Phreatic evaporation is a critical link in the vertical interactions between surface water and groundwater;accurate quantification is essential for assessing shallow groundwater resources in the Huaibei Plain.This study analyzes the influence of soil temperature at different water table depths(H)on phreatic evaporation using meteorological observations and data collected by large-scale lysimeter arrays at the Wudaogou hydrological and water resources experimental station between 2010 and 2019.A temperature coefficient is proposed to account for the relative difference between soil temperature at the surface and at water table depth.It is incorporated into three empirical formulas for phreatic evaporation to account for vertical soil temperature dynamics.Phreatic evaporation estimates derived from the original(E0)and improved(Ei)formulas are compared with observations(Eobs).Results indicate that:① Agreement between E1 and Eobs is higher than that between E0 and Eobs.Relative to E0,Ei improves overall accuracy by 5.89％—7.00％and reduces the relative error by 9.55％—10.13％.② Among the improved formulas,the power function outperforms the modified Averianov and Ye Shuiting formulas.Accuracy improvement is the largest for a water table depth of 0.6 m.③ Total-order and first-order sensitivity indices reveal that,in the improved formulas,the sensitivity of H is the highest,while the sensitivity of E0 is the lowest,further confirming the necessity of incorporating soil temperature effects into phreatic evaporation calculations.

Identifying the impacts of climate change and human activities on streamflow changes is essential for understanding hydrological processes and sustainable water resource management. This study investigated the spatiotemporal changes of annual streamflow in global 2264 catchments during 1961–2014, and quantified the contributions of precipitation (P), potential evapotranspiration (Ep) and landscape parameter (n) to streamflow changes using Choudhury-Yang equation based on Budyko hypothesis. The results indicated that significant increasing trends (p < 0.05) in annual streamflow (16.65

Abstract The temperature sensitivity ( Q 10 ) of soil respiration ( Rs ) is one of the major determinants of carbon emissions from Rs . Identifying the controlling factors of Q 10 can contribute to quantifying the response of Rs to climate change and assessing potential carbon emission risks. The Tibetan Plateau (TP), known as the Earth's third pole, plays a pivotal role in the global climate and carbon cycle system. However, the controlling factors of Q 10 and its responses to future climate change in this region are not well understood. Here, we introduce a method for calculating Q 10 from cumulative carbon emission data on Rs , which allows substantially expanding the available data set. Building on this, we integrate various statistical analyses with explainable machine learning techniques to examine the controlling factors regulating Q 10 on the TP. Our analyses consistently show that precipitation and soil potential of Hydrogen are the two most important factors regulating Q 10 . Precipitation is the dominant factor regulating Q 10 across nearly 70% of the plateau, while pH's dominance is about 30%, highlighting the primary role of precipitation as a regulator of Q 10 on the TP. Future precipitation increases (wetting) on the TP are predicted to markedly boost the warming‐induced increase in Rs , with this effect being particularly pronounced in arid areas. Our findings clearly demonstrate that under future warming and wetting on the TP, the compound effects of precipitation and temperature would enhance Rs , consequently accelerating the soil organic carbon release.

Understanding the impact of groundwater depth on actual crop evapotranspiration (ETc act) is essential for agricultural water management in shallow water table regions. A four-year field experiment (2019–2023) was conducted at the Wudaogou Hydrological Experimental Station to monitor the ETc act of winter wheat under different groundwater depth conditions, while simultaneously recording key meteorological variables, including air temperature, precipitation, wind speed, net radiation, relative humidity, sunshine duration, soil heat flux, and soil temperature. Based on the observed data, we analyzed the influence of groundwater depth on winter wheat ETc act and actual crop coefficients (Kc act). The results revealed that both ETc act and Kc act exhibited a clear exponential relationship with groundwater depth, showing a significant decreasing trend as the depth increased. Building upon this finding, we developed a Groundwater–Meteorology-Based Actual Crop Evapotranspiration Model (GW–M model) that incorporates both groundwater depth and meteorological factors. In the model, these variables affect ETc act by influencing the Kc act. The selection of meteorological factors was guided by the top three variables identified through geographical detector analysis. The model effectively reproduced the ETc act process of winter wheat in shallow groundwater areas. Comparative analysis with other evapotranspiration models demonstrated that the proposed model achieved higher accuracy. This study underscores the pivotal role of groundwater depth in regulating ETc act in shallow water table areas. The proposed modeling approach offers a flexible and scalable framework for simulating ETc act under variable groundwater conditions, thereby providing theoretical support for regional-scale agricultural water management.

