Content uploaded by Elise Bouchard
Author content
All content in this area was uploaded by Elise Bouchard on Jun 14, 2024
Content may be subject to copyright.
Trees, Forests and People 16 (2024) 100602
Available online 11 June 2024
2666-7193/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
Forecasting forest management impacts on regeneration traits of high
socio-economic value species in Senegal’s Sudanian zone, West Africa
Niang Fatimata
a
,
b
,
*
, Marchand Philippe
b
, Fenton Nicole J
b
, Sambou Bienvenu
a
,
Bouchard Elise
c
a
Institut des Sciences de l´Environnement, Universit´
e Cheikh Anta Diop de Dakar, Senegal
b
Institut de Recherches sur les Forˆ
ets, Universit´
e du Qu´
ebec en Abitibi-T´
emiscamingue, Canada
c
Universit´
e du Qu´
ebec `
a Montr´
eal, Canada
ARTICLE INFO
Keywords:
Saplings
Functional traits
Savannahs
Natural regeneration
Disturbance
ABSTRACT
Forest management can have major impacts on the dynamic of ecological communities, including shifts in
reproduction and survival strategies in newly recruited individuals. This study aims to predict the probability of
presence of saplings of various species in managed forests in the Sudanian domain of Senegal based on their
functional traits. Data on functional traits and the main commercial and domestic uses of twenty high-socio-
economic value species were collected, along with their presence-absence, in 12, 832 plots from twenty sites
in Senegal’s savannahs spanning four different types of forest management, including unmanaged forests, old
managed forests, recently managed forests, and community reserve forests. Mixed logistic regression models
were used to predict whether high-value species would be present in different forest management types,
depending on their main uses, and regeneration, growth and reproduction functional traits. We hypothesised that
forest management would favour species with more efcient colonising and competitive abilities, particularly at
high levels of disturbance. Our ndings demonstrated that unmanaged forests were more likely to host high-
value species, regardless of their functional traits and main uses. These protected forests also hosted the great-
est variety of regeneration functional traits, heights and uses. Old managed forests were more likely to host high-
value species compared to recently managed and community reserve forests. Moreover, high-value species
capable of vegetative regeneration and those with low maximum height were more likely to occur in all types of
forests, but this trend was more pronounced under higher management and disturbance intensities. This study
highlights that the availability of forest products can be optimised by encouraging management strategies that
promote a diversity of functional traits rather than the selective harvesting of certain high-value species.
Additionally, the temporal variation in the responses of tree species could be benecial for forest management,
with more diverse values of functional traits and more high-value species being present in older managed forests
relative to newly managed forests. These results indicate that it is essential to promote forest resilience after
management by protecting high-value species that have the potential to restore forest functional composition
over time in managed forests.
1. Introduction
Anthropogenic disturbance, such as selective logging, is a signicant
contributor to biodiversity loss in tropical ecosystems. This loss can be as
severe as deforestation and particularly detrimental to species of high
functional and conservation value (Barlow et al., 2016). Forest man-
agement is increasingly becoming the main strategy that can help
conserve or restore forest biodiversity and ecosystem services. However,
an increase in logging intensity and unsustainable harvesting practices
can affect the growth and survival of tree species at different points in
their life cycle (Paillet et al., 2010; Langridge et al., 2023). For instance,
logging causes canopy opening leading to changes in microclimatic
conditions, which in turn impact species regeneration potential (Wagner
et al., 2011). Pioneer tree species are especially prone to enhance their
recruitment and abundance in response to disturbances (Goodale et al.,
2012). Therefore, high degrees of disturbance resulting from increased
* Corresponding author.
E-mail address: fatimata.niang@uqat.ca (N. Fatimata).
Contents lists available at ScienceDirect
Trees, Forests and People
journal homepage: www.sciencedirect.com/journal/trees-forests-and-people
https://doi.org/10.1016/j.tfp.2024.100602
Trees, Forests and People 16 (2024) 100602
2
logging intensity are likely to select species with functional traits related
to efcient colonisation and survival strategies. The study of species
functional traits can help to predict the strategies employed by species
following disturbance as well as understanding trait-disturbance re-
lationships (Chave et al., 2009; Violle et al., 2007; Díaz et al., 2007;
Lavorel and Garnier, 2002). This is especially relevant in African sa-
vannahs, where regional trends between functional traits and the envi-
ronment do not mirror global trends and must be assessed within the
continent (Wigley et al., 2016, Bouchard et al.,2024).
Traits pertaining to the regeneration of woody species can help us to
understand the links between forest management practices and changes
in biodiversity (Tinya et al., 2023). These traits include seed
morphology, mode of dispersal and regeneration mode. Seed
morphology and dispersal mechanisms are important features that in-
uence recolonization after disturbance (Markl et al., 2012). For
instance, disturbances can affect the seed dispersal process by causing
loss or decline of animal dispersers, which may lead to the local
extinction of plant species (Wunderle, 1997; Rogers et al., 2021).
Additionally, seed dormancy, a physiological trait, remains a survival
strategy that allows seeds to control the timing of germination and to
persist over time (Yan and Chen, 2020). By detecting and responding to
environmental cues, dormant seeds can avoid germination under
unfavourable conditions, such as periods of water limitation in savan-
nahs. In contrast, the competitive ability of certain fruits like berries that
are not dormant is reduced in such an environment by their sensitivity to
desiccation.
Leaf and stem traits are also important for assessing trait-disturbance
relationships. Leaf area is coupled with photosynthetic activity and is
generally inuenced by a trade-off between resource acquisition and
conservation (Kazakou et al., 2016). For instance, acquisitive species in
savannahs tend to have larger leaves that increase uptake during the wet
season, as opposed to conservative species that exhibit smaller leaves
(Wigley et al., 2016). Moreover, tree growth is a key component of tree
demography that is reected through wood density (Wright et al., 2010;
Chave et al., 2006). There is a trade-off between growth rate and sur-
vival of a species modulated by resource availability. Species tend to
grow rapidly (lower wood density) when resources are abundant and
prioritize longevity over growth rate (higher wood density) when they
are scarce (Gonz´
alez-Melo et al., 2024). Species with high wood density
are more tolerant of competition and have a greater long-term
competitive capacity (Kunstler et al., 2016). Moreover, Keddy (2005)
suggests that competitive abilities for light can be predicted by the
height of plants with taller plants being more efcient in reaching light.
Therefore, species life forms, such as phanerophytes (Raunkier, 1934)
and species maximum height can provide insights on local and temporal
adaptation to environmental conditions.
In Senegal, despite conservation and management efforts, signicant
changes in forest structure and composition have been documented over
the past two decades (Solly et al., 2022; Niang et al., 2024). The use of
numerous tree species for domestic energy, food, and traditional medi-
cine, resulted in extensive logging (Ba et al., 2006; Sambou, 2004).
Primary forests have been gradually replaced by secondary forests due
to recurrent anthropogenic disturbances, leading to major shifts in
species composition and in vegetation structure (Faye et al., 2016;
Tappan et al., 2004). Sambou (2004) found that forest structure in sa-
vannahs is characterised by either a low abundance of adults or a lack of
juvenile trees (commonly saplings) under human disturbance. While
many factors can limit tree regeneration in Sudanian savannahs (Zida
et al., 2007; Bellefontaine, 1997), it is critical to maintain sufcient
natural regeneration, i.e future generations of canopy adults, for sus-
tainable forest management (Garcia_Florez et al., 2017). Saplings are an
important component of the forest structure that determines the resil-
ience and future dynamics of the ecosystem being managed (Frelich
et al., 2018).
The Sudanian bioclimatic zone is one of the largest biodiversity hot
spots in Senegal (USAID, 2008). This zone is characterized by a mosaic
of savannah types (wooded savannah, treed savannah, shrub savannah,
grassy savannah) and is highly susceptible to human disturbances. Bel-
lefontaine (2005) suggested that in Sudanian savannahs, acquisitive
pioneer species with high reproduction and growth rates are more likely
to thrive following disturbances than conservative late-successional
species. Moreover, several studies have demonstrated the prevalence
of vegetative regeneration strategies in various species inhabiting this
bioclimatic zone (Bationo, 2001; Bellefontaine, 2005; Niang Diop et al.,
2011; Faye et al., 2013). Disturbances may cause alterations in func-
tional diversity but not necessarily in taxonomic diversity, or vice versa
(Mayeld et al., 2010). However, research in the Sudanian region of
Senegal has primarily examined the impact of forest management on
taxonomic diversity (Niang et al., 2024). Therefore, there is a lack of
knowledge regarding the impact of human disturbances on the func-
tional composition of savannahs in the sudanian region, which is highly
sensitive to environmental changes.
