Journal of Hydrology最新文献

{"title":"How much historical data do we need? The role of data recency and training period length in LSTM-based rainfall-runoff modeling","authors":"Qiutong Yu,&nbsp;Bryan Tolson","doi":"10.1016/j.jhydrol.2026.135046","DOIUrl":"10.1016/j.jhydrol.2026.135046","url":null,"abstract":"<div><div>Rainfall-runoff models based on Long Short-Term Memory (LSTM) networks typically require extensive datasets for training. While increasing the number of watersheds generally improves model performance, the appropriate temporal scope of training data—especially under potential non-stationarity from climate change—remains unclear, and the importance of data recency remains underexplored. This study investigates whether decades of historical data are necessary when training with large numbers of watersheds, and whether shorter, more recent training periods can match the performance of extended records. Using hydrometeorological data from 1374 North American watersheds (spanning 1950–2023), we systematically evaluate the effect of data recency on LSTM performance through three complementary experimental designs: (1) backward-expanding training periods (fixed recent data + progressively older blocks), (2) forward-expanding periods (fixed old data + progressively newer blocks), and (3) sliding-window periods (fixed-length windows moving forward). Results reveal that newer data universally improves predictions, while older data contributes marginally or negatively to model performance. Notably, the benefit of increasing the number of watersheds is conditional on data recency, as Prediction in Ungauged Basins (PUB) performance improves most significantly with recent data. These findings provide the empirical evidence for the importance of data recency, demonstrating that data recency—not just data volume—is critical for LSTM-based streamflow prediction, underscoring the need for recency-aware training strategies.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"668 ","pages":"Article 135046"},"PeriodicalIF":6.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146072673","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impacts of spatial-temporal rainfall structures and antecedent wetness on flood variability at the catchment scale","authors":"Wencong Yang ,&nbsp;Changming Li ,&nbsp;Hanbo Yang","doi":"10.1016/j.jhydrol.2026.135019","DOIUrl":"10.1016/j.jhydrol.2026.135019","url":null,"abstract":"<div><div>Rainfall amounts largely determine flood magnitudes, but the influence of spatial–temporal rainfall structures and antecedent wetness remains unclear. This study presents a new approach to derive flood variability using different combinations of “event indicators”, which describe spatial–temporal rainfall dynamics and antecedent wetness at the catchment scale, while holding rainfall intensity fixed. Specifically, a multivariate distribution generates synthetic event indicators conditioned on rainfall intensity for historical high-rainfall events, and a data-driven runoff model predicts flood peaks based on these indicators. Applied to daily flood events in 140 catchments (304–86,475 km²) across the Eastern Monsoon Region of China, the approach uncovers the effects of event indicators on event flood variability and flood quantile uncertainty. Single-peak temporal rainfall and spatial uniform rainfall are prevalent across events. Antecedent wetness has the greatest impact on flood peak variability, with an attributable coefficient of variance (CV) of 0.23, followed by temporal (CV = 0.16) and spatial rainfall structures (CV = 0.05). Spatial-temporal rainfall correlation has minimal effects. Antecedent wetness and spatial rainfall structures exhibit greater impact in drier catchments, while temporal structures have less impact in larger, elongated catchments. For flood quantiles, variability in event indicators results in a mean CV of 0.15 across catchments for 50-year flood estimates, while uncertainty associated with unobserved hydrologic conditions accounts for 79% of total sampling uncertainty. These findings emphasize the need to incorporate spatial–temporal rainfall variability and antecedent wetness into flood risk estimation and process understanding.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"669 ","pages":"Article 135019"},"PeriodicalIF":6.