Search Results (7,760)

Winter food conditions can shape behavioral investment, yet measured food mass and short-term foraging behavior need not covary. We quantified the potentially available food dry mass (FD), foraging-attempt rate (FAR), and 15 min behavioral time allocation of wintering black-necked cranes (Grus nigricollis Przevalski, 1876) in five habitats across three sampling stages in Linzhou County, XiZang. Food was sampled in eight independent plots per stage × habitat combination, and 600 valid focal observations were analyzed under a working-independence assumption. Habitat, sampling stage, and their interaction were associated with both FD and FAR (all nominal p < 0.001 in the fixed-effects models). Unplowed farmland consistently had the highest FD and FAR, whereas riverbank and shallow-water areas generally had the lowest values. Foraging accounted for 59.2–75.3% of observed time. Because different plots and partly different valleys were sampled among stages, these stage-associated contrasts combine temporal and spatial variation. Plot-level FD–FAR relationships varied among habitats and stages, showing that measured dry mass and a 1 min behavioral rate were not interchangeable indicators of food accessibility, foraging success, or energetic payoff. Future work should revisit fixed plots, retain individual–flock–plot–session identifiers, and quantify food identity, burial depth, hydrological conditions, successful ingestion, and detection-corrected habitat use. Full article

Polders are typical semi-artificial and human-dominated ecosystems widely distributed in the middle and lower reaches of the Yangtze River Basin. They serve as important sinks and sources of nitrogen (N) and phosphorus (P) in agricultural watersheds. Long-term intensive human intervention substantially alters the hydrological and biogeochemical processes of polder ecosystems, resulting in complex and uncertain effects on water quality that remain insufficiently understood. This study conducts a systematic literature review and narrative synthesis of evidence on N and P transport in polders across the middle–lower Yangtze River plain. The evidence is derived from field monitoring, plot experiments, and numerical simulations. The review focuses on the spatiotemporal patterns of nutrient variation, sink–source conversion functions of ditches and small ponds, drivers of nutrient loss, and current research bottlenecks under artificial sluice-pump regulation. The synthesized results indicate that polders exhibit a distinctive nutrient transport pattern characterized by dispersed in situ retention under conventional water management and concentrated pulse export during drainage events. Artificial sluice-pump operation drives episodic nutrient export throughout the crop growth period, imposing persistent pressure on the water quality of downstream rivers and lakes. N and P transformations are jointly controlled by natural hydrological fluctuations and human activities. Within agricultural lands of polders, fertilizers account for 73.3% of total nitrogen inputs and 87.9–93.5% of total phosphorus inputs. Crop harvesting and regulated drainage constitute the two dominant pathways for nutrient export. Hydraulic regulation prolongs water residence time in polder ditches and ponds, resulting in retention efficiencies of 52–65% for allochthonous N and P. However, seasonal flooding and waterlogging can induce sediment hypoxia and endogenous nutrient release, thereby causing secondary internal pollution and increasing the eutrophication risk of adjacent receiving water bodies. Three major research gaps are identified: insufficient long-term continuous multi-indicator monitoring data, limited model applicability for simulating human-regulated hydrology–nutrient coupling, and poorly defined critical thresholds for polder sink–source functional reversal. This regional systematic review advances the understanding of human–hydrology–nutrient coupling mechanisms in Yangtze River polder systems. It also provides targeted theoretical support for agricultural non-point source pollution mitigation and water environment management in floodplain agricultural areas. Full article

