Water use efficiency responses in contrasting agroecosystems and land management practices in West Africa

Water use efficiency responses in contrasting agroecosystems and land management practices in West Africa | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Water use efficiency responses in contrasting agroecosystems and land management practices in West Africa Souleymane Sy, Jan Bliefernicht, Kiril Manevski, Samuel Guug, and 15 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8522874/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Water use efficiency (WUE) is a key indicator of ecosystem balance, reflecting how productivity responds to hydrological constraints under climate change. However, variability in WUE and its environmental drivers across West African agroecosystems remains poorly understood. Here, we integrate multi-year (2019–2024), half-hourly eddy-covariance observations of carbon and water-vapor fluxes from four contrasting land-use types in northern Ghana: a reserve savanna forest, rain-fed paddy rice, grassland, and rain-fed cropland. WUE exhibited pronounced diurnal and seasonal variability, shaped by hydrological, atmospheric, and land-management drivers. Diurnal patterns were bimodal, with morning and afternoon peaks shifting between wet and dry seasons. During the wet season, mean WUE was highest in the savanna forest (3.1 ± 0.26 g C kg⁻¹ H₂O), followed by paddy rice (2.08 ± 0.21), cropland (1.93 ± 0.20), and grassland (1.66 ± 0.18). Seasonal analyses highlighted ecosystem-specific controls, reflecting differences in radiation, soil moisture, and cultivation practices. Earth and environmental sciences/Climate sciences Biological sciences/Ecology Earth and environmental sciences/Ecology Earth and environmental sciences/Environmental sciences Earth and environmental sciences/Hydrology ecosystem-climate interactions hydrology carbon-water coupling eddy-covariance West Africa Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Carbon and water cycles are among the most important driving forces behind the coupled exchange of materials and energy in terrestrial ecosystems 1 , 2 . Ecosystem Water Use Efficiency (WUE), defined as the ratio of carbon assimilation via gross primary productivity (GPP), the total amount of CO₂ fixed by terrestrial ecosystems through photosynthesis per unit time, to water loss through evapotranspiration (ET), is a fundamental metric for assessing the coupling between carbon and water cycles in terrestrial ecosystems 3 – 5 . In essence, WUE provides critical insights into the trade-offs between ecosystem productivity and water consumption under varying environmental conditions 6 . Understanding its dynamics is essential in the context of climate change, increased weather variability, and water scarcity, as it provides valuable insights into how different ecosystems respond to both natural and anthropogenic drivers 3 , 7 . However, the extent to which GPP and ET regulate WUE across different ecosystem types and land management regimes remains poorly understood 3 , 8 – 11 . At the leaf level, WUE is primarily governed by stomatal regulation, which balances CO₂ uptake for photosynthesis with water vapor loss through transpiration 12 , 13 . At the ecosystem level, water availability becomes the dominant constraint on carbon sequestration 2 , 13 – 16 , particularly in semi-arid landscapes such as those in West Africa 11 , 17 , 18 . Research consistently reveals a strong gradient of WUE across West Africa's primary land covers, driven by a complex interplay of physiological, structural, and environmental factors 11 , 18 – 20 . For instance, forests in the humid region are known to have high absolute WUE due to large leaf area indices and deep root systems that access soil moisture year-round, allowing for sustained carbon assimilation 21 . However, high ET rates come at a cost of only moderate efficiency in conserving water per unit of carbon fixed 10 . Moreover, while WUE has been extensively investigated across various ecosystems, particularly in mid-latitude regions 6 , 8 , 22 , a substantial knowledge gap persists for semi-arid environments, such as the West African savanna 6 , 9 , 18 . The Sudanian region is dominated by savannas, which exhibit marked seasonal variability in WUE, with values often increasing during the wet season 11 , 17 , 18 . These ecosystems are composed of a mixture of woody C₃ species and herbaceous C₄ grasses, the latter being inherently more water-efficient due to their biochemical CO₂-concentrating mechanism and reduced photorespiration 23 , 24 . During the rainy season, the grass layer flourishes, driving high ecosystem WUE 25 . However, this situation reverses in the dry season as grasses senesce, leaving the deeper-rooted trees to maintain minimal gas exchange 11 , 23 . Furthermore, rainfed crops like millet and sorghum, which are often C₄ plants, can show high WUE during their short growing season 19 , 26 . However, their efficiency is heavily dependent on agricultural practices, soil quality, and the timing of precipitation 19 , 26 . Irrigated/flooded croplands, while boosting yields, typically exhibit very low WUE at the ecosystem level, as the additional water input leads to increased ET without a proportional increase in carbon sequestration 27 . Grasslands and Sahelian shrublands operate under extreme water stress with low WUE due to low biomass and GPP. These ecosystems have adapted to survive (rather than thrive) under arid conditions, with carbon and water fluxes being minimal. Despite a “hierarchy” of WUE in West Africa, it is essential to monitor these patterns amid weather variability, including recurrent droughts and shifting precipitation patterns, which complicate and often alter the water-carbon dynamics, highlighting the need for a deeper understanding of WUE in dynamic terms and across diverse agroecosystems 15 , 28 , 29 . High-resolution EC systems capture real-time variations in carbon and water fluxes, providing a robust method for evaluating WUE 4 , 27 . The density of EC monitoring stations in West Africa, despite previous research efforts 11 , 17 , 19 , 30 – 32 , remains considerably lower than in North America, Europe, or Asia 6 , 21 . Moreover, the application of EC systems across the diverse land-use types in the West African savanna is still limited. Expanding observational efforts is essential for enhancing our understanding of WUE patterns and supporting the development of more sustainable land and water management strategies in this climate-sensitive region. This study presents the first site-specific analysis of WUE across multiple time scales in the West African savanna. Using multi-year (2019–2024), half-hourly eddy-covariance measurements of carbon and water-vapor fluxes, we examine WUE dynamics across four representative agroecosystems in northern Ghana, including - for the first time - a rainfed paddy rice system, alongside a protected savanna forest, a semi-degraded grassland, and a rainfed cropland. The variations in WUE at instantaneous, daily, seasonal, and multi-annual timescales reveal environmental drivers underlying these dynamics for managed (rainfed croplands, paddy rice, grasslands) and undisturbed ecosystems (protected savanna forests) (see Methods). By quantifying both environmental controls on WUE across these diverse land-use types, this research provides critical empirical insights to guide sustainable land and water management, enhance ecosystem resilience, and support climate adaptation planning in the West African Savanna. Specifically, the study addresses two key research questions: (i) To what extent do WUE dynamics differ between managed and undisturbed ecosystems in the semi-arid West African Savanna? (ii) What are the primary environmental factors controlling WUE across different agroecosystem types? Results Environmental variables Table 2 presents the multi-year average values of key environmental drivers and limiting conditions influencing water use efficiency across different ecosystem types. Apart from air temperature being relatively consistent, the results reveal distinct microclimatic patterns among sites. Relative humidity was the highest for the forest site (~ 64.5%), reflecting a moist, shaded microclimate associated with dense canopy cover and transpiration, while cropland and grassland exhibited substantially lower RH (~ 47%), consistent with more open, sun- and wind-exposed surfaces, and rice field had intermediate RH (~ 57%), likely influenced by surface water from flooding. Soil water content showed the most pronounced variation across sites, with the rice field having higher values (30%), while forest, cropland, and grassland had low but measurable values (0.05–0.1%). Vapor pressure deficit, a key indicator of atmospheric water demand, was lowest in the forest, suggesting lower evaporative stress, and slightly higher but comparable across rice, cropland, and grassland, indicating drier atmospheric conditions in more open landscapes. Solar radiation was slightly higher at the forest and rice sites than at the cropland and grassland sites, likely due to differences in local atmospheric conditions. GPP and ET followed each other, with the forest showing the highest rates, consistent with mature vegetation and favorable microclimate conditions. Grassland had moderate GPP and ET, followed by rice and cropland, which exhibited the lowest values, likely reflecting crop-specific physiological traits and management effects. Table 2 Annual mean (± 95% CI) values of key meteorological and ecosystem exchange variables for the four study sites . Variables include air temperature (Ta), relative humidity (RH), soil water content (SWC) at 3 cm, vapor pressure deficit (VPD), Solar radiation (Rg), gross primary productivity (GPP), and evapotranspiration (ET). Sites Ta (°C) RH (%) SWC (%) VPD (hPa) Rg (W/m2) GPP (gC/m2/day) ET (kgH2O/m2/day) Reserve Forest 29.48 ± 0.28 64.51 ± 2.68 0.09± 001 21.64 ± 1.49 300.47 ± 7.78 8.35 ± 0.45 3.05 ± 0.19 Paddy rice 29.54 ± 0.27 57.05 ± 2.67 9.6 ± 1.25 23.09 ± 1.36 277.24 ± 7.22 3.45 ± 0.26 1.94 ± 0.15 Cropland 28.41 ± 0.24 46.91 ± 2.36 0.05 ± 00 22.79 ± 1.14 255.78 ± 3.61 2.82 ± 0.28 1.85 ± 0.13 Grassland 28.7 ± 0.22 47.81 ± 2.49 0.1 ± 0.01 22.70 ± 1.16 250.68 ± 2.74 3.53 ± 0.28 2.57 ± 0.15 Seasonally, air temperature was lowest during the wet season at all sites (Fig. 2 ). At cropland and grassland sites, temperatures can drop to around 25°C during the Harmattan months (December-January), likely due to their higher surface albedo compared to the forest site. Vapor pressure deficit and incoming solar radiation showed similar seasonal patterns to air temperature, with minima during the wet season and maxima during the dry season. In contrast, soil water content and relative humidity exhibited pronounced wet-dry seasonality, peaking in August-September at all sites and reaching their lowest values during the dry season. GPP and ET followed comparable seasonal dynamics, with maxima occurring in August at the forest and grassland sites and in September at the paddy rice and cropland sites (Fig. 2 ). These differences reflect site-specific phenology. At the forest site, LAI peaked in August, shortly after the July rainfall maximum, which provided sufficient moisture for full canopy development (see Figures S5 and S11). In contrast, LAI peaked in September at the rice, cropland, and grassland sites, reflecting their later phenological development (see Figures S1 , S8-10). Diurnal dynamics of WUE The diurnal dynamics of WUE across the four contrasting ecosystem types are presented in Fig. 3 a. Across both wet and dry seasons, WUE exhibited a distinct pattern with pronounced morning and evening peaks. During the wet season, the forest site exhibited a distinct diurnal pattern, with WUE reaching approximately 4.0 g C kg H₂O⁻¹ in the early morning and 4.5 g C kg H₂O⁻¹ in the evening. In contrast, although the paddy rice, cropland, and grassland sites also showed a dual-peak pattern, a single dominant morning peak was more pronounced, with WUE values of roughly 3.7, 3.2, and 2.7 g C kg H₂O⁻¹, respectively. WUE declined markedly toward midday as rising VPD and temperature constrained stomatal conductance and carbon assimilation, a reduction further likely amplified by increased soil evaporation and, in the case of the rice field, evaporation from ponded water, while GPP became increasingly limited under high VPD (see Figures S2-S4). During the dry season, all sites exhibited a more uniform diurnal trend, with WUE peaking in both the morning and late afternoon (18:00–19:00) before experiencing a sharp decline in the evening. The forest site again recorded the highest WUE, reaching up to 7.5 g C kg H₂O⁻¹ in the evening, whereas cropland and grassland values remained near 2.5 g C kg H₂O⁻¹. Compared to the wet season, daytime WUE values during the dry season were generally lower, likely due to reduced soil moisture availability (see Fig. 2 ) and increased atmospheric demand (see Figures S2-S4). However, in the late afternoon (16:00–18:00), WUE tended to increase more sharply under dry conditions, indicating improved carbon assimilation efficiency as evaporative demand and air temperature declined (see Figures S2-S4). Unlike WUE, which followed a double-peak pattern, uWUE exhibited a distinct evening maximum (between 18:00 and 19:00) during the dry season, exceeding 50 g C hPa 0. 5 kg H₂O⁻¹ in the savannah forest, and reaching approximately 16 and 14 g C hPa 0. 5 kg H₂O⁻¹ in cropland and grassland sites, respectively (Fig. 3 b). Savannah forest exhibited the highest WUE and uWUE throughout the day compared to the other ecosystems, reflecting its dense canopy structure, deep rooting system, and physiological capacity to regulate stomatal conductance efficiently, maintaining high photosynthetic rates while minimizing excessive water loss. Seasonal dynamics of WUE Seasonal patterns were evident in the WUEs (Fig. 4 ) and savannah forest exhibited a bimodal pattern with the highest values (exceeding 8 g C kgH₂O⁻¹) during the dry season (December-February), a secondary peak of about 4.5 g C kgH₂O⁻¹ in the mid-rainy season (July-August) and the lowest values (0.5 g C kgH₂O⁻¹) in the transitional months of March and November. WUE of paddy rice increased steadily from 0-0.5 g C kg H₂O⁻¹ during the early stages of crop establishment (June) to around 3 g C kg H₂O⁻¹ in July-August following the rice plant development (Figure S8). These dynamics reflect a shift in the carbon-water interactions with initially high soil evaporation from an open canopy and low photosynthetic activity, yielding low WUE. By the time the crop reached the reproductive stage, the dense canopy tends to minimize soil evaporation (i.e., in line with high SWC and low VPD) and maximize carbon uptake (see Fig. 2 b). At the cropland site, the bimodal pattern in WUE during the rainy season reflects the complex interaction of agricultural practices, crop phenology, and environmental conditions, with the lowest WUE (~ 0.5 g C kg ⁻¹ H₂O) typically in March due to fallow or senesced fields (high soil evaporation and low GPP), resulting in the lowest WUE. Later in the wet season (May-June), rapid vegetative growth and canopy development (Figure S9) enhance GPP, likely shifting ET from soil evaporation to plant transpiration. This is followed by a peak in WUE (~ 2.0 g C kg ⁻¹ H₂O), after which it declines sharply in July, typically coinciding with land preparation activities such as plowing and grass clearing (Figure S9). A second and more pronounced WUE peak (~ 2.5 g C kgH₂O⁻¹) occurs in September, during the crop’s reproductive stage (Fig. 4 a and S9). The grassland site displayed similar seasonal WUE patterns to the cropland site during the dry season (Fig. 4 a). During the rainy season, however, grassland exhibited a distinct and more pronounced WUE peak (~ 2 g C kg H₂O⁻¹) in August, aligning with the reproductive stage of the grass with a fully developed canopy, maximum photosynthetic capacity, and high GPP (see Fig. 2 b). However, this peak was followed by a sharp decline in WUE in September, primarily driven by a reduction in GPP, which decreased in response to reduced solar radiation (Fig. 2 ), despite continued adequate water availability. Overall, uWUE followed a seasonal pattern similar to WUE (Fig. 4 b), reflecting the combined effects of atmospheric demand and ecosystem productivity. However, uWUE peaked during the dry season across all sites, highlighting the influence of high VPD. Values exceeded 40 g C hPa⁰·⁵ kg H₂O⁻¹ in the forest in December-January, while cropland and grassland showed higher peaks around 10 g C hPa⁰·⁵ kg H₂O⁻¹. The lowest uWUE occurred in March, when soil moisture and vegetation activity were minimal, dropping to ~ 3 g C hPa⁰·⁵ kg H₂O⁻¹ in the forest and below 1 g C hPa⁰·⁵ kg H₂O⁻¹ in cropland and grassland. Inter-annual variability of WUE Table 3 presents an analysis of the five-year record of GPP, ET, VPD, WUE, and uWUE across ecosystems. Although data for the forest ecosystem were limited to a single year, it exhibited the highest GPP, which coincided with elevated VPD, as well as the highest WUE and uWUE. This suggests highly efficient stomatal control mechanisms that optimize the carbon-water trade-off through a conservative water-use strategy typical of mature forests. The rice ecosystem showed consistently low GPP alongside low VPD during 2022–2023, with moderate ET likely driven by significant non-stomatal water losses from soil evaporation and canopy interception. Notably, paddy rice WUE and uWUE declined the following year, independent of VPD fluctuations. The cropland site demonstrated moderate and relatively stable GPP over the period, ranging from 966 to 1098 g C m⁻² yr⁻¹, with consistently high VPD and only minor interannual variability in ET, WUE, and uWUE. In contrast, the grassland site exhibited greater interannual variability in both GPP and ET, while VPD remained relatively stable. Despite fluctuations in carbon uptake, WUE values remained consistent (1.3–1.4 g C kg⁻¹ H₂O), highlighting the grassland’s capacity to maintain WUE under variable weather conditions. Interestingly, uWUE increased from 6 to 7 g C hPa⁰·⁵ kg H₂O⁻¹ between 2019 and 2021 before stabilizing around 6.7 g C hPa⁰·⁵ kg H₂O⁻¹, suggesting enhanced physiological optimization in years with higher GPP or slightly more favorable evaporative conditions. Table 3 Inter-annual mean (± 95% CI) values of gross primary productivity (GPP), vapor pressure deficit (VPD), evapotranspiration (ET), water use efficiency (WUE), and underlying water use efficiency (uWUE) for the forest, paddy rice, cropland, and grassland sites across the five study years. Differences in ecosystem carbon and water fluxes relative to the forest ecosystem for the year 2023 are also shown, calculated as the forest-non-forest ecosystem (paddy rice, cropland, and grassland). Values are annual mean differences ± 95% CI. Years GPP (gCm⁻²y⁻¹) VPD (hPa) ET (kg H₂O m⁻² y⁻¹) WUE (gC/kg H₂O) uWUE (gC hPa 0.5 / kg H₂O) Reserve Forest 2023/2024 2910± 196 21.64± 1.49 1080± 95 2.6± 0.15 12.2± 0.85 Paddy Rice 2022 1013 ± 55 22.95± 0.82 547± 27 1.8± 0.13 6.1± 0.50 2023/2024 1016± 95 23.62± 1.23 618± 26 1.6± 0.17 5.6± 0.67 Cropland 2020 1050.9 ± 99 22.37 ± 1.15 714.7 ± 53 1.47 ± 0.09 6.95 ± 0.44 2021 1028.3 ± 116 23.49 ± 1.21 658.7 ± 55 1.56 ± 0.11 7.57 ± 0.55 2022 966.2 ± 102 22.52 ± 1.22 638.8 ± 51 1.51 ± 0.1 7.18 ± 0.50 2023 1098.1 ± 110 22.84 ± 1.21 701.5 ± 52 1.57 ± 0.1 7.48 ± 0.49 Grassland 2019 1184.9 ± 90 22.85 ± 1.25 947.6 ± 58 1.25 ± 0.06 5.98 ± 0.31 2020 1282.2 ± 100 22.84 ± 1.27 890.9 ± 60 1.44 ± 0.08 6.88 ± 0.37 2021 1679.9 ± 181 23.68 ± 1.36 1170.1 ± 90 1.44 ± 0.1 6.99 ± 0.48 2022 1211 ± 117 22.83 ± 1.20 866.9 ± 61 1.4 ± 0.09 6.67 ± 0.42 2023 1344.3 ± 105 21.41 ± 1.13 934.6 ± 61 1.44 ± 0.07 6.65 ± 0.36 Difference Relative to Forest Ecosystem (2023) Comparison GPP (gCm⁻²y⁻¹) VPD (hPa) ET (kg H₂O m⁻² y⁻¹) WUE (gC/kg H₂O) uWUE (gC hPa 0.5 / kg H₂O) Forest - Paddy Rice 1897± 132 -1.98±0.50 462± 42 1.0± 0.08 6.6± 0.25 Forest - Cropland 1812± 140 -1.20± 0.30 379± 75 1.03± 0.06 4.72± 0.34 Forest - Grassland 1565± 124 0.23± 0.32 146± 82 1.26± 0.09 5.55± 0.42 Mechanisms driving WUE responses To explore the drivers and limiting conditions of WUE under varying weather conditions, we analyzed its statistical relationships with key environmental variables (Fig. 5 ). In reserve forests during the dry season, WUE showed little response to temperature but increased under higher VPD and lower humidity and soil moisture, indicating that atmospheric dryness can enhance efficiency when water is limited. During the wet season, WUE increased with higher soil moisture, rainfall, and humidity, but declined with higher temperatures, VPD, and radiation. In the rice field, higher soil moisture and humidity increased WUE, whereas temperature, radiation, and VPD reduced it. Cropland and grassland showed similar patterns: dry-season WUE was constrained by temperature and low soil moisture, while wet-season WUE improved with humidity, soil moisture, and rainfall but decreased under high VPD and radiation. To further investigate the seasonal influence of environmental factors on WUE across all sites, Fig. 6 presents the results of an analysis of variance (see Methods). At the forest site, WUE exhibited strong seasonal shifts in drivers (Fig. 6 a, b). During the dry season, relative humidity (~ 55%), shallow soil moisture (~ 15%), and solar radiation (13%) explained the largest share of variance, although these contributions were not statistically significant, suggesting a multi-driver limitation system constrained by atmospheric demand and upper-layer soil water content. In contrast, the rainy season exhibited a pronounced shift toward radiation-dominated control, with solar radiation accounting for 65% of the variance. This indicates that, once water constraints are alleviated, energy availability becomes the primary driver of carbon-water coupling. Temperature remained a significant contributor (~ 30%), underscoring its persistent role in regulating stomatal behavior and water use in savannah forests. WUE in the rice ecosystem was shaped predominantly by solar radiation and soil moisture dynamics (Fig. 6 g). Solar radiation explained more than 55% of the variance in WUE, reflecting its central role in driving photosynthetic carbon uptake. Despite the flooded or near-saturated conditions typical of paddy systems, subsurface water availability remained a key regulator, with deep soil moisture (30 cm) accounting for ~ 35% of the variance and exerting a highly significant influence. At the cropland site (Figs. 6 c-d), dry-season WUE was primarily controlled by soil moisture, with shallow topsoil moisture explaining nearly 83% of the variance. Atmospheric drivers such as solar radiation and relative humidity contributed only marginally under these strongly water-limited conditions. During the rainy season, the system transitioned to an atmosphere-driven regime characterized by radiation and water-vapor demand (Figs. 6 d, g). Solar radiation accounted for more than 60% of the variability in WUE, highlighting the increasing importance of energy availability once water constraints are alleviated. The water vapor deficit remained a significant regulator during this period, accounting for over 25% of the variance. Concerning the grassland site, dry-season WUE was predominantly controlled by soil water availability, with shallow soil moisture explaining nearly 80% of the variance. This confirms that, as in the cropland system, soil moisture was the primary driver of WUE under water-limited conditions. During the rainy season, the regulation shifted toward atmospheric control, with air temperature becoming the strongest predictor of WUE (~ 34%, p < 0.05). Additionally, shallow soil moisture (~ 28%) and solar radiation (~ 29%) contributed, although without statistical significance. These patterns indicate that once water limitations are relieved, grassland WUE