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146071823","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"High-precision numerical simulation framework for the integrated modeling of urban “Source-Plant-Network-River” water environment","authors":"Guangxue Luan ,&nbsp;Jingming Hou ,&nbsp;Fuqiang Wang ,&nbsp;Tian Wang ,&nbsp;Donglai Li","doi":"10.1016/j.jhydrol.2026.135036","DOIUrl":"10.1016/j.jhydrol.2026.135036","url":null,"abstract":"<div><div>Urban water environments are facing increasingly severe pollution challenges, and rigorous numerical models have become indispensable for mitigating urban water pollution. Building on the “Gridding + GPU acceleration + Dynamic Link Library (DLL)” approach, this study develops an advanced coupled model that integrates (i) two-dimensional (2D) surface-water hydrodynamics and water-quality transport, (ii) 2D non-point-source pollutant (NPSP) build-up and wash-off, and (iii) one-dimensional (1D) pipe-network drainage and pollutant discharge, thereby enabling integrated simulation of the urban “Source-Plant-Network-River” (SPNR) system. The model employs high-resolution structured grids and a spatiotemporal flux scheme for multi-component pollutants in surface runoff, allowing accurate representation of NPSP wash-off and transport driven by coupled hydrological-hydrodynamic processes. DLL-based bidirectional coupling is implemented to dynamically link 2D surface processes and 1D pipe network hydraulics and water quality processes while reducing distortions in parameter transfer across modules. GPU acceleration, together with optimized water-quality flux computations and removal of redundant operations, significantly improves computational efficiency. The model is applied to the main urban area of Changzhi City under three spatially distributed rainfall scenarios. Performance is evaluated against observations of inundation depth, zoned drainage/sewage discharge, and combined sewer overflow (CSO) flow and water quality. The results show that the Nash-Sutcliffe efficiency (<em>NSE</em>) exceeds 0.7 for inundation depths at four flood-prone locations and for flow and pollutant concentrations at three representative drainage-outfall zones and four CSO outfalls. On an RTX 3070 workstation, the optimized model completes a 7.22 h simulation on 8,484,785 uniform structured grids coupled with 32,982 pipe-network nodes in 8.15 h, reducing runtime by 12.6% compared with the pre-optimization model. The proposed modeling framework is robust and efficient, offering strong potential for high-precision integrated simulations of urban water environments from source to receiving waters, as well as for evaluation, forecasting, early warning, and comprehensive water-environment management from a watershed perspective.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"668 ","pages":"Article 135036"},"PeriodicalIF":6.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146071824","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatiotemporal dynamics of meteorological and groundwater droughts in Southwest China and their cumulative and lagged impacts on vegetation","authors":"Yuanxiao Xing ,&nbsp;Shunsheng Wang ,&nbsp;Aili Wang ,&nbsp;Donglin Wang","doi":"10.1016/j.jhydrol.2026.135086","DOIUrl":"10.1016/j.jhydrol.2026.135086","url":null,"abstract":"<div><div>Drought is one of the most severe natural challenges in Southwest China, where complex karst hydrogeological conditions exacerbate its impacts on vegetation ecosystems. However, the mechanisms by which different drought types affect vegetation in this region remain unclear. This study integrated multiple datasets (GRACE, GLDAS, SPEI, and MODIS NDVI) to investigate the spatiotemporal evolution of meteorological drought, groundwater drought, and vegetation conditions across Southwest China from 2004 to 2023. Pearson correlation analysis was applied to quantify the cumulative and lag effects of the two drought types on vegetation in karst and non-karst areas. The results revealed that: (1) GRACE- and GLDAS-derived terrestrial water storage anomalies exhibited consistent patterns in Southwest China (r = 0.86, P &lt; 0.001); (2) both drought types showed a slight overall alleviation, yet drought intensification remained pronounced in the western high-altitude areas; (3) compared with cumulative effects, the lag effects of drought appeared to be relatively more pronounced; and (4) vegetation in karst regions responded more rapidly to groundwater drought, while shallow-rooted grasslands were more sensitive to the cumulative effects of meteorological drought, with a typical response time of approximately one month. These findings provide valuable insights for drought monitoring and ecological conservation in ecologically fragile regions and offer new scientific evidence for understanding drought–vegetation interactions in both karst and non-karst landscapes.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"669 ","pages":"Article 135086"},"PeriodicalIF":6.