zones of (ecological) tension
(Mittelbach et al., 2001), the shape (Guo and Berry, 1998; Mittelbach, (Veenendaal et al., 2015) or more generally as vegetation transitions or
2010; Mittelbach et al., 2001) and the role of succession gradient (Lasky ecotones (Gosz, 1993; Kark and van Rensburg, 2006). Although
et al., 2014) on the relationship. Ongoing discussions focus on envi- forest-savanna transitions occur across the tropics, the afro-tropical
ronmental determinants of this relationship; how does disturbance forest and savanna are probably the most remarkable, occupying pre-
affect the richness–biomass relationship across vegetation cover types? sumably 50 % of continental area and therefore represent a large
This is important because long-term disturbance is expected to impose a biomass carbon pool (Djiofack et al., 2024). Yet, studies examining the
successional shift in the diversity-biomass relationship (Lasky et al., relationship between diversity and ecosystem function in the
2014). forest-savanna transitions are hard to find. The origin and development
The relationship between species richness and aboveground biomass of these transitions in some African countries stems from human
describes the effects that differing number of tree species may have on disturbance (Bouvet et al., 2018; Youta-Happi et al., 2003), although
stand-level aboveground biomass (Rosenzweig, 1995). The prevalence ecological niche construction may also account for the emergence of
of the relationship depends on the spatial scale considered (Chisholm ecotones (Liautaud et al., 2020). Studying species richness–biomass
et al., 2013; Day et al., 2014), is influenced by climate (Ammer, 2019), within this sensitive ecosystem is key to estimating the impacts of
and by both climate and soil (Homeier and Leuschner, 2021; Li et al., climate change on ecosystem functions (Tilman et al., 2014).
2020). Losses in productivity is linked to concomitant decline in species The goal of this study was to evaluate the consistency of species
richness across global forest ecosystems (Liang et al., 2016). richness–aboveground biomass relationship across vegetation cover
Disturbance plays important roles in the richness–biomass relation- types while exploring the influence of climate and disturbance mecha-
ship. It has been demonstrated that diversity–biomass relationships are nisms. To do this, we established 197 0.5-ha plots across sites in a for-
dynamic across succession; correlations between species diversity and est–savanna transition ecosystem undergoing human pressure from
biomass change with succession in tropical forests (Lasky et al., 2014). agriculture (e.g. cocoa agroforests) and random fire events (e.g. for
Long-term disturbances may reduce biomass within the community hunting). Our approach of using fixed, medium-sized plots permitted us
while altering the species richness; from slow-growing shade-bearers to to exclude the weak richness–biomass relationship reported with large
fast-growing pioneers. As a result, disturbance fosters the establishment sized plots (Day et al., 2014) or the influence of varying plot sizes
and persistence of certain species successional guild, such as pioneers (Chisholm et al., 2013). We investigated the following three questions:
and non-pioneers light demanding (NPLD) trees, which increases the (1) Is the relationship between species richness and aboveground
richness of species in the community (Hawthorne, 1996, 1995). It is thus biomass consistent across vegetation cover types? (3) How does species
speculated that species richness would peak along the disturbance succession guilds mediate the response of aboveground biomass to
gradient and then decline once disturbance intensities get too high to species richness? (3) How does other factors such as climate and human
support the survival of shade-tolerant species in the system (Bongers disturbance affect the species richness–biomass relationship?
et al., 2009). However, at extremely high levels of disturbance, domi-
nated by pioneer tree species, the difference between old-growth and 2. Material and methods
disturbed forests would be smaller as all communities are expected to
have relatively few NPLD and pioneer species (Bongers et al., 2009). 2.1. Study area
The magnitude of the richness–biomass relationship is equally
important. Studies have recorded a positive, or negative, or hump- This study focused on a forest-savanna transition ecosystem in cen-
shaped or even no relationship at all between richness and biomass. In tral Cameroon between 4◦ 39 and 4◦ 49 N, and 11◦ 4 and 11◦ 19 E (Fig. 1).
the case of a positive relationship where higher species richness The forest-savanna transition covers about 35 % (14–17 million

2
M.C. Nyako et al. Forest Ecology and Management 569 (2024) 122196

hectares) of the approximately 47.6 million hectares of Cameroon’s the 58,835 trees sampled from the 197 0.5-ha plots, 1261 trees were
national territory (Mayaux et al., 1999; Onana, 2018) and is charac- identified as morpho-species (e.g. Drypetes sp., Uapaca sp.) and were
terized by transitional vegetation of the Guineo–Congolian and the used for the biomass calculations. We homogenised the nomenclature
Sudanian sectors (White, 1983). Different forest formations are known for genera and families according to the APG IV (2016) classification and
within the Guineo–Congolian sector, including the semi-deciduous for- species names were corrected using the Plants of the World Online
ests dominated by Malvaceaes and Cannabaceaes, and the Atlantic for- (POWO, 2023; [Link]
ests characterized by rare Caesalpiniaceae (Letouzey, 1985). The
Sudanian zone is characterised by saxicolous savannas with stunted
Combretaceaes (Letouzey, 1985). Both vegetation formations occur at 2.3. Vegetation cover types and succession
low to mid altitude (600–900 m). Co-existence of savannas, gallery
forests and tall canopy forests are separated by moderately sharp Being unsure of which vegetation cover type to allot the sampled
boundaries (Mitchard and Flintrop, 2013). The mean annual tempera- plots and consequently determined the succession state of the forest, we
ture of the study area is between 21.5◦ C and 25.3◦ C with mean annal employed a non-metric multidimensional scaling with stable solution
rainfall between 1513 mm and 2157 mm (Fick and Hijmans, 2017). Soils from random starts (NMDS) in combination with indicator species an-
are sesquioxide and ferralitics on acid rocks. The indurated ferralitic alyses to assign plots to different vegetation cover types. The NMDS was
soils are characterized by gravel outcrops and are especially frequent in run on a species × plot matrix while specifying the Bray-Curtis dissim-
savannas (Yerima and Ranst, 2005). ilarity as the optimal measure of ecological distance (Legendre and
Legendre, 1998). We used the metaMDS function from the vegan R-
package (Oksanen et al., 2012), allowing all the default commands but
2.2. Study design and field measurements specifying the number of dimensions (i.e. k = 3), minimum and
maximum numbers of random starts in search of stable solution (i.e.
We used a stratified random plot design (Gentry, 1988) for the trymax = 500) and maximum number of iterations in the single NMDS
establishment of small plots of 0.5 ha. Within each 0.5 ha small plot, all run (i.e. maxit = 999). We used the k = 3 dimensions obtained from the
trees with diameter at 1.3 m breast height (DBH) ≥ 5 cm were measured NMDS, as separate ecological groups upon which indicator species or
and identified in-situ. Large trees with buttresses or deformations were multi-level pattern analysis (De Cáceres and Legendre, 2009) was run.
measured at 30 cm below or above the deformation or buttress. Only We tested whether there was a subset of species (or indicator species)
woody non-liana species were sampled as sites were visibly scarce of showing a non-random association with each of the three ecological
lianas. Morpho-species codes were consistently assigned to unidentified groups using the multipatt function in the indicspecies R-package (De
species (e.g. Fabaceae 1, Beilschmiedia sp1, Unknown 1). Botanical Cáceres and Legendre, 2009). The multipatt function created combina-
specimens were collected for further confirmation of the in-situ identi- tions of the three predetermined ecological groups and compares each
fication at the mini-herbarium of the Plant Systematics and Ecology combination with the species matrix, chooses the combination with a
laboratory of the Higher Teachers’ Training College of the University of highest association value and test the statistical significance of the best
Yaoundé I and the National Herbarium of Cameroon in Yaoundé. Among matching associations (De Cáceres and Legendre, 2009).

Fig. 1. Location of the study. Maps showing: (A) Africa with Cameroon inside red square; (B) Cameroon with study site inside red rectangle and (C) study site with
different colours and symbols representing the plots and land use in the forest-savanna ecosystem in Cameroon.

