Article 1

Seasonal Variation in Transpiration and Stomatal Conductance 2

of Three Savanna Tree Species in Ruma National Park, Kenya 3

John Maina Nyongesa 1*, Wycliff Oronyi 1, Oyoo Lawrence 1, Ernest Kiplangat Ronoh 2,3, 4
Lindsay Sikuku Mwalati 4, Vincent Suba 4, Leopody Gayo 7, Jacques Nkengurutse 5, Denis Ochuodho Otieno 1 5
and Yuelin Li 4,6* 6

1 Department of Biological Sciences, Jaramogi Oginga Odinga University of Science and Technology, P.O. Box 7
210 – 40601, Bondo, Kenya. 8
2 Kenya Forestry Research Institute, P.O. Box 20412-00200, Nairobi, Kenya; 9
3 Vrije Universiteit Brussel, Department of Water and Climate, Brussels, Belgium. 10
4 South China Botanical Garden, Guangzhou, 510650, China 11
5 Center of Research in Natural and Environmental Sciences, Department of Biology, Faculty of Sciences, 12
University of Burundi, Bujumbura P.O. Box 2700, Burundi 13
6 University of Chinese Academy of Sciences, Beijing, 100039, China 14
7 Department of Biology, University of Dodoma, P. O. Box 338, Dodoma, Tanzania 15
* Correspondence: [email protected]/[email protected] 16

Abstract: Understanding the seasonal regulation of transpiration and stomatal conduct- 17

ance is critical for evaluating plant water-use strategies in response to environmental var- 18
iability. This study assessed these physiological traits in three dominant savanna tree spe- 19
cies (Combretum molle, Piliostigma thonningii, and Balanites aegyptiaca) in Ruma National 20
Park, Kenya. Measurements were taken during wet and dry seasons under varying can- 21
opy light conditions (light-exposed vs. shaded leaves) and soil moisture regimes. A ran- 22
domized design with four treatments and three replicates was employed. Results showed 23
significantly higher transpiration and stomatal conductance during wet seasons, espe- 24
cially in sunlit leaves (P < 0.05). P. thonningii exhibited the highest rates of transpiration 25
(19.8 µg cm⁻2 s⁻1) and stomatal conductance (2.9 µg cm⁻2 s⁻1) in light conditions, while B. 26
aegyptiaca maintained consistently low values, reflecting a drought-tolerant strategy. C. 27

Academic Editor(s): Name

molle demonstrated intermediate responses, suggesting a balance between water conser- 28
vation and resource use. Despite seasonal trends, low R2 values indicated that internal 29
Received: date
physiological regulation outweighed the influence of external climatic drivers. These find- 30
Revised: date
Accepted: date
ings reveal species-specific water-use strategies and highlight the ecological significance 31

Published: date of leaf-level responses to light and moisture availability in tropical savannas. The study 32
provides valuable insights for forest management and climate-resilient restoration plan- 33
Citation: Nyongesa, J.M.; Oronyi,
W.; Lawrence, O.; Ronoh, E.K.;
ning in water-limited ecosystems. 34

Mwalati, L.S.; Suba, V.; Gayo, L.;

Nkengurutse, J.; Otieno, D.O.; Li, Y. Keywords: Transpiration; Stomatal conductance; Plant water-use strategy; Canopy light gradient; 35
Seasonal Variation in Transpiration Tropical savanna; Climate adaptation 36
and Stomatal Conductance of Three 37
Savanna Tree Species in Ruma Na-
tional Park, Kenya. Forests 2025, 13,
x. https://doi.org/10.3390/xxxxx
1. Introduction 38
Copyright: © 2025 by the authors.
Leaf transpiration plays a vital role in the terrestrial water cycle, constituting a major 39
Submitted for possible open access
publication under the terms and
component of overall evapotranspiration (Ondier et al., 2020). Plant water use is predom- 40
conditions of the Creative Commons inantly regulated by the opening and closing of stomata, which in turn control carbon 41
Attribution (CC BY) license uptake and water loss across scales, from individual leaves to entire landscapes (Ghimire 42
(https://creativecommons.org/license et al., 2018; Haworth et al., 2021). The movement of stomata is influenced by both biotic 43
s/by/4.0/).