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146134977","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Developing a radar-rain gauge hourly blended precipitation dataset for Great Britain using the Gauss Blending Method","authors":"X. Qiu ,&nbsp;A.C. Green ,&nbsp;S. Blenkinsop ,&nbsp;H.J. Fowler","doi":"10.1016/j.jhydrol.2026.134954","DOIUrl":"10.1016/j.jhydrol.2026.134954","url":null,"abstract":"<div><div>Blended rainfall datasets take advantage of the strengths of multiple sources of rainfall data, producing higher accuracy and broader coverage. However, current blending methods either suffer from low accuracy or are overly complex. Some focus solely on improving statistical performance, but sacrifice the spatial structure of rainfall. Here we develop a Gauss Blending Method (GBM), based on an adaptation of the Gauss-Seidel method, to merge radar and gauge rainfall (970 gauges) in Great Britain to produce a high-resolution (1 km) hourly blended precipitation product. A comparison to 194 independent gauges (2006–2018) demonstrates that the blended product reduces the RMSE, MAE and MRE, compared with the only radar data, by 14.5%, 14.5% and 22.1% respectively, while improving its CC and NSE by 7.8% and 23.2%. GBM also enhances rainfall detectability without introducing new systematic bias.</div><div>GBM is compared with five standard approaches. The Multiplicative Adjustment Method and Mean Field Bias Adjustment Methods perform poorly and fail to improve radar rainfall. The Additive and Mixed Adjustment Methods show slightly worse performance and less stable rainfall detection skill, producing rainfall fields with artificial discontinuities and wedge-shaped artefacts. The only comparable method across most metrics is Kriging with External Drift. However, the GBM generates more complete data coverage, shows a better performance for extreme rainfall of ≥ 10 mm h⁻¹, better preserves rainfall structure and local variability, and is easy to apply. This is shown for several example extreme rainfall events. The resulting radar-gauge blended dataset will facilitate the analysis of spatial rainfall variability and extreme rainfall over GB, with significant advantages for extremes analysis, hydrological modeling, and flood risk assessment.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"668 ","pages":"Article 134954"},"PeriodicalIF":6.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146001381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact of changing recharge on a sole-source coastal aquifer: multi-model assessment for Virginia’s Eastern Shore","authors":"Farshad Hesamfar ,&nbsp;Sergio A. Barbosa ,&nbsp;Thanh-Nhan-Duc Tran ,&nbsp;Teresa Culver ,&nbsp;Lawrence Band ,&nbsp;Venkataraman Lakshmi","doi":"10.1016/j.jhydrol.2026.135003","DOIUrl":"10.1016/j.jhydrol.2026.135003","url":null,"abstract":"<div><div>Coastal aquifers face compound pressures from shifting meteorological patterns and sea-level rise (SLR), posing a direct threat to groundwater resources. We estimate future climate-driven recharge by forcing SWAT with downscaled, bias-corrected CMIP6 climate projections from four General Circulation Models (MIROC6, BCC-CSM2-MR, CanESM5, and MRI-ESM2-0) under SSP2-4.5 and SSP5-8.5. These recharge rates, in addition to local SLR, are input into SEAWAT, a variable-density groundwater flow and solute-transport model. Simulations then project changes in the coastal aquifer system over the Eastern Shore of Virginia (ESVA), quantifying groundwater availability and salinity risk from 2023 to 2080. Our findings are: (1) by 2080, the driest scenario (BCC-CSM2-MR; SSP2-4.5) reduces inland surficial groundwater heads by more than one meter and decreases freshwater volume by approximately 100 million m³ (−2.5%), an amount equivalent to 4.8 yrs of the current total water demand on ESVA. (2) The wettest scenario (CanESM5; SSP2-4.5) increases groundwater heads up to 1.9 m and expands freshwater volume by +2.7%, corresponding to 17 yrs of residential water supply. (3) While the equal-weight ensemble median head change remains small (|median Δh| ≲ 5 cm), the inter-model variability in freshwater storage, measured by the interquartile range (IQR), increases from ∼25 MCM in 2030 to ∼46 MCM in 2060, indicating growing uncertainty. (4) Groundwater head variation peaks along the central upland ridge (&gt;2 m) but drops below 0.25 m on barrier islands, where sea-level rise dominates. (5) High-emission scenarios show greater groundwater variability before 2060 but converge afterward, while intermediate scenarios show the opposite trend. These findings indicate the need for developing adaptive, climate-resilient groundwater management strategies to enhance water security and mitigate saltwater intrusion.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"668 ","pages":"Article 135003"},"PeriodicalIF":6.