3
M.C. Nyako et al. Forest Ecology and Management 569 (2024) 122196

2.4. Species richness and aboveground biomass values imply higher air humidity or lower aridity. It should be noted that
the HI has been used as indicator of the degree of climate humidity
We considered the total number of species as the species richness and (Tuhkanen, 1980) as well as a predictor of the relationship between
used the climate-based allometric equation 7 from Chave et al. (2014) species richness and biomass (Li et al., 2020).
which is similar to locally-derived allometric equations because of its
MAP
dependence on the local climatic factors (see Eq.1). We supplied as HI = (4)
(MAT + 10)
covariates of the allometric equation, the tree diameter (DBH, cm),
species mean wood density (WD, g cm− 3) and the environmental stress
index (E) as the main predictors of AGB (Mg ha− 1). We obtained WD for 2.6. Statistical analysis
tree species from the DRYAD global repository (Chave et al., 2009;
Zanne et al., 2009) using the following steps described by Gourlet-Fleury To explore the relationship between species richness and above-
et al. (2011): (i) for well identified tree species with corresponding WD ground biomass (AGB) across vegetation cover types (VCT) and whether
in the DRYAD database, we attributed the mean WD from the different disturbance and/or climate affect the richness–biomass relationship, we
values, and (ii) for trees identified as morpho-species (7.3 % of trees in fitted several non-mixed and mixed linear models involving additive and
the data), we attributed the means of the genus, family, order or simply interactive terms. The linear non-mixed models were fitted using a
plot-level mean WD if the previous was not found. We used equation 6a generalised least square (gls) as this minimizes the total error and yields
of Chave et al. (2014) to calculate E (Eq. 2), which is based on tem- an unbiased and precise prediction. Because species richness was
perature seasonality (TS) from the Worldclim dataset (bioclimatic vari- repeatedly measured in the different vegetation cover and that VCT
able 4, see Fick and Hijmans 2017), climatic water deficit (CWD, mm could vary randomly in space, another set of linear mixed-effect (lme)
yr− 1) and precipitation seasonality (PS) as defined in the Worldclim models were fitted introducing the VCT as the random effect terms. Both
dataset (bioclimatic variable 15, see Fick and Hijmans 2017). The above the gls and lme models were fitted using the Maximum Likelihood
procedures for computing the E and AGB were conducted using the method opposed to the Restricted Maximum Likelihood method because
computeE and computeAGB functions in the BIOMASS R-package of its robustness in comparing models with different fixed effect terms
(Réjou-Méchain et al., 2017). (Pinheiro et al., 2022). In all the models, AGB was the response while
species richness, climate and disturbance were the predictors, and these
AGB = exp( − 2.024 − 0.896 × E + 0.920 × ln(WD) + 2.795 were fitted as the fixed effect term and VCT as the random effect term.
× ln(D) − 0.0461 × [ln(D)2 ]) (1) In a series of gls models, we examined whether prediction of AGB
would be random, that is, none of species richness, disturbance or
E = (0.178 × TS − 0.938 × CWD − 6.61 × PS) × 10− 3
(2) climate factors significantly predicts AGB (null model, Table 1; M1);
whether species richness significantly predicts AGB (Table 1; M2). We
The total plot-level AGB was calculated as the sum of AGB for all
presumed that, and examined whether, species richness would have a
trees within a plot.
quadratic effect on AGB (Table 1; M3) but we discarded the quadratic or
null model if it was not significant at P ≤ 0.05. We equally examined
2.5. Disturbance and humidity indexes whether AGB and species richness were related to disturbance index (DI)
and/or humidity index (HI), as additive or interactive fixed-effect terms
We calculated a disturbance index (DI) or pioneer index (sensu (Table 1; M1-M11). Because we expected that the vegetation cover types
Hawthorne 1996) for each plot (see Eq. 3) as the weighted number of would have a random effect, we re-fitted M1 to M11 as lme models
species per succession guilds (pioneers, non-pioneer light-demanders, (Table 1; M12-M121) and used the VCT as the random effect terms. For
shade-bearers) over the total number of species (sumspp ) (Hawthorne, each of the gls and lme models, we tested a specific hypothesis (Table 1).
1996, 1995). The DI integrates the effects of a wide range of disturbance Finally, we compared the performances of the 21 gls and lme models
including windthrow, logging, farming or fire (Hawthorne, 1996; Sheil using the following coefficients of determination; Bayesian Information
and Burslem, 2003). The DI also highlights regeneration or succession of Criterion (BIC), Akaike Information Criterion (AIC), log-likelihood
a forest, as judged from the proportion of pioneers species and other (logLik), deviance and the variance explained (R2). We used the gls
non-pioneer light-demanders in the community (Hawthorne, 1996) and function from the nlme R-package (Pinheiro et al., 2022), and the lmer
has been used to predict disturbance in tropical vegetations (Bongers function from the lme4 R-package (Bates et al., 2015).
et al., 2009). Because our study involved agricultural crops, we added We selected the most parsimonious model as the one with the lowest
cultivated species to the calculation of DI. Assuming that pioneer species BIC, AIC, logLik, deviance and highest R2. We calculated the marginal R2
are the first colonizers and would be the strongest indicators of distur- (R2m; variance explained only by the fixed effect terms) and conditional
bance, we weighted the number of pioneers species (Pspp ) by a factor of 3 R2 (R2c ; variance explained by both the fixed effect and the random effect
and the number of non-pioneer light-demander species (NPLDspp ) by a terms) for the mixed effect models using the MuMIn R-package (Bartoń,
factor of 2 and number of cultivated species (Agricspp ) by a factor of 1 2009).
To explore the effect of long-time disturbance on species diversity,
(Hawthorne, 1996, 1995). Shade-bearer species can be indicator of
hence the richness–biomass relationship, we used the gls model to
undisturbed forest and were thus not weighted or simply weighted by a
determine the contributions of succession guilds on species richness.
factor of 0. Data for species succession guilds were obtained from the
Succession guilds of each species was determined as the proportion of
CoForTraits database (Bénédet et al., 2019) and complemented with our
the number of species belonging to one of pioneer, NPLD, cultivated or
expert knowledge of the local flora.
shade-bearer species.
( )
(Pspp × 3) + (NPLDspp × 2) + (Agricspp × 1) × 100 × 100 For each stage of the analyses, we explore the behaviour of the model
DI = (3) or relationship when all plots were pooled together and across each
sumspp
vegetation cover type. We repeated the original analyses with the locally
Humidity index (HI) was calculated from the plot’s mean annual weighted polynomial (loess) regressions and specified the smoothing
precipitation (MAP, mm) and mean annual temperature (MAT, ◦ C) (span = 1) and degree of polynomials (degree = 2).
based on Eq. 4 (Eq.4; Tuhkanen, 1980) below. MAP and MAT were
obtained from the WorldClim database version 2.1 ([Link] 3. Results
[Link]/data/[Link]) at a resolution of 30 arc sec, for
the period from 1970 to 2000 (Fick and Hijmans, 2017). Higher HI A total of 197 0.5-ha plots comprising 58 835 trees having diameter