is increasingly shaped by thermal and moisture conditions influencing canopy gas exchange. Overall, both grassland and cropland systems exhibited a clear seasonal transition, reflecting a shared ecohydrological response to the shifting availability of water and energy. Discussion This study provides the first long-term, site-specific assessment of WUE across four contrasting ecosystems in the semi-arid West African savanna, differing in the degree of land management, including, for the first time, a rainfed paddy rice system alongside savanna forest, cropland, and grassland. While WUE dynamics have been extensively examined using remote sensing, ecosystem modeling, and eddy covariance approaches, most studies have focused on mid-latitude regions 4 , 6 , 15 , 27 , 33 – 36 . In contrast, empirical evidence from African ecosystems remains limited due to the scarcity of EC measuring stations, with only a few eddy covariance studies in West Africa, typically restricted to short observation periods or single ecosystems 11 , 17 – 19 , 31 . Our results showed that WUE exhibited clear diurnal dynamics that differed markedly between the wet and dry seasons across all ecosystems. However, although most sites showed a bimodal pattern with a dominant morning peak, reaching roughly 3.7, 3.2, and 2.7 g C kg⁻¹ H₂O in the rice, cropland, and grassland systems, respectively, the savanna forest displayed a distinct and more pronounced afternoon enhancement. Specifically, WUE in the forest increased from about 4.0 g C kg⁻¹ H₂O in the early morning to roughly 4.5 g C kg⁻¹ H₂O later in the day, and this pattern became even stronger during the dry season. This divergence suggests ecosystem-specific regulation of carbon-water exchange: in cropland, rice, and grassland systems, morning peaks likely reflect optimal light and vapor-pressure conditions that favor stomatal opening, whereas the savanna forest’s stronger afternoon peak may indicate deeper rooting systems and sustained transpiration capacity under elevated evaporative demand 37 . In other terms, the amplified afternoon WUE in the forest during the dry season, reaching values nearly double those of the agricultural and grass systems, further indicates a competitive advantage in maintaining carbon uptake when atmospheric dryness intensifies 10 , 38 , 39 . Our findings are consistent with earlier observations of dual diurnal WUE peaks in West African ecosystems 18 , 19 , 31 , which identified similar diurnal patterns and magnitudes in a degraded savanna woodland in northern Benin with daytime WUE reaching 3.86 ± 1.03 g C kg⁻¹ H₂O during the dry season and 5.50 ± 1.29 g C kg⁻¹ H₂O in the wet season 31 . Comparable magnitudes were also observed by Boulain et al. (2009) 19 in their investigation of WUE over millet and fallow sites in the Wankama catchment of the AMMA-Niger observatory, underscoring the consistency of these WUE dynamics across semi-arid West African landscapes. In addition, although the ecosystem WUE observed in our savanna forests was similar to values reported for subtropical forests during the wet season 4 , 6 , 40 , 41 , it remained markedly higher in the late afternoon of the dry season (Fig. 3 a). This dry-season increase likely reflects short-term drought-avoidance strategies that limit water loss while sustaining carbon uptake under high vapor pressure deficits 42 . Differences can also be partly attributed to variations in water availability and physiological regulation 42 . The savanna forest is dominated by drought-adapted C₃ trees, which possess deep rooting systems, conservative hydraulic architecture, and tight stomatal control in response to VPD, enabling sustained carbon assimilation with minimal transpiration during the dry season 42 , 43 . Furthermore, the open canopy of the savanna forests exposes the trees to higher radiation and VPD, which, combined with their drought-adapted traits, may foster higher apparent WUE than in the more humid, shaded microclimates typical of closed-canopy temperate and subtropical forests 41 , 44 . In comparison, rice and cropland systems are primarily composed of C₃ vegetation, which tends to exhibit greater stomatal sensitivity to atmospheric dryness 26 , 45 , 46 . Specifically, maintaining photosynthesis in these systems often requires higher stomatal conductance 47 , 48 , resulting in lower ecosystem WUE during periods of peak evaporative demand. By contrast, although the grassland understory is dominated by C₄ grasses, which are known to exhibit high intrinsic WUE because of their CO₂-concentrating mechanism 24 , 49 , our ecosystem-scale measurements reveal WUE values slightly lower than those in C₃ croplands. This apparent discrepancy likely arises because ecosystem WUE integrates water losses not only through transpiration but also through soil evaporation, VPD, and canopy structural factors, which tend to diminish the physiological advantage of C₄ photosynthesis in semi-arid conditions 24 , 50 , 51 . Our results suggest that the observed dual-peak WUE patterns are driven by ecosystem-specific physiological and microclimatic factors, particularly stomatal optimization 52 – 54 . Across all ecosystems, diurnal WUE followed a consistent three-phase pattern: a morning peak under low VPD, a midday decline as evaporative demand increased, and a late-afternoon peak when photosynthesis remained active while evapotranspiration dropped sharply in line with previous studies 4 , 55 . Moreover, our findings show that uWUE remained consistently elevated during the dry season, with a pronounced rise that coincided with the late-afternoon peak in VPD (Figure S2a). These patterns reflect the strong responsiveness of ecosystem WUE to short-term changes in atmospheric dryness and align with stomatal optimization theory as well as previous evidence demonstrating the dominant influence of VPD on sub-daily WUE dynamics 56 . Importantly, the comparison between WUE and uWUE reveals the added value of using uWUE in ecohydrological assessments. While both indices capture similar seasonal trends, uWUE more effectively isolates physiological responses by reducing the confounding effects of VPD, thereby providing a clearer signal of canopy-level carbon-water exchange 57 . In contrast, traditional WUE amplifies VPD-driven variability, thereby reflecting the combined effects of environmental and physiological controls 57 . Finally, our results reveal clear ecosystem-specific shifts in the dominant controls of WUE. In the savanna forest, WUE regulation transitioned from a multi-factor limitation during the dry season, driven by relative humidity, shallow soil moisture, and solar radiation, which jointly explained ~ 85% of the variance, to a radiation-dominated control in the wet season, when water constraints were alleviated (Fig. 6 ). During this period, solar radiation accounted for 65% of the WUE variance, representing a 5-fold increase in explanatory power compared with the dry season (Fig. 6 b-a). This shift indicates that once soil moisture is sufficient, energy availability becomes the primary driver of carbon-water coupling within the savanna forest. In cropland and grassland ecosystems, WUE was largely regulated by soil moisture during the dry season (explaining ~ 80% of the total variance), but shifted toward atmospheric control under wet conditions, with solar radiation, VPD, and air temperature emerging as dominant factors. In contrast, the paddy rice system exhibited a distinct pattern, with solar radiation and deep soil moisture explaining most of the WUE variability (~ 90%), reflecting the hydrological dominance of flooded cultivation. This sensitivity to deep soil moisture is consistent with the hydrological functioning of paddy ecosystems, where root water uptake and aeration dynamics may depend not only on surface flooding but also on the moisture status of deeper soil layers, which govern subsurface oxygen diffusion, root activity, and water availability throughout the growing season 58 , 59 . Despite providing valuable insights, key limitations remain. While the eddy covariance technique is widely regarded as the most reliable approach for quantifying ecosystem carbon and water fluxes at the local scale (with a footprint of tens to hundreds of meters), our results are subject to uncertainties inherent to this method, as previously addressed in studies on the same sites 60 , 61 . Moreover, spatial heterogeneity within flux footprints, particularly in managed landscapes, can introduce additional uncertainty 60 . Despite these challenges, this study represents a significant advancement in understanding WUE dynamics from daily, seasonal, and annual scales in West Africa, a region where EC observations are exceptionally scarce, and field conditions are particularly challenging. The long-term micrometeorological measurements presented here provide critical insights into how carbon and water fluxes respond to contrasting ecosystems and management regimes. By identifying the dominant physical and physiological mechanisms governing ecosystem-scale carbon-water coupling, our results not only improve process-level understanding but also offer practical guidance for landscape management. Conclusion and policy implications At the regional scale, these findings carry important implications for climate assessment, land management, and policy planning across West Africa. Clear and consistent contrasts in water-use efficiency (WUE) emerge among ecosystem types. Savanna forests maintained the highest WUE in both seasons, reaching up to 3.1 ± 0.26 g C kg⁻¹ H₂O during the rainy season, followed by paddy rice (2.08 ± 0.21 g C kg⁻¹ H₂O), croplands (1.93 ± 0.20 g C kg⁻¹ H₂O), and grasslands (1.66 ± 0.18 g C kg⁻¹ H₂O). These ecosystem-level differences highlight the dominant role of forests in regulating regional carbon and water exchanges. The superior WUE of forests reflects their unique biophysical and physiological characteristics 62 , 63 . Overall, forests represent the moistest environments, benefiting from reduced aerodynamic drying and strong canopy-atmosphere coupling, which together support high productivity alongside substantial evapotranspiration. Forest gross primary productivity was approximately 2.5-3 times higher than that of non-forest systems, while evapotranspiration was 20–65% greater. This combination resulted in WUE values that exceeded those of other ecosystems by 60–80%, underscoring the efficiency with which forests convert water into biomass. In contrast, non-forest ecosystems exhibited distinct constraints that limited their WUE. Paddy rice systems exhibited relatively low GPP and WUE, likely due to flooded, anaerobic soil conditions and short vegetation cycles. Croplands were primarily constrained by strong seasonality and reduced radiation availability, leading to moderate productivity and WUE. Grasslands, while maintaining moderate GPP, experienced comparatively high evapotranspiration, resulting in the lowest WUE, an outcome consistent with open herbaceous systems that exert limited canopy control over water losses. Seasonal analyses further revealed a pronounced shift in controlling factors, from water limitation during the dry season to stronger energy and atmospheric regulation during the wet season. This transition highlights the adaptive coupling between hydrological availability and ecosystem physiological responses across the region. These insights point to several pathways for enhancing WUE in agroecosystems. Effective strategies include the adoption of water-saving irrigation technologies, improvements in soil-water retention through conservation tillage and mulching, as well as the integration of agroforestry practices. At the same time, indiscriminate afforestation should be avoided, as it may exacerbate water scarcity in already water-limited environments 59 , 62 – 65 . Nature-based solutions that prioritize the restoration of drought-adapted and water-efficient native species offer a more sustainable approach, enabling carbon sequestration gains while maintaining regional water balance. Finally, given the severe scarcity/absence of eddy covariance observations in West Africa, expanding multi-site flux monitoring networks is critical. Continuous measurements of carbon, water, and energy fluxes, integrated with soil-plant-atmosphere remote sensing and land-surface modeling, will be essential for reducing uncertainty and improving regional assessments. Such efforts will provide the evidence base needed to support climate-smart agriculture, drought mitigation strategies, and sustainable land-management policies. Data and Methods Study area and site descriptions In line with previous studies 60 , 61 , 63 , 66 , the study area (Fig. 1 ) is part of a wider research observatory established over the past decade to monitor CO₂, CH ₄, and H₂O fluxes 17 , 66 – 68 . This monitoring network was developed through a collaborative initiative (WASCAL) in the Sudanian savannah belt of West Africa 17 , 66 , 68 . Two eddy covariance (EC) flux towers were initially established in northern Ghana: one located in actively cultivated, rainfed cropland (Kayoro) within the Tono River catchment 17 , 67 , and another in semi-degraded grassland occasionally used as pasture by local communities (Gorigo) 68 . The latter site is located in the upper section of the Vea watershed, a sparsely populated area near the Ghana-Burkina Faso border (Fig. 1 ). The two towers were established in October 2013 and March 2017, respectively 68 . To enhance understanding of ecosystem-scale CO₂ and H₂O fluxes across various land-use systems, the observatory network has recently been expanded with two additional eddy covariance towers 66 . These new installations target rainfed paddy rice cultivation and protected savanna forest ecosystems 66 . The rice field site (Janga; installed in May 2022) is situated in a poorly drained floodplain prone to seasonal flooding 69 , 70 . Agricultural activities are concentrated in these low-lying areas, where rice and maize are the dominant crops cultivated during the rainy season 66 . The forest site (Mole; installed in June 2023) is situated in a protected reserve designated as a national park in 1958 and classified as a Category II protected area under the IUCN guidelines 66 , 71 , characterized by savanna woodland interspersed with wetlands and a floodplain ecosystem 66 . During the peak of the rainy season, the savanna forest can reach heights of up to six meters 71 . The soils are primarily iron-rich Ferralsols and structurally stable Nitisols, characterized by good drainage properties 72 . Key characteristics of the four EC monitoring sites are summarized in Table 1 . Table 1 Key characteristics of the eddy covariance monitoring sites used in this study . Note that for paddy rice and cropland sites, cropping and harvest periods vary annually with the rainy season, typically ranging from June 15 to July 15 for cropping and from October 5 to November 5 for harvest. Ecosystem characteristics Reserve forest Paddy rice Cropland Grassland Site names Mole National Park Janga Kayoro Gorigo Latitude (°N) 9.3388 10.1299 10.98 10.94016 Longitude (°W) −1.8688 −0.8838 −1.3210 −0.8264 EC Sensors IRGASON IRGASON LI-7500 LI-7500 Data coverage May 2023 to June 2024 July 2022 to May 2024 Jan-2020 to Dec-2023 Jan-2019 to Dec-2023 Land management Nature reserve Rainfed Rice Mix crops (soybean, millet) Highly degraded used for grazing Land cover type Pristine savanna Paddy Rice Mixture of fallow and cropland Grassland Land use intensity Very low High High High EC height (m) 12 4.3 3.1 4.82 Canopy height (m) 6 1 0.5 0.17 Elevation (m) 160.00 130.00 286.00 222.00 Soil texture Loamy sandy Loamy sandy Loamy sandy Loamy sandy Soil bulk density (g cm-3) 1.55 1.50 1.71 1.65 Sensor direction (Degrees) 337.00 337.00 132.00 139.00 References Guug et al., 2025 Guug et al., 2025 Bliefernicht et al., 2018) Bliefernicht et al., 2018) Climatically, all sites are influenced by the West African Monsoon (WAM) 73 , characterized by a wet season (May-October) and a dry season (November-April) (see Figure S5). Mean annual rainfall ranges from 700 to 1100 mm 74 , largely controlled by intermittent mesoscale convective systems that cause substantial spatiotemporal variability 73 . Average daily temperatures range between 22°C and 34°C, with pre-monsoon peaks frequently exceeding 40°C (March-April) 66 , 75 . Microclimatic Observations All four monitoring sites were equipped with nearly identical eddy covariance (EC) systems, following standardized measurement protocols but with site-specific installation heights 66 – 68 (see Table 1 ). At Kayoro, Gorigo, and Janga, hereinafter referred to as cropland, grassland, and paddy rice, turbulent fluxes of CO₂ and water vapor were measured at heights of 3.1 m, 4.8 m, and 4.3 m, respectively, using a sampling frequency of 20 Hz. Flux measurements at cropland and grassland sites were obtained using a LI-7500 open-path infrared gas analyzer (LI-COR Inc., Lincoln, USA) coupled with a three-dimensional ultrasonic anemometer (Campbell Scientific Inc., Logan, UT). In contrast, the paddy rice field utilized an integrated IRGASON system. At the Mole Forest Reserve, hereinafter referred to as the reserve savanna forest, flux data were collected using an IRGASON sensor installed at 12 m and operating at 10 Hz to account for the forest canopy structure and ensure the representativeness of the ecosystem at the forest scale. Meteorological variables, including air temperature, relative humidity, solar radiation, and precipitation, were continuously monitored and logged at 30-minute intervals in accordance with standard EC tower protocols 76 . Solar radiation was measured using a CNR4 net radiometer (Kipp & Zonen), and ground heat flux (G) was quantified using three self-calibrating heat flux plates installed at a depth of 8 cm to evaluate the components of the surface energy balance (see Figure S1 ). Soil parameters, including volumetric soil water content (SWC) and soil temperature, were measured at depths of 3, 10, and 30 cm and used to correct soil heat flux (G) for soil heat storage effects 60 , 77 . Precipitation was monitored by two complementary sensors: a tipping-bucket rain gauge (model 52203, R.M. Young) positioned 1 m above the ground, and a weighing gauge (Pluvio2, OTT). Although eddy covariance monitoring began at the cropland site in 2013 and at the grassland site in 2017 17,67,68 , this study focuses on recently collected data, i.e., between 2019 and 2024 (see Table 1 ). Earlier records contained substantial temporal gaps due to limited data-retrieval capacity, which impacted data quality and continuity. For the paddy rice field, following previous study 66 , CO₂ and water vapor flux measurements from the rainy season (May-October) were used, corresponding to the rice-growing period. This timeframe provides the most consistent and representative basis for comparing WUE across the different ecosystem types. However, during the dry season (November-March), the rice field and cropland sites are typically fallow, resulting in minimal GPP due to the absence of active vegetation, though soil evaporation may still occur from residual moisture 17 . Flux Data Processing High-frequency turbulent flux data were processed using the Turbulence Knight (TK3) software package 78 , 79 . Following the methodology of Grünwald and Bernhofer (2007) 80 , half-hourly fluxes of CO₂ and latent heat were computed. Data quality was evaluated according to the flagging criteria proposed by Mauder et al. (2013) 81 . Missing data were gap-filled following established procedures 17 , 66 to obtain daily, monthly, and annual flux aggregates. Missing precipitation records were supplemented with data from nearby meteorological stations 61 , 66 . Energy balance closure (EBC) was assessed to evaluate the consistency of eddy-covariance fluxes. Using 30-minute averages, EBC residuals ranged from 5% to 30% across sites (Figure S1 ), consistent with values reported for FLUXNET sites 82 , 83 . Notably, grassland and reserve forest sites showed the best closure (see Figure S1 ). To account for energy imbalance in subsequent evapotranspiration analyses, fluxes were corrected using the quantile mapping approach 60 . The eddy covariance flux footprint, which covered most of each ecosystem (see Figure S7), was calculated for each site using the method described by Kljun et al. (2015) 84 . Formal Analysis Water use efficiency was estimated as the ratio of GPP to ET (Eq. 1). GPP (g C m⁻² time⁻¹) represents the total carbon assimilated through vegetation photosynthesis 4 , 22 , 85 (Chapin et al., 2011), while ET (kgH₂O m⁻² time⁻¹) quantifies total water loss via transpiration, canopy interception, and soil evaporation. Because it is often impractical to separate these individual components in field observations, ET is generally used as an integrated measure of ecosystem water loss 6 , 9 , 15 , 16 . ET was derived from latent heat flux (LE): ET = LE / λ, where λ = 2454 J g⁻¹ following standard formulations 4 , 7 , 85 , 86 . The computed ET values were used to calculate WUE (g C kgH₂O⁻¹). $$\:\begin{array}{c}WUE=\frac{\text{G}\text{P}\text{P}}{\text{E}\text{T}}\:\#\left(1\right)\end{array}$$ GPP can be expressed as follows $$\:\begin{array}{c}GPP=-NEE+Re\:\#\left(2\right)\end{array}$$ where NEE represents the net flux of CO₂ between an ecosystem and the atmosphere (g C m⁻² time⁻¹), integrating carbon uptake via photosynthesis with carbon losses from respiration. This flux can be directly measured using eddy covariance techniques 4 , 17 , 67 . Total ecosystem respiration, Re, captures the combined respiratory carbon efflux from all ecosystem components (g C m⁻² time⁻¹). This study also evaluated underlying water use efficiency (uWUE; g C hPa 0. 