Assessing the role of gridded evapotranspiration products in improving streamflow simulation and reducing hydrological modeling uncertainty

Edge effects on evapotranspiration from tropical forest fragments

Accurate estimation of global terrestrial evapotranspiration (ET) is fundamental for understanding the Earth’s water and energy cycles, yet existing multi-source ET products inevitably contain uncertainties that require spatially explicit characterization for optimal data merging or data assimilation. While Quadruple Collocation Analysis (QCA) offers a robust and reference-free approach to quantify uncertainties, its reliability in the ET discipline remains underexplored, and algorithmic non-convergence frequently results in substantial spatial data gaps. To address these limitations, this study evaluated the accuracy of the QCA method using validation errors derived from high-quality FLUXNET sites (N = 55). Moreover, we employed a Random Forest (RF) framework that is driven by 17 environmental variables to generate spatially seamless error maps for four mainstream ET products, i.e., ERA5, GLDAS, GLEAM, and MERRA2, from 2000 to 2020. Results demonstrate that QCA-based errors strongly correlated with ground-based errors as Pearson’s correlation coefficient was >0.3 for all four ET products. Furthermore, the RF model successfully reconstructed the spatial gaps in QCA errors, achieving an exceptionally low mean prediction error of approximately 0.03 mm/day. Based on these seamless maps, the global mean ET error is estimated at roughly 0.3 mm/day, with pronounced high-error clusters emerging in regions such as central Canada and northern Argentina driven by underlying land cover heterogeneity. Ultimately, this seamless gap-filling redefined the global map of product with the lowest estimated collocation error. ERA5 emerged as the superior choice across approximately 45% of the land surface (predominantly in the tropics and mid-to-high latitudes). Meanwhile, before algorithmic gap-filling, GLEAM was optimal across approximately 28% of the valid land pixels; after spatial gap-filling, it proved most effective across approximately 30% of the globe, particularly within arid deserts and glaciated regions. Our work provides useful geographic guidance for optimizing multi-source data merging and land data assimilation frameworks in future global hydrological studies.

Globally, climate change is increasing stress on freshwater resources and rainfed agriculture, disrupting food systems and livelihoods, especially in regions where there is low adaptive capacity like West Africa. The Volta River Basin (VRB) is the focus of this study because it is a lifeline to over 24 million people. This study investigates how increasing temperatures, altered rainfall patterns and increased atmospheric water demand are affecting evapotranspiration (ET) and crop water stress in maize and rice production. Using over two decades of satellite data (2000–24) and future projections under RCP 4.5 and RCP 8.5, the results indicate a pronounced but non‐linear increase in reference evapotranspiration (ET₀), with peak values occurring around 2018, followed by a decline towards 2024. This trend coincides with a warming rate of +0.012 to +0.021°C yr ‐1 and dry‐season Crop Water Stress Index (CWSI) values exceeding 0.75, extending well into the severe stress regime. Spatial analysis identifies northern and transitional areas as endemic hotspots of water stress. By integrating climate data and crop physiology, the study offers prescriptive insight for building agricultural resilience. It prescribes site‐specific interventions like drought‐tolerant crop varieties, supplementary irrigation and soil water conservation. The study informs climate adaptation, risk zoning and policy planning, ensuring food security and rural livelihoods.