4
M.C. Nyako et al. Forest Ecology and Management 569 (2024) 122196

Table 1 Table 2
Models structure for the species richness–biomass relationship including cova- Summary of plot-level predictor and response variables for each vegetation
riates and hypotheses. AGB; Aboveground biomass, DI; Disturbance index, HI; cover types. CAF; cacao agroforests, OGF; old-growth forests, WLS; Woodland
Humidity index, VCT; Vegetation cover type. Models M1 to M11 were fitted as savanna. Mean and standard deviation are provided. AGB; Aboveground
non-mixed Generalised Least Square (gls) and refitted as models M12 to M21 biomass; Shadspp ; number of shade bearer species; Pspp ; number of pioneers
using mixed-effects Mixed effect models. The three vegetation cover types (Old- species; NPLDspp ; number of non-pioneer light-demander species; Agricspp ;
growth forests, Wood land savannas and Cocoa-agroforests) as the random effect number of cultivated species.
terms.
Vegetation cover types
Models Coded model structure Hypotheses
CAF OGF WLS
M1 AGB ~ 1 Prediction of AGB would be random
Number of plots (x 0.5-ha) 140 37 20
M2 AGB ~ Diversity Variation of AGB would be predicted by species
Mean number of species 4.4 ± 4.19 19.68 ± 8.39 4.9 ± 3.13
richness (positive or negative relationship)
Number of indicator 4 106 19
M3 AGB ~ Diversity + I Variation of AGB would be predicted by the
species
(Diversity^2) quadratic effect of species richness
AGB ([Link]− 1) 115.02 ± 78.7 217.99 ± 103.23 ±
M4 AGB ~ DI Variation of AGB would be linearly (but
119.1 117.0
negative) predicted by disturbance index
Shadspp 43 123 14
M5 AGB ~ Diversity + DI Disturbance index would directly affect the
Pspp 56 80 41
M6 AGB ~ Diversity × DI positive species richness–relationship;
- succession guild curves for pioneers and NPLDspp 67 99 28
cultivated species will be lower than those for Agricspp 15 7 0
non-pioneer light-demanding and shade- Disturbance index 173.06 ± 166.71 ± 236.67 ±
bearer species. 28.60 33.18 29.92
M7 AGB ~ HI Variation of AGB would be linearly (but Humidity index 49.06 ± 3.64 50.71 ± 6.82 49.72 ± 4.62
positive) predicted by humidity index Total trees 39,791 11,107 7,937
M8 AGB ~ Diversity + HI Humidity index would directly affect the
M9 AGB ~ Diversity × HI positive richness–biomass relationship
M10 AGB ~ Diversity + DI Humidity and disturbance indexes would (Table 3) but a rapid increase in biomass with species richness was
+ HI combine to directly affect the positive observed for the WLS (see inset of Fig. 2e). The curves of the species
M11 AGB ~ Diversity × DI richness–biomass relationship
richness-biomass relationship were unimodal indicating that above-
× HI
M12 AGB ~ 1 + (1 | VCT) M1 - M11 above is a function of the vegetation
ground biomass increases with species richness for each vegetation
M13 AGB ~ Diversity + (1 | cover types cover type (Fig. 2c, e, g).
VCT) The proportion of species’ succession guild mediated the effects of
M14 AGB ~ DI + (1 | VCT) species richness on aboveground biomass (Fig. 2). With all the vegeta-
M15 AGB ~ Diversity + DI
tion cover types pooled together, increase in species richness was
+ (1 | VCT)
M16 AGB ~ Diversity × DI matched by; a decrease in the proportion of cultivated species (Fig. 2b),
+ (1 | VCT) an initial increase then later decrease towards the end of the relationship
M17 AGB ~ HI + (1 | VCT) (monotonic curvilinear) of NPLD and pioneer species (Fig. 2b and
M18 AGB ~ Diversity + HI
Fig. S5) and an increase of shade-bearer species (Fig. 2b and Fig. S5). For
+ (1 | VCT)
M19 AGB ~ Diversity × HI
each of the three vegetation cover types, increase in species richness was
+ (1 | VCT) matched by different patterns of species succession guild: (i) a decrease
M20 AGB ~ Diversity + DI in the proportions of cultivated, pioneer and NPLD species but an in-
+ HI + (1 | VCT) crease in proportion of shade-bearer for OGF (Fig. 2d; Fig. S6b); (ii) a
M21 AGB ~ Diversity × DI
decrease in the proportions of pioneer and shade-bearer nut an increase
× HI × (1 | VCT)
in the proportion of NPLD species for WLS (Fig. 2f; Fig. S6c) and (iii) a
decrease in the proportions of cultivated species, and an increase in the
of ≥ 5 at 1.3 m breat height were used for this study. The response and proportions of NPLD, pioneer and shade-bearer species for CAF (Fig. 2h;
predictor variables were determined for each plot and are summarized Fig. S6a).
in Table 2 in below. Best statistically significant models explaining variation of above-
Three groups of vegetation cover types each consisting of several ground biomass, involved either one predictor only (e.g. richness or
plots were identified from the combination of NMDS analysis (Table 2; humidity index), two predictors (species richness and disturbance index
Fig. S1) and indicator species analysis (Table S1). The first vegetation or species richness and humidity index) or all the three predictors
cover was a group of 37 plots and was interpreted as old-growth forests (richness, disturbance, and humidity) when fitted either with non-mixed
(OGF; Fig. S1) because of the 106 indicator species (Table S1). The effect model or with mixed effects model (Table 3). A total of 12 models
second vegetation cover was composed of 20 plots and because of the 19 were significantly different thus rejecting the null hypotheses that their
indicator species (Table S1), this vegetation cover was interpreted as the variances or intercepts were different from zero. These models included
Woodland savannas (WLS; Fig. S1). The third vegetation cover five mixed effect models (M19, M18, M16, M21, M21; Table 3) and
comprised 140 plots and four indicator species (Table S1) and was seven non-mixed effect models (M11, M9, M5, M8, M7, M4, M2;
interpreted as the Cocoa agroforests (CAF; Fig. S1). Posthoc pair-wise Table 3). The vegetation cover types were significant as the random
comparisons of the three vegetation cover types showed significant effect terms across all the mixed effect models and the three predictors
differences between pairs of the vegetation cover types (Fig. S2). were equally significant as interaction or additive fixed effect terms in
The relationship between species richness and aboveground biomass some models (Table 3). The most parsimonious model among the 12
was positive and significant both when all vegetation cover types were selected models with the lowest BIC, AIC and logLik and highest R2
pooled together (F = 3.09, R2 = 0.46, P < 0.001; Fig. 2a) and when involved the species richness and the humidity index (i.e. model M19;
vegetation cover types were considered independently (gls for OGF: F = Table 3).
4.26, R2 = 0.57, P < 0.001; WLS: F = 7.32, R2 = 0.46, P = 0.001; CAF: F Disturbance index and humidity index had significant influence on
= 5.97, R2 = 0.42, P < 0.001; Fig. 2c, e, g), signifying that aboveground species richness and on aboveground biomass (Fig. 3, Figs. S3 & S4;
biomass increases with species richness irrespective of the vegetation Table 3). The disturbance index had a significant negative influence on
cover type. The strength of this relationship (expressed by R2) decreases the aboveground biomass (gls: F = − 0.61, R2 = 0.046, P = 0.002) and on
across the vegetation cover types from OGF, through WLS to CAF the species richness (gls: F = − 0.010, R2 = 0.026, P = 0.023) when all the

5
M.C. Nyako et al. Forest Ecology and Management 569 (2024) 122196

Fig. 2. Relationships between species richness and aboveground biomass mediated by species successional guilds. Left panel (a, c, e, g) with primary x-axis and y-axis
represents the relationships between species richness and aboveground biomass. Right panel (b, d, f, h) with secondary y-axis represents proportion of species
succession guilds. Coloured lines in right panel represent species succession guilds for cultivated (Agric), Non-pioneer light-demanding (NPLD), Shade-bearer and
pioneer species. Black dashed lines in both panels represent the same species richness – biomass regression models. Models were fitted for; All vegetation cover types
pooled (a, b); for Old-growth forests (OGF: c, d); for Wood land savanna (WLS: e, f) with inset showing log transformed scales and for Cocoa agroforest (CAF: g, h).

vegetation cover types were pooled (Fig. 3a; Table 3). Out of the three (Fig. S3b) but this relationship was not statistically significant (gls: F =
vegetation cover types, WLS had the highest average disturbance index 0.418, R2 = − 0.004, P = 0.519). In contrast to disturbance index, the
but lowest average aboveground biomass and moderate species richness humidity index rather had a significant positive influence on the
(Table 2) while the OGF had the lowest disturbance index but the aboveground biomass (gls: F = 4.50, R2 = 0.041, P = 0.004) and on the
highest species richness (Table 2; Fig. S3a) and highest average above- species richness (gls: F = 0.72, R2 = 0.022, P = 0.037) considering all
ground biomass (Fig. S3b). A slight increase in aboveground biomass vegetation cover types pooled (Fig. S4; Table 3). Out of the three
matching a small increase in disturbance index was observed for the CAF vegetation cover types, WLS and CAF had a similar lowest average

6
M.C. Nyako et al. Forest Ecology and Management 569 (2024) 122196

Table 3
Model results for the prediction of aboveground biomass from species diversity, climate and disturbance. AGB represents aboveground biomass, Diversity represents
species richness, DI represents Disturbance index, HI represents Humidity index, (1 | VCT) defines the three vegetation cover types (old-growth forest, woodland
savanna, and cocoa agroforest) as mixed-effect terms. Coefficients determination for each model included BIC; Bayesian Information Criterion, AIC; Akaike Infor-
mation Criterion, logLik; Maximum Likelihood parameter, R2m; marginal coefficient of determination, R2c conditional coefficient of determination; P; P-value significant
at 0.005 for the fixed effect term (Pm) and mixed effect (Pc). Results are ordered beginning with the most parsimonious models.
Models Coded model structure BIC AIC logLik R2m R2c Pm Pc

M19 AGB ~ Diversity × HI + (1 | VCT) 2243.178 2223.479 − 1105.739 0.634 0.771 <0.001 <0.001
M18 AGB ~ Diversity + HI + (1 | VCT) 2249.438 2233.022 − 1111.511 0.619 0.772 <0.001 <0.001
M13 AGB ~ Diversity + (1 | VCT) 2249.859 2236.726 − 1114.363 0.612 0.762 <0.001 <0.001
M16 AGB ~ Diversity × DI + (1 | VCT) 2252.196 2232.496 − 1110.248 0.621 0.755 0.002 <0.001
M15 AGB ~ Diversity + DI + (1 | VCT) 2253.918 2237.502 − 1113.751 0.611 0.762 1 <0.001
M20 AGB ~ Diversity + DI + HI + (1 | VCT) 2254.338 2234.638 − 1111.319 0.618 0.771 <0.001
M21 AGB ~ Diversity × DI × HI × (1 | VCT) 2256.592 2223.760 − 1101.880 0.642 0.766 0.252 <0.001
M9 AGB ~ Diversity × HI 2260.118 2243.702 − 1116.851 0.506 <0.001 -
M6 AGB ~ Diversity × DI 2264.092 2247.676 − 1118.838 0.496 -
M11 AGB ~ Diversity × DI × HI 2267.124 2237.575 − 1109.788 0.541 0.001 -
M2 AGB ~ Diversity 2267.881 2258.031 − 1126.016 0.458 <0.001 -
M5 AGB ~ Diversity + DI 2268.949 2255.816 − 1123.908 0.470 <0.001 -
M8 AGB ~ Diversity + HI 2269.195 2256.062 − 1124.031 0.469 <0.001 -
M10 AGB ~ Diversity + DI + HI 2271.351 2254.935 − 1122.467 0.477 0.090 -
M3 AGB ~ Diversity + I(Diversity^2) 2272.833 2259.701 − 1125.850 0.459 0.565 -
M14 AGB ~ DI + (1 | VCT) 2361.682 2348.549 − 1170.275 0.040 0.202 <0.001
M17 AGB ~ HI + (1 | VCT) 2363.102 2349.969 − 1170.984 0.023 0.229 <0.001
M12 AGB ~ 1 + (1 | VCT) 2363.449 2353.600 − 1173.800 0.000 0.221 <0.001
M4 AGB ~ DI 2379.315 2369.466 − 1181.733 0.046 <0.001 -
M7 AGB ~ HI 2380.334 2370.484 − 1182.242 0.041 0.004 -
M1 AGB ~ 1 2383.360 2376.794 − 1186.397 0.000 -