Drought–flood abrupt alternation (DFAA) events are sequential compound hydroclimatic extremes characterised by rapid transitions between dry and wet conditions and influenced by both climate change and human activities. They pose substantial challenges to water security in large river basins and have therefore attracted increasing scientific attention. However, the mechanism linking meteorological drivers to hydrological responses remains insufficiently understood. This study applied both long- and short-cycle Drought–Flood Abrupt Alternation Indices (DFAIs) to analyse the spatiotemporal evolution and meteorological–hydrological coupling of DFAA events in the Yangtze River Basin during 1980–2024. The results showed that meteorological DFAA events occurred at a high intensity in the middle Yangtze River Basin and high frequency in the lower reaches, averaging 2.98 and 37.04%, respectively. In contrast, hydrological responses showed an almost opposite spatial pattern, with relatively frequent flood-to-drought (FTD) events in the upper reaches and high-intensity but low-frequency FTD events downstream. The long-cycle DFAI generally declined, increasing only at Beibei and Hukou. Instead of a systematic rise in frequency under climate warming, DFAA changes were marked by intensified extremes in individual years, with mean and 90th-percentile intensities increasing by 9.28% and 25.17%, respectively. Danjiangkou station showed the strongest meteorological–hydrological DFAA coupling, with significant positive correlations in both intensity and frequency (r = 0.53–0.63). These findings reveal that DFAA events represent coupled climate–hydrology transition processes in which meteorological anomalies are transformed into hydrological responses through spatially heterogeneous basin processes. As such, the results advance our understanding of how flood–drought transitions evolve across large river basins.

Compound climate extremes, such as concurrent precipitation and temperature extremes, cause significant impacts on socioeconomics and ecosystems. Recent studies have made substantial progress in the specific type of compound extremes; however, characteristics of different types of compound precipitation and temperature extremes across the globe and their driving mechanisms remain limited understood. This study investigated characteristics of compound extremes including dry-warm (DW), wet-warm (WW), dry-cold (DC), and wet-cold (WC) combinations in annual, JJA (June, July, and August), and DJF (December, January, and February) occurrences during 1901-2024 across global land areas and their relationships with climate variability modes. Results indicated that the spatial distribution of the frequency of DW (WW) showed a similar pattern to that of WC (DC). The overall frequency of compound warm-related (cold-related) extremes successively increased (increased and then decreased) from the period 1901-1941 to the period 1983-2024. The spatial extent of compound warm-related (cold-related) extremes presented an increasing (decreasing) trend in all the continents over the past 124 years. The areas with positive (negative) precipitation-temperature correlation showed high frequency of WW and DC (DW and WC). The areas with positive (negative) dependence between precipitation and temperature extremes of DW and WC showed negative (positive) dependence between the two extremes of WW and DC, except for central Asia where the two extremes showed positive dependence in the four compound extremes. During 1901-2024, El Ni & ntilde;o (La Ni & ntilde;a) tended to induce annual high (low) DW and low (high) WC occurrences in northern South America, southern and central Africa, southern Asia, eastern Australia, and northwestern North America and high (low) WW and low (high) DC occurrences in South America, Africa, western and southern Asia, southern Europe, western North America, and western Australia. The positive (negative) Dipole Mode Index tended to induce high occurrences of compound warm-related (cold-related) extremes in most global land areas. The positive North Atlantic Oscillation (NAO) tended to induce high WW occurrences and low DC occurrences in central and northern Europe, northern and central-eastern Asia, and eastern North America, especially during DJF. This study provides scientific insights into the spatiotemporal characteristics and driving mechanisms of different compound extremes across the globe under a changing climate.