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146033579","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prediction of future landslide frequency and population impact in China based on machine learning model","authors":"Haoyu Jin ,&nbsp;Ke Zhang ,&nbsp;Chuntang Zhang ,&nbsp;Yi Xia ,&nbsp;Sheng Wang ,&nbsp;Moyang Liu","doi":"10.1016/j.jhydrol.2026.135007","DOIUrl":"10.1016/j.jhydrol.2026.135007","url":null,"abstract":"<div><div>Landslides are a common geological disaster, and due to their strong disaster-causing nature, they pose a serious threat to social and economic development. China is a mountainous country and landslide disasters occur frequently, so there is an urgent need to analyze their spatiotemporal variation patterns. In this study, six influencing factors were selected and their correlations with landslide frequency were analyzed. Multiple fitting models were then employed to simulate and predict landslide frequency. Ultimately, an analysis of the impact of landslide risks on the social-economic sphere was carried out. The results show that areas in China with higher landslide frequencies are mainly concentrated in Taiwan Province, the southeastern coast, and southwestern China. The correlation between landslides and slope length and slope steepness (LS) is the strongest, with an Spearman correlation coefficient (SCC) of 0.64, and the correlation with air temperature is the weakest, with an SCC of −0.04. The model with the best fitting effect for landslides frequency is random forest (RF), the second is bootstrap aggregating (Bagging), and the worst effect is support vector machine (SVM). Under the near-term (2027–2050) and long-term (2076–2099) scenarios of shared socioeconomic pathway (SSP) 2–4.5 and SSP5-8.5, the frequency of landslides in more than 80% area of China is increasing. The areas where the frequency of landslides has increased obviously are mainly in southwest China. Landslides pose a greater risk to the population and gross domestic product (GDP) of Taiwan Province, the southeast coastal areas, and the central Sichuan Region. The landslides risk to the urban and agricultural lands of Shaanxi, Shandong, and Hunan Provinces is greater. About 90% of the regions in China will have an increased risk of landslides to the population under the two scenarios in near-term and long-term. This study contributes to a deeper understanding of landslide disasters and presents an effective model for landslide prediction.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"668 ","pages":"Article 135007"},"PeriodicalIF":6.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146033581","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Land cover influences microclimate and non-rainfall water inputs in temperate agricultural environment","authors":"Jannis Groh ,&nbsp;Thomas Pütz ,&nbsp;Daniel Beysens ,&nbsp;Joan Cuxart ,&nbsp;Nurit Agam ,&nbsp;Werner Küpper ,&nbsp;Paul D. Colaizzi ,&nbsp;Harry Vereecken ,&nbsp;Horst H. Gerke ,&nbsp;Wulf Amelung","doi":"10.1016/j.jhydrol.2026.135040","DOIUrl":"10.1016/j.jhydrol.2026.135040","url":null,"abstract":"<div><div>Non-rainfall water inputs (NRWI) from dew, fog, frost, rime and soil water vapour adsorption (WVA) are key components of the terrestrial water cycle, but their individual contribution to the water budget is unclear due to a lack of suitable methods for identification and quantification. Here, we present a refined method to quantify and partition NRWI using weighing lysimeters. The method’s novelty lies in the variables used to determine when dew and frost (leaf-wetness), fog and rime (air visibility) occur. We applied this methodology to grassland and arable land lysimeters and compared it to the established method that relies on relative humidity and estimated dew-point temperature to quantify and partition NRWI.</div><div>Our results showed large differences between the refined and established method. The established method predicted larger amounts of dew (55<!--> <!-->%) and WVA (2522<!--> <!-->%), but no fog. NRWI mostly came from dew. Dew rates per event were generally higher on arable land, but the total amount of dew was larger on grassland due to higher frequency of dew formation. Dew amount differences could be attributed to land cover type, which promoted dew formation in the grassland while the arable ecosystem largely lost water through evapotranspiration (e.g. stomatal conductance, canopy-structure). We conclude that land cover type is a key control for microclimate (surface temperature, relative humidity), which significantly affects dew formation, while effects on fog or WVA are weaker. These effects are better determined by lysimeter assessment with a visibility and leaf-wetness device than by other existing NRWI identification systems.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"669 ","pages":"Article 135040"},"PeriodicalIF":6.