humidity index and lowest average species richness while the highest species of economic importance are conspicuous in Cocoa agroforests
average humidity index was recorded for the OGF with a corresponding (Sonwa et al., 2014). Considered as a vegetation cover type at the in-
highest average species richness (Table 2; Fig. S4a).and highest average termediate succession stage, the Woodland savannas with a diversity of
aboveground biomass (Table 2; Fig. S4b). 19 indicator species reflects a mix of grassland and scattered trees,
regaining its initial state after human disturbance such as fire and
4. Discussion grazing. A similar observation had rather interpreted these changes as
forest encroaching into savanna (Mitchard et al., 2009; Mitchard and
Understanding the relationship between species diversity and Flintrop, 2013; Youta-Happi et al., 2003). Our results equally showed
aboveground biomass across vegetation cover types is imperative for that both species richness and aboveground biomass decrease with the
devising strategies for managing biodiversity ecosystem functions intensity of disturbance (Fig. 3;Fig. S3a) and thus suggest that land use
(Lasky et al., 2014; Naeem et al., 2012). We found that aboveground can have opposing effects on the recovery of vegetation quality and
biomass increases with species richness consistently across three vege- structure, as also reported for neotropical forests (Marcano-Vega et al.,
tation cover types in a forest-savanna transition zone in Cameroon. 2002; Pascarella et al., 2000). We recorded substantial shift in species
Moreover, we found that the effect of species richness on aboveground succession guilds, from shade-bearer species (which are usually domi-
biomass weakens across these vegetation cover types with old-growth nant in intact old-growths), to pioneer species (which are prominent in
forest having the tightest and cocoa-agroforest having the weakest disturbed young vegetation). Indeed, the proportion of pioneers (low
relationship. Lastly, we found that among others, species richness in- quality) species was highest in the most disturbed woodland savannas
teracts with climate humidity to increase aboveground biomass while than any other vegetation cover types in this study. However, the pro-
interaction between species richness and disturbance decreases above- portion of pioneers was negatively related to species richness while the
ground biomass across the vegetation cover types. We discuss these re- proportion of shade-bearer species were rather positively related to
sults in the light of ecological succession and environmental species richness (Fig. S5), suggesting that a gradual succession from
determinism and anthropization. pioneers to shade-bearers is matched by an increase in species richness
Our results from the non-metric multidimensional scaling (NMDS) from woodland savanna to old-growth forests.
analysis revealed three distinct groups of vegetation cover types notably The relationship between species richness and ecosystem function
Old-growth forests, Woodland savannas, and Cocoa agroforests (as measured by biomass) have been reported be positive, negative, or
(Fig. S.1), each of these are clearly identifiable by the specific indicator curvilinear (Guo and Berry, 1998; Mittelbach et al., 2001). In this study,
species that characterized them (Table S1). These vegetation cover types we found aboveground biomass to increase with species richness (pos-
may be therefore indicative of an exceptional blend between ecological itive relationship) and this was consistent for the three vegetation cover
succession and human activities operating within this forest-savanna types (Fig. 2a, c, e, g). This result corroborates those from other tropical
transition ecosystem. The Old-growth forest with the highest diversity ecosystems (Chisholm et al., 2013; Day et al., 2014; Homeier and
of 106 indicator species may indicate a mature forest in the late suc- Leuschner, 2021; Steur et al., 2022). We postulate that this positive
cessional stage. As has been reported elsewhere and in this study relationships are consistent with niche complementarity hypothesis,
(Table 2), Old-growth forests are usually characterised by high pro- where higher species richness implies that species develop different
portions of shade-bearer species (Vleminckx et al., 2020). In contrast, niches and are therefore able to access more variable resources, thus
Cocoa agroforests, which has the lowest diversity of only four indicator enhancing the overall biomass (Poorter et al., 2015). This relationship
species, is an example of a human-impacted vegetation cover type. The has been observed for smaller plots of 0.04-ha and becomes inconsistent
Cocoa agroforests may be thought as a vegetation cover type in which for larger plots of 0.25-ha and 1-ha, suggesting that the species rich-
the succession process is regulated to balance agricultural productivity ness–biomass relationship is highly scale-dependent (Chisholm et al.,
with ecological concerns. As a result, high-value timber and other tree 2013; Day et al., 2014; Sullivan et al., 2017). Succession niche effect

7
M.C. Nyako et al. Forest Ecology and Management 569 (2024) 122196

Fig. 3. Relationships between aboveground biomass and species richness with disturbance index and with humidity index. (a) Disturbance index (x-axis) versus
aboveground biomass (y-axis) in grey and disturbance index versus species richness (secondary y-axis) in black symbols. (b) Humidity index (x-axis) versus
aboveground biomass (y-axis) in grey and humidity index versus species richness (secondary y-axis) in black symbols.

occurs when stands dominated by early-successional fast-growing spe- species declined from plots with low species richness to plots with
cies are more productive than those with more diverse communities high-species richness (Fig. S5). This pattern is consistent with increasing
which are usually composed of both early-successional and less pro- aboveground biomass, which we observed for the woodland savannas
ductive late-successional species (Lasky et al. 2014 & references (Fig. S.4).
therein). We noticed that the proportion of pioneers early-succession Our results equally showed that species richness–biomass

8
M.C. Nyako et al. Forest Ecology and Management 569 (2024) 122196

relationship weakens across the three vegetation cover types, suggesting N.E.N.N., S.T.M. and L.T.S., participated in data collection; M.B.L ana-
that environmental determinants mediate the relationship, consistent lysed the data; M.C.N, M.B.L. and R.L.P.B.K. wrote the first draft, S.T.M.,
with a recent global experimental analysis (Duffy et al., 2017). Specif- N.G., M.R.M.E., B.S., and L.Z read the draft and provided useful
ically, the effect of species richness on biomass was weakest for Cocoa feedback.
agroforest and strongest for Old-growth forests. A potential explanation
of this result is that the influence of species richness on biomass is a CRediT authorship contribution statement
response to environmental factors related to climate and disturbance.
Indeed, our analysis indicated that climate humidity index (a derivative Nicole M. Guedje: Writing – review & editing, Supervision, Meth-
of mean annual precipitation and temperature expressing climatic hu- odology. Louis Zapfack: Writing – review & editing, Supervision,
midity) had a strong positive effect on the richness and biomass. The Methodology. Laure B. Fabo: Writing – review & editing, Methodology.
vegetation cover types with higher humidity index (higher humidity) Alex B.T. Fomekong: Writing – review & editing, Methodology. Lise
also had more species and higher biomass. Thus, climate humidity may Zemagho: Writing – review & editing, Supervision, Methodology.
be the most important determinant of the species richness–biomass Cyntia K. Djouking: Writing – review & editing, Methodology. Bona-
relationship for this forest-savanna transition, as also reported for venture Sonké: Writing – review & editing, Supervision, Methodology.
Mongolian steppe communities (Li et al., 2020). Earlier, rainfall has Marius R.M. Ekué: Writing – review & editing, Methodology, Funding
been known to be a critical determinant of woody cover in mixed acquisition. Narcisse E.N. Njila: Writing – review & editing, Method-
woodland savannas (Sankaran et al., 2005). ology, Data curation. Stephane T. Momo: Writing – review & editing,
Considered as an independent determinant or a covariate of climate Writing – original draft, Methodology, Formal analysis, Data curation.
humidity, our results showed that disturbance also mediated the effect Le Bienfaiteur T. Sagang: Writing – review & editing, Methodology.
of species richness on biomass in this forest-savanna landscape. This is Marius G. Babonguen: Writing – review & editing, Methodology.
important because disturbance is expected to affect species diversity Marie C.S. Momo: Writing – review & editing, Supervision, Method-
over long time scales (Molino and Sabatier, 2001). As measured from the ology. Moses B. Libalah: Writing – review & editing, Writing – original
plant perspective in this study, disturbance may entail a complex draft, Supervision, Methodology, Funding acquisition, Formal analysis,
interaction of human activities at play: for example, fire that partially or Data curation, Conceptualization. Géraldine D. Nguemo: Writing –
completely consumes the forest cover can favour establishment of many review & editing, Methodology, Data curation. Louis-Paul-Roger B.
previously non-existent species (Bond et al., 2005; Gentry 1988; Leg- Kabelong: Writing – review & editing, Writing – original draft, Super-
endre, Legendre 1998), logging and windthrow open the forest canopy vision, Methodology, Data curation, Conceptualization. Tagnang N.
in favour of the establishment of fast-growing species (Chazdon, 2014). Madountsap: Writing – review & editing, Methodology. Adrian
The positive relationship between species richness and biomass has Medou: Writing – review & editing, Methodology. Melanie C. Nyako:
important co-benefits for biodiversity conservation and climate change Writing – review & editing, Writing – original draft, Methodology, Data
mitigation (Pan et al., 2011). Species-rich ecosystems can sequester curation, Conceptualization.
substantial amount of carbon; a major consideration for conservation
policies (Anderson-Teixeira, 2018; Gibbs et al., 2007). Old-growth for- Declaration of Competing Interest
ests with the high aboveground biomass also have high species richness,
with substantially more biomass than other vegetation cover with high The authors declare that they have no known competing financial
disturbance. Nevertheless, Woodland savanna and other less interests or personal relationships that could have appeared to influence
species-rich vegetation cover types with low biomass provide other the work reported in this paper.
important ecosystem services, including but not limited to belowground
biomass storage, habitat for faunal diversity, water regulation, and Data availability
nutrient cycling (Egoh et al., 2009). Woodland savannas and other
similar ecosystems with less species and biomass can benefit from Data will be made available on request.
restoration activities but must integrate ecological, economic and social
dimensions in both research and management policies (Bai et al., 2007). Acknowledgments