This study aims to assess the probability of presence of high-value
species saplings in Senegal’s Sudanian forests based on their func-
tional traits and the type of forest management. High value species are
dened as the woody species most commonly harvested for energy
(fuelwood), timber, as well as non-timber forest products like fruits,
leaves, roots and barks (Sambou, 2004; Diop et al., 2011; Dieng et al.,
2016). These species have been intensively exploited in the past, leading
to a reduction in their abundance (Niang et al., 2024), hence the ne-
cessity to assess their regeneration potential to ensure sustainable forest
management. This study is the rst to address this issue at the scale of
the entire Sudanian zone in Senegal, and one of a few to evaluate tree
functional traits in the country (Dendoncker et al., 2023; Di´
em´
e et al.,
2018; Wood, 2018; Marone, 2015).
We asked whether the presence of high-value species would vary
across four different forest management types and functional traits
associated with survival and reproduction strategies as well as species
main domestic and commercial uses. We expected that an increase in
forest management intensity would promote species with disturbance-
adapted traits, such as (i) vegetative reproduction, because cutting of
mature trees and non-sustainable fruit harvesting decrease the likeli-
hood of successful sexual reproduction; (ii) high maximum height and
low wood density, due to a competitive advantage under favourable
light conditions generated by tree felling; (iii) low leaf area, because
canopy openness might increase stomatal transpiration which affects
leaf area negatively (Sang et al., 2008). Finally, we predicted a negative
relationship between species occurrence probability and their domestic
and commercial uses, because illegal logging and the selective nature of
forest management practices may lead to intensied harvesting pressure
on certain key species and functional traits that meet the demand from
local populations (PROGEDE, 2008).
2. Materials and methods
2.1. Study area
The study area encompasses twenty forests of the Sudanian zone,
which lies between 12◦20
′
and 15◦10
′
north latitude and between 11◦20
′
and 16◦75
′
west longitude (Fig. 1). The climate is characterised by two
different seasons: the dry season, which lasts from November to May and
the rainy season, which lasts from June to October. Average annual
rainfall is estimated at 900 mm ±200 mm, while the average annual
temperature is 28 ◦C (ANACIM, 2020). The landscape of the area is
characterised by different vegetation types, the main one being the
savannah. Savannahs are grassy formations made up of a perennial
gramineous mat interspersed with trees and shrubs of varying density
(White, 1983). Soils are predominantly tropical ferruginous, leached
and slightly leached (Khouma, 2002).
N. Fatimata et al.
Trees, Forests and People 16 (2024) 100602
3
2.2. Forest management and sites
Four forest categories based on their management status were in-
ventoried. Formal forest management started in Senegal in 2004
(PROGEDE, 2008) and is based on coppice with standards for 8 years
coppicing rotation over harvesting period of 25 years (R´
epublique,
2018). Technically, a managed forest is divided into a specic number of
plots that are harvested on an annual rotation. For instance, dividing the
forest into eight separate plots would allow each plot to regenerate for
eight years before being cut again (Boutinot, 2014). Tree species are
targeted based on their use for energy wood and for lumber products and
services (further referred to as main use), and up to 50 % of trees with a
minimum diameter of 10 cm are removed during logging (R´
epublique,
2018).
Unmanaged forests (UMF), are ofcially protected against exploita-
tion but they are not primary forests since they are prone to clandestine
human intervention. These UMFs are more or less equivalent to ‘no
longer managed forests’ (Langridge et al., 2023). Older managed forests
(OMF) were managed at least ten years ago, whereas newly managed
forests (NMF) have been managed less than ten years ago. Finally,
community reserve forests (CRF) represent forest stands that have been
heavily disturbed in the past and are continuously managed by local
communities, unlike the UMF, OMF and NMF, which remain under the
control of the national forestry administration. In total, twenty forest
sites from 5 UMF, 4 OMF, 7 NMF and 4 CRF (Fig. 1) were included in the
study.
2.3. Disturbance levels
Forests in the region have been shaped by complex interactions be-
tween human activities and climate across multiple temporal and spatial
scales. Managed forests may be subject to various types, durations, and
intensities of human activities, while unmanaged forests experience
fewer, but irregular, human activities. To account for the undocumented
disturbances that occur in all forests, regardless of management status,
we also dened a level of disturbance based on the percent cover of
vegetation in each forest type (Table 1) measured on Landsat-ETM sat-
ellite images and eld data following Sambou (2004). Field data were
used to complement and validate Landsat images results. The distur-
bance level is dened as low when forest cover is at least 80 %, medium
if it lays between 50 % and 80 %, and high if forest cover is less than 50
% (Niang et al., 2024; Staver et al., 2011). Using this combined
approach, we were able to analyse both the formal administrative sys-
tem but also evaluate whether it represents what is found in reality on
the sites (Niang et al., 2024).
2.4. Forests inventories
Within the study area, a total of 12,832 sub-plots, covering an area of
5114 km
2
s, were employed to document tree species across twenty
forest sites (Fig. 1). The oristic survey is based on a stratied random
sampling technique (Sambou, 2004). In each site, 2017 Landsat images
were utilized to identify the different land-use types. For vegetation
land-use type, 250 metre grid cells were superimposed. A total of 401
mesh grids were randomly selected for each forest class, with the
number of grids allocated for each forest proportional to its surface area.
As part of the grid layout, 8 plots with dimensions of 20 m were posi-
tioned along their median and diagonal axes. Each plot was subdivided
into 4 sub-plots, measuring 5 m x 5 m, and located at the four corners to
count saplings, which are dened as individuals with a diameter at
breast height of less than 5 cm (Sambou, 2004).
Among the woody species, twenty species were selected as those
most commonly harvested for energy, timber and service wood, as well
as for non-timber forest product collection and they were tagged as high-
value species (Sambou, 2004; Diop et al., 2011; Dieng et al., 2016).
These species were Adansonia digitata L., Acacia macrostachya Reichenb.,
Afzelia africana Sm. ex Pers., Anogeissus leiocarpus (DC.) Guill. et Perrot.,
Bombax costatum Pellegr. & Vuillet, Combretum glutinosum Perr.Ex DC,
Combretum micranthum G. Don, Combretum nigricans Lepr., Cordyla pin-
nata (Lepr. ex A.Rich.) Milne-Redh., Daniellia oliveri (Rolfe) Hutch &
Dalz, Detarium microcarpum Guill. & Perr., Detarium senegalense J.F.
Gmel., Khaya senegalensis (desr.) A. Juss., Lannea acida A. Richard.,
Neocarya macrophylla (Sabine) Prance, Parkia biglobosa (Jacq.) G.Don,
Prosopis africana (G. et Perr.), Pterocarpus erinaceus Poir., Sterculia seti-
gera Del. and Terminalia avicennioides Guill and Perr..
2.5. Functional traits
Functional traits were selected based on their signicance in repro-
duction and survival strategies in the Sudanian region, as well as their
data availability (Table 2; Dawson et al., 2021). These functional traits
Fig. 1. Location of the different forests, color-coded based on their manage-
ment status (UMF =unmanaged forests, OMF =older managed forests >10
years, NMF =newly managed forests <10 years, CRF =community
reserve forests.
Table 1
Total area, forest cover area and disturbance level for the 20 forests sampled as
part of this study, grouped by management status (Source: PROGEDE, 2016;
Sambou, 2004).
