Droughts and floods are among the most damaging climate extremes, yet their abrupt alternations, rapid shifts from one to the other, remain poorly understood on a global scale. This study presents a comprehensive spatiotemporal analysis of abrupt drought-to-flood (DTF) and flood-to-drought (FTD) alternations worldwide from 1981 to 2023. Results show that these events are frequent, spatially widespread, and often occur concurrently with increasing levels of abrupt transitions in both directions. Regions with increasing FTD and DTF trends include the Yangtze Basin, Eastern India, Pakistan, and Southeast Africa, while California, Southern Africa, and Emilia-Romagna show declining trends, with California exhibiting a statistically significant decrease in FTD events (p < 0.1). In Southern Africa, the FTD coverage peaked at 5.5% in 1990, then declined to 4.6% in 2016. Emilia-Romagna (Italy) demonstrated greater temporal variability, with FTD areas rising from 7% in 1982 to 11% in 2005, before falling to 9% by 2022. DTF events were most frequent in Southern Africa (7.06 DTFs in 43 years) and Emilia-Romagna (7.13), while FTD events were highest in the Mississippi (8.73) and Yangtze (6.86) basins. Conversely, Eastern India and California exhibited consistently low frequencies of both DTF and FTD. These spatial disparities underscore the complex and region-specific dynamics of hydroclimatic extremes. Improved understanding of drought-flood alternations is essential for strengthening early warning systems, risk management, and adaptation strategies in vulnerable regions worldwide.

Abstract Understanding historical and future surface soil moisture (SSM) drying is pivotal due to its close links with droughts, heatwaves, and wildfires, yet debates regarding its evolution persist. In this study, we leverage advanced deep learning techniques to fill gaps of remote sensing‐based SSM data during 1983–2020 and therefore use these gap‐filled observations to constrain SSM estimates from 23 Earth System Models (ESMs) during 1901–2100. Our enhanced observations reveal that approximately half of Earth's landmass experienced SSM drying over the past four decades. However, in contrast to projections from current‐generation ESMs, observation‐constrained simulations indicate a less pronounced drying trend in dry‐wet transitions and monsoon margins during 2021–2100 compared to 1901–1980. Current ESMs may overestimate SSM drying in these regions, likely due to their limited representation of soil moisture‐atmosphere feedback. These findings highlight the need to integrate remote sensing and artificial intelligence into ESMs to improve projections of future droughts and their socio‐economic consequences.

El Niño Southern Oscillation (ENSO), encompassing El Niño and La Niña, significantly influences river flooding patterns, which is prompting to investigate its relationship with floods in Nepal's transboundary Karnali River Basin (KRB). Focusing on extreme flood events during El Niño and La Niña years, this research aims to enhance the understanding of ENSO's impact on hydrological and hydrodynamic processes. Precipitation and discharge data spanning from 1964 to 2020 sourced from the Department of Hydrology and Meteorology (DHM), Government of Nepal, were analyzed. Hydrodynamic modeling, employing HEC-HMS and HEC-RAS, identified four significant El Niño events in 1983, 2000, 2014 (La Niña years), and 2015 (a strong El Niño year) at the basin. The study identified ENSO events and examined basin characteristics, precipitation depth, river discharge, and gauge height, utilizing daily data for model estimation of flood discharge and depth. Analysis of ENSO-related variability, including Sea Surface Temperature (SST), Southern Oscillation Index (SOI), and Multivariate ENSO Index (MEI), alongside pressure, temperature, and discharge data across the KRB, was conducted using a three-year running mean. The Soil Conservation Service (SCS) method was integrated within the HEC-HMS and HEC-RAS models to evaluate rainfall duration and flood response considering terrain, soil, and land use. Model simulations revealed river channel shifts, particularly along the right bank, during the 2015 ENSO event. Intensity–duration–frequency (IDF) and correlation/regression analyses further elucidated the impact of ENSO, with the lowest recorded precipitation and discharge observed during the 2015 El Niño event despite localized heavy rainfall. Comparative analysis of flood discharge, gauge height, inundation extents, depths, and velocities across ENSO years highlighted a significant relationship between observed and modeled discharge during the monsoon season. The positive correlation between basin mean precipitation and discharge during ENSO event indicating precipitation as a key driver of the basin's hydrology. These findings offer valuable insights for water resource management and development, aiding in the anticipation of future strong ENSO and El Niño events in the region.