5. Conclusion This work was funded by the International Foundation for Science
(IFS; grant number D/5621–2), the British Ecological Society (Ecologists
We found a consistent pattern in the relationship between species in African grant number EA18 1601) and the CGIAR Excellence in
richness and aboveground biomass across three ecologically determined Agronomy Initiative (EiA). We thank the University of Yaoundé I for the
vegetation cover types. The relationship is positive linear for Old-growth quality training that they have offered to us to enable us successfully to
forests, Woodland savanna and Cocoa agroforests, indicating that spe- carry out this research study. We are grateful to the staff of the labora-
cies richness consistently increases aboveground biomass in this forest- tory of Botany and Ecology of the University of Yaoundé I for support
savanna ecosystem. We noted that climate humidity greatly promotes during field work and farmers and their respective cooperatives in
the establishment of species and sequestration of biomass, but distur- Ndikiniméki, Etoundou I & II, Nefand, Nomale and Kolokolo for their
bance tends to alter the quality and quantity of species richness and active collaboration during field work.
biomass. Thus, we recommend that both field and airborne models
developed to predict biomass from species richness for this landscape Appendix A. Supporting information
should directly incorporate measures of climate humidity and distur-
bance to increase their applicability. REDD policies that prioritize Supplementary data associated with this article can be found in the
species-rich vegetation cover types could incorporate ecological resto- online version at doi:10.1016/[Link].2024.122196.
ration of less species-rich vegetation to regain initial diversity and pro-
ductivity functions. References

Authors contributions statement Adler, P.B., Seabloom, E.W., Borer, E.T., Hillebrand, H., Hautier, Y., Hector, A., Stanley
Harpole, W., O’Halloran, L.R., Grace, J.B., Michael Anderson, T., Bakker, J.D.,
Biederman, L.A., Brown, C.S., Buckley, Y.M., Calabrese, L.B., Chu, C.J., Cleland, E.E.,
M.C.N, M.B.L. and R.L.P.B.K. designed the research; M.C.N, M.B.L., Collins, S.L., Cottingham, K.L., Crawley, M.J., Damschen, E.I., Davies, K.F.,
R.L.P.B.K., M.G.B., C.K.D., A.B.T.F., L.B.F., A.M., S.M., M.C.S.M., G.N., DeCrappeo, N.M., Fay, P.A., Firn, J., Frater, P., Gasarch, E.I., Gruner, D.S.,