5 kgH₂O⁻¹), a refined indicator that more accurately represents the relationship between GPP and ET and captures long-term variations in the photosynthesis-transpiration balance across ecosystem types 4 , 86 , 87 . UWUE was calculated as: $$\:\begin{array}{c}uWUE=\frac{\text{G}\text{P}\text{P}\:\times\:\:{\text{V}\text{P}\text{D}}^{0.5}}{\text{E}\text{T}}\#\left(3\right)\end{array}$$ where VPD is the vapor pressure deficit (hPa). The response of uWUE to environmental variability reflects potential physiological adjustments of ecosystems to changing weather conditions 4 . Compared with conventional WUE, uWUE provides a more robust description of plant biochemical processes by accounting for VPD influences 4 , 56 , 86 , 87 . To identify the environmental factors influencing WUE responses across different ecosystem types, the non-parametric Kendall rank correlation test 88 was employed. This method was selected because it is less affected by outliers and non-linear distributions than the Pearson or Spearman correlation tests 89 . Furthermore, analysis of variance (ANOVA) was implemented to quantify the proportion of variance in WUE explained by each potential explanatory variable, including key environmental factors across different ecosystem types. To account for temporal autocorrelation, which is often present in weather data measurements (e.g., temperature), the analysis incorporated a first-order autoregressive (AR(1)) model within a multivariable regression model 90 , expressed as follows: $$\:{y}_{t}\:=\:\alpha\:+{\beta\:}_{1}{x}_{t}^{1}+{\beta\:}_{2}{x}_{t}^{2}+\dots\:+{\beta\:}_{n}{x}_{t}^{n}+{\mu\:}_{t}\:,\:\:\:\:\:\:\:t=1,\dots\:.,T\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\left(4\right)$$ where 𝑦 𝑡 are the daily WUE values, 𝑡 is the time variable assigned to 𝑦 𝑡 , 𝛼 is a constant term, \(\:{x}_{t}^{1},\:{x}_{t}^{2},\:and\:\:{x}_{t}^{n}\) and denote key environmental variables such as temperature, relative humidity, vapor pressure deficit, soil water moisture at 3 cm and 30 cm depths, and solar radiation, across each ecosystem type. The coefficients, 𝛽 1 , 𝛽 2 , and 𝛽 n , represent the associated linear trends, and 𝜇 𝑡 represents the residual term that is assumed to be autoregressive of the order of 1 [AR (1)] and may be related as follows: $$\:{\mu\:}_{t}=\:⍴{\mu\:}_{t-1}+\:{\epsilon\:}_{t},\:\:\:t=2,\dots\:.,T\:\:\:\:\:\:\left(5\right)$$ Here, ρ represents the autocorrelation coefficient, and 𝜀 𝑡 ​denotes the white noise term, an independent random variable with zero mean and constant variance 90 . We also assumed that − 1 < ρ < 1, so the noise process 𝜇 𝑡 is stationary. This formulation allows for autocorrelation in the residuals of the daily WUE values, such that ρ = Corr (𝜇 𝑡 , 𝜇 𝑡 −1). It is essential to note that, due to limited data availability/information during the study period, factors such as field fertilization, plowing, and land preparation activities were not considered in the study. To reduce multicollinearity and ensure robust model estimates, we applied a hierarchical predictor selection procedure 91 , 92 . For structurally dependent predictors, such as air temperature, relative humidity, and vapor pressure deficit, only the variable most strongly associated with WUE is considered, allowing for the determination of the dominant atmospheric driver. For soil moisture, only the layer (3 cm or 30 cm) most correlated with WUE was retained. The remaining predictors were screened using pairwise correlations, removing variables with |r| > 0.7 that contributed less explanatory power to WUE. Finally, variance inflation factors (VIFs) were computed, and predictors with VIFs greater than 5 were iteratively excluded to ensure acceptable levels of collinearity 92 – 94 . Declarations Competing interests: The authors declare that they have no competing financial interests. Author Contribution Conceptualization: S.S.; Methodology: S.S., J.B., S.G., K.M., L.H.; Software: S.S., L.H., S.G.; Validation: S.S., J.B., R.K., R.S., F.N., T.J., P.L.; Formal analysis: S.S., B.Q., R.K.; Investigation: S.S., S.G., S.Sa., M.W., A.K.; Resources: F.E.O., E.Q., L.K.A.; Data curation: S.S., J.B., L.H., S.G., P.D.; Writing – original draft: S.S.; Writing – review & editing: All authors; Visualization: S.S.; Supervision: H.K., J.B.; Project administration: H.K.; Funding acquisition: H.K. Acknowledgement This research was supported by the German Federal Ministry of Education and Research (BMBF) through the West African Science Service Centre on Climate Change and Adapted Land Use (WASCAL) and the Concerted Regional Modelling and Observation Assessment for Greenhouse Gas Emissions and Mitigation Options under Climate and Land Use Change in West Africa (CONCERT-West Africa; grant no. 01LG2089A BMBF). The authors gratefully acknowledge the cooperation of the Mole National Park authorities, as well as the landowners of the Janga rice field, Kayoro, and Gorigo sites, for granting permission to install and operate the eddy covariance (EC) stations. Data Availability The datasets analyzed during the current study are not publicly available due to WASCAL’s Data Sharing Policy and Guidelines, but are available from the corresponding author upon reasonable request. References Gentine, P. et al. Coupling between the terrestrial carbon and water cycles—a review. Environ. Res. Lett. 14, 083003 (2019). Ponton, S. et al. Comparison of ecosystem water-use efficiency among Douglas-fir forest, aspen forest and grassland using eddy covariance and carbon isotope techniques. Glob. Change Biol. 12, 294–310 (2006). Fu, Y., Jian, S. & Yu, X. Water use efficiency in China is impacted by climate change and land use and land cover. Environ. Sci. Pollut. Res. 31, 42840–42856 (2024). Song, Q.-H. et al. Water use efficiency in a primary subtropical evergreen forest in Southwest China. Sci. Rep. 7, 43031 (2017). An, X. Responses of Water Use Efficiency to climate change in evapotranspiration and transpiration ecosystems. Ecol. Indic. 141, 109157 (2022). Ai, Z. et al. Variation of gross primary production, evapotranspiration and water use efficiency for global croplands. Agric. For. Meteorol. 287, 107935 (2020). Tang, J., Luyssaert, S., Richardson, A. D., Kutsch, W. & Janssens, I. A. Steeper declines in forest photosynthesis than respiration explain age-driven decreases in forest growth. Proc. Natl. Acad. Sci. 111, 8856–8860 (2014). Wang, T. et al. Differences in ecosystem water-use efficiency among the typical croplands. Agric. Water Manag. 209, 142–150 (2018). Wang, C. et al. Negative impacts of plant diversity loss on carbon sequestration exacerbate over time in grasslands. Environ. Res. Lett. 15, 104055 (2020). Liu, Y., Peñuelas, J., Cescatti, A., Zhang, Y. & Zhang, Z. Atmospheric Dryness Dominates Afternoon Depression of Global Terrestrial Photosynthesis. Geophys. Res. Lett. 51, e2024GL110954 (2024). Ago, E. E., Agbossou, E. K., Cohard, J.-M., Galle, S. & Aubinet, M. Response of CO2 fluxes and productivity to water availability in two contrasting ecosystems in northern Benin (West Africa). Ann. For. Sci. 73, 483–500 (2016). Keenan, T. F. et al. Increase in forest water-use efficiency as atmospheric carbon dioxide concentrations rise. Nature 499, 324–327 (2013). Medlyn, B. E. et al. How do leaf and ecosystem measures of water-use efficiency compare? New Phytol. 216, 758–770 (2017). Wang, Y. et al. Elucidating climatic drivers of photosynthesis by tropical forests. Glob. Change Biol. 29, 4811–4825 (2023). Yu, G. et al. Water-use efficiency of forest ecosystems in eastern China and its relations to climatic variables. New Phytol. 177, 927–937 (2008). Huang, M., Zhai, P. & Piao, S. Divergent responses of ecosystem water use efficiency to drought timing over Northern Eurasia. Environ. Res. Lett. 16, 045016 (2021). Quansah, E. et al. Carbon dioxide fluxes from contrasting ecosystems in the Sudanian Savanna in West Africa. Carbon Balance Manag. 10, 1 (2015). Ago, E. E., Serça, D., Agbossou, E. K., Galle, S. & Aubinet, M. Carbon dioxide fluxes from a degraded woodland in West Africa and their responses to main environmental factors. Carbon Balance Manag. 10, 22 (2015). Boulain, N. et al. Towards an understanding of coupled physical and biological processes in the cultivated Sahel – 2. Vegetation and carbon dynamics. J. Hydrol. 375, 190–203 (2009). Kayiranga, A. et al. Water Use Efficiency-Based Multiscale Assessment of Ecohydrological Resilience to Ecosystem Shifts Over the Continent of Africa During 1992–2015. J. Geophys. Res. Biogeosciences 125, e2020JG005749 (2020). Farooqi, T. J. A. et al. Advancing Knowledge in Forest Water Use Efficiency Under Global Climate Change Through Scientometric Analysis. Forests 15, 1893 (2024). Yang, Y. et al. Contrasting responses of water use efficiency to drought across global terrestrial ecosystems. Sci. Rep. 6, 23284 (2016). Tagesson, T. et al. Dynamics in carbon exchange fluxes for a grazed semi-arid savanna ecosystem in West Africa. Agric. Ecosyst. Environ. 205, 15–24 (2015). Taylor, S. H. et al. Physiological advantages of C4 grasses in the field: a comparative experiment demonstrating the importance of drought. Glob. Change Biol. 20, 1992–2003 (2014). Hutley, L. B., Beringer, J., Fatichi, S., Schymanski, S. J. & Northwood, M. Gross primary productivity and water use efficiency are increasing in a high rainfall tropical savanna. Glob. Change Biol. 28, 2360–2380 (2022). Friborg, T., Boegh, E. & Soegaard, H. Carbon dioxide flux, transpiration and light response of millet in the Sahel. J. Hydrol. 188–189, 633–650 (1997). Wang, T. et al. Differences in ecosystem water-use efficiency among the typical croplands. Agric. Water Manag. 209, 142–150 (2018). Tagesson, T. et al. Ecosystem properties of semiarid savanna grassland in West Africa and its relationship with environmental variability. Glob. Change Biol. 21, 250–264 (2015). Williams, C. A. et al. Africa and the global carbon cycle. Carbon Balance Manag. 2, 3 (2007). Bliefernicht, J. et al. Field and simulation experiments for investigating regional land–atmosphere interactions in West Africa: experimental set-up and first results. Ago, E. E. et al. Long term observations of carbon dioxide exchange over cultivated savanna under a Sudanian climate in Benin (West Africa). Agric. For. Meteorol. 197, 13–25 (2014). Ramier, D. et al. Towards an understanding of coupled physical and biological processes in the cultivated Sahel – 1. Energy and water. J. Hydrol. 375, 204–216 (2009). Chen, Y. et al. Quantifying the variability in water use efficiency from the canopy to ecosystem scale across main croplands. Agric. Water Manag. 262, 107427 (2022). Xue, Y., Liang, H., Zhang, B. & He, C. Vegetation restoration dominated the variation of water use efficiency in China. J. Hydrol. 612, 128257 (2022). Du, D. et al. Spatiotemporal dynamics of irrigated cropland water use efficiency and driving factors in northwest China’s Hexi Corridor. Ecol. Process. 13, 72 (2024). El Masri, B. et al. Carbon and Water Use Efficiencies: A Comparative Analysis of Ten Terrestrial Ecosystem Models under Changing Climate. Sci. Rep. 9, 14680 (2019). Franco, A. & Lüttge, U. Midday depression in savanna trees: coordinated adjustments in photochemical efficiency, photorespiration, CO2 assimilation and water use efficiency. Oecologia 131, 356–365 (2002). Liu, J., Wang, Q., Zhan, W., Lian, X. & Gentine, P. When and where soil dryness matters to ecosystem photosynthesis. Nat. Plants 11, 1390–1400 (2025). Wang, X. et al. Increasing atmospheric dryness exacerbates mangrove carbon–water decoupling. New Phytol. 249, 169–180 (2026). Li, Y., Zhang, K. & Liu, L. Water use efficiency at multi-time scales and its response to episodic drought and wet periods in a typical subtropical evergreen forest of Southeast China. Ecol. Indic. 151, 110254 (2023). Beer, C. et al. Temporal and among-site variability of inherent water use efficiency at the ecosystem level. Glob. Biogeochem. Cycles 23, (2009). Williams, C. A. & Albertson, J. D. Soil moisture controls on canopy-scale water and carbon fluxes in an African savanna. Water Resour. Res. 40, (2004). Machado, R. et al. Where do leaf water leaks come from? Trade-offs underlying the variability in minimum conductance across tropical savanna species with contrasting growth strategies. New Phytol. 229, 1415–1430 (2021). Hu, Z. Y. et al. Dissecting the Characteristics and Driver Factors on Global Water Use Efficiency Using GLASS Data Sets. Earths Future 12, e2024EF004630 (2024). Caine, R. S. et al. Rice with reduced stomatal density conserves water and has improved drought tolerance under future climate conditions. New Phytol. 221, 371–384 (2019). Govindaraj, M. et al. Evidence from simulated climatic conditions indicates rising CO2 levels impact pearl millet yield and nutritional traits. J. Agric. Food Res. 22, 102124 (2025). MEDRANO, H., ESCALONA, J. M., BOTA, J., GULÍAS, J. & FLEXAS, J. Regulation of Photosynthesis of C3 Plants in Response to Progressive Drought: Stomatal Conductance as a Reference Parameter. Ann. Bot. 89, 895–905 (2002). Roche, D. Stomatal Conductance Is Essential for Higher Yield Potential of C3 Crops. Crit. Rev. Plant Sci. 34, 429–453 (2015). Sales, C. R. G., Wang, Y., Evers, J. B. & Kromdijk, J. Improving C4 photosynthesis to increase productivity under optimal and suboptimal conditions. J. Exp. Bot. 72, 5942–5960 (2021). Nguyen, T. H. et al. Responses of winter wheat and maize to varying soil moisture: From leaf to canopy. Agric. For. Meteorol. 314, 108803 (2022). Novick, K. A. et al. The impacts of rising vapour pressure deficit in natural and managed ecosystems. Plant Cell Environ. 47, 3561–3589 (2024). Lin, Y.-S. et al. Optimal stomatal behaviour around the world. Nat. Clim. Change 5, 459–464 (2015). Katul, G. G., Palmroth, S. & Oren, R. Leaf stomatal responses to vapour pressure deficit under current and CO2-enriched atmosphere explained by the economics of gas exchange. Plant Cell Environ. 32, 968–979 (2009). Thomas, D. S., Eamus, D. & Bell, D. Optimization theory of stomatal behaviour: II. Stomatal responses of several tree species of north Australia to changes in light, soil and atmospheric water content and temperature. J. Exp. Bot. 50, 393–400 (1999). Antoniuk, V. et al. Diurnal and Seasonal Mapping of Water Deficit Index and Evapotranspiration by an Unmanned Aerial System: A Case Study for Winter Wheat in Denmark. Remote Sens. 13, 2998 (2021). Zhou, S., Yu, B., Huang, Y. & Wang, G. The effect of vapor pressure deficit on water use efficiency at the subdaily time scale. Geophys. Res. Lett. 41, 5005–5013 (2014). Knauer, J. et al. The response of ecosystem water-use efficiency to rising atmospheric CO2 concentrations: sensitivity and large-scale biogeochemical implications. New Phytol. 213, 1654–1666 (2017). Kim, H. K., Jang, T. I., Im, S. J. & Park, S. W. Estimation of irrigation return flow from paddy fields considering the soil moisture. Agric. Water Manag. 96, 875–882 (2009). Bwire, D., Saito, H., Sidle, R. C. & Nishiwaki, J. Water Management and Hydrological Characteristics of Paddy-Rice Fields under Alternate Wetting and Drying Irrigation Practice as Climate Smart Practice: A Review. Agronomy 14, 1421 (2024). Nadolski, L. et al. Exploring and Closing the Energy Balance of Eddy Covariance Measurements along a Land Use Gradient in the West African Sudanian Savanna. Front. Water 6, 1393884 (2024). Hingerl, L. et al. Comparative analysis of land–atmosphere interactions across three contrasting ecosystems in the West Sudanian Savanna. J. Hydrol. Reg. Stud. 61, 102751 (2025). Yahaya Seydou, A. N. et al. Biophysical effects of land cover changes in West Africa: a systematic review. Environ. Res. Lett. 20, 073001 (2025). Sy, S. et al. Impacts of Idealized Land Use and Land Management Changes on Weather Extremes in West Africa. Earths Future 13, e2025EF006094 (2025). Obahoundje, S., Tilahun, S. A. & Schmitter, P. Hydroclimatic extremes and food security hotspots in West Africa: evidence from rainfall–yield interactions. Front. Clim. 7, (2025). Sy, S. & Quesada, B. Anthropogenic land cover change impact on climate extremes during the 21st century. Environ. Res. Lett. 15, 034002 (2020). Guug, S. et al. Methane emissions from rice cultivation in West Africa and compensation options from nature reserve forests. Environ. Res. Lett. 20, 044050 (2025). Berger, S. et al. The impact of rain events on CO2 emissions from contrasting land use systems in semi-arid West African savannas. Sci. Total Environ. 647, 1478–1489 (2019). Bliefernicht, J. et al. The WASCAL Hydrometeorological Observatory in the Sudan Savanna of Burkina Faso and Ghana. Vadose Zone J. 17, 180065 (2018). Nyadzi, E., Bessah, E., Kranjac-Berisavljevic, G. & Ludwig, F. Hydro-climatic and land use/cover changes in Nasia catchment of the White Volta basin in Ghana. Theor. Appl. Climatol. 146, 1297–1314 (2021). Abdul-Ganiyu, S. Analysis of some hydrological parameters of the Nasia River Catchment in the Northern Region of Ghana for its socio-economic development. Afr. J. Agric. RESEEARCH 6, (2011). Schmitt, K. & Adu-Nsiah, M. The Vegetation of Mole National Park . (1993). Bowell, R. J. & Ansah, R. K. Mineral status of soils and forage in the Mole National Park, Ghana and implications for wildlife nutrition. Environ. Geochem. Health 16, 41–58 (1994). Nicholson, S. E. The West African Sahel: A Review of Recent Studies on the Rainfall Regime and Its Interannual Variability. Int. Sch. Res. Not. 2013, 453521 (2013). Bliefernicht, J. et al. The WASCAL hydrometeorological observatory in the Sudan Savanna of Burkina Faso and Ghana. Vadose Zone J. 17, 1–20 (2018). Sy, S. et al. Impacts of Idealized Land Use and Land Management Changes on Weather Extremes in West Africa. Earths Future 13, e2025EF006094 (2025). Charuchittipan, D., Babel, W., Mauder, M., Leps, J.-P. & Foken, T. Extension of the Averaging Time in Eddy-Covariance Measurements and Its Effect on the Energy Balance Closure. Bound.-Layer Meteorol. 152, 303–327 (2014). Kutikoff, S. et al. Heat storage and its effect on the surface energy balance closure under advective conditions. Agric. For. Meteorol. 265, 56–69 (2019). Mauder, M. & Foken, T. Documentation and Instruction Manual of the Eddy-Covariance Software Package TK3. Foken, T. et al. Post-Field Data Quality Control. in Handbook of Micrometeorology: A Guide for Surface Flux Measurement and Analysis (eds Lee, X., Massman, W. & Law, B.) 181–208 (Springer Netherlands, Dordrecht, 2005). doi: 10.1007/1-4020-2265-4_9 . GrüNwald, T. & Bernhofer, C. A decade of carbon, water and energy flux measurements of an old spruce forest at the Anchor Station Tharandt. Tellus B Chem. Phys. Meteorol. 59, 387–396 (2007). Mauder, M. et al. A strategy for quality and uncertainty assessment of long-term eddy-covariance measurements. Agric. For. Meteorol. 169, 122–135 (2013). Mauder, M., Jung, M., Stoy, P., Nelson, J. & Wanner, L. Energy balance closure at FLUXNET sites revisited. Agric. For. Meteorol. 358, 110235 (2024). Wilson, K. et al. Energy balance closure at FLUXNET sites. Agric. For. Meteorol. 113, 223–243 (2002). Kljun, N., Calanca, P., Rotach, M. W. & Schmid, H. P. A simple two-dimensional parameterisation for Flux Footprint Prediction (FFP). Geosci. Model Dev. 8, 3695–3713 (2015). Wang, D. et al. The lag and cumulative response of water use efficiency to the climate on the Shiyang river basin, Northwest of China. Sci. Rep. 15, 23659 (2025). Zhou, S., Yu, B., Zhang, Y., Huang, Y. & Wang, G. Partitioning evapotranspiration based on the concept of underlying water use efficiency. Water Resour. Res. 52, 1160–1175 (2016). Zhou, S., Yu, B., Huang, Y. & Wang, G. Daily underlying water use efficiency for AmeriFlux sites. J. Geophys. Res. Biogeosciences 120, 887–902 (2015). Croux, C. & Dehon, C. Influence functions of the Spearman and Kendall correlation measures. Stat. Methods Appl. 19, 497–515 (2010). Diouf, I. et al. Potential Contribution of Climate Conditions on COVID-19 Pandemic Transmission over West and North African Countries. Atmosphere 13, 34 (2022). Sy, S. et al. Sensitivity of trends to estimation methods and quantification of subsampling effects in global radiosounding temperature and humidity time series. Int. J. Climatol. 41, E1992–E2014 (2021). Kroll, C. N. & Song, P. Impact of multicollinearity on small sample hydrologic regression models. Water Resour. Res. 49, 3756–3769 (2013). Muthukrishnan, R. & Lowast, C. K. J. The Effect of Multicollinearity on Feature Selection. Indian J. Sci. Technol. 17, 3664–3668 (2024). Naseri, M. Y., Bernosky, G., Mayer, P. W. & Marston, L. T. Patterns and Predictors of Residential Indoor Water Use Across Major US Cities. Earths Future 13, e2024EF005467 (2025). Dertli, H. I., Hayes, D. B. & Zorn, T. G. Effects of multicollinearity and data granularity on regression models of stream temperature. J. Hydrol. 639, 131572 (2024). Additional Declarations No competing interests reported. Supplementary Files SyetalWUESupplementaryInformationv07012026.pdf Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 17 Feb, 2026 Reviews received at journal 29 Jan, 2026 Reviewers agreed at journal 19 Jan, 2026 Reviewers invited by journal 13 Jan, 2026 Editor assigned by journal 12 Jan, 2026 Submission checks completed at journal 08 Jan, 2026 First submitted to journal 05 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8522874","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":570898728,"identity":"fd31c10a-5498-43b7-a264-fa4791c4cc52","order_by":0,"name":"Souleymane Sy","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/0lEQVRIiWNgGAWjYHACxgMIdgUDAxuI5iGgB0nLGZK1MLZBGfi0yLf3GBz42LZN3py9gfkz77zDeXwMzA8Y3lTg1mJw5ozBwZlttw139hxgk+bddriYjYHNgHHOGTxaJHIMDvO23WbccCOBjRmoJbFN/oEBM28bbi3yM4Ba/rbdtt9w/wHQYXOAWhjYPzDz/sPjmRtALYxttxM33GBgkOZtAGnhAdrSgM8vxwoO9py7nbzhTGKb5Jxj6SAtBQfnHMPjsPbmjQ9+lN223XD88OEPb2qsE+c3sG988KYGj8NAgBEcfYwNTLDoOEBAAxD8gWr9QVjpKBgFo2AUjEAAAOYlV3I5TraWAAAAAElFTkSuQmCC","orcid":"","institution":"University of Augsburg","correspondingAuthor":true,"prefix":"","firstName":"Souleymane","middleName":"","lastName":"Sy","suffix":""},{"id":570898730,"identity":"d48b1545-3229-43c1-94eb-e8b3c208ad91","order_by":1,"name":"Jan Bliefernicht","email":"","orcid":"","institution":"University of Augsburg","correspondingAuthor":false,"prefix":"","firstName":"Jan","middleName":"","lastName":"Bliefernicht","suffix":""},{"id":570898734,"identity":"9e255628-f106-4df9-9fdb-d569ae9afaee","order_by":2,"name":"Kiril Manevski","email":"","orcid":"","institution":"Aarhus University","correspondingAuthor":false,"prefix":"","firstName":"Kiril","middleName":"","lastName":"Manevski","suffix":""},{"id":570898735,"identity":"405d84da-64f4-42c7-866c-edb15631bc39","order_by":3,"name":"Samuel Guug","email":"","orcid":"","institution":"West African Science Service Centre on Climate Change and Adapted Land Use, WASCAL Competence Centre, Ouagadougou, Burkina Faso","correspondingAuthor":false,"prefix":"","firstName":"Samuel","middleName":"","lastName":"Guug","suffix":""},{"id":570898737,"identity":"2ba48661-405f-4455-bea8-ac1706a7e6e6","order_by":4,"name":"Luitpold Hingerl","email":"","orcid":"","institution":"Karlsruhe Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Luitpold","middleName":"","lastName":"Hingerl","suffix":""},{"id":570898738,"identity":"378fe2e7-3439-4d24-930a-018099e03661","order_by":5,"name":"Benjamin Quesada","email":"","orcid":"","institution":"Universidad del Rosario","correspondingAuthor":false,"prefix":"","firstName":"Benjamin","middleName":"","lastName":"Quesada","suffix":""},{"id":570898739,"identity":"52610617-cbc4-4ac6-856b-41ed6ae59f32","order_by":6,"name":"Francis E. Oussou","email":"","orcid":"","institution":"Federal University of Technology","correspondingAuthor":false,"prefix":"","firstName":"Francis","middleName":"E.","lastName":"Oussou","suffix":""},{"id":570898740,"identity":"104639fb-8c14-4cf5-946f-8e0390e0bda9","order_by":7,"name":"Ralf Kiese","email":"","orcid":"","institution":"Karlsruhe Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Ralf","middleName":"","lastName":"Kiese","suffix":""},{"id":570898741,"identity":"653fedbc-aec3-45a4-8db6-3ea4612531af","order_by":8,"name":"Patrick Davies","email":"","orcid":"","institution":"Kwame Nkrumah University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Patrick","middleName":"","lastName":"Davies","suffix":""},{"id":570898742,"identity":"1eb200e2-1f4f-490e-bd69-e2e4546ed160","order_by":9,"name":"Thomas Jagdhuber","email":"","orcid":"","institution":"Deutsches Zentrum für Luft- und Raumfahrt (DLR)","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"","lastName":"Jagdhuber","suffix":""},{"id":570898743,"identity":"7f6278d4-baf9-4586-8419-beb3c76e9585","order_by":10,"name":"Patrick Laux","email":"","orcid":"","institution":"Karlsruhe Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Patrick","middleName":"","lastName":"Laux","suffix":""},{"id":570898744,"identity":"1765a459-50bf-4d84-b77b-e8144d0895ea","order_by":11,"name":"Moussa Woango","email":"","orcid":"","institution":"Agrhymet Regional Centre","correspondingAuthor":false,"prefix":"","firstName":"Moussa","middleName":"","lastName":"Woango","suffix":""},{"id":570898745,"identity":"991e5dec-8c04-4e07-9605-d6e12bd55e1b","order_by":12,"name":"Amadou Coulibaly","email":"","orcid":"","institution":"Agrhymet Regional Centre","correspondingAuthor":false,"prefix":"","firstName":"Amadou","middleName":"","lastName":"Coulibaly","suffix":""},{"id":570898746,"identity":"f6ccf1f6-f43b-4fe9-a550-d72479cd6135","order_by":13,"name":"Frank Neidl","email":"","orcid":"","institution":"Karlsruhe Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Frank","middleName":"","lastName":"Neidl","suffix":""},{"id":570898747,"identity":"2781cf83-79ed-492d-a997-a664129cc3fd","order_by":14,"name":"Rainer Steinbrecher","email":"","orcid":"","institution":"Karlsruhe Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Rainer","middleName":"","lastName":"Steinbrecher","suffix":""},{"id":570898748,"identity":"fc0c7c32-4bfe-4c81-84d7-5e2e1e44eac3","order_by":15,"name":"Emmanuel Quansah","email":"","orcid":"","institution":"Kwame Nkrumah University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Emmanuel","middleName":"","lastName":"Quansah","suffix":""},{"id":570898749,"identity":"77a10d04-6eab-45f8-996b-72ce26d05bbe","order_by":16,"name":"Seyni Salack","email":"","orcid":"","institution":"West African Science Service Centre on Climate Change and Adapted Land Use, WASCAL Competence Centre, Ouagadougou, Burkina Faso","correspondingAuthor":false,"prefix":"","firstName":"Seyni","middleName":"","lastName":"Salack","suffix":""},{"id":570898750,"identity":"82ab436c-607f-4ebc-9cae-8042b6373942","order_by":17,"name":"Leonard K. Amekudzi","email":"","orcid":"","institution":"Kwame Nkrumah University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Leonard","middleName":"K.","lastName":"Amekudzi","suffix":""},{"id":570898751,"identity":"19aeee11-1c91-4c40-8a68-9594db12b550","order_by":18,"name":"Harald Kunstmann","email":"","orcid":"","institution":"Karlsruhe Institute of Technology","correspondingAuthor":false,"prefix":"","firstName":"Harald","middleName":"","lastName":"Kunstmann","suffix":""}],"badges":[],"createdAt":"2026-01-05 15:23:36","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8522874/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8522874/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":99752858,"identity":"61eae0fa-a3e5-455c-bb72-698dc1046005","added_by":"auto","created_at":"2026-01-08 04:32:46","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5262173,"visible":true,"origin":"","legend":"","description":"","filename":"SyetalWUEWestAfrica05012026.docx","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/f8ffef900fe8d910aa674ead.docx"},{"id":99798162,"identity":"e58ca89c-af86-4ab2-863d-a093ce510742","added_by":"auto","created_at":"2026-01-08 13:47:28","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17565,"visible":true,"origin":"","legend":"","description":"","filename":"f833bdfaccb64a489e9c7ff15fe006a7.json","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/6d3c2fb89b0507b409b787cd.json"},{"id":99752859,"identity":"1e2f9eb6-d6e5-47da-a543-af9722a666bb","added_by":"auto","created_at":"2026-01-08 04:32:46","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8575999,"visible":true,"origin":"","legend":"","description":"","filename":"SyetalWUESupplementaryInformationSustainableAgriculture.