Forests 2025, 13, x. https://doi.org/10.3390/xxxxx www.mdpi.com/journal/forests

Forests 2025, 13, x FOR PEER REVIEW 2 of 13

and abiotic factors, with significant implications for plant physiological responses under 44
variable environmental conditions (Maina & Wang, 2013; Ghimire et al., 2018). 45
A mechanistic understanding of transpiration (Tr) and stomatal conductance (gs) is 46
essential for developing models that predict plant water use at ecosystem, regional, and 47
global scales, especially under projected climate change scenarios. In savanna ecosystems, 48
such models are particularly valuable due to the inherent variability in light, moisture and 49
vegetation structure. Previous studies have identified diverse strategies used by plants to 50
cope with fluctuating light conditions across canopy layers, with canopy position signifi- 51
cantly affecting photosynthetic capacity and water-use efficiency (Huo & Wang, 2007; 52
Maina et al., 2015; Niinemets, 2012). 53
Transpiration is known to increase with rising vapor pressure deficit (VPD), while 54
stomatal behavior is tightly coupled with soil moisture availability. Under dry conditions, 55
stomatal closure serves as a protective mechanism to prevent xylem embolism and con- 56
serve water (Ghimire et al., 2018; Quirk et al., 2019). Despite the importance of these pro- 57
cesses, empirical data on Tr and gs in East African humid savanna trees, especially at the 58
leaf level remain scarce. Limited studies suggest that both soil characteristics and atmos- 59
pheric conditions such as solar radiation, humidity and wind speed can influence transpi- 60
ration by affecting leaf temperature and vapor exchange dynamics (Allen & Grime, 1995; 61
Ondier et al., 2020). 62
In African savannas, several investigations have examined short-term patterns of 63
transpiration and seasonal gs (Holdo & Nippert, 2015; Marcelo et al., 2015). However, 64
many of these studies rely on indirect micrometeorological extrapolations rather than di- 65
rect measurements at the leaf or crown level. Additionally, the assumption of uniform 66
light distribution across tree canopies limits the ecological relevance of these assessments 67
(Casaroli et al., 2020; Dzikiti et al., 2022). The vertical stratification of canopies where 68
younger, light-exposed leaves differ physiologically from shaded, older leaves is seldom 69
considered in such analyses. 70
In Kenya, existing research on humid savanna species such as Combretum molle has 71
primarily focused on sap flow dynamics or canopy-scale CO2 fluxes (Otieno et al., 2015; 72
Ondier et al., 2020). Balanites aegyptiaca has been shown to exhibit strong physiological 73
sensitivity to soil moisture availability (Oronyi et al., 2023), though most data originate 74
from controlled experiments (Mukhtar et al., 2016). Similarly, studies in southern Africa 75
indicate that water availability may not directly trigger phenological changes in Pili- 76
ostigma thonningii (Chidumayo, 2015), underscoring the complex relationship between 77
plant physiology and environmental cues. 78
Leaf-level physiological traits such as Tr and gs can vary significantly by canopy po- 79
sition, leaf age and irradiance throughout the day. These fine-scale variations are critical 80
for understanding species' ecological strategies and their capacity to survive under fluc- 81
tuating water and light availability. However, how C. molle, B. aegyptiaca, and P. thonningii 82
physiologically respond to vertical light gradients within their canopies remains poorly 83
understood. This knowledge gap hinders our ability to predict carbon and water fluxes in 84
humid savanna systems, which are characterized by strong temporal variability in re- 85
source availability (Maina et al., 2012; Ondier et al., 2020). 86
Accordingly, this study addresses two key research questions: How do sunlit and 87
shaded leaves of savanna tree species respond to seasonal microclimatic conditions dur- 88
ing the growing season? Are there differences in transpiration and stomatal conductance 89
between top (light-exposed) and bottom (shaded) canopy leaves? These questions are crit- 90
ical for improving our understanding of canopy-level physiological heterogeneity. Prior 91
studies have shown that sun-exposed leaves generally exhibit higher photosynthetic rates, 92
modulated by leaf nitrogen content and light intensity (Dewar et al., 2012), while others 93
Forests 2025, 13, x FOR PEER REVIEW 3 of 13

have noted reduced biochemical sensitivity in shaded leaves of arid-adapted species 94

(Maina et al., 2015). 95
Although C. molle, B. aegyptiaca, and P. thonningii are widespread in Ruma National 96
Park, Kenya, their seasonal leaf-level gas exchange patterns under field conditions remain 97
undocumented. This study aims to fill that gap, providing insights critical for ecosystem 98
management, climate-resilient restoration and modeling of plant-water interactions in the 99
East African savannas. 100

2. Materials and methods 101

2.1. Study site 102

The study was conducted in Ruma National Park, located in the Lambwe Valley be- 103
tween the Kanyamwa Escarpment and the Gwasi Hills (00°36′28″ S, 34°15′24″ E) at ap- 104
proximately 1400 m above sea level (Fig. 1). The park is situated 23 km southwest of Homa 105
Bay Town and 425 km west of Nairobi (Kimanzi et al., 2015). A section within the park, 106
separated but managed by the National Youth Service (NYS) encompasses approximately 107
40 ha and has been used for long-term monitoring of savanna vegetation dynamics 108
(Otieno et al., 2010). Soils in the area are predominantly alkaline black Vertisols, as classi- 109
fied by the World Reference Base for Soil Resources (WRB). The region experiences a bi- 110
modal rainfall pattern, with annual precipitation ranging from 1200 to 1600 mm. The long 111
rainy season occurs from March to May, and the short rains fall between October and 112
December. The experimental work was carried out within a 70 m × 100 m fenced plot 113
previously subjected to grazing but protected during the study period. 114

115

Figure 1. Map of Ruma National Park and the surrounding (Kimanzi et al., 2015). 116