Forests by
management status
Total
area (ha)
Forest
cover (ha)
Percent
forest cover (
%)
Disturbance
levels
Unmanaged forests (UMF)
Welor 5469 5348 97.8 Low
Bala 20,891 19,220 92 Low
Patako 5638 5559 98.6 Low
Mampaye 8286 8226 100 Low
Kantora 22,314 21,645 97 Low
Older managed
forest (OMF)
Kandiator 54 179 39 635 73.16 Medium
N´
et´
eboulou 10 989 5 261 47.88 High
Koar 70,227 63,388 90.26 Low
Thi´
ewal 31 082 18 271 58.78 Medium
Newly managed
forests (NMF)
Mousdalifa 19 307 8 899 46.09 High
Mal`
eme Delby 12 125 4 575 37.73 High
Boussimballo 27 500 20 174 73.36 Medium
Mballocounda 26 518 20 203 76.19 Medium
Maka 54 174 33 603 62.03 Medium
Ouli 12 529 11 513 91.89 Low
Niani saloum 41,630 17,791 42.74 High
Community reserve
forests (CRF)
Ribot Escale 22 948 8 675 37.80 High
Guimara 52 596 48 014 91.29 Low
Mbirkilane 6 882 5 689 82.66 Low
Tomboronkoto 31 774 30 874 97.17 Low
N. Fatimata et al.
Trees, Forests and People 16 (2024) 100602
4
were documented for each of the twenty high value species. They
included leaf area, wood density, maximum stem height, fruit type,
dispersal mode, seed dormancy and regeneration mode. Functional
traits were gathered from the TRY database version 4, which is a global
data repository for plant functional traits published in July 2021 (Kattge
et al., 2020). Literature and other databases were reviewed to ll blanks
in the TRY database. These databases included the Conservatoire et
Jardin Botaniques de Gen`
eve (African Plant Database, 2022), Global
Wood Density Database (Zanne et al., 2009) and The World Agroforestry
Centre, as well as primary and secondary publications, and key resource
persons in the study region. For quantitative traits with multiple pub-
lished values, the mean value was used (Lamanna et al., 2014), while for
qualitative traits, the measurement from the closest location to the study
area was selected. If an estimate of the trait value for a given species was
not available, mean values from the same genus in the study area was
used, following previous studies (Paine et al., 2015; Lamanna et al.,
2014). Additionally, main domestic and commercial use of tree species
like fuelwood (FW), lumber and service timber (LST) and non-timber
forest products (NTFPs) were used as factors to reect the provision-
ing services from forests (Arbonnier, 2004; Sambou, 2004).
2.6. Data analysis
To determine whether functional trait values inuenced the presence
or absence of a species in a forest, we used mixed effects logistic
regression models with a binomial distribution of the response variable
(i.e. the number of plots where the saplings of high-value species was
present and absent in a given forest). The xed effects in the model
included the trait value, forest management status and disturbance level,
whereas the species identity was included as a random effect to account
for interspecic variation not captured by functional traits. The sites
were initially included as a random effect, but were removed in subse-
quent analyses because its estimated effect was 0. To determine whether
the effect of a trait on the sapling’ presence varied according to forest
management status, we specied an interaction between each trait and
management status, and used a likelihood-ratio test to test for the sig-
nicance (
α
=0.05) of this interaction. The same test was performed for
species main uses and also for the interaction between traits and forest
disturbance level. We also calculated the partial R
2
with the rsq. partial
package in R (Zhang, 2017) as a measure of the fraction of differences in
the response variables explained by each xed effect. For the main uses
and qualitative traits (life form, fruit type, seed dispersal mode, seed
dormancy, regeneration mode, vegetative reproduction ability, capacity
of resprouting), we used the emmeans package in R (Lenth, 2020) to
estimate the probability of presence of a species given main use and a
specic trait value and its 95 % condence interval for each forest. Based
on Tukey’s adjustment for multiple comparisons with a signicance
level of
α
=0.05, we determined for different forest statuses and each
trait value whether the differences between those probabilities are sig-
nicant or not. For the continuous quantitative traits (wood density,
mean leaf area and maximum height), it was more appropriate to esti-
mate the probability of presence or absence along gradients in functional
trait values rather than comparing those probabilities between discrete
levels of the trait. Predictions from the mixed logistic regression models
were used to explain differences in species presence odds across
different management statuses and disturbance levels. These predictions
show the mean linear trend as a function of the intercepts and the slope
(coefcient) estimated for the xed effects (forest management status,
functional trait and their interaction) and the random effect (species
identity). Parametric bootstrapping as implemented in the bootMer
function of the lmer package was used to get condence intervals for the
predicted trend lines.
3. Results
Management status and disturbance level had a signicant (P <
0.001) effect on probability of saplings presence and both explained 1 %
of the variation in the average species probability of presence when
considering the probability of presence of any species irrespective of its
traits. Note that a higher average probability of presence across all
saplings can be interpreted as a greater expected richness of those high-
value species at the plot level. The difference of probability of presence
in UMF compared to all types of managed forests is statistically higher,
while among the managed forests, species overall have a signicantly
higher probability of presence in OMF than in NMF and CRF, but the
difference between NMF and CRF is not signicant. When considering
disturbance levels, the average species presence is the highest in forests
with a medium disturbance level and the lowest in highly-disturbed
forests, with signicant differences between all three levels.
Across all forests, we found that traits related to regeneration mode,
vegetative reproductive ability and plant heights had signicant effects
when considering the probability of presence of any species irrespective
of forest types (Table 3). This result means that these traits inuence the
presence of saplings in different forest types regardless of their man-
agement status and disturbance level. Apart from species main use,
which also signicantly inuenced the probability of presence, we will
thus focus only on these traits whose additive effect is signicant, all of
which also have signicant interactions with both forest management
status and disturbance level (Table 3).
For the regeneration mode trait, the basic model (no interactions)
shows that species with sexual reproduction are less likely to be present
in all forest types than those with vegetative reproduction mode (odds
ratio of 0.06, with 95 % CI of 0.01 - 0.17). Species with either one of
these regeneration modes are signicatively more present in UMF than
in all three managed forest management statuses, but those that repro-
duce vegetatively are also signicatively more present in OMF and CRF
compared to NMF (Fig. 2a). Both regeneration modes have an increased
probability of presence in forests with a medium disturbance level
compared to high and low disturbance levels. However, it is only for
species reproducing vegetatively that the probability of presence is
signicantly less in highly-disturbed forests compared with those with
low levels of disturbance; the difference between medium and high
disturbance levels is also greater for those species. (Fig. 2b). For all these
models including regeneration mode, the R
2
is 0.08 for the xed effects
and the species identity as random effect has the greatest percentage of
variation (Table 3). The low value of xed effects is due to the fact that
there is a large part of the variability in the probability of presence that is
Table 2
Functional trait names and attributes and species main uses.
Species functional traits Traits attributes
Life form Megaphanerophyte
Mesophanerophyte
Microphanerophyte
Fruit type Dried
Fleshy
Seed dispersal mode Anemochory
Autochory
Zoochory
Seed dormancy No
Yes
Regeneration mode Vegetative
Sexual reproduction
Vegetative reproduction ability High
Moderate
Low
Capacity of resprouting No
Yes
Leaf area (mm
2
) -
Maximum height (m) -
Wood density (mg/mm
3
) -
Main uses* Fuelwood (FW)
Lumber and service timber (LST)
Non timber forest products (NTFPs)
*
Not a functional trait
N. Fatimata et al.
Trees, Forests and People 16 (2024) 100602
5
attributed to differences between species beyond the traits included in
the models.
Regarding the trait vegetative reproduction ability, we found that
the probability of presence of a species (across all forest management
statuses) increases from low to medium and from medium to high
vegetative reproduction ability, but only the difference between low and
high is signicant (Fig. 3a). The interactions of this trait with forest
management status and disturbance level are also signicant, with both
models having a R
2
of 0.08 (Table 3). For all values of the trait, the
probability of presence is signicantly greater in UMF compared with all
other management statuses. For species with high and moderate vege-
tative regeneration ability, the probability of presence is also signi-
cantly greater in OMF and CRF compared with NMF (Fig. 3b). Among
disturbance levels (Fig. 3c), species with moderate vegetative regener-
ation ability have the highest probability of presence in forests with
medium disturbance level and the lowest probability of presence in
forests with a high disturbance level; for species with low vegetative
regeneration ability, highly-disturbed forests show higher odds of
presence than those with low disturbance, but still less than those with
medium disturbance. Finally, there are no signicant differences
between disturbance levels for species with high vegetative regeneration
ability.
The interactions of plant maximum height with forest management
status and disturbance level are both signicant, with both models
having a R
2
of 0.07 (Table 3). For all types of forests, the probability of
presence decreases as the maximum height of a species increases.
However, the slope of this relationship becomes greater (so species of
high maximum height become relatively rarer) when going from UMF to
OMF and CRF, to NMF (Fig. 4a), and going from low to high disturbance
levels (Fig. 4b).
For species main uses, we found a signicant interaction between
main use and both management status and disturbance level, with an R
2
of 0.10 in both models (Table 3). In the basic model, species that are
used for NTFPs have a signicantly higher probability of presence than
those used for fuelwood FW (Fig. 5a). For all main uses, the species are
signicantly more present in UMF than in OMF, NMF and CRF; for
lumber and service timber and NTFPs species, there is a signicantly
higher probability of presence in CRF than NMF; and for lumber and
service timber only, there is a signicantly higher probability of pres-
ence in OMF than NMF (Fig. 5b). Species used for NTFPs are more
Table 3
Estimated p-value, xed and random effect R
2
from logistic mixed models to predict the presence of saplings of high-value species, as a function of trait, management
status and disturbance level (rst column), and interaction between the trait and management status (2nd column) or disturbance level (3rd column). The p-values
indicate signicance (signicative values are in bold) of a likelihood-ratio test comparing the model to one without any trait effect (for the 1st column) or with only the
additive trait effect (for the others).