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/1e9282b57b497cabd8afd375.pdf"},{"id":99752867,"identity":"a6835a03-0426-435c-b957-c0d256b53cdf","added_by":"auto","created_at":"2026-01-08 04:32:47","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":209404,"visible":true,"origin":"","legend":"","description":"","filename":"f833bdfaccb64a489e9c7ff15fe006a71enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/2e49f325acc0ef79e8eecc08.xml"},{"id":99752863,"identity":"42d44283-2f5b-4714-ae41-09a4f69150b3","added_by":"auto","created_at":"2026-01-08 04:32:46","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2453165,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/21271d7f5e0e946fe39b6a14.png"},{"id":99752876,"identity":"38834f72-e416-4477-92a3-ccb1d104768c","added_by":"auto","created_at":"2026-01-08 04:32:47","extension":"emf","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":11305072,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.emf","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/041af13c54c425fa50838817.emf"},{"id":99752860,"identity":"dad7aace-2bcb-402b-aea0-33336b2272b5","added_by":"auto","created_at":"2026-01-08 04:32:46","extension":"emf","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4998928,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.emf","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/0e8fc792cb44732d8a58d09a.emf"},{"id":99752866,"identity":"4dac17c3-aa7c-487e-9e65-38fb2507ae3a","added_by":"auto","created_at":"2026-01-08 04:32:47","extension":"emf","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6523464,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.emf","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/d779ba255fdddb9b1c801e06.emf"},{"id":99752877,"identity":"bb5d8033-76c3-4587-b050-854ccc7aaec3","added_by":"auto","created_at":"2026-01-08 04:32:47","extension":"emf","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6229984,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.emf","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/8c3b6ff6767f43ea29fb0c15.emf"},{"id":99798322,"identity":"afbd374c-10af-4c81-bead-ab680f6e99ce","added_by":"auto","created_at":"2026-01-08 13:47:58","extension":"emf","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4983084,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.emf","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/d7e637e35ac0291d79033701.emf"},{"id":99752874,"identity":"7812373e-20f5-4ae6-934d-15a2462c59cd","added_by":"auto","created_at":"2026-01-08 04:32:47","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":341727,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/69311e7d50c77459610ab813.png"},{"id":99797843,"identity":"dd4e7a1e-154b-4344-8c6c-3cc688d9b0fa","added_by":"auto","created_at":"2026-01-08 13:46:44","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":180732,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/c1aba10367c377dd3a198355.png"},{"id":99752861,"identity":"1e7a7fe4-e6cd-4402-a206-403ead6c04e0","added_by":"auto","created_at":"2026-01-08 04:32:46","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34982,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/01d26c94246ec7eef022fe0f.png"},{"id":99799325,"identity":"5866edb7-a516-497d-8117-8b6df3900e66","added_by":"auto","created_at":"2026-01-08 13:49:28","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":51754,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/02083931522062400fac44d5.png"},{"id":99752872,"identity":"c74e7a62-aad0-4e6f-b888-e2a9881a9847","added_by":"auto","created_at":"2026-01-08 04:32:47","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":38407,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/1557442be942dab2db453d38.png"},{"id":99752864,"identity":"fc84e535-9736-4071-8259-3ec900b139ec","added_by":"auto","created_at":"2026-01-08 04:32:47","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34007,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/0fff86aef08813e2ee4f03d5.png"},{"id":99752869,"identity":"edc2188b-c5f6-411a-a9e6-e83c8487ee54","added_by":"auto","created_at":"2026-01-08 04:32:47","extension":"xml","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":212858,"visible":true,"origin":"","legend":"","description":"","filename":"f833bdfaccb64a489e9c7ff15fe006a71structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/468bdbf87717bace1789bb38.xml"},{"id":99752875,"identity":"90ba8d2f-b645-43a5-bf1a-89e50df1ed85","added_by":"auto","created_at":"2026-01-08 04:32:47","extension":"html","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":230678,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/1578ebec462d780d6141cdd8.html"},{"id":99798259,"identity":"c640ec07-c1aa-4188-b0f5-164064958871","added_by":"auto","created_at":"2026-01-08 13:47:43","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":933831,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eStudy area and site characteristics\u003c/strong\u003e: (a) Regional topography (m above sea level) across West Africa (D1) and the WASCAL observatory network in the Sudanian savannas of Burkina Faso and Ghana, together with the dominant land-cover distribution within the study region (D2). Land-cover information is derived from the MODIS land-cover product based on the International Geosphere-Biosphere Programme (IGBP) 21-category classification (Sulla-Menashe et al., 2019). (b) Photographs of the eddy covariance (EC) monitoring sites and their associated ecosystem types: reserve forest (Mole National Park), paddy rice field (Janga), grassland (Gorigo), and cropland (Kayoro).\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/3405350a7c9baecab31f76d8.png"},{"id":99752854,"identity":"df136c81-fdea-4144-8668-71feee203bf9","added_by":"auto","created_at":"2026-01-08 04:32:46","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":710947,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDaily variation in key meteorological and ecosystem exchange variables at the four study sites\u003c/strong\u003e. Variables include air temperature (Ta), relative humidity (RH), soil water content (SWC) at 5 cm, vapor pressure deficit (VPD), solar radiation (Rg), gross primary productivity (GPP), and evapotranspiration (ET) at the reserve \u003cstrong\u003eforest (a), \u003c/strong\u003epaddy rice (b), cropland\u003cstrong\u003e (c) \u003c/strong\u003eand\u003cstrong\u003e grassland (d)\u003c/strong\u003e sites. Each dot represents a daily value from 2019 to 2024. Grey shading indicates the \u003cstrong\u003ewet season\u003c/strong\u003e (May to October). Annual mean values ± 95% CI for each variable are provided in the \u003cstrong\u003etop-left panel.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/d1bccf94ea0918649f74d76b.png"},{"id":99752856,"identity":"a22ccf02-e57f-48f6-9816-e152e1f85e66","added_by":"auto","created_at":"2026-01-08 04:32:46","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":191495,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDiurnal dynamics of (a) water use efficiency (WUE) and (b) underlying water use efficiency (UWUE), based on 30-minute data for the dry (red) and wet (blue) seasons\u003c/strong\u003e. Data are means (n=6; 2019-2024) and error bars represent the ±95% confidence interval (CI). Seasonal mean values ± 95% CI are provided in the \u003cstrong\u003etop-left panel. \u003c/strong\u003eDiurnal (30-min resolution) relationships between evapotranspiration (ET) and gross primary productivity (GPP) across different ecosystem types are shown in Figure S12.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/3d12623c474a9309afc6b3e0.png"},{"id":99797787,"identity":"8d91c595-073b-4bad-be12-3bed9c632de7","added_by":"auto","created_at":"2026-01-08 13:46:36","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":196011,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ea) Daily dynamics of water use efficiency (WUE) and b) underlying water use efficiency (uWUE) across different sites.\u003c/strong\u003e Each dot represents a daily mean value from 2019 to 2024. The grey shaded area denotes the wet season period (May to October).\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/8e30ecad1ad55a573402313f.png"},{"id":99797463,"identity":"69a73703-d78a-4b95-a87b-c947a471e525","added_by":"auto","created_at":"2026-01-08 13:45:51","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":143471,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAnalysis of environmental drivers of water use efficiency (WUE) across four agroecosystem types (reserve forest, paddy rice, cropland, and grassland) under dry and wet seasonal conditions.\u003c/strong\u003e The strength and direction of the correlations (Kendall’s tau, τ) are shown. Significance levels are indicated as follows: *** for \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01 (99% confidence), and ** for \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05 (95% confidence). Blue and red indicate positive and negative relationships, respectively. Grey color indicates no valid data.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/d77acd74271184b36a300bfa.png"},{"id":99752871,"identity":"8efcc445-5951-43e5-8332-40a5696ce180","added_by":"auto","created_at":"2026-01-08 04:32:47","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":104269,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEnvironmental drivers of variance in WUE across ecosystem types. \u003c/strong\u003eThe relative proportion of variance in WUE explained by each environmental variable is shown for all ecosystem types (see Methods). Variables include air temperature (Ta), relative humidity (Rh), vapour pressure deficit (VPD), soil moisture at 3 cm and 30 cm depths (SWC3 and SWC30), and solar radiation (Rg). Green bars denote statistically significant contributions (p \u0026lt; 0.05; 95% confidence interval), while red bars indicate statistically non-significant contributions (p \u0026gt; 0.05; 95% confidence interval). Grey bars represent the residual term, calculated as the difference between the left- and right-hand sides of Equation 4. Low residual values indicate that the model effectively captures the dominant environmental controls on WUE. See Figure S6 for the relative proportion of variance in uWUE explained by each environmental variable.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/1c641aed9a07a0efe80615b8.png"},{"id":99805396,"identity":"7e880851-8712-47ac-9f2c-78ffd8879a1d","added_by":"auto","created_at":"2026-01-08 14:16:30","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3971846,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/3728b699-9765-481c-9dca-5c7b404cb17c.pdf"},{"id":99797886,"identity":"adb4aec9-0772-4085-a700-ebd46aed27f6","added_by":"auto","created_at":"2026-01-08 13:46:51","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":8268352,"visible":true,"origin":"","legend":"","description":"","filename":"SyetalWUESupplementaryInformationv07012026.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8522874/v1/7ea559894871c180f1d32e86.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Water use efficiency responses in contrasting agroecosystems and land management practices in West Africa","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCarbon and water cycles are among the most important driving forces behind the coupled exchange of materials and energy in terrestrial ecosystems\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e. Ecosystem Water Use Efficiency (WUE), defined as the ratio of carbon assimilation via gross primary productivity (GPP), the total amount of CO₂ fixed by terrestrial ecosystems through photosynthesis per unit time, to water loss through evapotranspiration (ET), is a fundamental metric for assessing the coupling between carbon and water cycles in terrestrial ecosystems\u003csup\u003e\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. In essence, WUE provides critical insights into the trade-offs between ecosystem productivity and water consumption under varying environmental conditions\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. Understanding its dynamics is essential in the context of climate change, increased weather variability, and water scarcity, as it provides valuable insights into how different ecosystems respond to both natural and anthropogenic drivers\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. However, the extent to which GPP and ET regulate WUE across different ecosystem types and land management regimes remains poorly understood\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan additionalcitationids=\"CR9 CR10\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAt the leaf level, WUE is primarily governed by stomatal regulation, which balances CO₂ uptake for photosynthesis with water vapor loss through transpiration\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. At the ecosystem level, water availability becomes the dominant constraint on carbon sequestration\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan additionalcitationids=\"CR14 CR15\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e, particularly in semi-arid landscapes such as those in West Africa\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. Research consistently reveals a strong gradient of WUE across West Africa's primary land covers, driven by a complex interplay of physiological, structural, and environmental factors\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan additionalcitationids=\"CR19\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. For instance, forests in the humid region are known to have high absolute WUE due to large leaf area indices and deep root systems that access soil moisture year-round, allowing for sustained carbon assimilation\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. However, high ET rates come at a cost of only moderate efficiency in conserving water per unit of carbon fixed\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eMoreover, while WUE has been extensively investigated across various ecosystems, particularly in mid-latitude regions\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e, a substantial knowledge gap persists for semi-arid environments, such as the West African savanna\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. The Sudanian region is dominated by savannas, which exhibit marked seasonal variability in WUE, with values often increasing during the wet season\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. These ecosystems are composed of a mixture of woody C₃ species and herbaceous C₄ grasses, the latter being inherently more water-efficient due to their biochemical CO₂-concentrating mechanism and reduced photorespiration\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. During the rainy season, the grass layer flourishes, driving high ecosystem WUE\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. However, this situation reverses in the dry season as grasses senesce, leaving the deeper-rooted trees to maintain minimal gas exchange\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e. Furthermore, rainfed crops like millet and sorghum, which are often C₄ plants, can show high WUE during their short growing season\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. However, their efficiency is heavily dependent on agricultural practices, soil quality, and the timing of precipitation\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Irrigated/flooded croplands, while boosting yields, typically exhibit very low WUE at the ecosystem level, as the additional water input leads to increased ET without a proportional increase in carbon sequestration\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Grasslands and Sahelian shrublands operate under extreme water stress with low WUE due to low biomass and GPP. These ecosystems have adapted to survive (rather than thrive) under arid conditions, with carbon and water fluxes being minimal. Despite a \u0026ldquo;hierarchy\u0026rdquo; of WUE in West Africa, it is essential to monitor these patterns amid weather variability, including recurrent droughts and shifting precipitation patterns, which complicate and often alter the water-carbon dynamics, highlighting the need for a deeper understanding of WUE in dynamic terms and across diverse agroecosystems\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. High-resolution EC systems capture real-time variations in carbon and water fluxes, providing a robust method for evaluating WUE\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. The density of EC monitoring stations in West Africa, despite previous research efforts\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan additionalcitationids=\"CR31\" citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e, remains considerably lower than in North America, Europe, or Asia\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. Moreover, the application of EC systems across the diverse land-use types in the West African savanna is still limited. Expanding observational efforts is essential for enhancing our understanding of WUE patterns and supporting the development of more sustainable land and water management strategies in this climate-sensitive region.\u003c/p\u003e \u003cp\u003eThis study presents the first site-specific analysis of WUE across multiple time scales in the West African savanna. Using multi-year (2019\u0026ndash;2024), half-hourly eddy-covariance measurements of carbon and water-vapor fluxes, we examine WUE dynamics across four representative agroecosystems in northern Ghana, including - for the first time - a rainfed paddy rice system, alongside a protected savanna forest, a semi-degraded grassland, and a rainfed cropland. The variations in WUE at instantaneous, daily, seasonal, and multi-annual timescales reveal environmental drivers underlying these dynamics for managed (rainfed croplands, paddy rice, grasslands) and undisturbed ecosystems (protected savanna forests) (see Methods). By quantifying both environmental controls on WUE across these diverse land-use types, this research provides critical empirical insights to guide sustainable land and water management, enhance ecosystem resilience, and support climate adaptation planning in the West African Savanna. Specifically, the study addresses two key research questions: (i) To what extent do WUE dynamics differ between managed and undisturbed ecosystems in the semi-arid West African Savanna? (ii) What are the primary environmental factors controlling WUE across different agroecosystem types?\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eEnvironmental variables\u003c/h2\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e2\u003c/span\u003e presents the multi-year average values of key environmental drivers and limiting conditions influencing water use efficiency across different ecosystem types. Apart from air temperature being relatively consistent, the results reveal distinct microclimatic patterns among sites. Relative humidity was the highest for the forest site (~\u0026thinsp;64.5%), reflecting a moist, shaded microclimate associated with dense canopy cover and transpiration, while cropland and grassland exhibited substantially lower RH (~\u0026thinsp;47%), consistent with more open, sun- and wind-exposed surfaces, and rice field had intermediate RH (~\u0026thinsp;57%), likely influenced by surface water from flooding. Soil water content showed the most pronounced variation across sites, with the rice field having higher values (30%), while forest, cropland, and grassland had low but measurable values (0.05\u0026ndash;0.1%). Vapor pressure deficit, a key indicator of atmospheric water demand, was lowest in the forest, suggesting lower evaporative stress, and slightly higher but comparable across rice, cropland, and grassland, indicating drier atmospheric conditions in more open landscapes. Solar radiation was slightly higher at the forest and rice sites than at the cropland and grassland sites, likely due to differences in local atmospheric conditions. GPP and ET followed each other, with the forest showing the highest rates, consistent with mature vegetation and favorable microclimate conditions. Grassland had moderate GPP and ET, followed by rice and cropland, which exhibited the lowest values, likely reflecting crop-specific physiological traits and management effects.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cb\u003eAnnual mean (\u0026plusmn;\u0026thinsp;95% CI) values of key meteorological and ecosystem exchange variables for the four study sites\u003c/b\u003e. Variables include air temperature (Ta), relative humidity (RH), soil water content (SWC) at 3 cm, vapor pressure deficit (VPD), Solar radiation (Rg), gross primary productivity (GPP), and evapotranspiration (ET).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSites\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTa (\u0026deg;C)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRH (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSWC (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eVPD (hPa)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRg (W/m2)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eGPP (gC/m2/day)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eET (kgH2O/m2/day)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eReserve Forest\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e \u003cp\u003e29.48\u0026thinsp;\u0026plusmn;\u0026thinsp;0.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e64.51\u0026thinsp;\u0026plusmn;\u0026thinsp;2.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e0.09\u0026plusmn; 001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e \u003cp\u003e21.64\u0026thinsp;\u0026plusmn;\u0026thinsp;1.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e \u003cp\u003e300.47\u0026thinsp;\u0026plusmn;\u0026thinsp;7.