Accurate precipitation information is crucial for hydrological simulation and water resources management, especially in humid mountainous regions where complex terrain and spatially heterogeneous rainfall introduce considerable uncertainties. Satellite-based precipitation products provide important data sources for hydrological applications; however, their reliability and impacts on runoff simulations remain uncertain across different spatial and temporal scales. This study presents a multi-scale validation of four precipitation products, including GSMaP, PERSIANN, GPM IMERG, and CLDAS, and evaluates their effects on hydrological simulation in the Oujiang River Basin, a typical humid mountainous basin in southeastern China. Daily precipitation estimates from 2015 to 2020 were compared with gauge observations using statistical metrics and precipitation event indicators. The hydrological applicability of each product was further assessed by driving a semi-distributed Xin’anjiang model, with evaluations conducted at the basin outlet, seasonal periods, extreme rainfall events, and internal subbasins. Results showed that CLDAS achieved the best overall agreement with gauge observations, with lower systematic bias and higher capability in detecting precipitation variability. Satellite-only products exhibited larger uncertainties, particularly during extreme rainfall events and in areas with complex terrain. These precipitation uncertainties were further propagated into runoff simulations, leading to differences in hydrological performance among products. CLDAS-driven simulations showed the highest accuracy, achieving R², NSE, and KGE values of 0.706, 0.681, and 0.815, respectively, which were comparable to simulations driven by gauge-based precipitation. Multi-scale analysis revealed that product performance varied among subbasins due to differences in topography, rainfall characteristics, and human regulation. This study demonstrates the importance of multi-scale validation for quantifying uncertainties in satellite-based precipitation products and improving their application in hydrological modeling over mountainous regions. Full article

Accurate runoff forecasting serves as a fundamental basis for the scientific management of water resources and flood and drought risk mitigation. Owing to the nonlinearity, non-stationarity, and multi-scale temporal characteristics of runoff series, existing deep learning models still exhibit notable limitations in capturing long-term trends, responding to abrupt hydrological events, and ensuring model interpretability. The original Transformer relies on global self-attention, whose computational complexity increases quadratically with sequence length; it also has limited capacity to capture short-term local temporal dependencies such as rainfall–runoff relationships, and lacks prior constraints tailored to hydrological processes. To address these challenges, this study proposes a collaborative forecasting framework that integrates a gated convolutional Transformer (GCTrans) with an improved black kite algorithm (IBKA), enabling accurate, stable, and interpretable daily-scale runoff prediction. The GCTrans model consists of three customized modules: convolution-enhanced positional encoding (CEPE), which combines learnable positional encoding with local causal convolution to strengthen the temporal association of adjacent rainfall–runoff events, thereby providing a hydrologically meaningful positional reference for the attention mechanism; gated convolutional attention (GCA), which adopts a dual-path parallel architecture comprising global self-attention and local causal convolution to adaptively fuse long-term seasonal patterns with short-term storm-induced variations, thus capturing both baseflow evolution and flood peak responses; and a temporal gated output layer (TGOL), which performs adaptive feature weighting along the temporal dimension to selectively enhance the contribution of critical driving periods associated with extreme flood events, thereby improving the flood peak prediction accuracy. In addition, an improved black kite algorithm (IBKA) was developed by incorporating Tent chaotic initialization to enhance initial population diversity and introducing cosine adaptive inertia weights to dynamically balance global exploration and local exploitation, effectively alleviating premature convergence in high-dimensional hyperparameter spaces. Validation using data from the ME-Inland snowmelt-dominated watershed and the OR-Coastal storm-driven coastal watershed in the United States demonstrated that the GCTrans model consistently outperformed benchmark models including TCN, LSTM, Transformer, and Informer. After synergistic optimization with IBKA, both prediction accuracy and stability were further improved. SHAP-based interpretability analysis revealed that the model’s feature response patterns are statistically consistent with the rainfall–runoff generation mechanisms of the study basins: temperature-related drivers dominate in the inland watershed, while precipitation plays a dominant role in the coastal watershed, confirming the hydrological plausibility of the model’s decision-making logic. The integrated framework—encompassing model architecture, optimization algorithm, and interpretability—offers a valuable methodological reference for deep learning-based runoff forecasting in complex hydrological settings. Full article