Traits Additive effect of trait Trait* Management status interaction Trait* Disturbance level interaction
P Fixed
R
2
Random
R
2
P FixedR
2
Random R
2
P FixedR
2
Random R
2
Life form 0.269 0.03 36.9 <0.001 0.03 36.9 <0.001 0.04 36.9
Fruit type 0.356 0.02 38.3 <0.001 0.02 38.4 <0.001 0.34 38.3
Seed dispersal mode 0.414 0.04 36.1 <0.001 0.04 36.3 <0.001 0.04 36.4
Seed dormancy 0.745 0.05 32.0 <0.001 0.05 32.0 <0.001 0.05 32.0
Main regeneration mode 0.008 0.08 32.0 0.028 0.08 32.0 0.002 0.08 32.0
Vegetative reproduction ability 0.042 0.07 32.7 <0.001 0.08 32.8 <0.001 0.08 32.9
Leaf area 0.879 0.01 40.3 0.003 0.01 40.3 <0.001 0.01 40.5
Wood density 0.098 0.09 31.6 <0.001 0.09 31.2 <0.001 0.09 30.9
Plant height 0.000 0.07 32.9 <0.001 0.07 33.2 <0.001 0.07 33.3
Main uses* 0.029 0.10 30.4 <0.001 0.10 30.5 <0.001 0.10 30.5
*
Not a functional trait
Fig. 2. Differences in the mean probability of presence (measured as log odds ratio) of saplings of high-value species, estimated from linear mixed models with a
binomial distribution (species presence/absence in plots) of species regeneration mode as a function of (a) management status (a : UMF =unmanaged forests, OMF =
older managed forests NMF =newly managed forests CRF =community reserves forests) and (b) disturbance level (high, medium and low) with species identity as a
random effect. The Tukey-adjusted 95 % condence intervals (in blue) indicate signicant differences if they do not include 0.
N. Fatimata et al.
Trees, Forests and People 16 (2024) 100602
6
present in forests with a medium disturbance level, followed by forests
with a low and a high disturbance level, with all differences being sig-
nicant. However, the presence of species used for FW is not signi-
cantly different between forests with high and low disturbance levels
and the presence of species used for lumber and service timber is not
signicantly different between forests with medium and low disturbance
levels (Fig. 5c).
4. Discussion
We investigated whether functional traits, species main commercial
and domestic uses, forest management status, disturbance intensities
and their interactions predict the occurrence of saplings of high-value
tree species in Sudanian forests. We observed general trends across
forest management statuses and disturbance intensities that act inde-
pendently of functional traits and main uses. High-value species are
more likely to be present in unmanaged forests compared to managed
ones, whilst among managed forests, older managed forests showed a
greater probability of presence. This difference might arise from a re-
covery over time of high-value species in older managed forest relative
to newly managed forests (Niang et al. 2024) and a preservation of more
valuable species under reduced disturbance regimes of unmanaged
forests. Species identity explained a high proportion of the variation in
response to probability of saplings presence. This nding suggests
various possible mechanisms and factors underlying forest dynamics
following management that can operate at species levels and may vary
over time (Falk et al., 2022). Indeed, population dynamics processes
such as tree regeneration might be determined notably by stand
composition and age, the level, severity and type of disturbance as well
as the timing of disturbance (Dietze and Clark, 2008) with effects
varying from species to species. Among the studied traits, those related
to regeneration (mode and vegetative abilities) and competition abilities
(maximum height) were the most sensitive to forest disturbance and
management intensities. Unmanaged forests hosted the greatest
Fig. 3. Differences in the mean probability of presence (measured as log odds ratio) of saplings of high-value species, estimated from linear mixed models with a
binomial distribution (species presence/absence in plots) of (a) species vegetative regeneration ability (VRA) trait attributes only, (b) as a function of management
status (a : UMF =unmanaged forests, OMF =older managed forests, NMF =newly managed forests, CRF =community reserves forests) and (c) as a function of
disturbance level (high, medium and low) with species identity as a random effect. The Tukey-adjusted 95 % condence intervals (in blue) indicate signicant
differences if they do not include 0.
N. Fatimata et al.
Trees, Forests and People 16 (2024) 100602
7
diversity of these traits, conrming that selective logging might
constraint the regeneration and survival strategies deployed by species
in managed forests. The main commercial and domestic uses of species
also explained their prevalence in Sudanian savannahs.
4.1. Vegetative regeneration traits are prominent in predicting species
presence
We found that management status and disturbance level mainly
inuenced species traits linked to vegetative regeneration. This result is
in line with previous studies which indicated that the disturbance lter
acts mainly on regeneration traits (Garnier et al, 2016). At the species
level, one crucial mechanism to regenerate is the ability to establish new
seedlings or to regrow vegetative sprouts from stumps or roots (Hoff-
mann et al., 2000). Our ndings demonstrated that species that repro-
duce sexually are less prevalent than species that have vegetative
reproduction in any type of forest, irrespective of their management
status and level of disturbance. This low probability of establishment for
species that reproduce sexually may be due to selective cutting of
mature trees, fruit harvesting and their transportation outside their
natural habitat. Species that regenerate using both modes were signi-
cantly more prevalent in unmanaged forests, whereas those regenerat-
ing only by vegetative means were found in both unmanaged and
managed forests. In sudanian savannahs, vegetative regeneration is
common amongst the woody species (Bellefontaine, 2005). For instance,
ve of our twenty studied species (Anogeissus leiocarpus, Combretum
glutinosum, Combretum micranthum, Combretum nigricans and Terminalia)
belong to the Combretaceae. This botanical family dominates most West
African savannahs due to high vegetative reproduction abilities,
fast-growing capacity and adaptability to drier climatic conditions
(Osborne et al., 2018; N´
ek´
e et al. (2006).
Other factors such as drought stress, seed and seedling predation as
well as re are key drivers that promote vegetative regeneration (Bel-
lefontaine, 2005; Zida et al., 2007; Pare et al., 2009). Fire and canopy
openness following tree cutting are important factors in increasing
abundance of herbaceous plants in savannahs (Gold et al., 2023; Vadigi
and Ward, 2013). These herbaceous plants compete with tree saplings
for light, water and nutrients (Vadigi and Ward, 2013), which can limit
tree establishment. In addition, water scarcity is a major constraint in
savannahs so vegetative sprouts have a greater chance of survival than
new seedlings because they can grow faster with their established root
system (Chidumayo and Ellegaard, 1993). Thus, the lack of signicant
relationships between forest management status or disturbance level
and other traits related to seed germination such as dispersal mode, seed
dormancy and fruit type is not surprising, since seed germination in this
area is also constrained by abiotic factors such as climate as described
previously (Ou´
edraogo and Thiombiano, 2012). These harsh climatic
conditions could lead to the selection of tree species that reproduce
vegetatively, which is consistent with our results.
Many common species of sudanian savannahs are known to regen-
erate vegetatively following the removal of their aboveground parts by
tree logging and other disturbances such as drought, grazing or re
(Bellefontaine, 1997; Luoga et al., 2004; Ky-Demb´
el´
e et al., 2007;
Niang-Diop et al, 2011). However, among the species that are able to
reproduce vegetatively, the level of efciency of vegetative reproduc-
tion showed mixed results in explaining species presence within suda-
nian forests. Unmanaged forests host the highest probabilities of
presence of high-value species for any level of vegetative reproduction
ability (low, moderate or high). Across forest management statuses,
species with moderate vegetative abilities were the most likely to be
present, possibly because this vegetative reproductive ability is more
versatile and efcient across a wider range of environments. Our nd-
ings indicated that for any disturbance level, the probability presence of
species with high vegetative reproduction ability was less prevalent than
species with medium and low vegetative reproduction abilities. We
would expect higher vegetative abilities to be more prevalent in highly
disturbed forests because it is a known disturbance-adapted trait.
However, this hypothesis did not hold true in this study potentially
because we were not able to measure the real level of disturbance by
using the forest cover as a proxy of disturbance levels. A full forest cover
does not necessarily equal an absence of disturbance because it does not
take into account disturbance that might occur under cover as well as
the intensity of disruption at local scales (Potapov et al., 2011). Field
sampling to complement the vegetative cover indicator could provide
further insights on the impact of disturbance levels on functional traits.