117
Forests 2025, 13, x FOR PEER REVIEW 4 of 13

2.2. Experimental design and vegetation sampling 118

Three tree species common in the region; Combretum molle, Balanites aegyptiaca and 119
Piliostigma thonningii were selected based on their dominance and ecological relevance. 120
For each species, three individual trees were randomly selected, with heights ranging 121
from 2.0 to 2.8 m and diameters of approximately 0.17 m. All selected individuals were 122
estimated to be over 15 years old. 123
A completely randomized design was employed with three replicates per species and 124
four treatment combinations. Treatments were based on canopy position (upper vs. 125
lower) and light exposure (light-exposed vs. shaded leaves). Upper canopy leaves were 126
presumed to be younger and more light-exposed, while lower canopy leaves were older 127
and shaded. Leaves were carefully sampled from equivalent twig positions to minimize 128
variability due to age or damage. 129

Table 1. Description of tree species in Ruma National Park. 130

Characteristics P. thonningii B. aegyptiaca C. molle

woodland and wooded
Tropical, dry land African Throughout tropical Af-
Distribution grasslands of sub-humid
savanna rica
Africa
Variable shrub or small
Deciduous tree, 4-15 m Spiny shrub or Tree, up to
Life form Semi-deciduous tree, 3-
in height 10m I height
13 m.
Leaf type Broad-leaved Compound leaves Opposite and simple
Large, simple, two-
Reduced and spirally ar-
Leaf Characteristics lobed, leathery leave leathery leaves
ranged leaves
(camel's foot-like)
crooked or leaning
Stem Single stem tree multi-branched
trunk
complex system of strong, does not have an ag-
Root system Deep roots
diagonally radiating roots gressive root system

2.3. Data collection 131

2.3.1. Micrometeorology 132

Microclimatic variables were recorded using an automatic weather station 133

(AWSGP1, Delta-T Devices, Cambridge, UK) mounted 2 meters above ground. Parame- 134
ters recorded included wind direction and speed, precipitation, solar radiation, photosyn- 135
thetic photon flux density (PPFD), relative humidity (RH), and air temperature. Data were 136
logged every 5 minutes and averaged at 30-minute intervals using a Delta-T data logger. 137
Vapor pressure deficit (VPD) was computed using RH and temperature values as: 138

 100 − RH 
VPDair =    VPsat (1)
 100 
Where: VP air- Air vapour pressure 139
VP Sat-Saturation vapour pressure 140
RH-Relative Humidity 141
Vapor pressure deficit VPD was then be computed as the difference between VP sat 142
and VPair 143

2.3.2. Soil moisture content 144

Volumetric soil moisture content was measured using 5TE soil moisture sensors 145
(Decagon Devices, Washington, USA), installed at a depth of 60 cm and connected to an 146
EM50 data logger. Three sensors were installed around each tree, within a 50 cm radius. 147
Data were logged at 30-minute intervals and averaged daily. 148
Forests 2025, 13, x FOR PEER REVIEW 5 of 13

2.3.3. Leaf gas exchange 149

Leaf gas exchange was assessed using a steady-state porometer (LI-1600, LI-COR Bi- 150
osciences, Lincoln, NE, USA) fitted with a 2-cm circular aperture. Measurements were 151
taken on fully expanded leaves at both the top (upper canopy) and bottom (lower canopy) 152
positions. At each canopy level, five outer (light-exposed) and five inner (shaded) leaves 153
were measured. 154
Porometer readings included transpiration rate (Tr), stomatal conductance (gs), leaf 155
temperature, air temperature, RH, and PPFD on the adaxial surface. Measurements were 156
taken from sunrise to sunset. Under rainy or high-humidity conditions, diurnal readings 157
were taken at six intervals: 0700, 0900, 1100, 1300, 1500, and 1700 hours. Each measure- 158
ment lasted 20–30 seconds, and all leaves remained attached to the branch during meas- 159
urements. 160
Transpiration (Tr) was estimated using a semi-empirical approach based on leaf area 161
index and potential transpiration (Novak, 1995). Stomatal conductance (gs) was calculated 162
as: 163

E
gs = (2)
VPD
where VPD is the vapor pressure deficit and E is the evapotranspiration rate meas- 164
ured by the portable gas analyzer. 165

2.4. Data analysis 166

All statistical analyses were performed using SPSS v25 (IBM Corp.) and R v4.0.1. 167
Normality of data was tested using the Kolmogorov–Smirnov and Shapiro–Wilk tests. 168
Leaf gas exchange parameters (Tr and gs) were analyzed using one-way ANOVA and 169
differences were considered significant at P < 0.05. Linear regression analyses were used 170
to evaluate relationships between physiological parameters and environmental variables, 171
stratified by canopy position and species. 172

3. Results 173

3.1. Climatic conditions 174

Seasonal variability in climatic conditions at Ruma National Park followed the typi- 175
cal tropical bimodal rainfall pattern (Fig. 2). Rainfall peaked in April (261.5 mm) and 176

October (184.75 mm), corresponding to the long and short rainy seasons, 177

respectively. The driest months were January (95 mm) and July (84.8 mm), 178

marking distinct dry seasons. 179

Air temperature remained relatively stable throughout the year, ranging between 180
21.0°C and 26.4°C, with maximum temperatures recorded in February and lower values 181
between April and July. Relative humidity (RH) fluctuated from 54% to 79%, showing 182
higher values during the wet months (April and October) and lower values during dry 183
months (June and September). Soil water content (SWC) mirrored rainfall trends, peaking 184
in May (26.8%) and reaching its lowest level in February (22%). Vapor pressure deficit 185
(VPD) displayed an inverse relationship with RH, with higher values recorded during the 186
dry months (February and September) and lower values during the wet seasons. VPD 187
ranged from 0.55 kPa (April) to 1.38 kPa (February and September), indicating periods of 188
higher atmospheric demand and potential plant water stress. 189
Forests 2025, 13, x FOR PEER REVIEW 6 of 13