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e \u003cp\u003e8.35\u0026thinsp;\u0026plusmn;\u0026thinsp;0.45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c8\"\u003e \u003cp\u003e3.05\u0026thinsp;\u0026plusmn;\u0026thinsp;0.19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePaddy rice\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e \u003cp\u003e29.54\u0026thinsp;\u0026plusmn;\u0026thinsp;0.27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e57.05\u0026thinsp;\u0026plusmn;\u0026thinsp;2.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e9.6\u0026thinsp;\u0026plusmn;\u0026thinsp;1.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e \u003cp\u003e23.09\u0026thinsp;\u0026plusmn;\u0026thinsp;1.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e \u003cp\u003e277.24\u0026thinsp;\u0026plusmn;\u0026thinsp;7.22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e \u003cp\u003e3.45\u0026thinsp;\u0026plusmn;\u0026thinsp;0.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c8\"\u003e \u003cp\u003e1.94\u0026thinsp;\u0026plusmn;\u0026thinsp;0.15\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eCropland\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e \u003cp\u003e28.41\u0026thinsp;\u0026plusmn;\u0026thinsp;0.24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e46.91\u0026thinsp;\u0026plusmn;\u0026thinsp;2.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e0.05\u0026thinsp;\u0026plusmn;\u0026thinsp;00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e \u003cp\u003e22.79\u0026thinsp;\u0026plusmn;\u0026thinsp;1.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e \u003cp\u003e255.78\u0026thinsp;\u0026plusmn;\u0026thinsp;3.61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e \u003cp\u003e2.82\u0026thinsp;\u0026plusmn;\u0026thinsp;0.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c8\"\u003e \u003cp\u003e1.85\u0026thinsp;\u0026plusmn;\u0026thinsp;0.13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eGrassland\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c2\"\u003e \u003cp\u003e28.7\u0026thinsp;\u0026plusmn;\u0026thinsp;0.22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e47.81\u0026thinsp;\u0026plusmn;\u0026thinsp;2.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e0.1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c5\"\u003e \u003cp\u003e22.70\u0026thinsp;\u0026plusmn;\u0026thinsp;1.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c6\"\u003e \u003cp\u003e250.68\u0026thinsp;\u0026plusmn;\u0026thinsp;2.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c7\"\u003e \u003cp\u003e3.53\u0026thinsp;\u0026plusmn;\u0026thinsp;0.28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c8\"\u003e \u003cp\u003e2.57\u0026thinsp;\u0026plusmn;\u0026thinsp;0.15\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eSeasonally, air temperature was lowest during the wet season at all sites (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e). At cropland and grassland sites, temperatures can drop to around 25\u0026deg;C during the Harmattan months (December-January), likely due to their higher surface albedo compared to the forest site. Vapor pressure deficit and incoming solar radiation showed similar seasonal patterns to air temperature, with minima during the wet season and maxima during the dry season. In contrast, soil water content and relative humidity exhibited pronounced wet-dry seasonality, peaking in August-September at all sites and reaching their lowest values during the dry season. GPP and ET followed comparable seasonal dynamics, with maxima occurring in August at the forest and grassland sites and in September at the paddy rice and cropland sites (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These differences reflect site-specific phenology. At the forest site, LAI peaked in August, shortly after the July rainfall maximum, which provided sufficient moisture for full canopy development (see Figures S5 and S11). In contrast, LAI peaked in September at the rice, cropland, and grassland sites, reflecting their later phenological development (see Figures \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e, S8-10).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eDiurnal dynamics of WUE\u003c/h3\u003e\n\u003cp\u003eThe diurnal dynamics of WUE across the four contrasting ecosystem types are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003ea. Across both wet and dry seasons, WUE exhibited a distinct pattern with pronounced morning and evening peaks. During the wet season, the forest site exhibited a distinct diurnal pattern, with WUE reaching approximately 4.0 g C kg H₂O⁻\u0026sup1; in the early morning and 4.5 g C kg H₂O⁻\u0026sup1; in the evening. In contrast, although the paddy rice, cropland, and grassland sites also showed a dual-peak pattern, a single dominant morning peak was more pronounced, with WUE values of roughly 3.7, 3.2, and 2.7 g C kg H₂O⁻\u0026sup1;, respectively. WUE declined markedly toward midday as rising VPD and temperature constrained stomatal conductance and carbon assimilation, a reduction further likely amplified by increased soil evaporation and, in the case of the rice field, evaporation from ponded water, while GPP became increasingly limited under high VPD (see Figures S2-S4). During the dry season, all sites exhibited a more uniform diurnal trend, with WUE peaking in both the morning and late afternoon (18:00\u0026ndash;19:00) before experiencing a sharp decline in the evening. The forest site again recorded the highest WUE, reaching up to 7.5 g C kg H₂O⁻\u0026sup1; in the evening, whereas cropland and grassland values remained near 2.5 g C kg H₂O⁻\u0026sup1;. Compared to the wet season, daytime WUE values during the dry season were generally lower, likely due to reduced soil moisture availability (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e) and increased atmospheric demand (see Figures S2-S4). However, in the late afternoon (16:00\u0026ndash;18:00), WUE tended to increase more sharply under dry conditions, indicating improved carbon assimilation efficiency as evaporative demand and air temperature declined (see Figures S2-S4).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eUnlike WUE, which followed a double-peak pattern, uWUE exhibited a distinct evening maximum (between 18:00 and 19:00) during the dry season, exceeding 50 g C hPa\u003csup\u003e0.\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e kg H₂O⁻\u0026sup1; in the savannah forest, and reaching approximately 16 and 14 g C hPa\u003csup\u003e0.\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e kg H₂O⁻\u0026sup1; in cropland and grassland sites, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003eb). Savannah forest exhibited the highest WUE and uWUE throughout the day compared to the other ecosystems, reflecting its dense canopy structure, deep rooting system, and physiological capacity to regulate stomatal conductance efficiently, maintaining high photosynthetic rates while minimizing excessive water loss.\u003c/p\u003e\n\u003ch3\u003eSeasonal dynamics of WUE\u003c/h3\u003e\n\u003cp\u003eSeasonal patterns were evident in the WUEs (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003e) and savannah forest exhibited a bimodal pattern with the highest values (exceeding 8 g C kgH₂O⁻\u0026sup1;) during the dry season (December-February), a secondary peak of about 4.5 g C kgH₂O⁻\u0026sup1; in the mid-rainy season (July-August) and the lowest values (0.5 g C kgH₂O⁻\u0026sup1;) in the transitional months of March and November. WUE of paddy rice increased steadily from 0-0.5 g C kg H₂O⁻\u0026sup1; during the early stages of crop establishment (June) to around 3 g C kg H₂O⁻\u0026sup1; in July-August following the rice plant development (Figure S8). These dynamics reflect a shift in the carbon-water interactions with initially high soil evaporation from an open canopy and low photosynthetic activity, yielding low WUE. By the time the crop reached the reproductive stage, the dense canopy tends to minimize soil evaporation (i.e., in line with high SWC and low VPD) and maximize carbon uptake (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). At the cropland site, the bimodal pattern in WUE during the rainy season reflects the complex interaction of agricultural practices, crop phenology, and environmental conditions, with the lowest WUE (~\u0026thinsp;0.5 g C kg ⁻\u0026sup1; H₂O) typically in March due to fallow or senesced fields (high soil evaporation and low GPP), resulting in the lowest WUE. Later in the wet season (May-June), rapid vegetative growth and canopy development (Figure S9) enhance GPP, likely shifting ET from soil evaporation to plant transpiration. This is followed by a peak in WUE (~\u0026thinsp;2.0 g C kg ⁻\u0026sup1; H₂O), after which it declines sharply in July, typically coinciding with land preparation activities such as plowing and grass clearing (Figure S9). A second and more pronounced WUE peak (~\u0026thinsp;2.5 g C kgH₂O⁻\u0026sup1;) occurs in September, during the crop\u0026rsquo;s reproductive stage (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003ea and S9). The grassland site displayed similar seasonal WUE patterns to the cropland site during the dry season (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003ea). During the rainy season, however, grassland exhibited a distinct and more pronounced WUE peak (~\u0026thinsp;2 g C kg H₂O⁻\u0026sup1;) in August, aligning with the reproductive stage of the grass with a fully developed canopy, maximum photosynthetic capacity, and high GPP (see Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). However, this peak was followed by a sharp decline in WUE in September, primarily driven by a reduction in GPP, which decreased in response to reduced solar radiation (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e), despite continued adequate water availability.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eOverall, uWUE followed a seasonal pattern similar to WUE (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e4\u003c/span\u003eb), reflecting the combined effects of atmospheric demand and ecosystem productivity. However, uWUE peaked during the dry season across all sites, highlighting the influence of high VPD. Values exceeded 40 g C hPa⁰\u0026middot;⁵ kg H₂O⁻\u0026sup1; in the forest in December-January, while cropland and grassland showed higher peaks around 10 g C hPa⁰\u0026middot;⁵ kg H₂O⁻\u0026sup1;. The lowest uWUE occurred in March, when soil moisture and vegetation activity were minimal, dropping to ~\u0026thinsp;3 g C hPa⁰\u0026middot;⁵ kg H₂O⁻\u0026sup1; in the forest and below 1 g C hPa⁰\u0026middot;⁵ kg H₂O⁻\u0026sup1; in cropland and grassland.\u003c/p\u003e\n\u003ch3\u003eInter-annual variability of WUE\u003c/h3\u003e\n\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e3\u003c/span\u003e presents an analysis of the five-year record of GPP, ET, VPD, WUE, and uWUE across ecosystems. Although data for the forest ecosystem were limited to a single year, it exhibited the highest GPP, which coincided with elevated VPD, as well as the highest WUE and uWUE. This suggests highly efficient stomatal control mechanisms that optimize the carbon-water trade-off through a conservative water-use strategy typical of mature forests. The rice ecosystem showed consistently low GPP alongside low VPD during 2022\u0026ndash;2023, with moderate ET likely driven by significant non-stomatal water losses from soil evaporation and canopy interception. Notably, paddy rice WUE and uWUE declined the following year, independent of VPD fluctuations. The cropland site demonstrated moderate and relatively stable GPP over the period, ranging from 966 to 1098 g C m⁻\u0026sup2; yr⁻\u0026sup1;, with consistently high VPD and only minor interannual variability in ET, WUE, and uWUE. In contrast, the grassland site exhibited greater interannual variability in both GPP and ET, while VPD remained relatively stable. Despite fluctuations in carbon uptake, WUE values remained consistent (1.3\u0026ndash;1.4 g C kg⁻\u0026sup1; H₂O), highlighting the grassland\u0026rsquo;s capacity to maintain WUE under variable weather conditions. Interestingly, uWUE increased from 6 to 7 g C hPa⁰\u0026middot;⁵ kg H₂O⁻\u0026sup1; between 2019 and 2021 before stabilizing around 6.7 g C hPa⁰\u0026middot;⁵ kg H₂O⁻\u0026sup1;, suggesting enhanced physiological optimization in years with higher GPP or slightly more favorable evaporative conditions.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cb\u003eInter-annual mean (\u0026plusmn;\u0026thinsp;95% CI) values\u003c/b\u003e of gross primary productivity (GPP), vapor pressure deficit (VPD), evapotranspiration (ET), water use efficiency (WUE), and underlying water use efficiency (uWUE) for the forest, paddy rice, cropland, and grassland sites across the five study years. Differences in ecosystem carbon and water fluxes relative to the forest ecosystem for the year 2023 are also shown, calculated as the forest-non-forest ecosystem (paddy rice, cropland, and grassland). Values are annual mean differences\u0026thinsp;\u0026plusmn;\u0026thinsp;95% CI.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYears\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGPP (gCm⁻\u0026sup2;y⁻\u0026sup1;)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eVPD (hPa)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eET (kg H₂O m⁻\u0026sup2; y⁻\u0026sup1;)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eWUE (gC/kg H₂O)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003euWUE (gC hPa\u003csup\u003e0.5\u003c/sup\u003e / kg H₂O)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003eReserve Forest\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2023/2024\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2910\u0026plusmn; 196\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e21.64\u0026plusmn; 1.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1080\u0026plusmn; 95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2.6\u0026plusmn; 0.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e12.2\u0026plusmn; 0.85\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePaddy Rice\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2022\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1013 \u0026plusmn; 55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22.95\u0026plusmn; 0.82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e547\u0026plusmn; 27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.8\u0026plusmn; 0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.1\u0026plusmn; 0.50\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2023/2024\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1016\u0026plusmn; 95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23.62\u0026plusmn; 1.23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e618\u0026plusmn; 26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.6\u0026plusmn; 0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e5.6\u0026plusmn; 0.67\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eCropland\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2020\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1050.9\u0026thinsp;\u0026plusmn;\u0026thinsp;99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22.37\u0026thinsp;\u0026plusmn;\u0026thinsp;1.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e714.7\u0026thinsp;\u0026plusmn;\u0026thinsp;53\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.47\u0026thinsp;\u0026plusmn;\u0026thinsp;0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.95\u0026thinsp;\u0026plusmn;\u0026thinsp;0.44\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2021\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1028.3\u0026thinsp;\u0026plusmn;\u0026thinsp;116\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23.49\u0026thinsp;\u0026plusmn;\u0026thinsp;1.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e658.7\u0026thinsp;\u0026plusmn;\u0026thinsp;55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.56\u0026thinsp;\u0026plusmn;\u0026thinsp;0.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7.57\u0026thinsp;\u0026plusmn;\u0026thinsp;0.55\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2022\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e966.2\u0026thinsp;\u0026plusmn;\u0026thinsp;102\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22.52\u0026thinsp;\u0026plusmn;\u0026thinsp;1.22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e638.8\u0026thinsp;\u0026plusmn;\u0026thinsp;51\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.51\u0026thinsp;\u0026plusmn;\u0026thinsp;0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7.18\u0026thinsp;\u0026plusmn;\u0026thinsp;0.50\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2023\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1098.1\u0026thinsp;\u0026plusmn;\u0026thinsp;110\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22.84\u0026thinsp;\u0026plusmn;\u0026thinsp;1.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e701.5\u0026thinsp;\u0026plusmn;\u0026thinsp;52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.57\u0026thinsp;\u0026plusmn;\u0026thinsp;0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7.48\u0026thinsp;\u0026plusmn;\u0026thinsp;0.49\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eGrassland\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2019\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1184.9\u0026thinsp;\u0026plusmn;\u0026thinsp;90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22.85\u0026thinsp;\u0026plusmn;\u0026thinsp;1.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e947.6\u0026thinsp;\u0026plusmn;\u0026thinsp;58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.25\u0026thinsp;\u0026plusmn;\u0026thinsp;0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e5.98\u0026thinsp;\u0026plusmn;\u0026thinsp;0.31\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2020\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1282.2\u0026thinsp;\u0026plusmn;\u0026thinsp;100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22.84\u0026thinsp;\u0026plusmn;\u0026thinsp;1.27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e890.9\u0026thinsp;\u0026plusmn;\u0026thinsp;60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.44\u0026thinsp;\u0026plusmn;\u0026thinsp;0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.88\u0026thinsp;\u0026plusmn;\u0026thinsp;0.37\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2021\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1679.9\u0026thinsp;\u0026plusmn;\u0026thinsp;181\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23.68\u0026thinsp;\u0026plusmn;\u0026thinsp;1.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1170.1\u0026thinsp;\u0026plusmn;\u0026thinsp;90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.44\u0026thinsp;\u0026plusmn;\u0026thinsp;0.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.99\u0026thinsp;\u0026plusmn;\u0026thinsp;0.48\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2022\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1211\u0026thinsp;\u0026plusmn;\u0026thinsp;117\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22.83\u0026thinsp;\u0026plusmn;\u0026thinsp;1.20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e866.9\u0026thinsp;\u0026plusmn;\u0026thinsp;61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.4\u0026thinsp;\u0026plusmn;\u0026thinsp;0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.67\u0026thinsp;\u0026plusmn;\u0026thinsp;0.42\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003e2023\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1344.3\u0026thinsp;\u0026plusmn;\u0026thinsp;105\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e21.41\u0026thinsp;\u0026plusmn;\u0026thinsp;1.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e934.6\u0026thinsp;\u0026plusmn;\u0026thinsp;61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.44\u0026thinsp;\u0026plusmn;\u0026thinsp;0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.65\u0026thinsp;\u0026plusmn;\u0026thinsp;0.36\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e \u003cp\u003e\u003cb\u003eDifference Relative to Forest Ecosystem (2023)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eComparison\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eGPP (gCm⁻\u0026sup2;y⁻\u0026sup1;)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003eVPD (hPa)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003eET (kg H₂O m⁻\u0026sup2; y⁻\u0026sup1;)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003eWUE (gC/kg H₂O)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003euWUE (gC hPa\u003c/b\u003e\u003csup\u003e\u003cb\u003e0.5\u003c/b\u003e\u003c/sup\u003e \u003cb\u003e/ kg H₂O)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForest - Paddy Rice\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1897\u0026plusmn; 132\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-1.98\u0026plusmn;0.50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e462\u0026plusmn; 42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.0\u0026plusmn; 0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6.6\u0026plusmn; 0.25\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForest - Cropland\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1812\u0026plusmn; 140\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-1.20\u0026plusmn; 0.30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e379\u0026plusmn; 75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.03\u0026plusmn; 0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4.72\u0026plusmn; 0.34\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eForest - Grassland\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1565\u0026plusmn; 124\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.23\u0026plusmn; 0.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e146\u0026plusmn; 82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.26\u0026plusmn; 0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e5.55\u0026plusmn; 0.42\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e\n\u003ch3\u003eMechanisms driving WUE responses\u003c/h3\u003e\n\u003cp\u003eTo explore the drivers and limiting conditions of WUE under varying weather conditions, we analyzed its statistical relationships with key environmental variables (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e5\u003c/span\u003e). In reserve forests during the dry season, WUE showed little response to temperature but increased under higher VPD and lower humidity and soil moisture, indicating that atmospheric dryness can enhance efficiency when water is limited. During the wet season, WUE increased with higher soil moisture, rainfall, and humidity, but declined with higher temperatures, VPD, and radiation. In the rice field, higher soil moisture and humidity increased WUE, whereas temperature, radiation, and VPD reduced it. Cropland and grassland showed similar patterns: dry-season WUE was constrained by temperature and low soil moisture, while wet-season WUE improved with humidity, soil moisture, and rainfall but decreased under high VPD and radiation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo further investigate the seasonal influence of environmental factors on WUE across all sites, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003e presents the results of an analysis of variance (see Methods). At the forest site, WUE exhibited strong seasonal shifts in drivers (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003ea, b). During the dry season, relative humidity (~\u0026thinsp;55%), shallow soil moisture (~\u0026thinsp;15%), and solar radiation (13%) explained the largest share of variance, although these contributions were not statistically significant, suggesting a multi-driver limitation system constrained by atmospheric demand and upper-layer soil water content. In contrast, the rainy season exhibited a pronounced shift toward radiation-dominated control, with solar radiation accounting for 65% of the variance. This indicates that, once water constraints are alleviated, energy availability becomes the primary driver of carbon-water coupling. Temperature remained a significant contributor (~\u0026thinsp;30%), underscoring its persistent role in regulating stomatal behavior and water use in savannah forests. WUE in the rice ecosystem was shaped predominantly by solar radiation and soil moisture dynamics (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003eg). Solar radiation explained more than 55% of the variance in WUE, reflecting its central role in driving photosynthetic carbon uptake. Despite the flooded or near-saturated conditions typical of paddy systems, subsurface water availability remained a key regulator, with deep soil moisture (30 cm) accounting for ~\u0026thinsp;35% of the variance and exerting a highly significant influence.