Vegetative reproduction ability is typical of most sudanian tree species
(Bellefontaine, 2005) and its presence might also depend on forest initial
conditions (Dietze and Clark, 2008).
Fig. 4. Predicted mean probability presence of saplings of high-value species and its 95 % condence interval obtained by parametrized bootstrapping, from a
generalized linear mixed model with a binomial distribution of the response (presence/absence), (a) as a function of species maximum height and forest management
status (a : UMF =unmanaged forests, OMF =older managed forests, NMF =newly managed forests, CRF =community reserves forests) and (b) as a function of
disturbance level (high, medium and low) with species identity as random effect.
N. Fatimata et al.
Trees, Forests and People 16 (2024) 100602
8
4.2. Trend towards smaller trees
The maximum height of trees is a trait linked to growth and
competitive abilities that can help understand species performance
following disturbances (Rüger et al., 2012). We found that the proba-
bility of species presence decreased with an increase in species height for
all forest management statuses. This trend is more pronounced under
higher management and disturbance intensities. One possible explana-
tion for this negative relationship is related to the competitive ability of
species in disturbed environments, where pioneer species experience
rapid growth but fail to attain the same height as more high-value
species, which grow more slowly (DeArmond et al., 2022) as found in
unmanaged forests with taller trees. Another explanation could be
linked with frequent disturbances that might constrain taller species
from reaching their potential maximum height, which in turn can reduce
their competitive ability and lead to their displacement by smaller trees
or shrubs that are more competitive at smaller heights. Indeed, repeated
logging may transform single-stemmed trees into multiple-stems and
species with more than one stem tend to be shorter than those with one
stem (Shibata et al., 2016).
The processes involved in tree growth are known to be complex and
sensitive to environmental conditions (Gorgens et al., 2021). Our nd-
ings are consistent with changes in tree growth allocation to height due
to interactions with biotic and abiotic factors (Poorter et al., 2012). For
instance, late successional species allocate more resources for repro-
duction and survival, resulting in denser wood and larger fruits (Poorter
et al., 2004; Lohbeck et al., 2013). These characteristics are highly
desirable for human populations interested in these high-value species
that are frequently harvested. In this Sudanian bioclimatic zone, where
water availability is limiting, species may allocate fewer resources to
above-ground biomass and more to roots, resulting in increased water
uptake and a decrease in height (Chaves et al., 2003). The increasing
prevalence of small tree species implies that, in the long term, forest
management could be one of the drivers in the vegetation physiognomy
of the Sudanian zone like agriculture and re (Tappan et al., 2004;
Mbow et al., 2008; Sambou et al., 2015).
Fig. 5. Differences in the mean probability of presence (measured as log odds ratio) of saplings of high-value species estimated from linear mixed models with a
binomial distribution (species presence/absence in plots) of (a) species main uses attributes only (FW =Fuel Wood, LST =Lumber Service and Timber, NTFP =Nom
Timber Forest Products), (b) as a function of management status (UMF =unmanaged forests, OMF =older managed forests, NMF =newly managed forests, CRF =
community reserves forests) and (c) as a function of disturbance level (high, medium and low) with species identity as a random effect. The Tukey-adjusted 95 %
condence intervals (in blue) indicate signicant differences if they do not include 0.
N. Fatimata et al.
Trees, Forests and People 16 (2024) 100602
9
4.3. Seed, leaf and wood traits
Our results indicated that fruit type, seed dormancy and seed
dispersal mode do not inuence the probability presence of high-value
species. This lack of signicant contribution could be explained by the
low occurrence of sexual reproduction in nearly all forests, regardless of
their status. For mean leaf area and wood density, the weak effect can be
attributed to climate. There is limited water availability in Sudanian
savannahs, and the variability of these traits may be more closely linked
to abiotic lters such as water availability than biotic factors like human
disturbances. Indeed, the high median values of wood density (0.7 g/
cm
3
) encountered in our sites along with the low median of leaf area
(29.9 cm
2
) are both consistent with drought survival strategies (Bou-
chard et al., 2024; Wright et al., 2017), conrming that these functional
traits might be more limited by abiotic than anthropogenic disturbances
in Sudanian savannahs. Predicting tree species presence is not limited to
human disturbances, it is also relevant to consider climate projections
when foreseeing tree responses in West Africa. Moreover, some studied
traits may not fully capture specic plant functions which possibly
explain the low predictive power of these traits on species probability of
presence (Gonz´
alez-Melo et al., 2024). This is particularly true in West
Africa, where data scarcity makes predictive studies based on species
functional traits challenging.
4.4. Beyond functional traits; trends in forest products
Forests play a vital role in conserving biodiversity as well as in
providing ecosystem services like timber and non-timber forest products
(Lindenmayer, 2009; Brockerhoff et al., 2017). However, disturbances
have the potential to alter ecosystem services and functions due to
biomass loss (Seidl et al., 2016) and alterations in biodiversity (Tilman
et al., 1997; Brockerhoff et al., 2017) whereas species use may inuence
their abundance. Our ndings demonstrate a signicant relationship
between species probability of presence, their main use and forest
management status, as well as forest disturbance level. Unmanaged
forests are more likely to house high-value species, all uses considered.
Species that are used for non-timber forest products (NTFP) were
prevalent in all forests indicating that the harvesting of timber is more
intensive than that of non-timber products. This nding reects the
community´s need for fuelwood. Furthermore, the forests with the least
NTFP are newly managed forests and community forest reserves, with
newly managed forests being the worst. The demand for NTFP might be
more recent and closely related to local populations, whereas older
management focus mainly on long-lasting uses such as timber and
fuelwood products.
We also found that old managed forests host a variety of species that
are used for lumber service and timber. This result is in line with a
previous study that found a change in oristic composition of these high-
value species in managed forests (Niang et al. 2024). Competition for
access to forest products can lead to unsustainable harvesting practices
which increase trees loss (Gandon, 2003; Schumann et al., 2011). Na-
tional legislation includes the principle of sustainable provision of goods
and services from forests (R´
epublique, 2018). However, this principle is
often not implemented due to competition between local populations
and the forest administration for forest products, leading to unregulated
exploitation (Diop et al., 2011). Illegal logging, added to authorized
forest management, accelerates the loss of high-value species, which in
turn can alter the provision of ecosystem services. The loss of provi-
sioning services can lead to a change in regulating services as well as
cultural services (Seidl et al., 2016). Sustainable use of forests in West
Africa is very challenging due to the high dependence of local pop-
ulations on wood and non-timber forest products for commercial and
domestic uses such as food, medicine, and lumber (Arbonnier, 2004;
Lykke et al., 2004; Diop et al., 2011; Leßmeister et al., 2016). Never-
theless, legislation appear to be benecial since our results showed that
more uses and high value species are found in unmanaged forests.
5. Conclusion
This study highlighted that in the Sudanian zone, species functional
traits interact with forest management in various ways. The complexity
of these interactions makes it difcult to predict specic effects of forest
management, but we observed recurring patterns. Species that can
regenerate through vegetative means have an advantage over species
that are dependent on sexual reproduction. Due to the high prevalence
of vegetative regeneration and the limitation of the maximum potential
height, forest management appears to promote conservative strategies in
high-value species. The prevalence of species with low potential height
in managed forests suggest that, in the long term, forest management
could be a key driver of vegetation structure in the Sudanian zone. The
probability of presence of high-value species was also inuenced by
their commercial and domestic use in all type of forests, demonstrating
that fuelwood and timber remains the most common use of woody
species for the populations of Senegal. Moreover, older management
promotes more diverse values of functional traits and higher abundance
of high-value species. Even if management can lead to changes in the
composition of species functional traits, it appears to be modulated
when considering temporal variation in the responses of trees species
following forest management. Therefore, protecting managed forests
over the long term might be essential to promote forest resilience
following management, since our results demonstrated a potential for
forest functional composition to restore over time. Overall, our results
showed that predicting which species are likely to regenerate after forest
management is critical for planning management strategies of forests
and the ecosystem services they provide. Forest management can be
optimised by promoting different strategies aligned with specic func-
tional traits rather than solely exploitation. Finally, the great variation
observed across species identity indicated that more studies are needed
to improve the understanding of the underlying mechanisms of the re-
sponses of tree species functional traits to forest management in
savannahs.
CRediT authorship contribution statement
Niang Fatimata: Conceptualization, Methodology, Investigation,
Formal analysis, Writing – original draft, Writing – review & editing.