9
M.C. Nyako et al. Forest Ecology and Management 569 (2024) 122196

Hagenah, N., Lambers, J.H.R., Humphries, H., Jin, V.L., Kay, A.D., Kirkman, K.P., Duffy, J.E., Godwin, C.M., Cardinale, B.J., 2017. Biodiversity effects in the wild are
Klein, J.A., Knops, J.M.H., La Pierre, K.J., Lambrinos, J.G., Li, W., MacDougall, A.S., common and as strong as key drivers of productivity. Nature 549, 261–264. https://
McCulley, R.L., Melbourne, B.A., Mitchell, C.E., Moore, J.L., Morgan, J.W., [Link]/10.1038/nature23886.
Mortensen, B., Orrock, J.L., Prober, S.M., Pyke, D.A., Risch, A.C., Schuetz, M., Egoh, B., Reyers, B., Rouget, M., Bode, M., Richardson, D.M., 2009. Spatial congruence
Smith, M.D., Stevens, C.J., Sullivan, L.L., Wang, G., Wragg, P.D., Wright, J.P., between biodiversity and ecosystem services in South Africa. Biol. Conserv. 142,
Yang, L.H., 2011. Productivity is a poor predictor of plant species richness. Science 553–562. [Link]
80 (333), 1750–1753. [Link] Fick, S.E., Hijmans, R.J., 2017. WorldClim 2: new 1-km spatial resolution climate
Ammer, C., 2019. Diversity and forest productivity in a changing climate. N. Phytol. 221, surfaces for global land areas. Int. J. Climatol. 37, 4302–4315. [Link]
50–66. [Link] 10.1002/joc.5086.
Anderson-Teixeira, K.J., 2018. Prioritizing biodiversity and carbon. Nat. Clim. Chang. Foley, J.A., DeFries, R., Asner, G.P., Barford, C., Bonan, G., Carpenter, S.R., Chapin, F.S.,
[Link] Coe, M.T., Daily, G.C., Gibbs, H.K., Helkowski, J.H., Holloway, T., Howard, E.A.,
APG IV, 2016. An update of the Angiosperm Phylogeny group classification for the orders Kucharik, C.J., Monfreda, C., Patz, J.A., Prentice, I.C., Ramankutty, N., Snyder, P.K.,
and families of flowering plants: APG IV. Bot. J. Linn. Soc. 181, 1–20. [Link] 2005. Global consequences of land use. Science 80 (309), 570–574. [Link]
org/10.1111/boj.12385. 10.1126/science.1111772.
Bai, Y., Wu, J., Pan, Q., Huang, J., Wang, Q., Li, F., Buyantuyev, A., Han, X., 2007. Gentry, A.H., 1988. Changes in plant community diversity and floristic composition on
Positive linear relationship between productivity and diversity: Evidence from the environmental and geographical gradients. Ann. Missouri Bot. Gard. 75, 1–34.
Eurasian Steppe. J. Appl. Ecol. 44, 1023–1034. [Link] [Link]
2664.2007.01351.x. Gibbs, H.K., Brown, S., Niles, J.O., Foley, J.A., 2007. Monitoring and estimating tropical
Bartoń, K., 2009. MuMIn: Multi-model inference. CRAN 〈https//[Link]/we forest carbon stocks: making REDD a reality. Environ. Res. Lett. [Link]
b/packages/MuMIn/[Link]〉. 10.1088/1748-9326/2/4/045023.
Bates, D., Mächler, M., Bolker, B.M., Walker, S.C., 2015. Fitting linear mixed-effects Gosz, J.R., 1993. Ecotone hierarchies. Ecol. Appl. 3, 369–376. [Link]
models using lme4. J. Stat. Softw. 67, 1–48. [Link] 1941905.
Bénédet, F., Doucet, J.-L., Fayolle, A., Gillet, J.-F., Gourlet-Fleury, S., Vincke, D., 2019. Gourlet-Fleury, S., Rossi, V., Rejou-Mechain, M., Freycon, V., Fayolle, A., Saint-Andre, L.,
CoForTraits, African plant traits information database. version 1.0. [WWW Cornu, G., Gerard, J., Sarrailh, J.M., Flores, O., Baya, F., Billand, A., Fauvet, N.,
Document]. Gally, M., Henry, M., Hubert, D., Pasquier, A., Picard, N., 2011. Environmental
Bond, W.J., Woodward, F.I., Midgley, G.F., 2005. The global distribution of ecosystems filtering of dense-wooded species controls above-ground biomass stored in African
in a world without fire. N. Phytol. 165, 525–538. [Link] moist forests. J. Ecol. 99, 981–990. [Link]
8137.2004.01252.x. 2745.2011.01829.x.
Bongers, F., Poorter, L., Hawthorne, W.D., Sheil, D., 2009. The intermediate disturbance Guo, Q., Berry, W.L., 1998. Species richness and biomass: Dissection of the hump-shaped
hypothesis applies to tropical forests, but disturbance contributes little to tree relationships. Ecology 79, 2555–2559 [Link]
diversity. Ecol. Lett. 12, 798–805. [Link] 079[2555:SRABDO][Link];2.
0248.2009.01329.x. Hawthorne, W.D., 1995. Ecological profiles of Ghanaian forest trees. Trop. For. Pap. No.
Bouvet, A., Mermoz, S., Le Toan, T., Villard, L., Mathieu, R., Naidoo, L., Asner, G.P., 29, vi, vi.
2018. An above-ground biomass map of African savannahs and woodlands at 25 m Hawthorne, W.D., 1996. Holes and the sums of parts in Ghanaian forest: regeneration,
resolution derived from ALOS PALSAR. Remote Sens. Environ. 206, 156–173. scale and sustainable use. Proc. R. Soc. Edinb. Sect. B Biol. Sci. 104, 75–176. https://
[Link] [Link]/10.1017/S0269727000006126.
Cardinale, B.J., Wright, J.P., Cadotte, M.W., Carroll, I.T., Hector, A., Srivastava, D.S., Homeier, J., Leuschner, C., 2021. Factors controlling the productivity of tropical Andean
Loreau, M., Weis, J.J., 2007. Impacts of plant diversity on biomass production forests: climate and soil are more important than tree diversity. Biogeosciences 18,
increase through time because of species complementarity. Proc. Natl. Acad. Sci. U. 1525–1541. [Link]
S. A. 104, 18123–18128. [Link] Kark, S., van Rensburg, B.J., 2006. Ecotones: marginal or central areas of transition? Isr.
Chave, J., Coomes, D., Jansen, S., Lewis, S.L., Swenson, N.G., Zanne, A.E., 2009. Towards J. Ecol. Evol. 52, 29–53. [Link]
a worldwide wood economics spectrum. Ecol. Lett. 12, 351–366. [Link] Lasky, J.R., Uriarte, M., Boukili, V.K., Erickson, D.L., John Kress, W., Chazdon, R.L.,
10.1111/j.1461-0248.2009.01285.x. 2014. The relationship between tree biodiversity and biomass dynamics changes
Chave, J., Réjou-Méchain, M., Búrquez, A., Chidumayo, E., Colgan, M.S., Delitti, W.B.C., with tropical forest succession. Ecol. Lett. 17, 1158–1167. [Link]
Duque, A., Eid, T., Fearnside, P.M., Goodman, R.C., Henry, M., Martínez-Yrízar, A., ele.12322.
Mugasha, W.A., Muller-Landau, H.C., Mencuccini, M., Nelson, B.W., Ngomanda, A., Legendre, P., Legendre, L., 1998. Numerical Ecology, Numerical Ecology Second English
Nogueira, E.M., Ortiz-Malavassi, E., Pélissier, R., Ploton, P., Ryan, C.M., Edition. [Link]
Saldarriaga, J.G., Vieilledent, G., 2014. Improved allometric models to estimate the Letouzey, R., 1985. Carte phytogéographique du Cameroun, 1:500 000, 8 feuillets + 5
aboveground biomass of tropical trees. Glob. Chang. Biol. 20, 3177–3190. https:// notices. Institut de la Carte Internationale de la Végétation, Toulouse, France.
[Link]/10.1111/gcb.12629. Lewis, S.L., Edwards, D.P., Galbraith, D., 2015. Increasing human dominance of tropical
Chazdon, R.L., 2014. Forest Regeneration following Selective Logging and Land-Use forests. Science 80 (349), 827–832. [Link]
Synergisms. In: Second Growth: The Promise of Tropical Forest Regeneration in an Li, Zijing, Li, Zhiyong, Tong, X., Zhang, J., Dong, L., Zheng, Y., Ma, W., Zhao, L.,
Age of Deforestation. Chicago. University of Chicago Press, Chicago, pp. 146–166. Wang, L., Wen, L., Dang, Z., Tuvshintogtokh, I., Liang, C., Li, F.Y., 2020. Climatic
[Link] humidity mediates the strength of the species richness–biomass relationship on the
Chisholm, R.A., Muller-Landau, H.C., Abdul Rahman, K., Bebber, D.P., Bin, Y., Mongolian Plateau steppe. Sci. Total Environ. 718, 137252 [Link]
Bohlman, S.A., Bourg, N.A., Brinks, J., Bunyavejchewin, S., Butt, N., Cao, H., 10.1016/[Link].2020.137252.
Cao, M., Cárdenas, D., Chang, L.W., Chiang, J.M., Chuyong, G., Condit, R., Liang, J., Crowther, T.W., Picard, N., Wiser, S., Zhou, M., Alberti, G., Schulze, E.D.,
Dattaraja, H.S., Davies, S., Duque, A., Fletcher, C., Gunatilleke, N., Gunatilleke, S., McGuire, A.D., Bozzato, F., Pretzsch, H., De-Miguel, S., Paquette, A., Hérault, B.,
Hao, Z., Harrison, R.D., Howe, R., Hsieh, C.F., Hubbell, S.P., Itoh, A., Kenfack, D., Scherer-Lorenzen, M., Barrett, C.B., Glick, H.B., Hengeveld, G.M., Nabuurs, G.J.,
Kiratiprayoon, S., Larson, A.J., Lian, J., Lin, D., Liu, H., Lutz, J.A., Ma, K., Malhi, Y., Pfautsch, S., Viana, H., Vibrans, A.C., Ammer, C., Schall, P., Verbyla, D.,
Mcmahon, S., Mcshea, W., Meegaskumbura, M., Mohd. Razman, S., Morecroft, M.D., Tchebakova, N., Fischer, M., Watson, J.V., Chen, H.Y.H., Lei, X., Schelhaas, M.J.,
Nytch, C.J., Oliveira, A., Parker, G.G., Pulla, S., Punchi-Manage, R., Romero- Lu, H., Gianelle, D., Parfenova, E.I., Salas, C., Lee, E., Lee, B., Kim, H.S.,
Saltos, H., Sang, W., Schurman, J., Su, S.H., Sukumar, R., Sun, I.F., Suresh, H.S., Bruelheide, H., Coomes, D.A., Piotto, D., Sunderland, T., Schmid, B., Gourlet-
Tan, S., Thomas, D., Thomas, S., Thompson, J., Valencia, R., Wolf, A., Yap, S., Ye, W., Fleury, S., Sonké, B., Tavani, R., Zhu, J., Brandl, S., Vayreda, J., Kitahara, F.,
Yuan, Z., Zimmerman, J.K., 2013. Scale-dependent relationships between tree Searle, E.B., Neldner, V.J., Ngugi, M.R., Baraloto, C., Frizzera, L., Bałazy, R.,
species richness and ecosystem function in forests. J. Ecol. 101, 1214–1224. https:// Oleksyn, J., Zawiła-Niedźwiecki, T., Bouriaud, O., Bussotti, F., Finér, L.,
[Link]/10.1111/1365-2745.12132. Jaroszewicz, B., Jucker, T., Valladares, F., Jagodzinski, A.M., Peri, P.L.,
Day, M., Baldauf, C., Rutishauser, E., Sunderland, T.C.H., 2014. Relationships between Gonmadje, C., Marthy, W., O’Brien, T., Martin, E.H., Marshall, A.R., Rovero, F.,
tree species diversity and above-ground biomass in Central African rainforests: Bitariho, R., Niklaus, P.A., Alvarez-Loayza, P., Chamuya, N., Valencia, R.,
implications for REDD. Environ. Conserv. 41, 64–72. [Link] Mortier, F., Wortel, V., Engone-Obiang, N.L., Ferreira, L.V., Odeke, D.E., Vasquez, R.
S0376892913000295. M., Lewis, S.L., Reich, P.B., 2016. Positive biodiversity-productivity relationship
De Cáceres, M., Legendre, P., 2009. Associations between species and groups of sites: predominant in global forests. Science 80 (354). [Link]
indices and statistical inference. Ecology 90, 3566–3574. [Link] aaf8957.
08-1823.1. Liautaud, K., Barbier, M., Loreau, M., 2020. Ecotone formation through ecological niche
Djiofack, B.Y., Beeckman, H., Bourland, N., Belanganayi, B.L., Laurent, F., Ilondea, B.A., construction: the role of biodiversity and species interactions. Ecography 714–723.
Nsenga, L., Huart, A., Longwwango, M.M., Deklerck, V., Lejeune, G., Verbiest, W.W. [Link]
M., Van den Bulcke, J., Van Acker, J., De Mil, T., Hubau, W., 2024. Protecting an Loreau, M., Hector, A., 2001. Partitioning selection and complementarity in biodiversity
artificial savanna as a nature-based solution to restore carbon and biodiversity in the experiments. Nature 412, 72–76. [Link]
Democratic Republic of the Congo. Glob. Chang. Biol. 30, 1–18. [Link] Marcano-Vega, H., Aide, T.M., Báez, D., 2002. Forest regeneration in abandoned coffee
10.1111/gcb.17154. plantations and pastures in the Cordillera Central of Puerto Rico. Plant Ecol. 161,
Djuikouo, M.N.K., Doucet, J.L., Nguembou, C.K., Lewis, S.L., Sonké, B., 2010. Diversity 75–87. [Link]
and aboveground biomass in three tropical forest types in the Dja Biosphere Reserve, Mayaux, P., Richards, T., Janodet, E., 1999. A vegetation map of Central Africa derived
Cameroon. Afr. J. Ecol. 48, 1053–1063. [Link] from satellite imagery. J. Biogeogr. 26, 353–366.
2028.2010.01212.x. Mitchard, E.T.A., Flintrop, C.M., 2013. Woody encroachment and forest degradation in
sub-Saharan Africa’s woodlands and savannas 1982-2006. Philos. Trans. R. Soc. B
Biol. Sci. 368, 1–7. [Link]