Evapotranspiration (ET) is a major component of the water and energy cycle, influencing hydrologic processes, agricultural management, groundwater recharge, and land surface–atmosphere interactions. Water balance methods for estimating ET are widely used because of their direct connection to the conservation of mass and their applicability across scales. This review examines the theoretical basis and recent developments in water balance approaches for estimating ET across different hydroclimatic regimes. It summarizes classic soil water balance methods, physically based hydrologic models, remote-sensing approaches, and integrated machine learning techniques. Major themes include uncertainty in precipitation, runoff, and storage estimates; groundwater flow; water balance closure; spatial heterogeneity; and the integration of Moderate Resolution Imaging Spectroradiometer (MODIS), Landsat, and ground-based observations. More recently, hybrid physics-based and machine learning approaches have advanced ET estimation by combining process-based understanding with data-driven methods. Advances in computational hydrology, data assimilation, and Earth observation datasets are improving applications related to irrigation management, drought assessment, climate adaptation, and water-resource planning. Challenges remain in quantifying uncertainty, assessing model transferability, and representing groundwater and storage dynamics under changing hydroclimatic conditions. Overall, the review highlights the continued importance of water balance approaches for understanding ET dynamics and supporting sustainable water-resource management. Full article

Graph neural networks are increasingly used for river water-quality forecasting because they can represent interactions among monitoring locations that conventional site-wise time-series models cannot capture. However, the usefulness of graph learning depends strongly on whether graph structure reflects the hydrological processes governing pollutant transport. This review examines recent advances in river-network graph learning from the perspectives of graph construction, dynamic connectivity, propagation delay, physical constraints, and model validation. Existing approaches are organized into geometric, correlation-based, hydrological topology, transport-weighted, dynamic, and physics-informed graphs. To clarify their increasing physical content, we propose a graph physical-fidelity ladder from H0 to H6 and evaluate validation strength independently on a V0–V3 axis, thereby separating what a graph represents from how rigorously that representation is tested. Current studies commonly rely on fixed adjacency structures, while variations in discharge, flow direction, travel time, tributary contributions, reservoir regulation, and pollutant-specific transformation remain incompletely represented. Particular attention is given to dynamic edge updating, event- and pollutant-adaptive graphs, travel-time-aware message passing, mass-conserving architectures, and cross-basin representation learning. We further argue that predictive accuracy alone is insufficient for establishing hydrological credibility. Learned connectivity and edge importance should be tested against flow direction, transport time, mass balance, and structural counterfactuals, including edge reversal, deletion, weight perturbation, and dynamic-graph freezing. Future progress will depend on matching the complexity of graph representations to the physical claims they support and on evaluating predictive skill together with structural validity, physical consistency, uncertainty, and transferability. Such hydrologically faithful graph learning could provide more reliable and operationally defensible river water quality forecasts. Full article

The SWAT model is a fundamental tool for water resource management and is widely used in hydrological monitoring. In this context, a scientometric analysis was conducted to characterize the use of the SWAT model in Brazil, a country with significant water availability. A total of 119 scientific articles published between 2011 and 2024 in journals indexed in the Scopus and Web of Science databases were analyzed. These publications were distributed across 62 journals and involved 396 authors affiliated with 128 institutions from 19 countries. The study revealed a concentration of research in specific hydrographic regions and thematic areas, with 40% of the studies focusing on the impacts of land-use changes on hydrology and sediment yield. The main gaps identified relate to water quality modeling and the application of SWAT in urban basins. Furthermore, limitations regarding the transparency and availability of input data were observed, hindering study replicability and knowledge dissemination. The unequal geographic distribution of these applications represents a challenge for Brazilian science. Therefore, for SWAT research to mature and effectively integrate science with water sustainability, it is imperative to promote transparent data sharing and foster inter-regional collaborations. By overcoming these barriers, it will be possible to ensure study reproducibility and support the formulation of equitable public policies across the entire national territory. Full article