Marchand Philippe: Conceptualization, Methodology, Formal analysis,
Writing – review & editing. Fenton Nicole J: Conceptualization,
Methodology, Validation, Writing – review & editing. Sambou Bien-
venu: Conceptualization, Methodology, Writing – review & editing.
Bouchard Elise: Validation, Writing – review & editing.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Data availability
The authors do not have permission to share data.
Acknowledgments
We thank Institut de Recherches sur les Forˆ
ets (IRF), Universit´
e du
Qu´
ebec en Abitibi T´
emiscamingue, UQAT (Canada), Fondation J.A.
DeS`
eve & Fondation de l’UQAT and Unit´
e de Recherches sur les
´
Ecosyst`
emes Naturels et l´Environnement (URENE), Institut des Sciences
de l´Environnement (ISE), Universit´
e Cheikh Anta Diop de Dakar, UCAD,
(S´
en´
egal). We are also grateful to TRY database for functional traits data
access and all people who participated in eld inventories as well as to
all our colleagues and friends for their constructive comments and we
N. Fatimata et al.
Trees, Forests and People 16 (2024) 100602
10
also thank the two anonymous reviewers for their valuable comments in
the earlier version of this manuscript.
References
African Plant Database (version 4.0.0). 2022. Conservatoire et Jardin botaniques de la
Ville de Gen`
eve and South African National Biodiversity Institute, Pretoria, "acc`
es
[mois, 2022]", http://africanplantdatabase.ch.
ANACIM, 2020. Services D´Informations m´
et´
eorologiques, Agence Nationale De
L’aviation Civile et De La M´
et´
eorologie (ANACIM). S´
en´
egal.
Arbonnier, M. 2004. Trees, shrubs and lianas of West African dry zones. Mus´
eum
National 419 d’Histoire Naturelle, CIRAD.
Ba, C., Bishop, J., Deme, M., Diadhiou, H.D., Dieng, A.B., 2006. The economic value of
wild resources in Senegal: a preliminary evaluation of non-timber forest products,
game and freshwater sheries. IUCN.
Bationo, B.A., Ouedraogo, S.J., Guinko, S., 2001. Strat´
egie de r´
eg´
en´
eration naturelle de
Detarium microcarpum Guill.& Perr. dans la forˆ
et class´
ee de Nazinon (Burkina Faso).
Fruits 56, 271–285.
Barlow, J., Lennox, G.D., Ferreira, J., Berenguer, E., Lees, A.C., Nally, R.M., Gardner, T.
A., 2016. Anthropogenic disturbance in tropical forests can double biodiversity loss
from deforestation. Nature 535 (7610), 144–147.
Bellefontaine, R., 1997. Synth`
ese des esp`
eces des domaines sah´
elien et soudanien qui se
multiplient naturellement par voie v´
eg´
etative, pp. 95-104. Fonctionnement Et
Gestion Des ´
ecosyst`
emes forestiers Contract´
es Sah´
eliens. Ed. John Libbey Eurotext,
Paris, p. 274. Niamey, nov. 1995 (J.M. d’Herb`
es, J.M.K. Ambouta, R. Peltier Eds.).
Bellefontaine, R., 2005. Sexual reproduction is not the only way for a lot of woody trees:
analysis of 875 cases Introduction, table and bibliography. S´
echeresse 16, 315–317.
Bouchard, E., Searle, E.B., Drapeau, P., Liang, J., Gamarra, J.G., Abegg, M., Paquette, A.,
2024. Global patterns and environmental drivers of forest functional composition.
Global Ecol. Biogeogr. 33, 303–324.
Boutinot, L., 2014. L’enjeu des am´
enagements forestiers au S´
en´
egal. Entre le hasard
´
ecologique et la n´
ecessit´
e politique. Anthropol. Develop. 157–184 (37-38-39).
Brockerhoff, E.G., Barbaro, L., Castagneyrol, B., Forrester, D.I., Gardiner, B., Gonz´
alez-
Olabarria, J.R., Jactel, H., 2017. Forest biodiversity, ecosystem functioning and the
provision of ecosystem services. Biodivers. Conserv. 26, 3005–3035.
Chave, J., Coomes, D., Jansen, S., Lewis, S.L., Swenson, N.G., Zanne, A.E., 2009. Towards
a worldwide wood economics spectrum. Ecol. Lett. 12 (4), 351–366.
Chave, J., Muller-landau, H.C., Baker, T.R., Easdale, T.A., Ter, S.H., Webb, C.O., 2006.
Regional and phylogenetic variation of wood density across 2456 neotropical tree
species. Ecolog. Applicat. 16 (6), 2356–2367.
Chaves, M.M., Maroco, J.P., Pereira, J.S., 2003. Understanding plant responses to
drought – from genes to the whole plant. Funct. Plant Biol. 30, 239–264 pg.
Chidumayo, E.N., & Ellegaard, A. 1993. Responses of miombo to harvesting: ecology and
management.
Dawson, S.K., Carmona, C.P., Gonz´
alez-Su´
arez, M., J¨
onsson, M., Chichorro, F., Mallen-
Cooper, M., Duthie, A.B., 2021. The traits of “trait ecologists”: An analysis of the use
of trait and functional trait terminology. Ecol. Evol. 11 (23), 16434–16445.
DeArmond, D., Ferraz, J.B., Marra, D.M., Amaral, M.R., Lima, A.J., Higuchi, N., 2022.
Logging intensity affects growth and lifespan trajectories for pioneer species in
Central Amazonia. For. Ecol. Manage 522, 120450.
Dendoncker, M., Taugourdeau, S., Messier, C., Vincke, C., 2023. A functional trait-based
approach to evaluate the resilience of key ecosystem functions of tropical savannas.
Forests. 14 (2), 291.
Díaz, S., Lavorel, S., de Bello, F., Quetier, F., Grigulis, K., Robson, T.M., 2007.
Incorporating plant functional diversity effects in ecosystem service assessments.
Proc. Natl. Acad. Sci. u S. a 104, 20684–20689.
Di´
em´
e, J.S., Diouf, M., Armas, C., Rusch, G.M., Pugnaire, F.I., 2018. Functional groups of
Sahelian trees in a semiarid agroforestry system of Senegal. J. Plant Ecol. 11 (3),
375–384.
Dieng, S.D., Diop, M., Goudiaby, A., Niang-Diop, F., Faye, L.C., Guiro, I., Sambou, S.,
Lykke, A.M., Sambou, B., 2016. Caract´
erisation des services ´
ecosyst´
emiques fournis
par Cordyla pinnata dans la p´
eriph´
erie de la Forˆ
et class´
ee de Patako au S´
en´
egal.
VertigO 16 (2), 1–17.
Dietze, M.C., Clark, J.S., 2008. Changing the gap dynamics paradigm: vegetative
regeneration control on forest response to disturbance. Ecol. Monogr. 78 (3),
331–347.
Diop, M., Sambou, B., Goudiaby, A., Guiro, I., Niang-Diop, F., 2011. Ressources v´
eg´
etales
et pr´
ef´
erences sociales en milieu rural s´
en´
egalais. Bois & Forets Des. Tropiques 310,
57–68.
Falk, D.A., van Mantgem, P.J., Keeley, J.E., Gregg, R.M., Guiterman, C.H., Tepley, A.J.,
Marshall, L.A., 2022. Mechanisms of forest resilience. For. Ecol. Manage 512,
120129.
Faye, E.H., Diallo, H., Samba, S., Tour´
e, M., Dram´
e, A., Fall, A.B.K., Lejoly, J., Diatta, M.,
Kaïr´
e, M., De Canni`
ere, C., Mahy, G., Bogaert, J., 2013. Importance de la m´
ethode de
coupe sur la r´
eg´
en´
eration de Combretaceae du Bassin arachidier s´
en´
egalais.
Tropicultura 31 (1), 44–52.
Faye, L.C., Sambou, H., Kyereh, B., Sambou, B., 2016. Land use and land cover change in
a community-managed forest in South-Eastern Senegal under a formal forest
management regime. Am. J. Environm. Protect. 5 (1), 1. https://doi.org/10.11648/j.
ajep.20160501.11.
Frelich, L.E., J˜
ogiste, K., Stanturf, J.A., Parro, K., Baders, E., 2018. Natural disturbances
and forest management: interacting patterns on the landscape. Ecosystem Services
from Forest Landscapes Broadscale Considerations, pp. 221–248.
Gandon, B., 2003. L’emondage Du Pterocarpus erinaceus (v`
ene) : ´
etude des Pratiques Et
De Leurs Impacts Sur L’arbre Sur 4 Terroirs Agro-Sylvo-Pastoraux Du S´
en´
egal
oriental, DES5 Gestion des Syst`
emes Agro-Sylvo-Pastoraux En Zone Tropicale.