Peatland function and ecosystem services are increasingly at risk from climate and land-use change. At high altitudes climate warming is enhanced, while large peatland areas have undergone drainage, yet there is little knowledge of their interaction. Such information is essential for informing future management and restoration decisions. To address this gap, we investigated the effects of warming on ecohydrological function of intact (wet) and drained (dry) high-altitude peatlands. Our experiment compared the response of water table, pore-water chemistry, litter decomposition and vegetation composition to drainage and warming in a factorial experiment, utilising open top chambers to simulate warming. Our results showed that shallow peat (8 cm depth) warmed by 0.75 degrees C and 0.17 degrees C in the dry and wet site respectively, over one year of warming. However, we found limited effects of warming on peatland function, attributed to the short-term nature of the experiment, where the ecosystem showed a certain resilience to one-year of increased temperatures. Drainage significantly affected ecosystem function. A mean difference of 10.2 cm in water table level between the dry and wet sites, increased shallow pore-water dissolved organic carbon in the dry site with a greater contribution from recent shallow peat decomposition. Further, drainage also enhanced litter decomposition rates and altered vegetation composition, increasing graminoid abundance. We found small differences in water table have large impacts on function, therefore rewetting drained high-altitude peatlands by restoration, may help improve ecosystem services, while enhancing resilience to warming.

Transboundary water resources pose significant governance challenges as climate change, population growth, and geopolitical tensions increase competition over shared waters. International river basins spanning 150 countries and supporting approximately 40% of the global population require frameworks that move beyond event-based conflict tracking to understand the underlying structural dynamics that generate hydropolitical tensions. This study develops a novel framework for systematically mapping Hydropolitical System Archetypes (HSAs) across 286 international river basins, addressing a critical gap in global water governance research. By integrating system archetypes theory with hydropolitical analysis, this research identifies recurring patterns of conflict and cooperation, categorizing them into five HSAs based on power asymmetries, institutional fragility, resource dependency, and geopolitical factors. Our findings reveal the uneven distribution of HSAs worldwide, demonstrating how physical geography, historical treaties, and socio-political conditions shape water-sharing interactions. Critically, this framework shifts analytical perspective from geopolitical temperature, the observable heat of diplomatic events captured by event-based indices, to structural process, the underlying feedback mechanisms that generate hydropolitical behavior. A comparative analysis with the Basin at Risk (BAR) event-based index reveals that several basins classified as cooperative by BAR metrics simultaneously exhibit all five HSAs, indicating high structural risk. This approach enables early detection of conflict preconditions even during periods of diplomatic calm, supports adaptive governance strategies, and promotes more equitable transboundary water-sharing mechanisms in an era of increasing environmental uncertainty.

Surface albedo feedback (SAF) amplifies warming in northern high latitudes, affecting the Arctic climate system, ecosystems, infrastructure, and global trade routes. However, Earth system models (ESMs) exhibit large uncertainties in SAF projections, complicating future Arctic warming estimates. Here, we develop a machine-learning method based on emergent constraints (ECs) and use in-situ observations to constrain SAF projections over Arctic land regions. Our approach leverages a physical relationship between historical albedo-temperature dynamics (1985-2014) and future SAF (2070-2099) across ESM ensembles. The constrained SAF is reduced by 0.29-0.52 W m-2 K-1 across emission scenarios, with uncertainties decreased by 45-55% compared to unconstrained projections. These findings enhance confidence in regional climate projections, offering more precise insights for climate adaptation and policy in vulnerable high-latitude communities.