As water resources are becoming increasingly scarce in the North China Plain, irrigation strategies that simultaneously improve grain yield and reduce irrigation water input are needed for winter wheat (Triticum aestivum L.) production. Current irrigation decision rules are based either on fixed soil moisture thresholds or on evapotranspiration (ET)-based ratios applied uniformly across the growing season, limiting their flexibility for growth stage-specific irrigation management. In this study, a multi-objective simulation optimization framework was developed to jointly optimize soil moisture lower control limits (irrigation trigger thresholds) and evapotranspiration-based irrigation replenishment ratios across key winter wheat growth stages. The framework integrated the AquaCrop-OSPy crop model with the PyFAO56 soil moisture balance, irrigation scheduling model and the NSGA-II evolutionary optimization algorithm. A field experiment was conducted during the 2024–2025 growing season in Laoling City, Shandong Province, China, employing a four-dense–one-sparse strip cropping pattern with two irrigation treatments: T1 (subsurface sprinkler irrigation) and T2 (shallow subsurface drip irrigation). The AquaCrop-OSPy model was calibrated and validated using measured canopy cover, aboveground biomass, grain yield, and soil moisture content in the 0–60 cm soil layer. Simulated canopy cover and grain yield showed good agreement with observations, with the coefficient of determination (R2) ranging from 0.87 to 0.94. For grain yield, the normalized root mean square error (NRMSE) ranged from 2.24% to 3.75%, and the root mean square error (RMSE) ranged from 0.29 to 0.54 t·ha−1. For aboveground biomass, R2 was 0.99, while RMSE ranged from 1.02 to 1.11 t·ha−1, and NRMSE ranged from 14.25% to 15.49%. The PyFAO56 irrigation strategy model simulated average root-zone soil-moisture dynamics with satisfactory accuracy, with an R2 of 0.86 and an RMSE of 5%. Multi-objective optimization (maximizing yield while minimizing irrigation volume) generated 23 Pareto-optimal irrigation strategies, with irrigation volumes ranging from 51 to 128 mm, corresponding yields ranging from 9.8 to 10.8 t·ha−1, and irrigation water use efficiency (IWUE) ranging from 0.08 to 0.19 t·ha−1·mm−1. Correlation analysis within the Pareto set indicated that soil-moisture control lower limits during the regreening–jointing stage and higher soil-moisture control lower limits during the flowering–maturity stage were key controlling factors for achieving high yields and irrigation water use efficiency. The Entropy-Weighted Ranked Minimum Distance method identified an optimal irrigation scheme involving two irrigations (one at the end of the jointing stage and another at the beginning of the grain filling stage) involving an irrigation depth of 75 mm, achieving a simulated yield of 10.4 t·ha−1 and an IWUE of 0.16 t·ha−1·mm−1. The proposed AquaCrop-PyFAO56-NSGA-II framework provides a flexible, process-based workflow for jointly optimizing irrigation control thresholds and evapotranspiration-based irrigation replenishment ratios across different winter wheat growth stages. Under the monitored conditions of the 2024–2025 wet season, the framework identified a two-irrigation strategy that balanced grain yield and irrigation input. This study should, therefore, be regarded as a proof-of-concept evaluation conducted in a well-instrumented single-site field setting rather than as a universally transferable recommendation. Because model calibration, within-season validation, and optimization were all based on one wet growing season at one site, the derived stage-specific thresholds, Pareto front, and S5 recommendation are most applicable to hydro-climatic conditions similar to the study year and should be further tested across contrasting year-types and locations before broader extrapolation.

In Mediterranean and semi-arid regions, climate change imposes significant pressure on water resources and hydrological systems. Accurate simulation and measurement of water availability require precise hydrological models, usually calibrated using streamflow data. However, the increasing availability of remote sensing data, such as evapotranspiration (ET), offers new opportunities to enhance calibration, particularly in data-scarce regions. This study used the SWAT+ model, implemented via the R-SWAT parallel computing framework, to analyze calibration strategies in the Upper Oum Er Rbia watershed, Morocco. Two sensitivity analyses were conducted to identify parameters affecting streamflow and ET, which were then used to evaluate five calibration strategies using streamflow observations and GLDAS ET data. Results showed that the single-variable calibrations achieved high Q or ET performance but reduced performance in the other variable. In contrast, a multi-variable calibration approach (Multi_Q&ET) optimized streamflow and ET, achieving satisfactory results. Streamflow calibration and validation performance metrics were NSE = 0.75 and 0.67, KGE = 0.77 and 0.70, and PBIAS = 19.2% and 9.1%. For ET, calibration and validation yielded NSE = 0.51 and 0.55, KGE = 0.64 and 0.71, and PBIAS = 5.0% and 2.9%. While slightly less accurate than single-objective calibrations, the multi-variable approach achieved balanced and acceptable performance across variables. These findings highlight the significant contribution of open-access remote sensing evapotranspiration data in refining model parameters, offering a realistic and effective strategy to enhance hydrological model calibration. This approach is particularly valuable for improving water resource management in regions vulnerable to climate change and limited by observational data. • Integrating GLDAS evapotranspiration (ET) data improves hydrological model calibration. • Multi-objective calibration in SWAT+ balances accuracy across variables. • Parallel computing via R-SWAT enhances calibration efficiency, enabling robust sensitivity analysis. • RS data improves hydrological modeling offering a scalable solution for water management.