Accurate water level observations are essential for flood forecasting, hydrological analysis, and water resource management; however, sensor malfunctions and telemetry errors frequently introduce anomalous observations that compromise data quality. This study proposes a reconstruction-oriented Bidirectional Long Short-Term Memory (BiLSTM) Autoencoder for river water level anomaly correction and compares its performance with conventional first-, second-, and third-order polynomial and exponential regression models. The proposed framework incorporates a simplified encoder–decoder architecture, a dynamic block masking strategy to emulate contiguous sensor failures in highly autocorrelated water level series, and a threshold-based peak-oriented training scheme to improve reconstruction during high-flow events. Model hyperparameters were optimized using Gaussian process-based Bayesian optimization. The methodology was evaluated using hourly observed water level data from the Han River, Republic of Korea. Results showed that the proposed BiLSTM Autoencoder achieved reconstruction accuracy comparable to conventional regression models during calibration while exhibiting superior generalization to unseen validation datasets and better preserving the temporal continuity and dynamic characteristics of downstream hydrographs. Furthermore, a model calibrated using a relatively short but hydrologically representative period successfully reconstructed a substantially longer unseen record. Synthetic outlier injection experiments further demonstrated that reconstruction accuracy gradually deteriorated with increasing training data contamination, emphasizing the importance of high-quality training data for reliable sequence reconstruction. The proposed framework demonstrates potential as an effective sequence-reconstruction approach for offline river water level quality control. Full article

Flooding is one of the most dangerous natural hazards and is expected to occur more frequently due to the impacts of climate change. By altering the hydrologic characteristics of basins, urbanization exacerbates flooding risk. The Storm Water Management Model (SWMM) is a widely used urban hydrology model. Discretization of sub-catchments is a critical phase in constructing an SWMM model because it influences the effort of model construction and accuracy of the results. In the current review, we examine sub-catchment delineation in SWMM-based modeling by analyzing information from 43 journal papers encompassing 123 modeling scenarios. A novel categorization of sub-catchment delineation techniques with five primary and three secondary methods is proposed. Despite the hydrologic significance of external routing, where runoff is routed between sub-catchments, and internal routing, where runoff is routed from impervious to pervious portion of a sub-catchment or vice versa, such connections were considered in only 27% and 15% of the reviewed scenarios, respectively. Results also show that 69% of the scenarios used sub-catchments that were coarser than parcel (CP) resolution, but they were outperformed in calibration and validation by scenarios that used sub-catchments that were finer than or equal to parcel (FEP) resolution. Calibration was attempted in only 40% of the scenarios. Depression storage and surface roughness were commonly calibrated model parameters, but the tuned values suggest that in many cases calibration was achieved at the expense of physical plausibility. Future SWMM modeling applications can be improved by (i) delineating sub-catchments at FEP scale, (ii) utilizing SWMM routing options to better represent the physical flow pathways, and (iii) avoiding physically unrealistic calibration scenarios. Continued research on secondary delineation methods, which combine multiple primary methods, is recommended to offset the limitations of primary methods. Full article

Global warming has intensified the hydrological cycle and increased the frequency of extreme precipitation events. However, the localized impacts of rapid physical urbanization on extreme precipitation remain insufficiently quantified, particularly at the metropolitan scale. This study investigates the spatiotemporal dynamics and driving mechanisms of extreme precipitation in the Zhengzhou Metropolitan Area, a rapidly urbanizing region in central China. Daily precipitation records from 51 meteorological stations during 1985–2021 were used to calculate 11 extreme precipitation indices, including annual total wet-day precipitation (PRCPTOT) and maximum 5-day precipitation (Rx5day). Eight physical urbanization indicators and two topographic indicators were selected as explanatory variables. Spearman correlation analysis and Geographically and Temporally Weighted Regression (GTWR) were applied to examine the spatially and temporally heterogeneous relationships between physical urbanization, topography, and extreme precipitation. The results show that PRCPTOT increased by 1.12 mm per decade, while Rx5day increased by 0.43 mm per decade. Spatially, prolonged but relatively low-frequency precipitation events were mainly distributed in the western mountainous areas, whereas high-intensity extreme precipitation was concentrated in the eastern plains. Across the five temporally matched cross-sections, the associations between physical urbanization-related landscape indicators and precipitation intensity became more evident beginning in 2005, although their magnitude and statistical significance varied among years. The GTWR model explained 89% of the variation in PRCPTOT, indicating strong explanatory power for the combined effects of physical urbanization and topography. Physical urbanization-related landscape patterns and topographic factors showed distinct and temporally varying associations with extreme precipitation across the Zhengzhou Metropolitan Area. These findings deepen the understanding of physical urbanization–precipitation interactions and provide scientific support for flood-resilient urban planning and climate-adaptive spatial governance in rapidly urbanizing metropolitan regions. Full article