Abstract. The development of the Variable Infiltration Capacity (VIC) model to version 5 has endowed it with enhanced land-surface-modelling features, establishing it as a modern and cutting-edge tool for hydrological research. However, its adoption is limited by challenges in complex data processing and parameter preparation. To address this issue, this study proposes the Easy VIC Build (EVB) framework, an open-source Python package with an object-oriented, modular design, developed for efficient and flexible VIC deployment. Two real-world basin setups were selected for model application to validate the effectiveness of the EVB framework. The results indicate that, based on the EVB framework, reliable spatial parameters with robust transferability can be derived, while VIC models can be conveniently deployed and achieve satisfactory simulation performance. The EVB framework provides a well-suited workbench for advanced applications of the VIC model, promising to further leverage the value of VIC-5 in hydrology and related fields.

Accurate prediction of flood events is important for flood control and risk management. Machine learning techniques contributed greatly to advances in flood predictions, and existing studies mainly focused on predicting flood resource variables using single or hybrid machine learning techniques. However, class-based flood predictions have rarely been investigated, which can aid in quickly diagnosing comprehensive flood characteristics and proposing targeted management strategies. This study proposed a prediction approach of flood regime metrics and event classes coupling machine learning algorithms with clustering-deduced membership degrees. Five algorithms were adopted for this exploration. Results showed that the class membership degrees accurately determined event classes with class hit rates up to 100%, compared with the four classes clustered from nine regime metrics. The nonlinear algorithms (Multiple Linear Regression, Random Forest, and least squares-Support Vector Machine) outperformed the linear techniques (Multiple Linear Regression and Stepwise Regression) in predicting flood regime metrics. The proposed approach well predicted flood event classes with average class hit rates of 66.0%-85.4% and 47.2%-76.0% in calibration and validation periods, respectively, particularly for the slow and late flood events. The predictive capability of the proposed prediction approach for flood regime metrics and classes was considerably stronger than that of hydrological modeling approach.

Most existing attribution assessments of runoff changes mainly focused on direct effects of climate change and human activities on runoff changes, and their indirect effect assessments were not widely reported, i.e., impacts on underlying surface conditions and their consequences to runoff changes. In this study, we adopted an improved Budyko framework, employing the elasticity coefficient method and principal component regression, to comprehensively assess both the direct and indirect impacts of climate changes and human activities on runoff changes. Thirty-one source catchments were selected in the upper and middle reaches of the Yellow River Basin, comprising the Yellow River Source Region (YRSB), Wei River Basin (WRB) and Yiluo River Basin (YRB). Results showed that during 1995-2022, the mean annual runoff changes in the post-change period ranged from -24.0 to 129.9 mm in 31 source catchments compared to the pre-change period. The underlying surface parameter n in the Budyko model ranged from 0.4 to 3.8, with lower values mainly in the YRSB. Sensitivity analysis revealed that parameter n was most sensitive to potential evapotranspiration, particularly in the YRB, and runoff changes exhibited the greatest sensitivity to precipitation, with the highest sensitivity observed in the WRB. Changes in parameter n were primarily attributed to human activities (77.4%) rather than climatic factors (22.6%), with irrigation water (22.6%) and temperature (8.4%) exhibiting the largest impacts, respectively. Contributions to runoff changes were 54.7% from the direct impact of climate change, highest in the YRSB at 59.8%, while 12.2% from its indirect impact, and 33.1% from human activities, with the greatest influence both in the YRB at 16.2% and 42.4%, respectively.

High-accuracy long-term precipitation data are essential to hydrometeorological and cryohydrological studies, especially in alpine regions where observations are sparse. However, existing precipitation datasets exhibit substantial discrepancies in magnitude and spatiotemporal patterns, posing a major challenge to water cycle research in these areas. This study first evaluated five datasets, including the China Meteorological Forcing Dataset (CMFD), its updated version (CMFDv2.0), the gridded daily observation dataset over China (CN05.1), the enhanced global dataset for the land component of the fifth generation European Centre for Medium-Range Weather Forecasts (ECMWF) atmospheric reanalysis (ERA5-Land), and the high-resolution near-surface meteorological forcing dataset for the Third Pole (TPMFD) region, against gauge observations in the source region of the Yellow River (SRYR), a typical alpine region. A new gridded precipitation dataset (named CCLI) with a spatial resolution of 0.1 degrees 3 0.1 degrees and a time series spanning 1951-2020 were subsequently constructed based on the evaluations of five datasets and using the delta correction method. The CCLI and five other datasets were then evaluated locally by using gauge observations and regionally by considering the spatiotemporal consistency and uncertainty. The CCLI outperformed the other five datasets across monthly, seasonal, and annual scales at the local scale. It also showed strong spatiotemporal consistency with the other five datasets and the lowest uncertainty across multiple temporal scales, with grid uncertainty ranging from 0.66 to 7.89 mm month-1. The CMFDv2.0 followed closely behind. In contrast, the ERA5-Land, CN05.1, and TPMFD datasets displayed considerable biases and high uncertainties. Further analysis based on the CCLI revealed a significant and consistent increase in precipitation of 2.03 mm yr-1 in the SRYR since the 1950s. This study provides a high-quality precipitation dataset and lays the groundwork for further exploration of water cycle studies.