190

Figure 2. Seasonal variability of climate in Ruma National Park. 191

3.2. Stomatal conductance in light and shaded leaves 192

Stomatal conductance varied markedly between light-exposed (TL) and shaded (TD) 193
leaves (Fig. 3a). In all species, TL conditions resulted in significantly higher gs values, 194
especially during the wet seasons. These increases reflect improved soil moisture availa- 195
bility and higher photosynthetic activity. 196
Among the three species, P. thonningii demonstrated the most dynamic seasonal gs 197
response, with pronounced increases during the rainy periods. C. molle exhibited moder- 198
ate but steady conductance across seasons, while B. aegyptiaca showed consistently low 199
and stable gs values, indicating its drought-adapted physiology. Under shaded conditions 200
(TD), all species exhibited low stomatal conductance with minimal seasonal fluctuation. 201
Forests 2025, 13, x FOR PEER REVIEW 7 of 13

202

Figure 3a. Seasonal trends in stomatal conductance for P. thonningii, C. molle, and B. aegyptiaca under 203
light (TL) and shaded (TD) canopy conditions. 204

3.3. Transpiration patterns in light and shaded leaves 205

Transpiration rates were generally higher under TL conditions compared to TD (Fig. 206
3b). All species exhibited increased transpiration during wet seasons (April–June and Oc- 207
tober–December), coinciding with greater water availability and favorable microclimatic 208
conditions. 209
P. thonningii consistently recorded the highest transpiration rates, particularly during 210
the rainy periods. C. molle showed moderate seasonal variability, while B. aegyptiaca main- 211
tained low and relatively stable transpiration rates across all seasons, consistent with a 212
conservative water-use strategy. During the dry seasons (January–March and July–Sep- 213
tember), transpiration declined in all species, reflecting reduced soil moisture and in- 214
creased atmospheric demand. 215
Forests 2025, 13, x FOR PEER REVIEW 8 of 13

216

Figure 3b. Seasonal trends in transpiration for P. thonningii, C. molle, and B. aegyptiaca under light 217
(TL) and shaded (TD) canopy conditions. 218

3.4. Influence of climate on transpiration and stomata conductance 219

Regression analysis revealed species-specific relationships between gas exchange pa- 220
rameters and key climate variables (Table 2). Under TL conditions, P. thonningii showed 221
a moderately strong positive correlation between transpiration and temperature (R2 = 222
0.353, P = 0.026), suggesting that warmer conditions significantly enhance its water loss. 223
Relationships with other climatic variables (rainfall, RH, SWC, VPD) were weak or non- 224
significant for this species. 225
In contrast, C. molle and B. aegyptiaca demonstrated weaker correlations overall, in- 226
dicating more conservative or stable transpiration patterns regardless of external climate 227
variation. Notably, C. molle showed a significant relationship between gs and VPD (R 2 = 228
0.502, P = 0.006) and RH (R2 = 0.421, P = 0.027) under light conditions. This indicates its 229
stomatal function is moderately sensitive to atmospheric moisture demand. 230
P. thonningii also showed significant correlations between gs and rainfall (R2 = 0.345, 231
P = 0.022) and VPD (R2 = 0.295, P = 0.013) under shaded conditions. However, across all 232
species, shaded leaves generally displayed lower correlations, reflecting limited stomatal 233
responsiveness due to reduced photosynthetic demand. 234
The low R2 values for many variables suggest that intrinsic plant mechanisms (e.g., 235
hormonal regulation, leaf anatomy, hydraulic traits) may play a stronger role than exter- 236
nal climatic drivers in regulating leaf-level gas exchange. 237

Table 2. Coefficients of determination (R2) and P-values for relationships between climate variables 238
and leaf gas exchange parameters in P. thonningii, C. molle, and B. aegyptiaca under light and shaded 239
canopy conditions. 240
Forests 2025, 13, x FOR PEER REVIEW 9 of 13