Urban heat islands (UHIs) are widely recognized as one of the clearest indicators of anthropogenic modification of the local climate system, resulting from the transformation of natural land surfaces into dense built environments characterized by altered thermal, radiative, and hydrological properties. The replacement of vegetated surfaces with impervious materials such as asphalt and concrete significantly modifies the surface energy balance, promoting heat storage during daytime and delayed release during nighttime, thereby enhancing urban–rural thermal contrasts. This study investigates the urban heat island (UHI) over the Berlin metropolitan region using a long-term high-resolution gridded dataset of daily minimum temperature from the EMO-1 (European Meteorological Observations gridded meteorological dataset with a spatial resolution of 1 arcmin × 1 arcmin) dataset, combined with Lamb weather types (LWTs) representing large-scale atmospheric circulation patterns. The analysis covers the period 1990–2023, enabling robust detection of long-term trends and variability. UHI intensity is quantified relative to a dynamic rural baseline defined by the 10th percentile temperature, allowing improved robustness compared to fixed rural references and reducing biases associated with spatial heterogeneity. Results demonstrate a persistent UHI centered over the urban core, with pronounced seasonal variability and statistically significant warming trends ranging from 0.004 to 0.018 °C yr⁻¹ across circulation regimes. Anticyclonic and Unclassified conditions dominate both the frequency and magnitude of UHI trends, confirming the dominant role of synoptic forcing controlling urban thermal dynamics. These findings underscore the necessity of integrating large-scale atmospheric circulation into UHI assessments and urban climate modeling. Full article

The Yangtze River Basin is a critical water supply region in China. To quantify the contributions of climate change and human activities to water resources across multiple temporal scales, this study focuses on actual evapotranspiration (AET), which directly influences water availability. Monthly runoff data from the Zhimenda Hydrological Station in the source region of the Yangtze River for the period 1982–2019 were analyzed using the Mann–Kendall (M-K) abrupt change test and the Bernaola–Galván (B-G) segmentation algorithm to identify the change point in runoff depth. The study period was subsequently divided into a baseline period and a post-change period. The ABCD hydrological model was employed to simulate monthly runoff variations during both periods, and the simulated results were used to calculate intra-annual (seasonal and monthly) AET. The Trend-Free Pre-Whitening Mann–Kendall (TFPW-MK) test was then applied to the AET series derived from the ABCD model to analyze temporal trends and intra-annual distribution characteristics. Finally, a multi-time scale Budyko framework was constructed to conduct attribution analysis based on AET data, quantifying the respective contributions of climate change and human activities to AET variation. The results indicate that: (1) The change point in the runoff series was detected in 2008. The Nash–Sutcliffe efficiency coefficients for both the baseline and post-change periods exceeded 0.88. (2) At the monthly scale, AET showed an increasing trend in January, February, July, September, November, and December. At the seasonal scale, AET exhibited a decreasing trend in spring and summer and an increasing trend in autumn and winter, though these trends were not statistically significant. Despite the varying directional trends observed across individual months and seasons, AET showed no statistically significant changes at either the seasonal or monthly intra-annual scale. (3) The intra-annual distribution of AET in the source region of the Yangtze River was highly consistent with precipitation patterns, showing significant synchronous variation. (4) Attribution analysis revealed that human activities played a dominant role in driving the observed trends in AET. Full article