Benchmarking multi-model evapotranspiration in the Yangtze River Basin using GNSS/GRACE fusion and anthropogenic disturbance quantification

Influence pathways of hydrological processes: Perspectives from the pathway probability in different types of watersheds.

Terrestrial ecosystems play a vital role in sequestering carbon dioxide (CO2); however, growing both agri-horticultural intensifications can influence CO2 sequestration potential. This study was conducted in an agri-horticultural-intensive ecosystem of Central India to investigate the net ecosystem exchange (NEE) of CO2, water (H2O), and energy fluxes and assess the influence of meteorological factors using the eddy covariance datasets acquired from January to June 2024. The highest monthly mean diurnal CO2 flux was recorded in January (-7.98µmolm-2s-1), and the lowest was in June (-3.03µmolm-2s-1), resulting in a cumulative CO2 uptake of -94.5g C m-2 half-year-1. Evapotranspiration (ET) over the January-to-June period was recorded at 273.3mm, with sensible heat (H) dominating over latent heat (LE) flux, indicating water-stressed semi-arid ecosystem. Over the 6-month period, the ecosystem exhibited a net ecosystem exchange (NEE) of -177.59g C m-2 half-year-1, a gross primary productivity (GPP) of 567.13g C m-2 half-year-1, and an ecosystem respiration (Reco) of 389.54g C m-2 half-year-1. Despite agri-horticultural intensification, the ecosystem functioned as a strong sink for atmospheric CO2. The findings provide critical insights into carbon, water, and energy exchanges in agri-horticultural landscapes and are essential for understanding the market-driven agri-horticultural intensification and developing sustainable land restoration strategies to address groundwater depletion, climate vulnerability and ecosystem functioning.

Abstract Due to its location at the interface between land surface and atmosphere, soil moisture (SM) plays an important role in modulating energy, water and carbon fluxes. During periods of decreasing SM, SM loss is dependent on evapotranspiration (ET), drainage and changes in plant water storage. Investigating SM loss can give important insights into these processes. Here we use 25 years of global remote sensing data to investigate how SM loss is controlled by vegetation and temperature. We find that positive vegetation anomalies lead to slower SM loss in most areas, except for cold boreal forests. We hypothesize that these effects arise from competing effects of soil shading, transpiration and root water uptake by the vegetation. The effect whereby positive vegetation anomalies increase SM loss is limited to high SM conditions and disappears at lower SM, likely due to water stress limiting transpiration. By analyzing temperature and vegetation anomalies jointly we find that the relationship between SM loss and temperature varies between regions, but vegetation cover effects persist across the full range of temperature anomalies. Using a simple energy and moisture budget model, we can reproduce observed vegetation and temperature effects, supporting the interpretation that vegetation controls topsoil SM loss through shading and transpiration. We also find widespread positive SM loss trends which indicates accelerated topsoil water cycling, likely due to higher atmospheric water demand driven by increasing temperatures.