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAt the cropland site (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003ec-d), dry-season WUE was primarily controlled by soil moisture, with shallow topsoil moisture explaining nearly 83% of the variance. Atmospheric drivers such as solar radiation and relative humidity contributed only marginally under these strongly water-limited conditions. During the rainy season, the system transitioned to an atmosphere-driven regime characterized by radiation and water-vapor demand (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003ed, g). Solar radiation accounted for more than 60% of the variability in WUE, highlighting the increasing importance of energy availability once water constraints are alleviated. The water vapor deficit remained a significant regulator during this period, accounting for over 25% of the variance. Concerning the grassland site, dry-season WUE was predominantly controlled by soil water availability, with shallow soil moisture explaining nearly 80% of the variance. This confirms that, as in the cropland system, soil moisture was the primary driver of WUE under water-limited conditions. During the rainy season, the regulation shifted toward atmospheric control, with air temperature becoming the strongest predictor of WUE (~\u0026thinsp;34%, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Additionally, shallow soil moisture (~\u0026thinsp;28%) and solar radiation (~\u0026thinsp;29%) contributed, although without statistical significance. These patterns indicate that once water limitations are relieved, grassland WUE is increasingly shaped by thermal and moisture conditions influencing canopy gas exchange. Overall, both grassland and cropland systems exhibited a clear seasonal transition, reflecting a shared ecohydrological response to the shifting availability of water and energy.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study provides the first long-term, site-specific assessment of WUE across four contrasting ecosystems in the semi-arid West African savanna, differing in the degree of land management, including, for the first time, a rainfed paddy rice system alongside savanna forest, cropland, and grassland. While WUE dynamics have been extensively examined using remote sensing, ecosystem modeling, and eddy covariance approaches, most studies have focused on mid-latitude regions\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan additionalcitationids=\"CR34 CR35\" citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. In contrast, empirical evidence from African ecosystems remains limited due to the scarcity of EC measuring stations, with only a few eddy covariance studies in West Africa, typically restricted to short observation periods or single ecosystems\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan additionalcitationids=\"CR18\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOur results showed that WUE exhibited clear diurnal dynamics that differed markedly between the wet and dry seasons across all ecosystems. However, although most sites showed a bimodal pattern with a dominant morning peak, reaching roughly 3.7, 3.2, and 2.7 g C kg⁻\u0026sup1; H₂O in the rice, cropland, and grassland systems, respectively, the savanna forest displayed a distinct and more pronounced afternoon enhancement. Specifically, WUE in the forest increased from about 4.0 g C kg⁻\u0026sup1; H₂O in the early morning to roughly 4.5 g C kg⁻\u0026sup1; H₂O later in the day, and this pattern became even stronger during the dry season. This divergence suggests ecosystem-specific regulation of carbon-water exchange: in cropland, rice, and grassland systems, morning peaks likely reflect optimal light and vapor-pressure conditions that favor stomatal opening, whereas the savanna forest\u0026rsquo;s stronger afternoon peak may indicate deeper rooting systems and sustained transpiration capacity under elevated evaporative demand\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. In other terms, the amplified afternoon WUE in the forest during the dry season, reaching values nearly double those of the agricultural and grass systems, further indicates a competitive advantage in maintaining carbon uptake when atmospheric dryness intensifies\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOur findings are consistent with earlier observations of dual diurnal WUE peaks in West African ecosystems\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e, which identified similar diurnal patterns and magnitudes in a degraded savanna woodland in northern Benin with daytime WUE reaching 3.86\u0026thinsp;\u0026plusmn;\u0026thinsp;1.03 g C kg⁻\u0026sup1; H₂O during the dry season and 5.50\u0026thinsp;\u0026plusmn;\u0026thinsp;1.29 g C kg⁻\u0026sup1; H₂O in the wet season\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Comparable magnitudes were also observed by Boulain et al. (2009)\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e in their investigation of WUE over millet and fallow sites in the Wankama catchment of the AMMA-Niger observatory, underscoring the consistency of these WUE dynamics across semi-arid West African landscapes. In addition, although the ecosystem WUE observed in our savanna forests was similar to values reported for subtropical forests during the wet season\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e,\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e, it remained markedly higher in the late afternoon of the dry season (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003ea). This dry-season increase likely reflects short-term drought-avoidance strategies that limit water loss while sustaining carbon uptake under high vapor pressure deficits\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. Differences can also be partly attributed to variations in water availability and physiological regulation\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. The savanna forest is dominated by drought-adapted C₃ trees, which possess deep rooting systems, conservative hydraulic architecture, and tight stomatal control in response to VPD, enabling sustained carbon assimilation with minimal transpiration during the dry season\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. Furthermore, the open canopy of the savanna forests exposes the trees to higher radiation and VPD, which, combined with their drought-adapted traits, may foster higher apparent WUE than in the more humid, shaded microclimates typical of closed-canopy temperate and subtropical forests\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e,\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. In comparison, rice and cropland systems are primarily composed of C₃ vegetation, which tends to exhibit greater stomatal sensitivity to atmospheric dryness\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e,\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u003c/sup\u003e. Specifically, maintaining photosynthesis in these systems often requires higher stomatal conductance\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e,\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e, resulting in lower ecosystem WUE during periods of peak evaporative demand. By contrast, although the grassland understory is dominated by C₄ grasses, which are known to exhibit high intrinsic WUE because of their CO₂-concentrating mechanism\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e, our ecosystem-scale measurements reveal WUE values slightly lower than those in C₃ croplands. This apparent discrepancy likely arises because ecosystem WUE integrates water losses not only through transpiration but also through soil evaporation, VPD, and canopy structural factors, which tend to diminish the physiological advantage of C₄ photosynthesis in semi-arid conditions\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e,\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOur results suggest that the observed dual-peak WUE patterns are driven by ecosystem-specific physiological and microclimatic factors, particularly stomatal optimization\u003csup\u003e\u003cspan additionalcitationids=\"CR53\" citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e. Across all ecosystems, diurnal WUE followed a consistent three-phase pattern: a morning peak under low VPD, a midday decline as evaporative demand increased, and a late-afternoon peak when photosynthesis remained active while evapotranspiration dropped sharply in line with previous studies\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e. Moreover, our findings show that uWUE remained consistently elevated during the dry season, with a pronounced rise that coincided with the late-afternoon peak in VPD (Figure S2a). These patterns reflect the strong responsiveness of ecosystem WUE to short-term changes in atmospheric dryness and align with stomatal optimization theory as well as previous evidence demonstrating the dominant influence of VPD on sub-daily WUE dynamics\u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. Importantly, the comparison between WUE and uWUE reveals the added value of using uWUE in ecohydrological assessments. While both indices capture similar seasonal trends, uWUE more effectively isolates physiological responses by reducing the confounding effects of VPD, thereby providing a clearer signal of canopy-level carbon-water exchange\u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e. In contrast, traditional WUE amplifies VPD-driven variability, thereby reflecting the combined effects of environmental and physiological controls\u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eFinally, our results reveal clear ecosystem-specific shifts in the dominant controls of WUE. In the savanna forest, WUE regulation transitioned from a multi-factor limitation during the dry season, driven by relative humidity, shallow soil moisture, and solar radiation, which jointly explained\u0026thinsp;~\u0026thinsp;85% of the variance, to a radiation-dominated control in the wet season, when water constraints were alleviated (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003e). During this period, solar radiation accounted for 65% of the WUE variance, representing a 5-fold increase in explanatory power compared with the dry season (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e6\u003c/span\u003eb-a). This shift indicates that once soil moisture is sufficient, energy availability becomes the primary driver of carbon-water coupling within the savanna forest. In cropland and grassland ecosystems, WUE was largely regulated by soil moisture during the dry season (explaining\u0026thinsp;~\u0026thinsp;80% of the total variance), but shifted toward atmospheric control under wet conditions, with solar radiation, VPD, and air temperature emerging as dominant factors. In contrast, the paddy rice system exhibited a distinct pattern, with solar radiation and deep soil moisture explaining most of the WUE variability (~\u0026thinsp;90%), reflecting the hydrological dominance of flooded cultivation. This sensitivity to deep soil moisture is consistent with the hydrological functioning of paddy ecosystems, where root water uptake and aeration dynamics may depend not only on surface flooding but also on the moisture status of deeper soil layers, which govern subsurface oxygen diffusion, root activity, and water availability throughout the growing season\u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e,\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eDespite providing valuable insights, key limitations remain. While the eddy covariance technique is widely regarded as the most reliable approach for quantifying ecosystem carbon and water fluxes at the local scale (with a footprint of tens to hundreds of meters), our results are subject to uncertainties inherent to this method, as previously addressed in studies on the same sites\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e,\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e. Moreover, spatial heterogeneity within flux footprints, particularly in managed landscapes, can introduce additional uncertainty\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e. Despite these challenges, this study represents a significant advancement in understanding WUE dynamics from daily, seasonal, and annual scales in West Africa, a region where EC observations are exceptionally scarce, and field conditions are particularly challenging. The long-term micrometeorological measurements presented here provide critical insights into how carbon and water fluxes respond to contrasting ecosystems and management regimes. By identifying the dominant physical and physiological mechanisms governing ecosystem-scale carbon-water coupling, our results not only improve process-level understanding but also offer practical guidance for landscape management.\u003c/p\u003e"},{"header":"Conclusion and policy implications","content":"\u003cp\u003eAt the regional scale, these findings carry important implications for climate assessment, land management, and policy planning across West Africa. Clear and consistent contrasts in water-use efficiency (WUE) emerge among ecosystem types. Savanna forests maintained the highest WUE in both seasons, reaching up to 3.1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.26 g C kg⁻\u0026sup1; H₂O during the rainy season, followed by paddy rice (2.08\u0026thinsp;\u0026plusmn;\u0026thinsp;0.21 g C kg⁻\u0026sup1; H₂O), croplands (1.93\u0026thinsp;\u0026plusmn;\u0026thinsp;0.20 g C kg⁻\u0026sup1; H₂O), and grasslands (1.66\u0026thinsp;\u0026plusmn;\u0026thinsp;0.18 g C kg⁻\u0026sup1; H₂O). These ecosystem-level differences highlight the dominant role of forests in regulating regional carbon and water exchanges. The superior WUE of forests reflects their unique biophysical and physiological characteristics\u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e,\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e. Overall, forests represent the moistest environments, benefiting from reduced aerodynamic drying and strong canopy-atmosphere coupling, which together support high productivity alongside substantial evapotranspiration. Forest gross primary productivity was approximately 2.5-3 times higher than that of non-forest systems, while evapotranspiration was 20\u0026ndash;65% greater. This combination resulted in WUE values that exceeded those of other ecosystems by 60\u0026ndash;80%, underscoring the efficiency with which forests convert water into biomass. In contrast, non-forest ecosystems exhibited distinct constraints that limited their WUE. Paddy rice systems exhibited relatively low GPP and WUE, likely due to flooded, anaerobic soil conditions and short vegetation cycles. Croplands were primarily constrained by strong seasonality and reduced radiation availability, leading to moderate productivity and WUE. Grasslands, while maintaining moderate GPP, experienced comparatively high evapotranspiration, resulting in the lowest WUE, an outcome consistent with open herbaceous systems that exert limited canopy control over water losses. Seasonal analyses further revealed a pronounced shift in controlling factors, from water limitation during the dry season to stronger energy and atmospheric regulation during the wet season. This transition highlights the adaptive coupling between hydrological availability and ecosystem physiological responses across the region. These insights point to several pathways for enhancing WUE in agroecosystems. Effective strategies include the adoption of water-saving irrigation technologies, improvements in soil-water retention through conservation tillage and mulching, as well as the integration of agroforestry practices. At the same time, indiscriminate afforestation should be avoided, as it may exacerbate water scarcity in already water-limited environments\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e,\u003cspan additionalcitationids=\"CR63 CR64\" citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e. Nature-based solutions that prioritize the restoration of drought-adapted and water-efficient native species offer a more sustainable approach, enabling carbon sequestration gains while maintaining regional water balance. Finally, given the severe scarcity/absence of eddy covariance observations in West Africa, expanding multi-site flux monitoring networks is critical. Continuous measurements of carbon, water, and energy fluxes, integrated with soil-plant-atmosphere remote sensing and land-surface modeling, will be essential for reducing uncertainty and improving regional assessments. Such efforts will provide the evidence base needed to support climate-smart agriculture, drought mitigation strategies, and sustainable land-management policies.\u003c/p\u003e"},{"header":"Data and Methods","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eStudy area and site descriptions\u003c/h2\u003e \u003cp\u003eIn line with previous studies\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e,\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e,\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e,\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e, the study area (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e1\u003c/span\u003e) is part of a wider research observatory established over the past decade to monitor CO₂, CH\u003csub\u003e₄,\u003c/sub\u003e and H₂O fluxes\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan additionalcitationids=\"CR67\" citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e. This monitoring network was developed through a collaborative initiative (WASCAL) in the Sudanian savannah belt of West Africa\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e,\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e. Two eddy covariance (EC) flux towers were initially established in northern Ghana: one located in actively cultivated, rainfed cropland (Kayoro) within the Tono River catchment\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e, and another in semi-degraded grassland occasionally used as pasture by local communities (Gorigo)\u003csup\u003e\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e. The latter site is located in the upper section of the Vea watershed, a sparsely populated area near the Ghana-Burkina Faso border (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The two towers were established in October 2013 and March 2017, respectively\u003csup\u003e\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e. To enhance understanding of ecosystem-scale CO₂ and H₂O fluxes across various land-use systems, the observatory network has recently been expanded with two additional eddy covariance towers\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e. These new installations target rainfed paddy rice cultivation and protected savanna forest ecosystems\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e. The rice field site (Janga; installed in May 2022) is situated in a poorly drained floodplain prone to seasonal flooding\u003csup\u003e\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e,\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e. Agricultural activities are concentrated in these low-lying areas, where rice and maize are the dominant crops cultivated during the rainy season\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e. The forest site (Mole; installed in June 2023) is situated in a protected reserve designated as a national park in 1958 and classified as a Category II protected area under the IUCN guidelines\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e,\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e, characterized by savanna woodland interspersed with wetlands and a floodplain ecosystem\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e. During the peak of the rainy season, the savanna forest can reach heights of up to six meters\u003csup\u003e\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e. The soils are primarily iron-rich Ferralsols and structurally stable Nitisols, characterized by good drainage properties\u003csup\u003e\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e\u003c/sup\u003e. Key characteristics of the four EC monitoring sites are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cb\u003eKey characteristics of the eddy covariance monitoring sites used in this study\u003c/b\u003e. Note that for paddy rice and cropland sites, cropping and harvest periods vary annually with the rainy season, typically ranging from June 15 to July 15 for cropping and from October 5 to November 5 for harvest.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEcosystem characteristics\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eReserve forest\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePaddy rice\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCropland\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGrassland\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSite names\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMole National Park\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eJanga\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eKayoro\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGorigo\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLatitude (\u0026deg;N)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9.3388\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10.1299\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e10.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e10.94016\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLongitude (\u0026deg;W)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026minus;1.8688\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026minus;0.8838\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026minus;1.3210\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026minus;0.8264\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEC Sensors\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIRGASON\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIRGASON\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLI-7500\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLI-7500\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eData coverage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMay 2023 to June 2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eJuly 2022 to May 2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eJan-2020 to Dec-2023\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eJan-2019 to Dec-2023\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLand management\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNature reserve\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRainfed Rice\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMix crops (soybean, millet)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHighly degraded used for grazing\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLand cover type\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePristine savanna\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePaddy Rice\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMixture of fallow and cropland\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGrassland\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLand use intensity\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVery low\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHigh\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHigh\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eHigh\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEC height (m)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4.82\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCanopy height (m)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eElevation (m)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e160.