Universit´
e Paris XII-Val-de-Marne, p. 57.
Garcia_Florez, L., Vanclay, J.K., Glencross, K., Nichols, J.D., 2017. Understanding 48
years of changes in tree diversity, dynamics and species responses since logging
disturbance in a subtropical rainforest. For. Ecol. Manag. 393, 29–39.
Garnier, E., Navas, M.L., Grigulis, K., 2016. Plant functional diversity: organism traits,
community structure, and ecosystem properties. Oxford University Press.
Gold, Z.J., Pellegrini, A.F., Refsland, T.K., Andrioli, R.J., Bowles, M.L., Brockway, D.G.,
Staver, A.C., 2023. Herbaceous vegetation responses to experimental re in savannas
and forests depend on biome and climate. Ecol. Lett.
Gonz´
alez-Melo, A., Posada, J.M., Beauchˆ
ene, J., Lehnebach, R., Levionnois, S.,
Derroire, G., Clair, B., 2024. The links between wood traits and species demography
change during tree development in a lowland tropical rainforest. AoB Plants. 16 (1),
plad090.
Goodale, U.M., Ashton, M.S., Berlyn, G.P., Gregoire, T.G., Singhakumara, B.M.P.,
Tennakoon, K.U., 2012. Disturbance and tropical pioneer species: Patterns of
association across life history stages. For. Ecol. Manage 277, 54–66.
Gorgens, E.B., Nunes, M.H., Jackson, T., Coomes, D., Keller, M., Reis, C.R., Ometto, J.P.,
2021. Resource availability and disturbance shape maximum tree height across the
Amazon. Glob. Chang. Biol. 27 (1), 177–189.
Hoffmann, W.A., Bazzaz, F.A., Chatterton, N.J., Harrison, P.A., Jackson, R.B., 2000.
Elevated CO2 enhances resprouting of a tropical savanna tree. Oecologia 123,
312–317.
Kattge, J., B¨
onisch, G., Díaz, S., Lavorel, S., Prentice, I.C., Leadley, P., Cuntz, M., 2020.
TRY plant trait database–enhanced coverage and open access. Glob. Chang. Biol. 26
(1), 119–188.
Kazakou, E., Fried, G., Richarte, J., Gimenez, O., Violle, C., Metay, A., 2016. A plant trait-
based response-and-effect framework to assess vineyard inter-row soil management.
Bot. Lett. 163 (4), 373–388.
Keddy, P., 2005. Putting the plants back into plant ecology: Six pragmatic models for
understanding and conserving plant diversity. Ann. Bot. 96, 177–189.
Khouma, M., 2002. ISRA CNRA De Bambey, Les grands Types De Sols Au S´
en´
egal.
Rapport sur les Ressources en Sols du Monde (FAO). http://www.fao.
org/docrep/005/y3948f/y3948f07.htm#P0_0.
Kunstler, G., Falster, D., Coomes, D.A., Hui, F., Kooyman, R.M., Laughlin, D.C.,
Poorter, L., et al., 2016. Plant functional traits haveglobally consistent effects on
competition. Nature 529, 204–207.
Ky-Dembele, C., Tigabu, M., Bayala, J., Ou´
edraogo, S.J., Od´
en, P.C., 2007. The relative
importance of different regeneration mechanisms in a selectively cut savanna-
woodland in Burkina Faso. West Africa. Forest Ecol. Manage. 243, 28–38.
Lamanna, C., Blonder, B., Violle, C., Kraft, N.J., Sandel, B., ˇ
Símov´
a, I., Enquist, B.J.,
2014. Functional trait space and the latitudinal diversity gradient. Proceed. Nat.
Acad. Sci. 111 (38), 13745–13750.
Langridge, J., Delabye, S., Gilg, O., Paillet, Y., Reyjol, Y., Sordello, R., Gosselin, F., 2023.
Biodiversity responses to forest management abandonment in boreal and temperate
forest ecosystems: a meta-analysis reveals an interactive effect of time since
abandonment and climate. Biolog. Conserv. 287, 110296.
Lavorel, S., Garnier, E., 2002. Predicting changes in community composition and
ecosystem functioning from plant traits: revisiting the Holy Grail. Funct. Ecol. 16,
545–556.
Lenth, R., 2020. emmeans: estimated marginal means, aka least-squared means [accessed
20 February 2021]. Available from https://cran.r-project.org/web/packages/emme
ans/index.html.
Leßmeister, A., Heubach, K., Lykke, A.M., Thiombiano, A., Wittig, R., Hahn, K., 2016.
The contribution of non-timber forest products (NTFPs) to rural household revenues
in two villages in south-eastern Burkina Faso. Agrofor. Syst. 92, 139–155.
Lindenmayer, D.B., 2009. Forest wildlife management and conservation. Ann. N. Y.
Acad. Sci. 1162 (1), 284–310.
Lohbeck, M., Poorter, L., Lebrija-Trejos, E., Martínez-Ramos, M., Meave, J.A., Paz, H.,
Bongers, F., 2013. Successional changes in functional composition contrast for dry
and wet tropical forest. Ecology. 94 (6), 1211–1216.
Luoga, E.J., Witkowski, E.T., Balkwill, K., 2004. Regeneration by coppicing (resprouting)
of miombo (African savanna) trees in relation to land use. For. Ecol. Manage 189
(1–3), 23–35.
Lykke, A.M., Kristensen, M.K., Ganaba, S., 2004. Valuation of local use and dynamics of
56 woody species in the Sahel. Biodivers. Conserv. 13, 1961–1990.
Markl, J.S., Schleuning, M., Forget, P.M., Jordano, P., Lambert, J.E., Traveset, A.,
B¨
ohning-Gaese, K., 2012. Meta-analysis of the effects of human disturbance on seed
dispersal by animals. Conservat. Biol. 26 (6), 1072–1081.
Marone, D., 2015. ´
Etude du potentiel de stock de carbone d’esp`
eces agroforesti`
eres et de
leurs traits fonctionnels en lien avec les syst`
emes d’utilisation des terres au S´
en´
egal.
Universit´
e Laval, Canada, p. 166.
Mayeld, M.M., Bonser, S.P., Morgan, J.W., Aubin, I., McNamara, S., Vesk, P.A., 2010.
What does species richness tell us about functional trait diversity? Predictions and
evidence for responses of species and functional trait diversity to land-use change.
Global Ecol. Biogeogr. 19, 423–431.
Mbow, C., Mertz, O., Diouf, A., Rasmussen, K., Reenberg, A., 2008. The history of
environmental change and adaptation in eastern Saloum–Senegal—Driving forces
and perceptions. Glob. Planet. Change 64 (3-4), 210–221.
Neke, K.S., Owen-Smith, N., Witkowski, E.T., 2006. Comparative resprouting response of
Savanna woody plant species following harvesting: the value of persistence. For.
Ecol. Manage 232 (1-3), 114–123.
Niang, F., Marchand, P., Sambou, B., Fenton, N., 2024. Exploring the effects of forest
management on tree diversity, community composition, population structure and
carbon stocks in sudanian domain of Senegal, West Africa. For. Ecol. Manage 559,
121821. https://doi.org/10.1016/j.foreco.2024.121821.
N. Fatimata et al.
Trees, Forests and People 16 (2024) 100602
11
Niang-Diop, F., Lykke, A.M., Sambou, B., 2011. R´
eg´
en´
eration naturelle de Cordyla
pinnata (Lepr. Ex. A. Rich.) Milne-Redh. dans une savane soumise au feu et au
pˆ
aturage. Sci. et Changements Plan´
etaires/S´
echeresse 22 (3), 186–191.
Osborne, C.P., Charles-Dominique, T., Stevens, N., Bond, W.J., Midgley, G., Lehmann, C.
E., 2018. Human impacts in African savannas are mediated by plant functional traits.
New Phytolog. 220 (1), 10–24.
Ou´
edraogo, A., Thiombiano, A., 2012. Regeneration pattern of four threatened tree
species in Sudanian savannas of Burkina Faso. Agrofor. Syst. 86, 35–48.
Paillet, Y., Berg`
es, L., Hj¨
alt´
en, J., Odor, P., Avon, C., Bernhardt-R¨
omermann, M., et al.,
2010. Biodiversity differences between managed and unmanaged forests: meta-
analysis of species richness in Europe. Conserv. Biol. 24 (1), 101–112.