10
M.C. Nyako et al. Forest Ecology and Management 569 (2024) 122196

Mitchard, E.T.A., Saatchi, S.S., Gerard, F., Lewis, S.L., Meir, P., 2009. Measuring Woody Chen, Y., Colwell, R.K., Chazdon, R.L., Robin, C.L., Clark, C., Clark, D.B., Clark, D.A.,
Encroachment along a Forest – Savanna Boundary in Central Africa. Earth inter. 13, Culmsee, H., Damas, K., Dattaraja, H.S., Dauby, G., Davidar, P., DeWalt, S.J.,
1–29. [Link] Doucet, J.-L., Duque, A., Durigan, G., Eichhorn, K.A.O., Eisenlohr, P.V., Eler, E.,
Mittelbach, G.G., 2010. Understanding species richness-productivity relationships: the Ewango, C., Farwig, N., Feeley, K.J., Ferreira, L., Field, R., de Oliveira Filho, A.T.,
importance of meta-analyses. Ecology 91, 2540–2544. [Link] Fletcher, C., Forshed, O., Franco, G., Fredriksson, G., Gillespie, T., Gillet, J.-F.,
1029.1. Amarnath, G., Griffith, D.M., Grogan, J., Gunatilleke, N., Harris, D., Harrison, R.,
Mittelbach, G.G., Steiner, C.F., Scheiner, S.M., Gross, K.L., Reynolds, H.L., Waide, R.B., Hector, A., Homeier, J., Imai, N., Itoh, A., Jansen, P.A., Joly, C.A., de Jong, B.H.J.,
Willig, M.R., Dodson, S.I., Gough, L., 2001. What is the observed relationship Kartawinata, K., Kearsley, E., Kelly, D.L., Kenfack, D., Kessler, M., Kitayama, K.,
between species richness and productivity? Ecology 82, 2381–2396 [Link] Kooyman, R., Larney, E., Laumonier, Y., Laurance, S., Laurance, W.F., Lawes, M.J.,
10.1890/0012-9658(2001)082[2381:WITORB][Link];2. Amaral, I.L., do, Letcher, S.G., Lindsell, J., Lu, X., Mansor, A., Marjokorpi, A.,
Molino, J., Sabatier, D., 2001. Tree diversity in tropical rain forests: a validation of the Martin, E.H., Meilby, H., Melo, F.P.L., Metcalfe, D.J., Medjibe, V.P., Metzger, J.P.,
intermediate disturbance hypothesis. Science 294 (80), 1702–1704. [Link] Millet, J., Mohandass, D., Montero, J.C., de Morisson Valeriano, M., Mugerwa, B.,
10.1126/science.1060284. Nagamasu, H., Nilus, R., Ochoa-Gaona, S., Onrizal, Page, N., Parolin, P., Parren, M.,
Naeem, S., Duffy, J.E., Zavaleta, E., 2012. The functions of biological diversity in an age Parthasarathy, N., Paudel, E., Permana, A., Piedade, M.T.F., Pitman, N.C.A.,
of extinction. Science 336 (80), 1401–1406. [Link] Poorter, L., Poulsen, A.D., Poulsen, J., Powers, J., Prasad, R.C., Puyravaud, J.-P.,
science.1215855. Razafimahaimodison, J.-C., Reitsma, J., Dos Santos, J.R., Roberto Spironello, W.,
Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., O’Hara, R.B., Simpson, G.L., Romero-Saltos, H., Rovero, F., Rozak, A.H., Ruokolainen, K., Rutishauser, E.,
Solymos, P., Stevens, M.H.H., Wagner, H., 2012. vegan: Community Ecology Saiter, F., Saner, P., Santos, B.A., Santos, F., Sarker, S.K., Satdichanh, M., Schmitt, C.
Package. [Link] B., Schöngart, J., Schulze, M., Suganuma, M.S., Sheil, D., da Silva Pinheiro, E.,
Oliveras, I., Malhi, Y., 2016. Many shades of green: The dynamic tropical Sist, P., Stevart, T., Sukumar, R., Sun, I.-F., Sunderland, T., Sunderand, T., Suresh, H.
forest–savannah transition zones. Philos. Trans. R. Soc. B Biol. Sci. [Link] S., Suzuki, E., Tabarelli, M., Tang, J., Targhetta, N., Theilade, I., Thomas, D.W.,
10.1098/rstb.2015.0308. Tchouto, P., Hurtado, J., Valencia, R., van Valkenburg, J.L.C.H., Van Do, T.,
Onana, J.M., 2018. Mapping the ecosystems of Cameroon. Int. J. Biol. Chem. Sci. 12, Vasquez, R., Verbeeck, H., Adekunle, V., Vieira, S.A., Webb, C.O., Whitfeld, T.,
940–957. Wich, S.A., Williams, J., Wittmann, F., Wöll, H., Yang, X., Adou Yao, C.Y., Yap, S.L.,
Pan, Y., Birdsey, R.A., Fang, J., Houghton, R., Kauppi, P.E., Kurz, W.A., Phillips, O.L., Yoneda, T., Zahawi, R.A., Zakaria, R., Zang, R., de Assis, R.L., Garcia Luize, B.,
Shvidenko, A., Lewis, S.L., Canadell, J.G., Ciais, P., Jackson, R.B., Pacala, S.W., Venticinque, E.M., 2015. An estimate of the number of tropical tree species. Proc.
McGuire, A.D., Piao, S., Rautiainen, A., Sitch, S., Hayes, D., 2011. A large and Natl. Acad. Sci. U. S. A 112. [Link]
persistent carbon sink in the world’s forests. Science 333 (80), 988–993. [Link] Sonwa, D.J., Weise, S.F., Schroth, G., Janssens, M.J.J., Howard-Yana Shapiro, 2014.
org/10.1126/science.1201609. Plant diversity management in cocoa agroforestry systems in West and Central
Pascarella, J.B., Aide, T.M., Serrano, M.I., Zimmerman, J.K., 2000. Land-use history and Africa-effects of markets and household needs. Agrofor. Syst. 88, 1021–1034.
forest regeneration in the Cayey Mountains, Puerto Rico. Ecosystems 3, 217–228. [Link]
[Link] Steur, G., ter Steege, H., Verburg, R.W., Sabatier, D., Molino, J.F., Bánki, O.S.,
Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., R. Core Team, 2022. “nlme” Title Linear Castellanos, H., Stropp, J., Fonty, É., Ruysschaert, S., Galbraith, D.,
and Nonlinear Mixed Effects Models. R Packag. version 3.1-164, 〈https//CRAN. Kalamandeen, M., van Andel, T.R., Brienen, R., Phillips, O.L., Feeley, K.J.,
[Link]/package=nlme〉. Terborgh, J., Verweij, P.A., 2022. Relationships between species richness and
Poorter, L., van der Sande, M.T., Thompson, J., Arets, E.J.M.M., Alarcón, A., Álvarez- ecosystem services in Amazonian forests strongly influenced by biogeographical
Sánchez, J., Ascarrunz, N., Balvanera, P., Barajas-Guzmán, G., Boit, A., Bongers, F., strata and forest types. Sci. Rep. 12, 1–11. [Link]
Carvalho, F. a, Casanoves, F., Cornejo-Tenorio, G., Costa, F.R.C., de Castilho, C.V., 09786-6.
Duivenvoorden, J.F., Dutrieux, L.P., Enquist, B.J., Fernández-Méndez, F., Sullivan, M.J.P., Talbot, J., Lewis, S.L., Phillips, O.L., Qie, L., Begne, S.K., Chave, J., Cuni-
Finegan, B., Gormley, L.H.L., Healey, J.R., Hoosbeek, M.R., Ibarra-Manríquez, G., Sanchez, A., Hubau, W., Lopez-Gonzalez, G., Miles, L., Monteagudo-Mendoza, A.,
Junqueira, a B., Levis, C., Licona, J.C., Lisboa, L.S., Magnusson, W.E., Martínez- Sonké, B., Sunderland, T., Ter Steege, H., White, L.J.T., Affum-Baffoe, K., Aiba, S.I.,
Ramos, M., Martínez-Yrizar, A., Martorano, L.G., Maskell, L.C., Mazzei, L., Meave, J. De Almeida, E.C., De Oliveira, E.A., Alvarez-Loayza, P., Dávila, E.Á., Andrade, A.,
a, Mora, F., Muñoz, R., Nytch, C., Pansonato, M.P., Parr, T.W., Paz, H., Pérez- Aragão, L.E.O.C., Ashton, P., Aymard, G.A., Baker, T.R., Balinga, M., Banin, L.F.,