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e130.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e286.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e222.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSoil texture\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLoamy sandy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLoamy sandy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLoamy sandy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLoamy sandy\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSoil bulk density (g cm-3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.71\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1.65\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSensor direction (Degrees)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e337.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e337.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e132.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e139.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReferences\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGuug et al., 2025\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGuug et al., 2025\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eBliefernicht et al., 2018)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eBliefernicht et al., 2018)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eClimatically, all sites are influenced by the West African Monsoon (WAM)\u003csup\u003e\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e\u003c/sup\u003e, characterized by a wet season (May-October) and a dry season (November-April) (see Figure S5). Mean annual rainfall ranges from 700 to 1100 mm\u003csup\u003e74\u003c/sup\u003e, largely controlled by intermittent mesoscale convective systems that cause substantial spatiotemporal variability\u003csup\u003e\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e\u003c/sup\u003e. Average daily temperatures range between 22\u0026deg;C and 34\u0026deg;C, with pre-monsoon peaks frequently exceeding 40\u0026deg;C (March-April)\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e,\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eMicroclimatic Observations\u003c/h2\u003e \u003cp\u003eAll four monitoring sites were equipped with nearly identical eddy covariance (EC) systems, following standardized measurement protocols but with site-specific installation heights\u003csup\u003e\u003cspan additionalcitationids=\"CR67\" citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e (see Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e1\u003c/span\u003e). At Kayoro, Gorigo, and Janga, hereinafter referred to as cropland, grassland, and paddy rice, turbulent fluxes of CO₂ and water vapor were measured at heights of 3.1 m, 4.8 m, and 4.3 m, respectively, using a sampling frequency of 20 Hz. Flux measurements at cropland and grassland sites were obtained using a LI-7500 open-path infrared gas analyzer (LI-COR Inc., Lincoln, USA) coupled with a three-dimensional ultrasonic anemometer (Campbell Scientific Inc., Logan, UT). In contrast, the paddy rice field utilized an integrated IRGASON system. At the Mole Forest Reserve, hereinafter referred to as the reserve savanna forest, flux data were collected using an IRGASON sensor installed at 12 m and operating at 10 Hz to account for the forest canopy structure and ensure the representativeness of the ecosystem at the forest scale. Meteorological variables, including air temperature, relative humidity, solar radiation, and precipitation, were continuously monitored and logged at 30-minute intervals in accordance with standard EC tower protocols\u003csup\u003e\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e\u003c/sup\u003e. Solar radiation was measured using a CNR4 net radiometer (Kipp \u0026amp; Zonen), and ground heat flux (G) was quantified using three self-calibrating heat flux plates installed at a depth of 8 cm to evaluate the components of the surface energy balance (see Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Soil parameters, including volumetric soil water content (SWC) and soil temperature, were measured at depths of 3, 10, and 30 cm and used to correct soil heat flux (G) for soil heat storage effects\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e,\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e\u003c/sup\u003e. Precipitation was monitored by two complementary sensors: a tipping-bucket rain gauge (model 52203, R.M. Young) positioned 1 m above the ground, and a weighing gauge (Pluvio2, OTT).\u003c/p\u003e \u003cp\u003eAlthough eddy covariance monitoring began at the cropland site in 2013 and at the grassland site in 2017\u003csup\u003e17,67,68\u003c/sup\u003e, this study focuses on recently collected data, i.e., between 2019 and 2024 (see Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Earlier records contained substantial temporal gaps due to limited data-retrieval capacity, which impacted data quality and continuity. For the paddy rice field, following previous study\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e, CO₂ and water vapor flux measurements from the rainy season (May-October) were used, corresponding to the rice-growing period. This timeframe provides the most consistent and representative basis for comparing WUE across the different ecosystem types. However, during the dry season (November-March), the rice field and cropland sites are typically fallow, resulting in minimal GPP due to the absence of active vegetation, though soil evaporation may still occur from residual moisture\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eFlux Data Processing\u003c/h2\u003e \u003cp\u003eHigh-frequency turbulent flux data were processed using the \u003cem\u003eTurbulence Knight\u003c/em\u003e (TK3) software package\u003csup\u003e\u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e78\u003c/span\u003e,\u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e79\u003c/span\u003e\u003c/sup\u003e. Following the methodology of Gr\u0026uuml;nwald and Bernhofer (2007)\u003csup\u003e\u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e80\u003c/span\u003e\u003c/sup\u003e, half-hourly fluxes of CO₂ and latent heat were computed. Data quality was evaluated according to the flagging criteria proposed by Mauder et al. (2013)\u003csup\u003e\u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e81\u003c/span\u003e\u003c/sup\u003e. Missing data were gap-filled following established procedures\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e to obtain daily, monthly, and annual flux aggregates. Missing precipitation records were supplemented with data from nearby meteorological stations\u003csup\u003e\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e,\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e. Energy balance closure (EBC) was assessed to evaluate the consistency of eddy-covariance fluxes. Using 30-minute averages, EBC residuals ranged from 5% to 30% across sites (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e), consistent with values reported for FLUXNET sites\u003csup\u003e\u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e82\u003c/span\u003e,\u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e83\u003c/span\u003e\u003c/sup\u003e. Notably, grassland and reserve forest sites showed the best closure (see Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). To account for energy imbalance in subsequent evapotranspiration analyses, fluxes were corrected using the quantile mapping approach\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e. The eddy covariance flux footprint, which covered most of each ecosystem (see Figure S7), was calculated for each site using the method described by Kljun et al. (2015)\u003csup\u003e\u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e84\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eFormal Analysis\u003c/h2\u003e \u003cp\u003eWater use efficiency was estimated as the ratio of GPP to ET (Eq.\u0026nbsp;1). GPP (g C m⁻\u0026sup2; time⁻\u0026sup1;) represents the total carbon assimilated through vegetation photosynthesis\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e85\u003c/span\u003e\u003c/sup\u003e (Chapin et al., 2011), while ET (kgH₂O m⁻\u0026sup2; time⁻\u0026sup1;) quantifies total water loss via transpiration, canopy interception, and soil evaporation. Because it is often impractical to separate these individual components in field observations, ET is generally used as an integrated measure of ecosystem water loss\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. ET was derived from latent heat flux (LE): ET\u0026thinsp;=\u0026thinsp;LE / λ, where λ\u0026thinsp;=\u0026thinsp;2454 J g⁻\u0026sup1; following standard formulations\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e85\u003c/span\u003e,\u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e86\u003c/span\u003e\u003c/sup\u003e. The computed ET values were used to calculate WUE (g C kgH₂O⁻\u0026sup1;).\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:\\begin{array}{c}WUE=\\frac{\\text{G}\\text{P}\\text{P}}{\\text{E}\\text{T}}\\:\\#\\left(1\\right)\\end{array}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eGPP can be expressed as follows\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:\\begin{array}{c}GPP=-NEE+Re\\:\\#\\left(2\\right)\\end{array}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere NEE represents the net flux of CO₂ between an ecosystem and the atmosphere (g C m⁻\u0026sup2; time⁻\u0026sup1;), integrating carbon uptake via photosynthesis with carbon losses from respiration. This flux can be directly measured using eddy covariance techniques\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e. Total ecosystem respiration, Re, captures the combined respiratory carbon efflux from all ecosystem components (g C m⁻\u0026sup2; time⁻\u0026sup1;).\u003c/p\u003e \u003cp\u003eThis study also evaluated underlying water use efficiency (uWUE; g C hPa\u003csup\u003e0.\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e kgH₂O⁻\u0026sup1;), a refined indicator that more accurately represents the relationship between GPP and ET and captures long-term variations in the photosynthesis-transpiration balance across ecosystem types\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e86\u003c/span\u003e,\u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e87\u003c/span\u003e\u003c/sup\u003e. UWUE was calculated as:\u003cdiv id=\"Equc\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equc\" name=\"EquationSource\"\u003e\n$$\\:\\begin{array}{c}uWUE=\\frac{\\text{G}\\text{P}\\text{P}\\:\\times\\:\\:{\\text{V}\\text{P}\\text{D}}^{0.5}}{\\text{E}\\text{T}}\\#\\left(3\\right)\\end{array}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere VPD is the vapor pressure deficit (hPa). The response of uWUE to environmental variability reflects potential physiological adjustments of ecosystems to changing weather conditions\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e. Compared with conventional WUE, uWUE provides a more robust description of plant biochemical processes by accounting for VPD influences\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e,\u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e86\u003c/span\u003e,\u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e87\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eTo identify the environmental factors influencing WUE responses across different ecosystem types, the non-parametric Kendall rank correlation test\u003csup\u003e\u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e88\u003c/span\u003e\u003c/sup\u003e was employed. This method was selected because it is less affected by outliers and non-linear distributions than the Pearson or Spearman correlation tests\u003csup\u003e\u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e89\u003c/span\u003e\u003c/sup\u003e. Furthermore, analysis of variance (ANOVA) was implemented to quantify the proportion of variance in WUE explained by each potential explanatory variable, including key environmental factors across different ecosystem types. To account for temporal autocorrelation, which is often present in weather data measurements (e.g., temperature), the analysis incorporated a first-order autoregressive (AR(1)) model within a multivariable regression model\u003csup\u003e\u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e90\u003c/span\u003e\u003c/sup\u003e, expressed as follows:\u003cdiv id=\"Equd\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equd\" name=\"EquationSource\"\u003e\n$$\\:{y}_{t}\\:=\\:\\alpha\\:+{\\beta\\:}_{1}{x}_{t}^{1}+{\\beta\\:}_{2}{x}_{t}^{2}+\\dots\\:+{\\beta\\:}_{n}{x}_{t}^{n}+{\\mu\\:}_{t}\\:,\\:\\:\\:\\:\\:\\:\\:t=1,\\dots\\:.,T\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\:\\left(4\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere \u0026#119910;\u003csub\u003e\u0026#119905;\u003c/sub\u003e are the daily WUE values, \u0026#119905; is the time variable assigned to \u0026#119910;\u003csub\u003e\u0026#119905;\u003c/sub\u003e, \u0026#120572; is a constant term, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{x}_{t}^{1},\\:{x}_{t}^{2},\\:and\\:\\:{x}_{t}^{n}\\)\u003c/span\u003e\u003c/span\u003e and denote key environmental variables such as temperature, relative humidity, vapor pressure deficit, soil water moisture at 3 cm and 30 cm depths, and solar radiation, across each ecosystem type. The coefficients, \u0026#120573;\u003csub\u003e1\u003c/sub\u003e, \u0026#120573;\u003csub\u003e2\u003c/sub\u003e, and \u0026#120573;\u003csub\u003en\u003c/sub\u003e, represent the associated linear trends, and \u0026#120583;\u003csub\u003e\u0026#119905;\u003c/sub\u003e represents the residual term that is assumed to be autoregressive of the order of 1 [AR (1)] and may be related as follows:\u003cdiv id=\"Eque\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Eque\" name=\"EquationSource\"\u003e\n$$\\:{\\mu\\:}_{t}=\\:⍴{\\mu\\:}_{t-1}+\\:{\\epsilon\\:}_{t},\\:\\:\\:t=2,\\dots\\:.,T\\:\\:\\:\\:\\:\\:\\left(5\\right)$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eHere, \u003cem\u003eρ\u003c/em\u003e represents the autocorrelation coefficient, and \u0026#120576;\u003csub\u003e\u0026#119905;\u003c/sub\u003e ​denotes the white noise term, an independent random variable with zero mean and constant variance\u003csup\u003e\u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e90\u003c/span\u003e\u003c/sup\u003e. We also assumed that \u0026minus;\u0026thinsp;1\u0026thinsp;\u0026lt;\u0026thinsp;\u003cem\u003eρ\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;1, so the noise process \u0026#120583;\u003csub\u003e\u0026#119905;\u003c/sub\u003e is stationary. This formulation allows for autocorrelation in the residuals of the daily WUE values, such that \u003cem\u003eρ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;Corr (\u0026#120583;\u003csub\u003e\u0026#119905;\u003c/sub\u003e, \u0026#120583;\u003csub\u003e\u0026#119905;\u003c/sub\u003e\u0026minus;1). It is essential to note that, due to limited data availability/information during the study period, factors such as field fertilization, plowing, and land preparation activities were not considered in the study.\u003c/p\u003e \u003cp\u003eTo reduce multicollinearity and ensure robust model estimates, we applied a hierarchical predictor selection procedure\u003csup\u003e\u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e91\u003c/span\u003e,\u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e92\u003c/span\u003e\u003c/sup\u003e. For structurally dependent predictors, such as air temperature, relative humidity, and vapor pressure deficit, only the variable most strongly associated with WUE is considered, allowing for the determination of the dominant atmospheric driver. For soil moisture, only the layer (3 cm or 30 cm) most correlated with WUE was retained. The remaining predictors were screened using pairwise correlations, removing variables with |r| \u0026gt; 0.7 that contributed less explanatory power to WUE. Finally, variance inflation factors (VIFs) were computed, and predictors with VIFs greater than 5 were iteratively excluded to ensure acceptable levels of collinearity\u003csup\u003e\u003cspan additionalcitationids=\"CR93\" citationid=\"CR92\" class=\"CitationRef\"\u003e92\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e94\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eCompeting interests:\u003c/h2\u003e \u003cp\u003eThe authors declare that they have no competing financial interests.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eConceptualization: S.S.; Methodology: S.S., J.B., S.G., K.M., L.H.; Software: S.S., L.H., S.G.; Validation: S.S., J.B., R.K., R.S., F.N., T.J., P.L.; Formal analysis: S.S., B.Q., R.K.; Investigation: S.S., S.G., S.Sa., M.W., A.K.; Resources: F.E.O., E.Q., L.K.A.; Data curation: S.S., J.B., L.H., S.G., P.D.; Writing \u0026ndash; original draft: S.S.; Writing \u0026ndash; review \u0026amp; editing: All authors; Visualization: S.S.; Supervision: H.K., J.B.; Project administration: H.K.; Funding acquisition: H.K.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThis research was supported by the German Federal Ministry of Education and Research (BMBF) through the West African Science Service Centre on Climate Change and Adapted Land Use (WASCAL) and the Concerted Regional Modelling and Observation Assessment for Greenhouse Gas Emissions and Mitigation Options under Climate and Land Use Change in West Africa (CONCERT-West Africa; grant no. 01LG2089A BMBF). The authors gratefully acknowledge the cooperation of the Mole National Park authorities, as well as the landowners of the Janga rice field, Kayoro, and Gorigo sites, for granting permission to install and operate the eddy covariance (EC) stations.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe datasets analyzed during the current study are not publicly available due to WASCAL\u0026rsquo;s Data Sharing Policy and Guidelines, but are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eGentine, P. \u003cem\u003eet al.\u003c/em\u003e Coupling between the terrestrial carbon and water cycles\u0026mdash;a review. \u003cem\u003eEnviron. Res. Lett.\u003c/em\u003e 14, 083003 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePonton, S. \u003cem\u003eet al.\u003c/em\u003e Comparison of ecosystem water-use efficiency among Douglas-fir forest, aspen forest and grassland using eddy covariance and carbon isotope techniques. \u003cem\u003eGlob. Change Biol.\u003c/em\u003e 12, 294\u0026ndash;310 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFu, Y., Jian, S. \u0026amp; Yu, X. Water use efficiency in China is impacted by climate change and land use and land cover. \u003cem\u003eEnviron. Sci. Pollut. Res.\u003c/em\u003e 31, 42840\u0026ndash;42856 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSong, Q.-H. \u003cem\u003eet al.\u003c/em\u003e Water use efficiency in a primary subtropical evergreen forest in Southwest China. \u003cem\u003eSci. Rep.\u003c/em\u003e 7, 43031 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAn, X. Responses of Water Use Efficiency to climate change in evapotranspiration and transpiration ecosystems. \u003cem\u003eEcol. Indic.\u003c/em\u003e 141, 109157 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAi, Z. \u003cem\u003eet al.\u003c/em\u003e Variation of gross primary production, evapotranspiration and water use efficiency for global croplands. \u003cem\u003eAgric. For. Meteorol.\u003c/em\u003e 287, 107935 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang, J., Luyssaert, S., Richardson, A. D., Kutsch, W. \u0026amp; Janssens, I. A. Steeper declines in forest photosynthesis than respiration explain age-driven decreases in forest growth. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 111, 8856\u0026ndash;8860 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, T. \u003cem\u003eet al.\u003c/em\u003e Differences in ecosystem water-use efficiency among the typical croplands. \u003cem\u003eAgric. Water Manag.\u003c/em\u003e 209, 142\u0026ndash;150 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, C. \u003cem\u003eet al.\u003c/em\u003e Negative impacts of plant diversity loss on carbon sequestration exacerbate over time in grasslands. \u003cem\u003eEnviron. Res. Lett.\u003c/em\u003e 15, 104055 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, Y., Pe\u0026ntilde;uelas, J., Cescatti, A., Zhang, Y. \u0026amp; Zhang, Z. Atmospheric Dryness Dominates Afternoon Depression of Global Terrestrial Photosynthesis. \u003cem\u003eGeophys. Res. Lett.\u003c/em\u003e 51, e2024GL110954 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAgo, E. E., Agbossou, E. K., Cohard, J.-M., Galle, S. \u0026amp; Aubinet, M. Response of CO2 fluxes and productivity to water availability in two contrasting ecosystems in northern Benin (West Africa). \u003cem\u003eAnn. For. Sci.\u003c/em\u003e 73, 483\u0026ndash;500 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKeenan, T. F. \u003cem\u003eet al.\u003c/em\u003e Increase in forest water-use efficiency as atmospheric carbon dioxide concentrations rise. \u003cem\u003eNature\u003c/em\u003e 499, 324\u0026ndash;327 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMedlyn, B. E. \u003cem\u003eet al.\u003c/em\u003e How do leaf and ecosystem measures of water-use efficiency compare? \u003cem\u003eNew Phytol.\u003c/em\u003e 216, 758\u0026ndash;770 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, Y. \u003cem\u003eet al.\u003c/em\u003e Elucidating climatic drivers of photosynthesis by tropical forests. \u003cem\u003eGlob. Change Biol.\u003c/em\u003e 29, 4811\u0026ndash;4825 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu, G. \u003cem\u003eet al.\u003c/em\u003e Water-use efficiency of forest ecosystems in eastern China and its relations to climatic variables. \u003cem\u003eNew Phytol.\u003c/em\u003e 177, 927\u0026ndash;937 (2008).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang, M., Zhai, P. \u0026amp; Piao, S. Divergent responses of ecosystem water use efficiency to drought timing over Northern Eurasia. \u003cem\u003eEnviron. Res. Lett.\u003c/em\u003e 16, 045016 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQuansah, E. \u003cem\u003eet al.\u003c/em\u003e Carbon dioxide fluxes from contrasting ecosystems in the Sudanian Savanna in West Africa. \u003cem\u003eCarbon Balance Manag.\u003c/em\u003e 10, 1 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAgo, E. E., Ser\u0026ccedil;a, D., Agbossou, E. K., Galle, S. \u0026amp; Aubinet, M. Carbon dioxide fluxes from a degraded woodland in West Africa and their responses to main environmental factors. \u003cem\u003eCarbon Balance Manag.\u003c/em\u003e 10, 22 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoulain, N. \u003cem\u003eet al.\u003c/em\u003e Towards an understanding of coupled physical and biological processes in the cultivated Sahel \u0026ndash; 2. Vegetation and carbon dynamics. \u003cem\u003eJ. Hydrol.\u003c/em\u003e 375, 190\u0026ndash;203 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKayiranga, A. \u003cem\u003eet al.