Lake hot events (months with abnormally high lake surface water temperature) and dry events (months with abnormally low lake surface extent) can individually cause significant stress to lake ecosystems. When these events occur simultaneously, as compound hot-dry events, their impacts can be even more severe, particularly in dryland regions. Yet, their frequency, long-term dynamics, and driving factors remain poorly understood. Here, we leverage model-derived lake surface water temperature and satellite-derived surface area datasets to identify and analyze hot events, dry events, and compound hot-dry events in 2,338 lakes across global drylands from 1985 to 2020. We find that the occurrence of compound hot-dry event has significantly increased in 520 lakes, while 240 lakes have shown a significant decrease. Hot events and dry events individually increased in 1007 and 767 lakes, respectively. Furthermore, in most lakes, the temporal changes in compound hot-dry events are primarily driven by changes in dry event frequency, rather than hot events or their co-occurrence. These findings underscore the growing compound stress on dryland lake systems and emphasize the need for targeted adaptation and resilience strategies.

Groundwater stress has become a significant constraint on sustainable development. In China, various adaptation measures have been implemented, such as water withdrawal restriction, inter-basin water transfer, and unconventional water resources utilization, yet their effectiveness in alleviating future groundwater stress remains unclear. Here, we combine survey data and large-scale hydrological projections and reveal that increased non-agricultural water withdrawal will aggravate future groundwater stress in China when adaptation measures are neglected, with the population exposed to severe groundwater stress rising from 23.1% (22.3%–23.9%) currently to 30.9% (29.9%–33.0%) by 2050. While existing measures may mitigate some future stress, 19.3% (17.1%–21.4%) of the nationwide population will still face severe risk. We demonstrate that the synergistic effects of adaptation solutions exceed their individual effects, and only more stringent water withdrawal management combined with enhanced water supply can substantially alleviate future groundwater stress. Our findings provide essential insights and solutions for regional groundwater security and achieving Sustainable Development Goals.

Study region: The Vakhsh River Basin, a major tributary of Amu Darya River Basin (ADRB), an essential source of irrigation and hydropower, has experienced substantial alterations in streamflow under climate change. Understanding how glacier– and snow–fed streamflow will evolve under continued warming is critical for ensuring future water security. Study focus: This study assesses historical and future changes in high-flow extremes using the Spatial Processes in Hydrology model and downscaled bias-corrected CMIP6 data. New hydrological insights: High-flow extremes exhibited declining magnitude and progressively earlier timing during the baseline period (2000–2023). Future projections show consistent declines in annual mean high-flow extreme, with relative decreases of 11–13% in far future (2076–2100) across the scenarios. High-flow extreme events are likely to occur earlier, advancing by 2–3.5 weeks for central date of event occurrences in the far future. Early-season high flow extremes remain dominated by snowmelt (77–96%), with glacier melt as a secondary contributor. The projected advance in high-flow extreme timing is primarily driven by earlier snowmelt peak, shifting from June in baseline (with ∼50% snowmelt contribution) to May in future periods (70–80% snowmelt contribution), along with the compression of glacier-melt-driven extremes into late summer. Overall, the results indicate an earlier, shorter, and weaker melt–driven flood season. These changes have important implications for hydropower operations, irrigation planning, and downstream climate adaptation in the Amu Darya River system.