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146095902","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AVFF-RI: an improved rainfall intensity measurement using common cameras","authors":"Manuel Fiallos-Salguero ,&nbsp;Soon-Thiam Khu ,&nbsp;Mingna Wang","doi":"10.1016/j.jhydrol.2026.135045","DOIUrl":"10.1016/j.jhydrol.2026.135045","url":null,"abstract":"<div><div>Rainfall measurements and modelling have faced a critical bottleneck due to data scarcity and limited performance of conventional methods, hindering precision in urban hydrology applications. In contrast, surveillance cameras have emerged as a cost-effective alternative for rainfall monitoring, although continuous detection remains computationally intensive due to the increasing variability and intensity of rainfall events. To tackle this, we propose AVFF-RI, a novel audio-visual fusion framework that leverages temporally resolved information from spatially distributed surveillance cameras for reliable rainfall estimation. The proposed framework is based on a two-stage approach for modality-specific pre-processing, followed by an adaptive fusion stage, which addresses the challenges of ambient noise, visual obstructions, and environmental conditions. The pre-processing stage includes a denoising network to isolate rainfall acoustic features from audio recordings, and a dual-stream detection scheme to extract rain streaks while removing visual distractors. Thus, these enhanced features are fused using a tailored multimodal regression model, improved with a bidirectional GRU-based adaptive fusion strategy that dynamically weights modality relevance and captures spatial and temporal dependencies. The framework evaluation was across varying rainfall events and conditions at two locations, demonstrating the AVFF-RI robustness, with R² ranging from 0.87 to 0.91 and mean absolute percentage errors ranging from 5.9% to 13.7%. Compared to unimodal and alternative fusion models, AVFF-RI exhibited lower error metrics and better generalization, underlining its effectiveness through late adaptive fusion and attention-guided weighting. This study highlights the viability of scalable rainfall estimation using widely deployed surveillance infrastructure, contributing to urban hydrology and disaster response systems.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"668 ","pages":"Article 135045"},"PeriodicalIF":6.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146095901","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Four decades of terrestrial dissolved organic matter in lakes across the Yangtze River Delta: Spatiotemporal dynamics and driving factors","authors":"Song Miao ,&nbsp;Heng Lyu ,&nbsp;Huaiqing Liu ,&nbsp;Yuan Li ,&nbsp;Xiaoguang Ruan ,&nbsp;Wenyu Liu","doi":"10.1016/j.jhydrol.2026.135037","DOIUrl":"10.1016/j.jhydrol.2026.135037","url":null,"abstract":"<div><div>Terrestrial dissolved organic matter (tDOM), a key regulator of lake ecosystems and the global carbon cycle, significantly influences underwater light regimes, thermal structure, and biogeochemical cycles. However, quantifying the fluorescence intensity and spatiotemporal patterns of tDOM is a persistent challenge for traditional monitoring techniques, which creates a critical gap in regional carbon budget estimations and the understanding of large-scale ecosystem effects. To overcome this limitation, a novel remote sensing algorithm was developed to retrieve tDOM, which is characterized by the fluorescence intensity of fulvic acid-like and humic acid-like components, based on the absorption characteristics of colored dissolved organic matter (CDOM). This approach was subsequently applied to assess the spatiotemporal dynamics and long-term trends of tDOM across Yangtze River Delta (YRD) lakes from 1986 to 2024. The robustness of the algorithm was confirmed using both independent in-situ and satellite-ground match-up datasets (median absolute percentage error &lt; 30%). The satellite-derived tDOM revealed significant spatiotemporal heterogeneity in the YRD. Trend analysis indicated that tDOM in the most lakes (63.46%) significantly increased, whereas it decreased in 17.31% and remained stable in 19.23% of the lakes. Furthermore, the random forest model results revealed that temperature and wind speed were the key drivers influencing the tDOM dynamics, acting synergistically with anthropogenic pressure which was associated with elevated tDOM levels. This study provides a feasible algorithm for large-scale tDOM monitoring and highlights the critical influences of climate change and anthropogenic activities on the estimation of lake carbon storage in rapidly developing regions.</div></div>","PeriodicalId":362,"journal":{"name":"Journal of Hydrology","volume":"668 ","pages":"Article 135037"},"PeriodicalIF":6.3,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146071694","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}