Interferometric synthetic aperture radar (InSAR) maps ground motion in permafrost regions, but deformation does not uniquely identify the underlying process. This systematic review assessed how far the field has progressed from deformation detection toward validated process inference. A Web of Science Core Collection search on 16 July 2026 returned 409 records; 262 publications met the core scope, and 175 contributed 372 coherent scientific results and 727 companion metrics. Research expanded rapidly after 2020 and diversified from seasonal mapping toward active-layer thickness (ALT), ground ice, hydrology, infrastructure, and predictive modelling. Evidence maturity nevertheless declined along the inference chain. Descriptive deformation/quality and model/algorithm results comprised 58.3% of the evidence, whereas direct validation against field observations, global navigation satellite system (GNSS) measurements, or levelling comprised 4.6%. Only 12.4% of results reported a defensible analytic sample size and 33.9% provided usable uncertainty. ALT evaluations using probing, ground-penetrating radar (GPR), or model references estimated non-equivalent quantities and could not support one pooled accuracy measure. The radar line-of-sight (LOS) displacement is also a projected, non-unique response whose process interpretation depends on motion geometry and thermal, hydrological, and mechanical assumptions. No evidence family met the prespecified requirements for global meta-analysis. Progress toward cumulative inference requires explicit estimands, matched independent validation, uncertainty propagation, and transparent dataset dependence. Full article

Endogenous pollution released by riverbed sediments has become a key limiting factor for the sustainable improvement of river water quality. In this study, endogenous pollution specifically refers to the pollutants released into the overlying water by the resuspension process of riverbed sediments after hydrodynamic disturbances caused by rainfall runoff. To quantify this process, we modified the nutrient module of SWAT_2012 by introducing a rainfall-driven nonlinear resuspension term. Studies have shown that due to the gradual increase in disturbance, the release of pollutants in sediment exhibits a process from weak release to threshold activation and then to release saturation. In this study, two S-shaped response functions, Sigmoid and Gompertz, were selected to characterize this process. Model performance was evaluated using the Nash–Sutcliffe efficiency (NSE) and percent bias (PBIAS) against observed runoff and CBOD flux. Runoff simulation remained essentially unaffected, whereas CBOD simulation improved substantially: NSE increased from 0.607 in SWAT_2012 to 0.815 in SWAT_ZS and 0.744 in SWAT_ZG, with corresponding PBIAS values improving from −41.19% to 11.49% and 15.81%, respectively. Overall, SWAT_ZS performed the best in the Songhua River Basin. The results further indicate that rainfall-induced sediment resuspension may represent an important contribution of CBOD variability during wet periods, highlighting the substantial role of endogenous pollution under strong hydrological disturbance. By explicitly incorporating rainfall-driven nonlinear sediment resuspension into the SWAT nutrient module, this study provides a framework for representing episodic endogenous pollution and offers a transferable approach for quantifying sediment-derived pollution at the watershed scale. Full article

Reliable streamflow forecasting with quantified uncertainty is essential for water resource management, flood mitigation, and climate adaptation in semi-arid regions. This research introduces a framework that combines three conformal prediction techniques, including Split Conformal Prediction (SplitCP), Cross Validation Plus (CV+), and conformal quantile regression, with two probabilistic machine learning algorithms, namely Natural Gradient Boosting (NGBoost) and Probabilistic Gradient Boosting Machines (PGBM), to quantify uncertainty in hydrological modeling of the Sattarkhan Dam in East Azerbaijan Province, located in north-eastern Iran. The data collected ranged from 21 March 1996 to 22 September 2022 and were divided into two chronological groups: training (70%) and testing (30%) for modeling. Probabilistic prediction quality was evaluated using the continuous ranked probability score (CRPS) and negative log-likelihood (NLL). In contrast, for conformal prediction, we used the mean predicted interval width, effective coverage, and coverage width criteria. Results in terms of correlation coefficient (CC), mean absolute error (MAE), and root mean square error (RMSE) with the test dataset suggest improved performance by NGBoost (RMSE: 0.833 m³/s, CC: 0.918, MAE: 0.375) using optimal values of user-defined parameters in comparison to PGBM (RMSE: 0.909 m³/s, CC: 0.902, MAE: 0.388). NGBoost outperforms PGBM in probabilistic prediction. Its higher coverage indicates CV+ as the most effective uncertainty estimation method for this dataset. These findings support model reliability and inform future decision-making. These findings support operational forecasting and risk-informed decision-making in semi-arid regions. Also, the framework provides a transferable template for similar hydrological uncertainty studies. Full article