ABSTRACT The isotope altitude effect is a key indicator of mountain hydrological processes, yet its ecohydrological regulation in complex terrain remains poorly understood. This study investigates the spatial and seasonal variations in δ 2 H, δ 18 O and d ‐excess in surface water across three slope aspects of Cangshan Mountain, Southwest China, based on 116 surface water samples collected during the 2022 rainy season and 2023 dry season. Integrating isotope data with remotely sensed ecological indicators, we examine how ecological factors modulate the δ 18 O–elevation relationship. Results show surface water isotopes are more enriched in the dry season, with δ 18 O following a North > East > West spatial pattern and d ‐excess opposite. The local surface water line (δ 2 H = 4.09δ 18 O‐40.98, R 2 = 0.82) indicates substantial evaporative enrichment. The δ 18 O lapse rate exhibits strong slope‐dependence: steepest on the North (−6.56‰ to −3.67‰/km), weaker on the East (−0.35‰ to −1.47‰/km) and reversed on the West (+0.45‰ to +1.25‰/km). Structural equation modelling indicates a physical‐driver–ecological‐modulation framework mechanism: altitude affects δ 18 O indirectly through land surface temperature (LST), while normalized difference vegetation index (NDVI), modified normalized difference water index (MNDWI) and evapotranspiration (ET) jointly determine lapse rate amplification, attenuation, or reversal across slope aspects and seasons. These results demonstrate that isotopic altitudinal gradients in mountain streams are highly sensitive to slope aspects and ecohydrological heterogeneity, with implications for water‐source identification and isotope‐informed hydrological assessment in complex mountain terrain.

Global agricultural water scarcity necessitates intelligent irrigation management systems capable of optimizing water allocation while preserving crop productivity. This paper proposes a reinforcement learning (RL) framework for adaptive irrigation scheduling under seasonal water budget constraints, formulated as a constrained Markov decision process (MDP) and solved using a deep Q-network (DQN) agent. The DQN agent processes multi-layer soil water content, crop developmental stage, accumulated evapotranspiration (ET) deficit, and short-term precipitation forecasts as state inputs, selecting from five discrete daily irrigation volumes. A calibrated AquaCrop crop simulation model serves as the training environment, providing physically realistic state transitions across soil compartments. A dual-component reward function jointly penalizes water application cost and crop water stress. Evaluated across ten held-out growing seasons, the DQN agent reduced total seasonal irrigation by 18.4% relative to a conventional threshold-based scheduler, while maintaining mean grain yield at 97.2% of the benchmark. These results confirm that RL-driven adaptive policies can effectively navigate the yield-water tradeoff under resource-constrained agricultural conditions.

Unraveling the drivers and synergistic mechanisms of selenium distribution in cultivated soils across China: A quantitative analysis using explainable machine learning.

Soil moisture constraints override atmospheric aridity in governing vegetation dynamics on the warming Qinghai-Tibetan plateau

Yellow River Basin (YRB) This study aims to systematically investigate the spatiotemporal evolution patterns and driving mechanisms of ecological drought in the Yellow River Basin (YRB) from 1982 to 2022. A novel Standardized Ecological Water Deficit Index (SEWDI) integrating vegetation dynamics and hydrological processes is developed. Using multi-source data and an integrated methodology including improved run theory, Copula joint probability models, Modified Mann-Kendall method (MMK) trend analysis, and XGBoost-SHAP machine learning framework, the research quantitatively assesses drought characteristics across multiple time scales and identifies key driving factors. A significant basin-wide drying trend (–0.017/10a) with strongest intensification in western upstream regions (–0.043/10a). Distinct westward migration of drought centers, with over 91% of western areas affected in the 2010s. The July 2019–April 2020 event was the most severe on record, peaking in February 2020 with 98.08% of the basin affected, a severity of 9.14, and a 10-month duration. Drought intensification is particularly pronounced in December (Z s = –1.33) and winter (Z s = –1.46), with 97.79% and 94.25% of the basin showing aggravating trends, respectively. Evapotranspiration (ET) emerges as dominant climatic driver, while Atlantic Multidecadal Oscillation (AMO) is primary circulation factor exacerbating drought. These findings provide crucial insights for ecological drought early warning and adaptive water resource management in the YRB. • Ecological drought intensified significantly in the Yellow River Basin (1982–2022). • The most severe drought lasted 10 months and affected 98% of the basin. • Evapotranspiration was the dominant climatic driver of ecological drought. • Atlantic Multidecadal Oscillation primarily exacerbated drought conditions.