Transpiration Stomata Conductance

Species Condition Climate Variable
R2 P-value R2 P-value
P. thonningii Light Temperature (°C) 0.353 0.026 0.020 0.820
P. thonningii Light Rainfall (mm) 0.026 0.542 0.137 0.633
P. thonningii Light Relative Humidity (%) 0.188 0.158 0.056 0.966
P. thonningii Light Soil Water Content (%) 0.285 0.167 0.010 0.966
P. thonningii Light Vapor Pressure Deficit (kPa) 0.019 0.729 0.014 0.957
P. thonningii Dark Temperature (°C) 0.040 0.527 0.026 0.778
P. thonningii Dark Rainfall (mm) 0.229 0.308 0.345 0.022
P. thonningii Dark Relative Humidity (%) 0.286 0.009 0.112 0.463
P. thonningii Dark Soil Water Content (%) 0.000 0.618 0.003 0.966
P. thonningii Dark Vapor Pressure Deficit (kPa) 0.295 0.013 0.129 0.230
C. molle Light Temperature (°C) 0.003 0.617 0.115 0.166
C. molle Light Rainfall (mm) 0.140 0.379 0.280 0.183
C. molle Light Relative Humidity (%) 0.000 0.564 0.421 0.027
C. molle Light Soil Water Content (%) 0.039 0.542 0.003 0.863
C. molle Light Vapor Pressure Deficit (kPa) 0.001 0.549 0.502 0.006
C. molle Dark Temperature (°C) 0.014 0.914 0.138 0.409
C. molle Dark Rainfall (mm) 0.002 0.471 0.000 0.618
C. molle Dark Relative Humidity (%) 0.241 0.051 0.236 0.264
C. molle Dark Soil Water Content (%) 0.014 0.880 0.322 0.112
C. molle Dark Vapor Pressure Deficit (kPa) 0.251 0.066 0.067 0.820
B. aegyptiaca Light Temperature (°C) 0.073 0.564 0.055 0.696
B. aegyptiaca Light Rainfall (mm) 0.001 0.863 0.417 0.006
B. aegyptiaca Light Relative Humidity (%) 0.227 0.111 0.076 0.534
B. aegyptiaca Light Soil Water Content (%) 0.127 0.331 0.003 0.880
B. aegyptiaca Light Vapor Pressure Deficit (kPa) 0.297 0.082 0.107 0.372
B. aegyptiaca Dark Temperature (°C) 0.056 0.571 0.000 0.914
B. aegyptiaca Dark Rainfall (mm) 0.030 0.499 0.042 0.649
B. aegyptiaca Dark Relative Humidity (%) 0.123 0.075 0.000 0.422
B. aegyptiaca Dark Soil Water Content (%) 0.015 0.795 0.022 0.391
B. aegyptiaca Dark Vapor Pressure Deficit (kPa) 0.182 0.077 0.000 0.778

4. Discussion 241

4.1. Seasonal responses and species adaptations 242

This study examined the seasonal dynamics of transpiration and stomatal conduct- 243
ance in three dominant savanna tree species, Piliostigma thonningii, Combretum molle and 244
Balanites aegyptiaca, under varying light and soil moisture conditions. The findings re- 245
vealed distinct species-specific adaptations, largely shaped by differences in light expo- 246
sure, canopy position, and water availability. All three species exhibited increased tran- 247
spiration and stomatal conductance during the wet seasons, particularly in light-exposed 248
leaves, due to enhanced water availability and favorable environmental conditions for 249
photosynthesis (Ondier et al., 2021; Oronyi et al., 2023; Otieno et al., 2010). Conversely, 250
reductions in both parameters during the dry season reflected adaptive water-conserva- 251
tion strategies. These seasonal trends were most pronounced in P. thonningii, which 252
demonstrated strong physiological plasticity and the ability to maximize water use during 253
favorable periods. 254

4.2. Influence of light conditions on gas exchange 255

Light availability played a central role in regulating gas exchange, with consistently 256
higher rates of transpiration and stomatal conductance recorded in sunlit leaves across all 257
species. This is consistent with the role of light in stimulating stomatal opening to enable 258
CO2 uptake for photosynthesis (Maina & Wang, 2015; Nyongesah et al., 2016). Under 259
shaded conditions, all species exhibited reduced gas exchange, indicating efficient 260
Forests 2025, 13, x FOR PEER REVIEW 10 of 13

stomatal regulation aimed at minimizing water loss in the absence of high photosynthetic 261
demand (Sadok et al., 2021). Interestingly, C. molle maintained relatively higher transpira- 262
tion under shade compared to the other species, suggesting a physiological mechanism 263
supporting minimal hydration or nutrient transport even under low light. 264

4.3. Climatic drivers of gas exchange 265

Regression analyses revealed that temperature was the most consistent climatic 266
driver of transpiration, especially in P. thonningii under light conditions. This aligns with 267
previous research linking higher evaporative demand with increased stomatal activity 268
(Sadok et al., 2021). In contrast, other environmental variables such as rain- 269

fall, relative humidity and soil water content exhibited weaker or non-sig- 270

nificant relationships with gas exchange. These results suggest that, in this 271

humid savanna system, intrinsic physiological controls may override ex- 272

ternal climatic cues in determining transpiration rates. 273

The relationship between stomatal conductance and climate variables was gener- 274

ally weak across species, particularly under shaded conditions. This finding indicates 275

that stomatal behavior may be governed more by internal regulatory mech- 276

anisms, such as abscisic acid signaling or hydraulic feedbacks (Rodríguez 277

et al., 2016), than by direct responses to environmental fluctuations. Similar 278