García, E. a, Rentería, L.Y., Rodríguez-Velazquez, J., Rozendaal, D.M. a, Ruschel, a Baraloto, C., Bastin, J.F., Berry, N., Bogaert, J., Bonal, D., Bongers, F., Brienen, R.,
R., Sakschewski, B., Salgado-Negret, B., Schietti, J., Simões, M., Sinclair, F.L., Camargo, J.L.C., Cerón, C., Moscoso, V.C., Chezeaux, E., Clark, C.J., Pacheco, Á.C.,
Souza, P.F., Souza, F.C., Stropp, J., ter Steege, H., Swenson, N.G., Thonicke, K., Comiskey, J.A., Valverde, F.C., Coronado, E.N.H., Dargie, G., Davies, S.J., De
Toledo, M., Uriarte, M., van der Hout, P., Walker, P., Zamora, N., Peña-Claros, M., Canniere, C., Djuikouo, M.N., Doucet, J.L., Erwin, T.L., Espejo, J.S., Ewango, C.E.N.,
2015. Diversity enhances carbon storage in tropical forests. Glob. Ecol. Biogeogr. 24, Fauset, S., Feldpausch, T.R., Herrera, R., Gilpin, M., Gloor, E., Hall, J.S., Harris, D.J.,
1314–1328. [Link] Hart, T.B., Kartawinata, K., Kho, L.K., Kitayama, K., Laurance, S.G.W., Laurance, W.
POWO, 2023. Plants of the World Online. Facilitated by the Royal Botanic Gardens, Kew F., Leal, M.E., Lovejoy, T., Lovett, J.C., Lukasu, F.M., Makana, J.R., Malhi, Y.,
[WWW Document]. R. Bot. Gard. Kew. URL [Link] Maracahipes, L., Marimon, B.S., Junior, B.H.M., Marshall, A.R., Morandi, P.S.,
(accessed 3.9.23). Mukendi, J.T., Mukinzi, J., Nilus, R., Vargas, P.N., Camacho, N.C.P., Pardo, G., Peña-
Rajpar, M.N., 2018. Tropical Forests Are An Ideal Habitat for Wide Array of Wildlife Claros, M., Pétronelli, P., Pickavance, G.C., Poulsen, A.D., Poulsen, J.R., Primack, R.
Species, in: Tropical Forests - New Edition. [Link] B., Priyadi, H., Quesada, C.A., Reitsma, J., Réjou-Méchain, M., Restrepo, Z.,
intechopen.73315. Rutishauser, E., Salim, K.A., Salomão, R.P., Samsoedin, I., Sheil, D., Sierra, R.,
Raven, P., 1988. Our Diminishing Tropical Forests in Biodiversity, in: Biodiversity. p. Silveira, M., Slik, J.W.F., Steel, L., Taedoumg, H., Tan, S., Terborgh, J.W., Thomas, S.
504. C., Toledo, M., Umunay, P.M., Gamarra, L.V., Vieira, I.C.G., Vos, V.A., Wang, O.,
Réjou-Méchain, M., Tanguy, A., Piponiot, C., Chave, J., Hérault, B., 2017. Biomass: an R Willcock, S., Zemagho, L., 2017. Diversity and carbon storage across the tropical
package for estimating above-ground biomass and its uncertainty in tropical forests. forest biome. Sci. Rep. 7, 1–12. [Link]
Methods Ecol. Evol. 8, 1163–1167. [Link] Tilman, D., Isbell, F., Cowles, J.M., 2014. Biodiversity and ecosystem functioning. Annu.
Rosenzweig, M.L., 1995. Species Diversity in Space and Time, Species Diversity in Space Rev. Ecol. Evol. Syst. 45, 471–493. [Link]
and Time. Cambridge University Press, UK. [Link] 120213-091917.
cbo9780511623387. Tuhkanen, S., 1980. Climatic parameters and indices in plant geography, Acta
Roswell, M., Harrison, T., Genung, M.A., 2023. Biodiversity-ecosystem function Phytogeographica Suecica. Almqvist & Wiksell International, Stockolm.
relationships change in sign and magnitude across the Hill diversity spectrum. Veenendaal, E.M., Torello-Raventos, M., Feldpausch, T.R., Domingues, T.F., Gerard, F.,
Philos. Trans. R. Soc. B Biol. Sci. 378 [Link] Schrodt, F., Saiz, G., Quesada, C.A., Djagbletey, G., Ford, A., Kemp, J., Marimon, B.
Sankaran, M., Hanan, N.P., Scholes, R.J., Ratnam, J., Augustine, D.J., Cade, B.S., S., Marimon-Junior, B.H., Lenza, E., Ratter, J.A., Maracahipes, L., Sasaki, D.,
Gignoux, J., Higgins, S.I., Roux, X., Le, Ludwig, F., Ardo, J., Banyikwa, F., Bronn, A., Sonké, B., Zapfack, L., Villarroel, D., Schwarz, M., Yoko Ishida, F., Gilpin, M.,
Bucini, G., Caylor, K.K., Coughenour, M.B., Diouf, A., Ekaya, W., Feral, C.J., Nardoto, G.B., Affum-Baffoe, K., Arroyo, L., Bloomfield, K., Ceca, G., Compaore, H.,
February, E.C., Frost, P.G.H., Hiernaux, P., Worden, J., Zambatis, N., 2005. Davies, K., Diallo, A., Fyllas, N.M., Gignoux, J., Hien, F., Johnson, M., Mougin, E.,
Determinants of woody cover in African savannas. Nature 438, 8–11. [Link] Hiernaux, P., Killeen, T., Metcalfe, D., Miranda, H.S., Steininger, M., Sykora, K.,
org/10.1038/nature04070. Bird, M.I., Grace, J., Lewis, S., Phillips, O.L., Lloyd, J., 2015. Structural,
Schuldt, A., Liu, X., Buscot, F., Bruelheide, H., Erfmeier, A., He, J.S., Klein, A.M., Ma, K., physiognomic and above-ground biomass variation in savanna-forest transition
Scherer-Lorenzen, M., Schmid, B., Scholten, T., Tang, Z., Trogisch, S., Wirth, C., zones on three continents - How different are co-occurring savanna and forest
Wubet, T., Staab, M., 2023. Carbon–biodiversity relationships in a highly diverse formations? Biogeosciences 12, 2927–2951. [Link]
subtropical forest. Glob. Chang. Biol. 29, 5321–5333. [Link] 2015.
gcb.16697. Vleminckx, J., Bauman, D., Demanet, M., Hardy, O.J., Doucet, J.L., Drouet, T., 2020. Past
Sheil, D., Burslem, D.F.R.P., 2003. Disturbing hypotheses in tropical forests. Trends Ecol. human disturbances and soil fertility both influence the distribution of light-
Evol. 18, 18–26. [Link] demanding tree species in a Central African tropical forest. J. Veg. Sci. 31, 440–453.
Slik, J.W.F., Arroyo-Rodríguez, V., Aiba, S.-I., Alvarez-Loayza, P., Alves, L.F., Ashton, P., [Link]
Balvanera, P., Bastian, M.L., Bellingham, P.J., van den Berg, E., Bernacci, L., da White, F., 1983. The Vegetation of Africa, A Descriptive Memoir to Accompany the
Conceição Bispo, P., Blanc, L., Böhning-Gaese, K., Boeckx, P., Bongers, F., Boyle, B., UNESCO/AETFAT/UNSO Vegetation Map of Africa, UNESCO. [Link]
Bradford, M., Brearley, F.Q., Breuer-Ndoundou Hockemba, M., Bunyavejchewin, S., 10.2307/2260340.
Calderado Leal Matos, D., Castillo-Santiago, M., Catharino, E.L.M., Chai, S.-L.,

11
M.C. Nyako et al. Forest Ecology and Management 569 (2024) 122196

Yerima, B.P.U.K., Ranst, V.E., 2005. Major soil classification systems used in the tropics: quarante ans ( 1950-1990) Dans la région du confluent du Mbam et du Kim, Centre-
soils of Cameroon. Trafford Publishing. Cameroun, in: Peuplements Anciens et Actuels Des Forêts Tropicales. p. 380 p.
Youta-Happi, J., Bonvallot, J., Hotyat, M., Guillet, B., Peltre, P., Schwartz, D., Servant, Zanne, A.E., Lopez-Gonzalez, G., Coomes, D.A.A., Ilic, J., Jansen, S., Lewis, S.L.S.L.,
M., Simonneaux, V., 2003. Bilan de la dynamique du contact forêt-savane en Miller, R.B.B., Swenson, N.G.G., Wiemann, M.C.C., Chave, J., 2009. Global wood
density database. Dryad 235, 33. [Link]

12