Achieving a sustainable future for water resources demands accurate models that address the interdisciplinary nature of water dynamics. The eco-hydrological-socioeconomic (ECHO) framework integrates physics-based hydrological models with data-driven machine learning techniques, leveraging reanalysis and multi-source remote sensing data. This enables dynamic estimation of sector-specific water demand and interaction with hydrological estimates. ECHO's modular structure allows coupling with grid-based models and includes modules for runoff, evapotranspiration (ET), groundwater flow, surface water routing, and water demand estimation. Calibration and validation demonstrate robust performance in simulating rainfall-runoff processes, with strong agreement observed for monthly ET estimates and gravity recovery and climate experiment-follow on (GRACE-FO) data on total water storage changes. The model accurately estimates total water demand across sectors and aligns with recorded water use data. Simulation outputs of water stress closely match findings from the China Water Resources Bulletin, while also showing promise to enhance projections aligned with sustainable development goals (SDGs) for global water management strategies. By providing high-resolution, dynamic assessments, ECHO offers a scalable tool for policymakers to identify water stress hotspots and optimize allocation strategies essential for meeting SDG targets.

The accurate estimation of reference evapotranspiration (ETo) on seasonal and annual scale is vital for determination of water requirement of crops and impact of climate change on irrigated agriculture. This study investigates the influence of climatic variables (temperature, relative-humidity, solar-radiation, and wind speed) on ETo and comparison of the performance evaluation of temperature and radiation-based models over Cross River basin on seasonal and annual basis. Estimation of ETo was done using temperature-based; Schendel, Samani, Trajkovic, Droogers & Allen-2, Dorji, Hadria, Hargreaves-samani, Blaney-Morin-Nigeria, and radiation-based models; Jensen-Haise, Stephens and Stewart, Oudin, Abtew, Irmak-1, Copais, Tabari & Talaee-4, and Hargreaves under humid-tropic condition, with Penman–Monteith (PM-ETo) as a reference. Remotely-sensed meteorological variables from 1987 to 2017 were sourced from the Climatic Research Unit (CRU) database, over 22 stations. These variables were used for estimating ETo. Models were evaluated with coefficient of determination, a root-mean-square-error, Willmott’s index of agreement, Percentage-Bias and Nash–Sutcliffe Efficiency. Sensitivity analysis (± 5%) revealed that PM-ETo was most sensitive to solar-radiation and temperature, compared to relative-humidity and wind speed. Blaney-Morin-Nigeria demonstrated seasonal estimation bias. Further analysis revealed that radiation-based models out-performed the temperature-based models across all categories. Blaney-Morin-Nigeria and Copais recorded the best performance in dry season, while Hargreaves–Samani and Abtew recorded the best performance in rainy season. Abtew and Hargreaves–Samani recorded the best performance on annual basis. An increasing trend (slope = 0.0029 mm/day) of PM-ETo further suggests global warming scenario. This demonstrates the ability of radiation-driven models for crop water requirement estimation in data-scarce regions.