patterns have been observed in other woody species, where drought toler- 279

ance is linked to xylem resistance, plant water storage, and root architec- 280

ture (Chen et al., 2022). 281

4.4. Interspecific variation in water-use strategies 282

The three species showed contrasting ecological strategies that reflect their spatial 283
distribution and functional adaptations. For instance, B. aegyptiaca demonstrated a highly 284
conservative water-use strategy, maintaining low transpiration and stomatal conductance 285
regardless of season or light condition. This species is typically found in savanna plains 286
characterized by low soil moisture and high VPD (Otieno et al., 2010; Oronyi et al., 2023). 287
Its drought-resilience is supported by morphological traits such as deep root systems and 288
reduced leaf area, which limit water loss and allow survival under prolonged dry condi- 289
tions. 290
C. molle, dominant on the slopes, exhibited an intermediate strategy. Its moderate gas 291
exchange rates suggest a balance between water conservation and opportunistic resource 292
use. Slopes in this ecosystem provide better drainage and intermittent moisture, which 293
likely favor this flexible response (Ondier et al., 2020). The species’ moderate sensitivity to 294
relative humidity and soil moisture suggests a capacity to optimize performance across a 295
range of conditions (Bordoloi & Ng, 2020; Qiao et al., 2024). 296
P. thonningii, also occupying slope habitats, displayed a highly responsive strategy, 297
maximizing water use and gas exchange during favorable periods. Its high transpiration 298
and stomatal conductance in light conditions reflect its fast growth and competitive ad- 299
vantage during wet seasons. However, this reliance on external moisture makes it less 300
drought-resilient than B. aegyptiaca, likely confining its success to microenvironments 301
with more consistent water availability (Oronyi et al., 2024). 302

4.5. Canopy position and physiological plasticity 303

The observed differences between top (light-exposed) and bottom (shaded) canopy 304
leaves across all species underscore the physiological plasticity associated with light gra- 305
dients. In all cases, transpiration and stomatal conductance were significantly higher in 306
Forests 2025, 13, x FOR PEER REVIEW 11 of 13

the upper canopy, where irradiance and photosynthetic demand are greatest (Wang et al., 307
2020). P. thonningii showed the highest rates under these conditions, reinforcing its light- 308
dependent strategy. By contrast, C. molle maintained moderate gas exchange in both can- 309
opy layers, demonstrating its adaptability to fluctuating light and moisture availability. 310
B. aegyptiaca showed minimal variation between canopy positions, which is consistent 311
with its conservative water-use strategy and tight stomatal regulation (Gupta et al., 2023; 312
Querejeta et al., 2022). 313
Shaded leaves across all species exhibited significantly reduced stomatal conduct- 314
ance, reflecting lower photosynthetic activity and an emphasis on water conservation. P. 315
thonningii experienced the steepest decline in gs under shade, further highlighting its sen- 316
sitivity to irradiance (Xiong et al., 2018). C. molle showed moderate reductions, while B. 317
aegyptiaca remained largely unresponsive, underscoring its low physiological plasticity. 318

4.6. Ecological implications and broader relevance 319

The species-specific responses observed in this study reflect evolutionary adaptations 320
to distinct ecological niches within the humid savanna. These physiological strategies 321
ranging from opportunistic to conservative enable coexistence and resilience in an envi- 322
ronment characterized by intermittent resource availability. Importantly, the study high- 323
lights the dominant role of intrinsic traits and microhabitat preferences in shaping gas 324
exchange dynamics. While climate variables do influence transpiration and stomatal be- 325
havior, their impact is modulated by internal regulation and species-specific architecture. 326
These insights are critical for developing species-specific models of plant water use and 327
for informing management practices aimed at restoring or conserving savanna ecosys- 328
tems under climate variability. Understanding how native species partition water re- 329
sources across microenvironments can guide selection for afforestation, ecological resto- 330
ration, and climate-resilient landscape planning. 331

5. Conclusion 332

This study demonstrates the critical role of light availability and seasonal moisture 333
variation in shaping leaf-level physiological responses among three key savanna tree spe- 334
cies. Piliostigma thonningii exhibited strong light dependency, with marked differences in 335
transpiration and stomatal conductance between sunlit and shaded leaves. This suggests 336
a resource-acquisitive strategy that enables it to thrive under favorable conditions with 337
ample light and moisture. In contrast, Combretum molle showed moderate physiological 338
differentiation between canopy positions, reflecting its capacity to balance water use and 339
conservation. This intermediate response indicates ecological flexibility, enabling survival 340
in environments with fluctuating light and soil water availability. Balanites aegyptiaca ex- 341
hibited the least variation between light conditions, consistent with its conservative water- 342
use strategy. Its low and stable transpiration and stomatal conductance, regardless of en- 343
vironmental variation, suggest strong physiological control that supports survival in arid 344
and high-stress habitats. 345
These species-specific differences underscore the importance of intrinsic adaptations 346
in regulating water use strategies in tropical savannas. Light availability emerged as a 347
dominant factor influencing gas exchange, while seasonal water dynamics played a sup- 348
porting role. Collectively, these traits determine each species’ ecological niche and resili- 349
ence under climate variability. Future research should integrate additional functional 350
traits, such as leaf hydraulic properties, xylem vulnerability and rooting depth to deepen 351
our understanding of plant water regulation. Long-term, cross-ecosystem studies are also 352
needed to assess the adaptability and persistence of these species under shifting climatic 353
regimes. The findings of this study provide valuable insights for forest management, 354
Forests 2025, 13, x FOR PEER REVIEW 12 of 13

conservation planning, and restoration efforts in water-limited ecosystems, particularly 355

those facing increased variability in light and moisture due to climate change. 356