ABSTRACT The transboundary Sun-Koshi River basin, characterized by intricate topography and geo-climatic diversity, has encountered distinct periods of droughts significantly impacting downstream river discharge. This study focuses on assessing temporal and spatial patterns of meteorological drought using SPI and SPEI indices, gauging their influence on river discharge through hydro-meteorological station data and the HEC–HMS model. Severe drought episodes were notably observed in 2010 and 2015. In 2010, prominent drought occurrences extended beyond Nepal's border into China. Conversely, in 2015, Jiri, Okhaldhunga, and Salleri experienced heightened drought conditions. Spatially, over 99% of the basin area experienced drought, varying from moderate to extreme magnitudes during 2010 and 2015. The estimated annual rainfall and basin outlet discharge were 1,800 mm, 1,907 mm, and 1,899 m3/s, 1,086 m3/s in 2010, and 2015, respectively. During drought periods, the stations indicated significantly reduced discharge, indicating a marked departure from normal conditions across the basin. Ultimately, station data performance and the HEC–HMS SCS curve number model showed that discharge in the Sunkoshi River basin is profoundly impacted by drought, notably influencing rainfall intensity on monthly, seasonal, and annual scales. The smaller basins discharge more accurate results compared to the larger outlet basins.

Extreme floods pose escalating risks in a changing climate, yet forecasting remains challenging due to peak flow underestimation and high uncertainty. We introduce diffusion‐based runoff model (DRUM), a probabilistic deep learning (DL) approach that advances extreme flood forecasting across representative basins in the contiguous United States. DRUM outperforms state‐of‐the‐art benchmarks, enhancing nowcasting skill for the top 1‰ of flows in 72.3% of studied basins. Under operational scenarios, DRUM extends reliable lead times by nearly a full day for 20‐ and 50‐year floods. When evaluated with measured precipitation, an ideal condition, recall improves by 0.3–0.4 and the early warning window extends by 2.3 days for 50‐year floods. The enhancement potential varies regionally, with precipitation‐driven flood zones in the eastern and northwestern US benefiting most, gaining 3–7 days in lead time. These findings highlight the transformative potential of diffusion models as a cutting‐edge generative AI technique for advancing hydrology and broader Earth system sciences.

High-altitude peatlands (HAPs; defined as > 1,500 m) provide important ecosystem services including soil carbon (C) storage. However, temperatures in high-altitude regions have been rising rapidly in recent decades, while HAPs are increasingly affected by human activities such as intensive drainage and grazing. Collectively, climate change and land management may strongly affect the HAP C cycle. Here, we synthesise current global progress on the HAP C cycle, focussing on the impacts of climate change and land management. Warming increased both ecosystem respiration (ER) and methane (CH4) emissions (26 %–86 %), while impacts on net ecosystem exchange (NEE) of CO2 were still unclear. However, short-term drought decreased ER and CH4 emissions (7 %–96 %), along with NEE (12 %–52 %). Snow, permafrost, and glacier decline may also impact the C cycle in HAPs, although a limited number of studies have been conducted. Grazing and vegetation degradation impacts on HAP C cycling were related to grazing and degradation intensity, while generally decreasing soil organic C stocks (3 %–51 %). Moving from shallower to deeper WTLs stimulated ER (9 %–812 %), while reducing CH4 emissions (13 %–100 %), with variable effects on NEE (-53 %–700 %). Restoration by rewetting began to reverse the trend of drainage. We highlight several knowledge gaps, including limited understanding of climate change and land-management effects on gross primary productivity and dissolved organic carbon, while there is still limited knowledge of regional differences in HAP C cycling. Future research should focus on the interaction of land-use and climate change in HAPs, including HAP restoration, which may help future conservation of these valuable ecosystems.