Ecosystem transpiration (T) is the core process in terrestrial water and carbon cycles. Accurately estimating T is critical to improving evapotranspiration (ET) models and understanding global ecosystem responses to climate change. In this study, we evaluated four ET partitioning methods (TEA, Z16, L19, and Y21) using 368 global eddy covariance (EC) sites and 15 sap flow sites. Intercomparison results showed that TEA, Z16, and Y21 maintained good consistency, whereas L19 exhibited lower agreement, primarily due to its high sensitivity to energy closure errors and poor non-linear fitting accuracy under extreme conditions. Validation against sap flow data indicated that Z16 performed best (R2 = 0.45, KGE = 0.52), followed by Y21, while TEA had the lowest accuracy due to systematic overestimation driven by unremoved persistent background soil evaporation in its training dataset. Global analysis revealed that mean annual T ranged from 213 mm yr−1 (Z16) to 294 mm yr−1 (TEA), with annual T/ET varying between 0.45 (Z16) and 0.63 (TEA). Trend analysis further showed consistent increasing trends across all four methods for both annual T (0.33–0.83 mm·yr−2) and annual T/ET (0.0015–0.0019 yr−1). Additionally, a notably stronger relationship was found between gross primary productivity (GPP) and T than between GPP and ET. Despite substantial differences in model structures, these methods effectively capture the temporal dynamics of T and the coupled relationships between ecosystem carbon and water fluxes. Our findings provide critical benchmarks for terrestrial water cycle modeling and sustainable water resource management under a changing climate.

The 2022 Chongqing wildfires, occurring during an unprecedented heatwave, severely degraded subtropical forest ecosystems and disrupted hydrological cycling. We developed an integrated artificial intelligence framework combining Long Short-Term Memory and Transformer architectures to simulate post-fire evapotranspiration (ET) dynamics using 37 months of field observations (2022–2025) across 24 plots with four burn severities. The Penman–Monteith–Leuning model provided physically based benchmarks. Results revealed three distinct recovery phases: destruction/stagnation (0–7 months, ET at 6%–10% of pre-fire levels), rapid recovery (8–19 months), and stabilization (20–37 months, reaching 100% ET recovery). The coupled LSTM–Transformer ensemble achieved superior performance (RMSE = 0.10 mm·day−1, NSE = 0.98), outperforming single models by 31% in uncertainty reduction. SHAP analysis identified phase-dependent factor shifts: soil water content dominated Stage I (42.5%), while leaf area index (LAI) controlled Stages II–III (>48%). A bimodal LAI time-lag effect emerged: 4–7 days (leaf water potential equilibrium, 27.7% contribution) and 8–14 days (root uptake compensation, 21.7%). Burn severity significantly extended time-lags (severe burns: 12/21 days vs. unburned: 5/12 days), indicating hydraulic system reconstruction requirements. Despite equivalent LAI recovery, severe burns maintained 12%–15% ET reduction, suggesting lasting hydraulic limitations. This study demonstrates that physics-constrained AI models effectively capture complex post-fire ecohydrological dynamics while providing mechanistic interpretability, advancing understanding of vegetation–water coupling reconstruction under increasing fire frequency.

As competition for water resources intensifies, especially in water-scarce regions, there is a growing need to manage water usage effectively, particularly in irrigated agriculture. However, data on agricultural water use and abstractions are often unavailable or only at coarse spatial resolutions. Water use by crops can be physically characterised by the landscape’s rate of evapotranspiration (ET). Satellite-based monitoring of actual ET provides one potential solution to this data gap, but significant knowledge gaps remain about the uncertainty in satellite-based estimates of irrigation water use (IWU) and their associated implications for policy and management. In this model-intercomparison study, we attempt to address the relevance of model choice on satellite-based estimates of IWU by assessing the variability resulting from different model estimates of satellite-based IWU. We utilised six satellite-based ET datasets from OpenET and five precipitation datasets to estimate field-level IWU over 6 years in the high plains aquifer, United States. Results reveal substantial variability in IWU estimates, particularly at field and seasonal scales, which reduces when aggregated spatially or temporally. ET, rather than precipitation, was the primary driver of variability in IWU estimates. These findings highlight the challenges of using satellite data to estimate IWU at fine spatial and temporal scales or in areas where irrigation supplements rainfall. Aggregating IWU estimates reduces variability but emphasises the importance of model choices when monitoring irrigation water usage both at the farm and at regional levels.