\u003c/em\u003e Water Use Efficiency-Based Multiscale Assessment of Ecohydrological Resilience to Ecosystem Shifts Over the Continent of Africa During 1992\u0026ndash;2015. \u003cem\u003eJ. Geophys. Res. Biogeosciences\u003c/em\u003e 125, e2020JG005749 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFarooqi, T. J. A. \u003cem\u003eet al.\u003c/em\u003e Advancing Knowledge in Forest Water Use Efficiency Under Global Climate Change Through Scientometric Analysis. \u003cem\u003eForests\u003c/em\u003e 15, 1893 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang, Y. \u003cem\u003eet al.\u003c/em\u003e Contrasting responses of water use efficiency to drought across global terrestrial ecosystems. \u003cem\u003eSci. Rep.\u003c/em\u003e 6, 23284 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTagesson, T. \u003cem\u003eet al.\u003c/em\u003e Dynamics in carbon exchange fluxes for a grazed semi-arid savanna ecosystem in West Africa. \u003cem\u003eAgric. Ecosyst. Environ.\u003c/em\u003e 205, 15\u0026ndash;24 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaylor, S. H. \u003cem\u003eet al.\u003c/em\u003e Physiological advantages of C4 grasses in the field: a comparative experiment demonstrating the importance of drought. \u003cem\u003eGlob. Change Biol.\u003c/em\u003e 20, 1992\u0026ndash;2003 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHutley, L. B., Beringer, J., Fatichi, S., Schymanski, S. J. \u0026amp; Northwood, M. Gross primary productivity and water use efficiency are increasing in a high rainfall tropical savanna. \u003cem\u003eGlob. Change Biol.\u003c/em\u003e 28, 2360\u0026ndash;2380 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFriborg, T., Boegh, E. \u0026amp; Soegaard, H. Carbon dioxide flux, transpiration and light response of millet in the Sahel. \u003cem\u003eJ. Hydrol.\u003c/em\u003e 188\u0026ndash;189, 633\u0026ndash;650 (1997).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, T. \u003cem\u003eet al.\u003c/em\u003e Differences in ecosystem water-use efficiency among the typical croplands. \u003cem\u003eAgric. Water Manag.\u003c/em\u003e 209, 142\u0026ndash;150 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTagesson, T. \u003cem\u003eet al.\u003c/em\u003e Ecosystem properties of semiarid savanna grassland in West Africa and its relationship with environmental variability. \u003cem\u003eGlob. Change Biol.\u003c/em\u003e 21, 250\u0026ndash;264 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilliams, C. A. \u003cem\u003eet al.\u003c/em\u003e Africa and the global carbon cycle. \u003cem\u003eCarbon Balance Manag.\u003c/em\u003e 2, 3 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBliefernicht, J. \u003cem\u003eet al.\u003c/em\u003e Field and simulation experiments for investigating regional land\u0026ndash;atmosphere interactions in West Africa: experimental set-up and first results.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAgo, E. E. \u003cem\u003eet al.\u003c/em\u003e Long term observations of carbon dioxide exchange over cultivated savanna under a Sudanian climate in Benin (West Africa). \u003cem\u003eAgric. For. Meteorol.\u003c/em\u003e 197, 13\u0026ndash;25 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRamier, D. \u003cem\u003eet al.\u003c/em\u003e Towards an understanding of coupled physical and biological processes in the cultivated Sahel \u0026ndash; 1. Energy and water. \u003cem\u003eJ. Hydrol.\u003c/em\u003e 375, 204\u0026ndash;216 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, Y. \u003cem\u003eet al.\u003c/em\u003e Quantifying the variability in water use efficiency from the canopy to ecosystem scale across main croplands. \u003cem\u003eAgric. Water Manag.\u003c/em\u003e 262, 107427 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXue, Y., Liang, H., Zhang, B. \u0026amp; He, C. Vegetation restoration dominated the variation of water use efficiency in China. \u003cem\u003eJ. Hydrol.\u003c/em\u003e 612, 128257 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDu, D. \u003cem\u003eet al.\u003c/em\u003e Spatiotemporal dynamics of irrigated cropland water use efficiency and driving factors in northwest China\u0026rsquo;s Hexi Corridor. \u003cem\u003eEcol. Process.\u003c/em\u003e 13, 72 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEl Masri, B. \u003cem\u003eet al.\u003c/em\u003e Carbon and Water Use Efficiencies: A Comparative Analysis of Ten Terrestrial Ecosystem Models under Changing Climate. \u003cem\u003eSci. Rep.\u003c/em\u003e 9, 14680 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFranco, A. \u0026amp; L\u0026uuml;ttge, U. Midday depression in savanna trees: coordinated adjustments in photochemical efficiency, photorespiration, CO2 assimilation and water use efficiency. \u003cem\u003eOecologia\u003c/em\u003e 131, 356\u0026ndash;365 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, J., Wang, Q., Zhan, W., Lian, X. \u0026amp; Gentine, P. When and where soil dryness matters to ecosystem photosynthesis. \u003cem\u003eNat. Plants\u003c/em\u003e 11, 1390\u0026ndash;1400 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, X. \u003cem\u003eet al.\u003c/em\u003e Increasing atmospheric dryness exacerbates mangrove carbon\u0026ndash;water decoupling. \u003cem\u003eNew Phytol.\u003c/em\u003e 249, 169\u0026ndash;180 (2026).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, Y., Zhang, K. \u0026amp; Liu, L. Water use efficiency at multi-time scales and its response to episodic drought and wet periods in a typical subtropical evergreen forest of Southeast China. \u003cem\u003eEcol. Indic.\u003c/em\u003e 151, 110254 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBeer, C. \u003cem\u003eet al.\u003c/em\u003e Temporal and among-site variability of inherent water use efficiency at the ecosystem level. \u003cem\u003eGlob. Biogeochem. Cycles\u003c/em\u003e 23, (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilliams, C. A. \u0026amp; Albertson, J. D. Soil moisture controls on canopy-scale water and carbon fluxes in an African savanna. \u003cem\u003eWater Resour. Res.\u003c/em\u003e 40, (2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMachado, R. \u003cem\u003eet al.\u003c/em\u003e Where do leaf water leaks come from? Trade-offs underlying the variability in minimum conductance across tropical savanna species with contrasting growth strategies. \u003cem\u003eNew Phytol.\u003c/em\u003e 229, 1415\u0026ndash;1430 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu, Z. Y. \u003cem\u003eet al.\u003c/em\u003e Dissecting the Characteristics and Driver Factors on Global Water Use Efficiency Using GLASS Data Sets. \u003cem\u003eEarths Future\u003c/em\u003e 12, e2024EF004630 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCaine, R. S. \u003cem\u003eet al.\u003c/em\u003e Rice with reduced stomatal density conserves water and has improved drought tolerance under future climate conditions. \u003cem\u003eNew Phytol.\u003c/em\u003e 221, 371\u0026ndash;384 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGovindaraj, M. \u003cem\u003eet al.\u003c/em\u003e Evidence from simulated climatic conditions indicates rising CO2 levels impact pearl millet yield and nutritional traits. \u003cem\u003eJ. Agric. Food Res.\u003c/em\u003e 22, 102124 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMEDRANO, H., ESCALONA, J. M., BOTA, J., GUL\u0026Iacute;AS, J. \u0026amp; FLEXAS, J. Regulation of Photosynthesis of C3 Plants in Response to Progressive Drought: Stomatal Conductance as a Reference Parameter. \u003cem\u003eAnn. Bot.\u003c/em\u003e 89, 895\u0026ndash;905 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoche, D. Stomatal Conductance Is Essential for Higher Yield Potential of C3 Crops. \u003cem\u003eCrit. Rev. Plant Sci.\u003c/em\u003e 34, 429\u0026ndash;453 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSales, C. R. G., Wang, Y., Evers, J. B. \u0026amp; Kromdijk, J. Improving C4 photosynthesis to increase productivity under optimal and suboptimal conditions. \u003cem\u003eJ. Exp. Bot.\u003c/em\u003e 72, 5942\u0026ndash;5960 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNguyen, T. H. \u003cem\u003eet al.\u003c/em\u003e Responses of winter wheat and maize to varying soil moisture: From leaf to canopy. \u003cem\u003eAgric. For. Meteorol.\u003c/em\u003e 314, 108803 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNovick, K. A. \u003cem\u003eet al.\u003c/em\u003e The impacts of rising vapour pressure deficit in natural and managed ecosystems. \u003cem\u003ePlant Cell Environ.\u003c/em\u003e 47, 3561\u0026ndash;3589 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin, Y.-S. \u003cem\u003eet al.\u003c/em\u003e Optimal stomatal behaviour around the world. \u003cem\u003eNat. Clim. Change\u003c/em\u003e 5, 459\u0026ndash;464 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKatul, G. G., Palmroth, S. \u0026amp; Oren, R. Leaf stomatal responses to vapour pressure deficit under current and CO2-enriched atmosphere explained by the economics of gas exchange. \u003cem\u003ePlant Cell Environ.\u003c/em\u003e 32, 968\u0026ndash;979 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThomas, D. S., Eamus, D. \u0026amp; Bell, D. Optimization theory of stomatal behaviour: II. Stomatal responses of several tree species of north Australia to changes in light, soil and atmospheric water content and temperature. \u003cem\u003eJ. Exp. Bot.\u003c/em\u003e 50, 393\u0026ndash;400 (1999).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAntoniuk, V. \u003cem\u003eet al.\u003c/em\u003e Diurnal and Seasonal Mapping of Water Deficit Index and Evapotranspiration by an Unmanned Aerial System: A Case Study for Winter Wheat in Denmark. \u003cem\u003eRemote Sens.\u003c/em\u003e 13, 2998 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, S., Yu, B., Huang, Y. \u0026amp; Wang, G. The effect of vapor pressure deficit on water use efficiency at the subdaily time scale. \u003cem\u003eGeophys. Res. Lett.\u003c/em\u003e 41, 5005\u0026ndash;5013 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKnauer, J. \u003cem\u003eet al.\u003c/em\u003e The response of ecosystem water-use efficiency to rising atmospheric CO2 concentrations: sensitivity and large-scale biogeochemical implications. \u003cem\u003eNew Phytol.\u003c/em\u003e 213, 1654\u0026ndash;1666 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim, H. K., Jang, T. I., Im, S. J. \u0026amp; Park, S. W. Estimation of irrigation return flow from paddy fields considering the soil moisture. \u003cem\u003eAgric. Water Manag.\u003c/em\u003e 96, 875\u0026ndash;882 (2009).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBwire, D., Saito, H., Sidle, R. C. \u0026amp; Nishiwaki, J. Water Management and Hydrological Characteristics of Paddy-Rice Fields under Alternate Wetting and Drying Irrigation Practice as Climate Smart Practice: A Review. \u003cem\u003eAgronomy\u003c/em\u003e 14, 1421 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNadolski, L. \u003cem\u003eet al.\u003c/em\u003e Exploring and Closing the Energy Balance of Eddy Covariance Measurements along a Land Use Gradient in the West African Sudanian Savanna. \u003cem\u003eFront. Water\u003c/em\u003e 6, 1393884 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHingerl, L. \u003cem\u003eet al.\u003c/em\u003e Comparative analysis of land\u0026ndash;atmosphere interactions across three contrasting ecosystems in the West Sudanian Savanna. \u003cem\u003eJ. Hydrol. Reg. Stud.\u003c/em\u003e 61, 102751 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYahaya Seydou, A. N. \u003cem\u003eet al.\u003c/em\u003e Biophysical effects of land cover changes in West Africa: a systematic review. \u003cem\u003eEnviron. Res. Lett.\u003c/em\u003e 20, 073001 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSy, S. \u003cem\u003eet al.\u003c/em\u003e Impacts of Idealized Land Use and Land Management Changes on Weather Extremes in West Africa. \u003cem\u003eEarths Future\u003c/em\u003e 13, e2025EF006094 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eObahoundje, S., Tilahun, S. A. \u0026amp; Schmitter, P. Hydroclimatic extremes and food security hotspots in West Africa: evidence from rainfall\u0026ndash;yield interactions. \u003cem\u003eFront. Clim.\u003c/em\u003e 7, (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSy, S. \u0026amp; Quesada, B. Anthropogenic land cover change impact on climate extremes during the 21st century. \u003cem\u003eEnviron. Res. Lett.\u003c/em\u003e 15, 034002 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuug, S. \u003cem\u003eet al.\u003c/em\u003e Methane emissions from rice cultivation in West Africa and compensation options from nature reserve forests. \u003cem\u003eEnviron. Res. Lett.\u003c/em\u003e 20, 044050 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBerger, S. \u003cem\u003eet al.\u003c/em\u003e The impact of rain events on CO2 emissions from contrasting land use systems in semi-arid West African savannas. \u003cem\u003eSci. Total Environ.\u003c/em\u003e 647, 1478\u0026ndash;1489 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBliefernicht, J. \u003cem\u003eet al.\u003c/em\u003e The WASCAL Hydrometeorological Observatory in the Sudan Savanna of Burkina Faso and Ghana. \u003cem\u003eVadose Zone J.\u003c/em\u003e 17, 180065 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNyadzi, E., Bessah, E., Kranjac-Berisavljevic, G. \u0026amp; Ludwig, F. Hydro-climatic and land use/cover changes in Nasia catchment of the White Volta basin in Ghana. \u003cem\u003eTheor. Appl. Climatol.\u003c/em\u003e 146, 1297\u0026ndash;1314 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbdul-Ganiyu, S. Analysis of some hydrological parameters of the Nasia River Catchment in the Northern Region of Ghana for its socio-economic development. \u003cem\u003eAfr. J. Agric. RESEEARCH\u003c/em\u003e 6, (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchmitt, K. \u0026amp; Adu-Nsiah, M. \u003cem\u003eThe Vegetation of Mole National Park\u003c/em\u003e. (1993).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBowell, R. J. \u0026amp; Ansah, R. K. Mineral status of soils and forage in the Mole National Park, Ghana and implications for wildlife nutrition. \u003cem\u003eEnviron. Geochem. Health\u003c/em\u003e 16, 41\u0026ndash;58 (1994).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNicholson, S. E. The West African Sahel: A Review of Recent Studies on the Rainfall Regime and Its Interannual Variability. \u003cem\u003eInt. Sch. Res. Not.\u003c/em\u003e 2013, 453521 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBliefernicht, J. \u003cem\u003eet al.\u003c/em\u003e The WASCAL hydrometeorological observatory in the Sudan Savanna of Burkina Faso and Ghana. \u003cem\u003eVadose Zone J.\u003c/em\u003e 17, 1\u0026ndash;20 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSy, S. \u003cem\u003eet al.\u003c/em\u003e Impacts of Idealized Land Use and Land Management Changes on Weather Extremes in West Africa. \u003cem\u003eEarths Future\u003c/em\u003e 13, e2025EF006094 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCharuchittipan, D., Babel, W., Mauder, M., Leps, J.-P. \u0026amp; Foken, T. Extension of the Averaging Time in Eddy-Covariance Measurements and Its Effect on the Energy Balance Closure. \u003cem\u003eBound.-Layer Meteorol.\u003c/em\u003e 152, 303\u0026ndash;327 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKutikoff, S. \u003cem\u003eet al.\u003c/em\u003e Heat storage and its effect on the surface energy balance closure under advective conditions. \u003cem\u003eAgric. For. Meteorol.\u003c/em\u003e 265, 56\u0026ndash;69 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMauder, M. \u0026amp; Foken, T. Documentation and Instruction Manual of the Eddy-Covariance Software Package TK3.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFoken, T. \u003cem\u003eet al.\u003c/em\u003e Post-Field Data Quality Control. in \u003cem\u003eHandbook of Micrometeorology: A Guide for Surface Flux Measurement and Analysis\u003c/em\u003e (eds Lee, X., Massman, W. \u0026amp; Law, B.) 181\u0026ndash;208 (Springer Netherlands, Dordrecht, 2005). doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/1-4020-2265-4_9\u003c/span\u003e\u003cspan address=\"10.1007/1-4020-2265-4_9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGr\u0026uuml;Nwald, T. \u0026amp; Bernhofer, C. A decade of carbon, water and energy flux measurements of an old spruce forest at the Anchor Station Tharandt. \u003cem\u003eTellus B Chem. Phys. Meteorol.\u003c/em\u003e 59, 387\u0026ndash;396 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMauder, M. \u003cem\u003eet al.\u003c/em\u003e A strategy for quality and uncertainty assessment of long-term eddy-covariance measurements. \u003cem\u003eAgric. For. Meteorol.\u003c/em\u003e 169, 122\u0026ndash;135 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMauder, M., Jung, M., Stoy, P., Nelson, J. \u0026amp; Wanner, L. Energy balance closure at FLUXNET sites revisited. \u003cem\u003eAgric. For. Meteorol.\u003c/em\u003e 358, 110235 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilson, K. \u003cem\u003eet al.\u003c/em\u003e Energy balance closure at FLUXNET sites. \u003cem\u003eAgric. For. Meteorol.\u003c/em\u003e 113, 223\u0026ndash;243 (2002).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKljun, N., Calanca, P., Rotach, M. W. \u0026amp; Schmid, H. P. A simple two-dimensional parameterisation for Flux Footprint Prediction (FFP). \u003cem\u003eGeosci. Model Dev.\u003c/em\u003e 8, 3695\u0026ndash;3713 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang, D. \u003cem\u003eet al.\u003c/em\u003e The lag and cumulative response of water use efficiency to the climate on the Shiyang river basin, Northwest of China. \u003cem\u003eSci. Rep.\u003c/em\u003e 15, 23659 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, S., Yu, B., Zhang, Y., Huang, Y. \u0026amp; Wang, G. Partitioning evapotranspiration based on the concept of underlying water use efficiency. \u003cem\u003eWater Resour. Res.\u003c/em\u003e 52, 1160\u0026ndash;1175 (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, S., Yu, B., Huang, Y. \u0026amp; Wang, G. Daily underlying water use efficiency for AmeriFlux sites. \u003cem\u003eJ. Geophys. Res. Biogeosciences\u003c/em\u003e 120, 887\u0026ndash;902 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCroux, C. \u0026amp; Dehon, C. Influence functions of the Spearman and Kendall correlation measures. \u003cem\u003eStat. Methods Appl.\u003c/em\u003e 19, 497\u0026ndash;515 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiouf, I. \u003cem\u003eet al.\u003c/em\u003e Potential Contribution of Climate Conditions on COVID-19 Pandemic Transmission over West and North African Countries. \u003cem\u003eAtmosphere\u003c/em\u003e 13, 34 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSy, S. \u003cem\u003eet al.\u003c/em\u003e Sensitivity of trends to estimation methods and quantification of subsampling effects in global radiosounding temperature and humidity time series. \u003cem\u003eInt. J. Climatol.\u003c/em\u003e 41, E1992\u0026ndash;E2014 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKroll, C. N. \u0026amp; Song, P. Impact of multicollinearity on small sample hydrologic regression models. \u003cem\u003eWater Resour. Res.\u003c/em\u003e 49, 3756\u0026ndash;3769 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuthukrishnan, R. \u0026amp; Lowast, C. K. J. The Effect of Multicollinearity on Feature Selection. \u003cem\u003eIndian J. Sci. Technol.\u003c/em\u003e 17, 3664\u0026ndash;3668 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNaseri, M. Y., Bernosky, G., Mayer, P. W. \u0026amp; Marston, L. T. Patterns and Predictors of Residential Indoor Water Use Across Major US Cities. \u003cem\u003eEarths Future\u003c/em\u003e 13, e2024EF005467 (2025).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDertli, H. I., Hayes, D. B. \u0026amp; Zorn, T. G. Effects of multicollinearity and data granularity on regression models of stream temperature. \u003cem\u003eJ. Hydrol.\u003c/em\u003e 639, 131572 (2024).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-sustainable-agriculture","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [npj Sustainable Agriculture](https://www.nature.com/npjsustainagric/)","snPcode":"44264","submissionUrl":"https://submission.springernature.com/new-submission/44264/3","title":"npj Sustainable Agriculture","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"ecosystem-climate interactions, hydrology, carbon-water coupling, eddy-covariance, West Africa","lastPublishedDoi":"10.21203/rs.3.rs-8522874/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8522874/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWater use efficiency (WUE) is a key indicator of ecosystem balance, reflecting how productivity responds to hydrological constraints under climate change. However, variability in WUE and its environmental drivers across West African agroecosystems remains poorly understood. Here, we integrate multi-year (2019\u0026ndash;2024), half-hourly eddy-covariance observations of carbon and water-vapor fluxes from four contrasting land-use types in northern Ghana: a reserve savanna forest, rain-fed paddy rice, grassland, and rain-fed cropland. WUE exhibited pronounced diurnal and seasonal variability, shaped by hydrological, atmospheric, and land-management drivers. Diurnal patterns were bimodal, with morning and afternoon peaks shifting between wet and dry seasons. During the wet season, mean WUE was highest in the savanna forest (3.1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.26 g C kg⁻\u0026sup1; H₂O), followed by paddy rice (2.08\u0026thinsp;\u0026plusmn;\u0026thinsp;0.21), cropland (1.93\u0026thinsp;\u0026plusmn;\u0026thinsp;0.20), and grassland (1.66\u0026thinsp;\u0026plusmn;\u0026thinsp;0.18). Seasonal analyses highlighted ecosystem-specific controls, reflecting differences in radiation, soil moisture, and cultivation practices.\u003c/p\u003e","manuscriptTitle":"Water use efficiency responses in contrasting agroecosystems and land management practices in West Africa","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-01-08 04:32:40","doi":"10.21203/rs.3.rs-8522874/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"7044960352609557254037094459352954412","date":"2026-02-17T09:39:36+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-29T15:30:29+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"335803756359129719886923676298064264255","date":"2026-01-19T06:01:16+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-14T03:55:06+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-12T05:00:51+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-08T13:38:46+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Sustainable Agriculture","date":"2026-01-05T15:03:05+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-sustainable-agriculture","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [npj Sustainable Agriculture](https://www.nature.com/npjsustainagric/)","snPcode":"44264","submissionUrl":"https://submission.springernature.com/new-submission/44264/3","title":"npj Sustainable Agriculture","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"cca0dee6-b0a5-48d9-896f-43533344f976","owner":[],"postedDate":"January 8th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":60751721,"name":"Earth and environmental sciences/Climate sciences"},{"id":60751722,"name":"Biological sciences/Ecology"},{"id":60751723,"name":"Earth and environmental sciences/Ecology"},{"id":60751724,"name":"Earth and environmental sciences/Environmental sciences"},{"id":60751725,"name":"Earth and environmental sciences/Hydrology"}],"tags":[],"updatedAt":"2026-01-14T04:08:21+00:00","versionOfRecord":[],"versionCreatedAt":"2026-01-08 04:32:40","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8522874","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8522874","identity":"rs-8522874","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}