Paine, C.T., Amissah, L., Auge, H., Baraloto, C., Baruffol, M., Bourland, N., Hector, A.,
2015. Globally, functional traits are weak predictors of juvenile tree growth, and we
do not know why. J. Ecol. 103 (4), 978–989.
Pare, S., Savadogo, P., Tigabu, M., Oden, P.C., Ouadba, J.M., 2009. Regeneration and
spatial distribution of seedling populations in Sudanian dry forests in relation to
conservation status and human pressure. Trop. Ecol. 50 (2), 339–353.
Poorter, H., Niklas, K.J., Reich, P.B., Oleksyn, J., Poot, P., Mommer, L., 2012. Biomass
allocation to leaves, stems and roots: meta-analyses of interspecic variation and
environmental control. New Phytologist 193 (1), 30–50.
Poorter, L., Van de Plassche, M., Willems, S., Boot, R.G.A., 2004. Leaf traits and
herbivory rates of tropical tree species differing in successional status. Plant Biol. 6
(6), 746–754.
Potapov, P., Turubanova, S., Hansen, M.C., 2011. Regional-scale boreal forest cover and
change mapping using Landsat data composites for European Russia. Remote Sens.
Environ. 115 (2), 548–561.
PROGEDE, 2008. Bilan Des R´
ealisations Du PROGEDE, Janvier 1998- d´
ecembre 2008,
Dakar, Unit´
e de Coordination Du programme, Minist`
ere de L´Environnement Et De La
Protection de La Nature, Des Bassins de R´
etention et Des Lacs Articiels. Ministere
de l´Energie, p. 35.
PROGEDE, 2016. Rapport scientique sur l´
´
evaluation des stocks de carbone dans les
massifs forestiers de la zone d´intervention du PROGEDE 2. Minist`
ere de
l´Environnement et de la Protection de la Nature, p. 61.
Raunkiaer, C., 1934. The Life Forms of Plants and Statistical Plant geography; Being the
Collected Papers of C. Raunkiaer. Oxford Clarendon Press, p. 632.
R´
epublique du S´
en´
egal. 2018. Loi n◦2018/25 du 12 novembre 2018 portant code
forestier du S´
en´
egal, 68 p.
Rogers, H.S., Donoso, I., Traveset, A., Fricke, E.C., 2021. Cascading impacts of seed
disperser loss on plant communities and ecosystems. Annu Rev. Ecol. Evol. Syst. 52,
641–666.
Rüger, N., Wirth, C., Wright, S.J., Condit, R., 2012. Functional traits explain light and
size response of growth rates in tropical tree species. Ecology. 93 (12), 2626–2636.
Sambou, B., 2004. Evaluation De l’´
etat, De La Dynamique Et Des Tendances ´
evolutives
de La Flore Et De La V´
eg´
etation Ligneuse Dans Les Domaines Soudanien Et Sub-
Guin´
een Au S´
en´
egal. Doctorat d’Etat. Universit´
e Cheikh Anta Diop de Dakar
(UCAD), Dakar [Google Scholar].
Sambou, S., Lykke, A.M., Sambou, H., Guiro, I., Sambou, B., Mbow, C., 2015. Land use-
land cover change and drivers of deforestation in the Patako protected area (Center-
West of Senegal). Am. J. Environm. Protect. 4 (6), 306–317.
Sang, W., Chen, S., Li, G., 2008. Dynamics of leaf area index and canopy openness of
three forest types in a warm temperate zone. Front. For. China 3, 416–421. https://
doi.org/10.1007/s11461-008-0062-3.
Schumann, K., Wittig, R., Thiombiano, A., Becker, U., Hahn, K., 2011. Impact of land-use
type and harvesting on population structure of a non-timber forest product-
providing tree in a semi-arid savanna, West Africa. Biol. Conserv. 144, 2369–2376.
Seidl, R., Spies, T.A., Peterson, D.L., Stephens, S.L., Hicke, J.A., 2016. Searching for
resilience: addressing the impacts of changing disturbance regimes on forest
ecosystem services. J. Appl. Ecol. 53 (1), 120–129.
Shibata, R., Kurokawa, H., Shibata, M., Tanaka, H., Iida, S., Masaki, T., Nakashizuka, T.,
2016. Relationships between resprouting ability, species traits and resource
allocation patterns in woody species in a temperate forest. Funct. Ecol. 30 (7),
1205–1215.
Solly, B., Andrieu, J., Di`
eye, E.H.B., Jarju, A.M., 2022. Dynamiques contrast´
ees de
reverdissement et d´
egradation de la couverture v´
eg´
etale au S´
en´
egal r´
ev´
el´
ees par
analyse de s´
erie temporelle du NDVI MODIS. VertigO 22 (1), 1–24.
Staver, A.C., Archibald, S., Levin, S., 2011. Tree cover in sub-Saharan Africa: rainfall and
re constrain forest and savanna as alternative stable states. Ecology. 92 (5),
1063–1072.
Tappan, G.G., Sall, M., Wood, E.C., Cushing, M., 2004. Ecoregions and land cover trends
in Senegal. J. Arid. Environ. 59 (3), 427–462.
Tilman, D., Knops, J., Wedin, D., Reich, P., Ritchie, M., Siemann, E., 1997. The Inuence
of functional diversity and composition on ecosystem processes. Science (1979) 277
(5330), 1300–1302. https://doi.org/10.1126/science.277.5330.1300.
bcda Tinya, F, Doerer, Inken, de Groot, Maarten, Heilman-Clausen, Jacob,
Kov´acs, Bence, Mårell, Anders, ´Odor, Peter, 2023. A synthesis of multi-Taxa
management experiments to guide forest biodiversity conservation in Europe. Global
Ecology and Conservation.
USAID, 2008. Senegal biodiversity and tropical forests assessment/senegal under
prosperity, livelihoods and conserving ecosystems. ECODIT, p. 96.
Vadigi, S., Ward, D., 2013. Shade, nutrients, and grass competition are important for tree
sapling establishment in a humid savanna. Ecosphere, 4 (11), 1–27.
Violle, C., Navas, M.L., Vile, D., Kazakou, E., Fortunel, C., Hummel, I., Garnier, E., 2007.
Let the concept of trait be functional! Oikos. 116, 882–892.
Wagner, S., Fischer, H., Huth, F., 2011. Canopy effects on vegetation caused by
harvesting and regeneration treatments. Eur. J. Forest Res. 130, 17–40. https://doi.
org/10.1007/s10342-010-0378-z.
White, F. 1983. The vegetation map of Africa: a descriptive memoir to accompany the
Wiley & Sons, pp. 417, New York, USA.
Wigley, B.J., Slingsby, J.A., Diaz, S., Bond, W.J., Fritz, H., Coetsee, C., 2016. Leaf traits of
African woody savanna species across climate and soil fertility gradients: evidence
for conservative versus acquisitive resource-use strategies. J. Ecol. 104 (5),
1357–1369.
Wright, I.J., Dong, N., Maire, V., Prentice, I.C., Westoby, M., Díaz, S., Wilf, P., 2017.
Global climatic drivers of leaf size. Science 357 (6354), 917–921.
Wright, S.J., Kitajima, K., Kraft, N.J.B., Reich, P.B., Wright, I.J., Bunker, D.E., Condit, R.,
et al., 2010. Functional traits and the growth-mortality trade-ofn tropical trees.
Ecology. 91, 3664–3674.
Wunderle Jr, J.M., 1997. The role of animal seed dispersal in accelerating native forest
regeneration on degraded tropical lands. For. Ecol. Manage 99 (1-2), 223–235.
Wood, S.A., 2018. Nutritional functional trait diversity of crops in south-eastern Senegal.
J. Appl. Ecol. 55 (1), 81–91.
Yan, A., Chen, Z., 2020. The control of seed dormancy and germination by temperature,
light and nitrate. Botan. Rev. 86 (1), 39–75.
Zanne, A.E., Lopez-Gonzalez, G., Coomes, D.A., Ilic, J., Jansen, S., Lewis, S.L., Miller, R.
B., Swenson, N.G., Wiemann, M.C., Chave, J., 2009. Global wood density database
Dryad. Identier 235 (2009). http://hdl.handle.net/10255/dryad.
Zhang, D., 2017. A coefcient of determination for generalized linear models. Am. Stat.
71 (4), 310–316.
Zida, D., Sawadogo, L., Tigabu, M., Tiveau, D., Od´
en, P.C., 2007. Dynamics of sapling
population in savanna woodlands of Burkina Faso subjected to grazing, early re and
selective tree cutting for a decade. For. Ecol. Manage 243 (1), 102–115.
N. Fatimata et al.