Author Contributions: J.M.N.: Conceptualization, Writing – Original Draft, Review & Editing, 357
Funding Acquisition, Supervision. W.O., O.L., E.R.K., L.S.M., V.S., L.G.: Field Investigations, Data 358
Collection, Writing – Initial Drafts. J.N., Y.L.: Review & Editing, Methodological Refinement; Y.L. 359
Funding Acquisition. D.O.O.: Writing – Review & Editing, Project Supervision, Final Approval of 360
the Manuscript. 361

Funding: 362

Data Availability Statement: The datasets used and analyzed during the current study will be avail- 363
able from the corresponding author on reasonable request. 364

Acknowledgments: This research was supported in part by the Jaramogi Oginga Odinga University 365
of Science and Technology (JOOUST) Small Grants for Research (JSGR) 2015/16 and the National 366
Natural Science Foundation of China (Grant Nos. 31961143023 and 31670453). The authors grate- 367
fully acknowledge the National Youth Service (NYS) section of the Kenya Wildlife Service (KWS) at 368
Ruma National Park for providing access to the study site and facilitating fieldwork activities. 369

Conflicts of Interest: The authors declare that there is no conflict of interest regarding the publica- 370
tion of this manuscript. 371

References 372

1. Allen, S.J., & Grime, V.L. (1995). Measurements of transpiration from savannah shrubs using sap flow gauges. Agricultural and 373
Forest Meteorology, 75(1–3), 23–41. 374
2. Casaroli, D., Sérvulo, A.C.O., Vellame, L.M., Alves Júnior, J., Evangelista, A.W.P., Mesquita, M., & Flores, R.A. (2020). Transpi- 375
ration and growth of young African mahogany plants subject to different water regimes. International Journal of Biometeorol- 376
ogy, 64, 1–13. 377
3. Chen, Z., Li, S., Wan, X., & Liu, S. (2022). Strategies of tree species to adapt to drought from leaf stomatal regulation and stem 378
embolism resistance to root properties. Frontiers in Plant Science, 13, 926535. 379
4. Chidumayo, E. (2015). Dry season watering alters the significance of climate factors influencing phenology and growth of sap- 380
lings of savanna woody species in central Zambia, southern Africa. Austral Ecology, 40, 794–805. 381
5. Dewar, R.C., Tarvainen, L., Parker, K., Wallin, G., & McMurtrie, R.E. (2012). Why does leaf nitrogen decline within tree canopies 382
less rapidly than light? An explanation from optimization subject to a lower bound on leaf mass per area. Tree Physiology, 32, 383
520–534. 384
6. Dzikiti, S., Ntuli, N.R., Nkosi, N.N., Ntshidi, Z., Ncapai, L., Gush, M.B., ... Pienaar, H.H. (2022). Contrasting water use patterns 385
of two drought-adapted native fruit tree species growing on nutrient-poor sandy soils in northern KwaZulu-Natal. South Afri- 386
can Journal of Botany, 147, 197–207. 387
7. Ghimire, C.P., Bruijnzeel, L.A., Lubczynski, M.W., Zwartendijk, B.W., Odongo, V.O., Ravelona, M., & van Meerveld, H.J. (2018). 388
Transpiration and stomatal conductance in a top secondary tropical montane forest: Contrasts between native trees and invasive 389
understorey shrubs. Tree Physiology, 38, 1053–1070. 390
8. Gupta, S.R., Dagar, J.C., Sileshi, G.W., & Chaturvedi, R.K. (2023). Agroforestry for climate change resilience in degraded land- 391
scapes. In Agroforestry for Sustainable Intensification of Agriculture in Asia and Africa (pp. 121–174). 392
9. Haworth, M., Marino, G., Loreto, F., & Centritto, M. (2021). Integrating stomatal physiology and morphology: Evolution of 393
stomatal control and development of future crops. Oecologia, 197(4), 867–883. 394
10. Holdo, R.M., & Nippert, J.B. (2015). Transpiration dynamics support resource partitioning in African savanna trees and grasses. 395
Ecology, 96, 1466–1472. 396
11. Huo, H., & Wang, C.K. (2007). Effects of canopy position and leaf age on photosynthesis and transpiration of Pinus koraiensis. 397
Journal of Applied Ecology, 18(6), 1181–1186. 398
12. Kimanzi, J.K., Sanderson, R.A., Rushton, S.P., & Mugo, M.J. (2015). Spatial distribution of snares in Ruma National Park, Kenya, 399
with implications for management of the roan antelope Hippotragus equinus langheldi and other wildlife. Oryx, 49(2), 295– 400
302. https://doi.org/10.1017/S0030605313000689 401
Forests 2025, 13, x FOR PEER REVIEW 13 of 13