Land use and cover change (LUCC) could directly lead to a local drying of surface soils, threatening food security, water availability, and ecosystems. However, LUCC can also trigger indirect effects via land-atmosphere interactions that can impact soil moisture locally and remotely. To date, the extent to which LUCC’s indirect effects modulate their direct effects on soil moisture remain unclear at the global scale, which limits our ability to make informed land-use decisions toward sustainable development. Here, we combine data-driven and coupled Earth system models to disentangle LUCC’s direct and indirect effects on global soil moisture over the past century. We found that LUCC’s indirect feedback has mitigated their direct drying effect for global land on average by regulating atmospheric circulation, moisture budget, and precipitation patterns. By identifying regions where land-atmosphere feedbacks have offset or exacerbated the drying, this study can guide targeted land-use policies to achieve Sustainable Development Goals (SDGs).

Freshwater is increasingly abstracted beyond sustainable levels in many watersheds worldwide. Is the public aware of this water scarcity challenge? This study investigated public perceptions of water scarcity and how individual characteristics influence these perceptions using an ordinal logistic regression model. Based on 3262 online survey responses, we found that participants tend to underestimate water scarcity, contrasting the fact that over half of the cities in China face water scarcity. High-income tap water users are more likely to underestimate water scarcity than low-income non-tap water users. Understanding these perceptions is critical for promoting water-saving practices and developing effective mitigation strategies.

Many terminal lakes in Central Asia have witnessed concerning rates of shrinkage in recent decades. These lakes are particularly sensitive to both climate change and human water withdrawals. Although human water withdrawals are acknowledged as a major factor influencing long-term lake changes, previous studies often fail to distinguish the specific contributions of different sectors such as irrigation, livestock, industry, and domestic water usage. This knowledge gap is largely due to the absence of observed multi-sectoral water withdrawals. Recognizing the value of machine learning methods in predicting water withdrawals through complex, non-linear relationships between water uses and potential explanatory factors, we developed an innovative approach that integrates a hydrological model and a machine learning-based water use model. This methodology was applied to simulate the long-term changes in the area of Ebinur Lake, a large terminal lake in Central Asia. Water withdrawals estimated by Random Forest based on meteorological (temperature and precipitation) and socio-economic data (e.g., population, multi-sectoral GDP, per capita income, etc.) and allocated by irrigated area and population were extracted from the river route in the hydrological model which in turn affects the inflow into the lake. Finally, integrated model simulations were validated using remotely sensed lake areas and streamflow data from mountain hydrologic stations. Several experiments, including and excluding different sectoral water uses, were conducted to isolate factors influencing lake dynamics. The results indicated irrigation water withdrawal not only caused lake shrinkage, but also increased seasonal variability, thereby increasing the uncertainty of water supply to lake ecosystems. The proposed modelling approach provided a framework for quantifying the responses of terminal lake area changes to different sectoral water withdrawals in arid basins, especially in the absence of specific water withdrawal data.

At present, station observation data are widely utilized for rainfall erosivity estimation while the spatiotemporal coverage of the data is limited. High-resolution satellite precipitation data offer the possibility of estimating rainfall erosivity with global coverage in a real time manner. A few studies have attempted to use satellite precipitation to derive rainfall erosivity and found that satellite data systematically underestimate rainfall erosivity compared to station observations. Thus, a Global Erosive Rainfall Database (GERD) was constructed using half-hourly satellite rainfall data from 2001 to 2020. Rainfall erosivity was calculated using the rainfall erosivity estimation method in RUSLE2 and bias corrected by a station-based annual average rainfall erosivity map. Then, the spatiotemporal variations of global rainfall erosivity and erosive rainfall event are revealed. The results showed that: (I) The 20-year average rainfall erosivity was 2,538.6 MJ mm ha-1 h- 1 yr- 1. The 20-year average number of erosive rainfall events was 67 events per annum. (II) There has been a discernible downward trend in the global rainfall erosivity anomaly from 2001 to 2020, with an average change rate of -22.09 MJ mm ha-1 h- 1 yr- 2, particularly pronounced in the Southern Hemisphere, where the decline rate reached -68.21 MJ mm ha-1 h- 1 yr- 2. In contrast, the number of erosive rainfall events has exhibited an upward trend during the same period. (III) Seasonal rainfall erosivity and number of erosive rainfall events during the period of June to August were obviously different from other seasons. On 34.1 % and 24.6 % of the global area, rainfall erosivity and number of erosive rainfall events in this period constituted more than half of the whole year.