13. Maina, J.N., & Wang, Q. (2015). Seasonal response of chlorophyll a/b ratio to stress in a typical desert species: Haloxylon am- 402
modendron. Arid Land Research and Management, 29(3), 321–334. 403
14. Maina, J.N., & Quan, W. (2013). Variation of photosynthesis and pigment concentration relative to irradiance and nitrogen 404
content for two coexisting desert shrubs. Ecological Engineering, 58, 238–248. 405
15. Maina, J.N., Quan, W., Otieno, D.O., Oula, D.A., Mupenzi, J., & Onyango, J.C. (2012). Interactive effects of livestock and trees 406
on nutrients and herbaceous layer production in a humid Kenyan savanna. East African Journal of Science and Technology, 407
1(1), 9–26. 408
16. Maina, J.N., Quan, W., & Xu, L. (2016). Remote sensing of assimilating branch light use efficiency using the photochemical 409
reflectance index in Haloxylon ammodendron forest. Journal of Applied Remote Sensing, 10(2), 026017. 410
https://doi.org/10.1117/1.JRS.10.026017 411
17. Maina, J.N., Quan, W., & Li, P. (2015). Effectiveness of photochemical reflectance index to trace vertical and seasonal chlorophyll 412
a/b ratio in Haloxylon ammodendron. Acta Physiologiae Plantarum, 37(2), Article 2. 413
18. Mukhtar, R.B., Aliero, M.M., Abdullahi, S., & Bunza, M.R. (2016). The growth of Balanites aegyptiaca (L.) seedlings under varied 414
watering intervals in the nursery. Agro-Science, 15(3), 30–33. 415
19. Nyongesah, M.J., Quan, W., & Lu, X. (2016). Remote sensing of assimilating branch light use efficiency using the photochemical 416
reflectance index in Haloxylon ammodendron forest. Journal of Applied Remote Sensing, 10(2), 026017. 417
20. Ondier, J., Otieno, D., Okatch, D., & Onyango, J. (2020). Sap flow dynamics and responses to grazing and seasonal changes in 418
soil moisture for Acacia ancistroclada and Combretum molle in a Kenyan savanna. https://doi.org/10.21203/rs.3.rs-48464/v1 419
21. Ondier, J.O., Okach, D.O., Onyango, J.C., & Otieno, D.O. (2021). Ecosystem productivity and CO₂ exchange response to the 420
interaction of livestock grazing and rainfall manipulation in a Kenyan savanna. Environmental and Sustainability Indicators, 9, 421
100095. 422
22. Otieno, D.O., K’Otuto, G.O., Maina, J.N., Kuzyakov, Y., & Onyango, J.C. (2010). Responses of ecosystem carbon dioxide fluxes 423
to soil moisture fluctuations in a moist Kenyan savanna. Journal of Tropical Ecology, 26(6), 605–618. 424
23. Otieno, D., Ondier, J., Sebastian, A., Okach, D., Ruidisch, M., Lee, B., Kolb, A., Onyango, J., & Huwe, B. (2015). Patterns of CO₂ 425
exchange and productivity of the herbaceous vegetation and trees in a humid savanna in western Kenya. Plant Ecology, 216, 426
1441–1456. https://doi.org/10.1007/s11258-015-0523-3 427
24. Querejeta, J.I., Prieto, I., Armas, C., Casanoves, F., Diémé, J.S., Diouf, M., ... Rusch, G.M. (2022). Higher leaf nitrogen content is 428
linked to tighter stomatal regulation of transpiration and more efficient water use across dryland trees. New Phytologist, 235(4), 429
1351–1364. 430
25. Quirk, J., Bellasio, C., Johnson, D.A., & Beerling, D.J. (2019). Response of photosynthesis, growth and water relations of a sa- 431
vannah-adapted tree and grass grown across high to low CO₂. Annals of Botany, 124(1), 77–90. 432
https://doi.org/10.1093/aob/mcz048 433
26. Rodríguez, H.G., Maiti, R., & Kumari, A. (2016). Research advances on leaf and wood anatomy of woody species of a Tamauli- 434
pan thorn scrub forest and its significance in taxonomy and drought resistance. Forest Research, 5(3), 183–191. 435
27. Sadok, W., Lopez, J.R., & Smith, K.P. (2021). Transpiration increases under high-temperature stress: Potential mechanisms, 436
trade-offs, and prospects for crop resilience in a warming world. Plant, Cell & Environment, 44(7), 2102–2116. 437
28. Sigdel, S.R., Liang, E., Rokaya, M.B., Rai, S., Dyola, N., Sun, J., ... Peñuelas, J. (2023). Functional traits of a plant species fingerprint 438
ecosystem productivity along broad elevational gradients in the Himalayas. Functional Ecology, 37(2), 383–394. 439
29. Wang, B., Cai, W., Li, J., Wan, Y., Li, Y., Guo, C., ... Liu, K. (2020). Leaf photosynthesis and stomatal conductance acclimate to 440
elevated [CO₂] and temperature, thus increasing dry matter productivity in a double rice cropping system. Field Crops Re- 441
search, 248, 107735. 442
30. Xiong, D., Douthe, C., & Flexas, J. (2018). Differential coordination of stomatal conductance, mesophyll conductance, and leaf 443
hydraulic conductance in response to changing light across species. Plant, Cell & Environment, 41(2), 436–450. 444

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual au- 445
thor(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to 446
people or property resulting from any ideas, methods, instructions or products referred to in the content. 447