Warming across the Tibetan Plateau is reorganizing snow dynamics and the resulting runoff regime, with increasingly direct implications for regional freshwater security. However, the distinct roles of rapid surface versus delayed subsurface snowmelt in regulating river flow remain poorly understood. Here, we apply a dual-phase tracking framework to decipher surface and subsurface snowmelt contributions to runoff across ten headwater basins from 1968-2018. Results show that surface snowmelt delivers a short spring pulse, contributing over 25% of total runoff during the melt peak, whereas subsurface snowmelt provides sustained recharge year-round and primarily constitutes winter discharge. Over the past five decades, the centroid timing of surface snowmelt advances significantly at ~6.8 days per decade across ~93% of grid cells, while subsurface snowmelt timing remains nearly stationary. Consequently, the timing of total runoff shifts only modestly (0–3 days per decade), demonstrating strong timing buffering by subsurface pathways that retain and gradually release meltwater. This buffering softens late-season low-flow declines in some basins, yet basin dependence limits its capacity against warming-driven drying. These results underscore the critical role of subsurface meltwater in regulating river flow timing in a warming climate, with implications for water resource management in cryospheric regions.

Study region: This study focuses on six major urban agglomerations (UAs) in China: the BeijingTianjin-Hebei (BTH), Yangtze River Delta (YRD), Pearl River Delta (PRD), Chengdu-Chongqing (CC), Central China (CCR), and Harbin-Changchun (HC) regions. Study focus: Leveraging high-resolution hourly precipitation records from 1985 to 2021, this study examines the intensity (maximum and average), amount, duration, and frequency of extreme precipitation events (EPEs) and their spatiotemporal evolution across these six UAs, highlighting regional contrasts shaped by distinct climatic and environmental contexts. New hydrological insights for the region: Across the six UAs (1985-2021), event amount increases consistently and frequency shows a weak but generally positive rise, whereas intensity diverges regionally-average intensity strengthens in humid PRD/CC/YRD but weakens or remains stable in semi-humid BTH/HC/CCR. Maximum intensity can decouple from average intensity, with PRD showing a pronounced increase in peaks. Duration further differentiates regions, decreasing in PRD but increasing or remaining near-stationary elsewhere. Urbanization effects are metric- and UA-dependent, strongest for amount, weaker for duration and intensity, and minimal for frequency, underscoring coupled climate-urban controls and the need for region-specific risk assessment.

Rapid urbanisation and resource exploitation have reshaped land-use patterns in Bohai Bay (BB), altered hydrological processes, and placed increasing pressure on ecological functions and restoration planning. Most existing studies have primarily relied on static assessments or single ecosystem services, whereas scenario-based evaluations of coastal carbon storage (CS) and habitat quality (HQ), together with their hydrological drivers, remain limited. This study assessed CS and HQ in BB from 2000 to 2020, analysed their dominant hydrological drivers, and projected their trajectories to 2040 under three land-use scenarios: natural development (Q1), ecological protection (Q2), and urban development (Q3). Between 2000 and 2020, CS decreased by 58.00 million tons, while mean HQ declined from 0.50 to 0.48, primarily due to the expansion of impervious and the reduction of cropland and grassland. Compared to 2020, Q2 was associated with the least deterioration in CS (2.72%) and HQ (3.45%), indicating that coordinated protection and restoration can partially buffer development-driven losses. Shapley additive explanations (SHAP) analysis showed that digital elevation model (DEM), distance to rivers, and plant available water content (PAWC) were the dominant drivers of both CS and HQ, jointly contributing more than 80% to their spatial variation. Groundwater storage provided an additional positive effect, indicating that groundwater support is an important foundation for maintaining coastal ecological functions. Integrated CS-HQ zoning revealed a distinct pattern of 'higher in the west and north, lower in the east and south', broadly corresponding to areas with higher elevation, lower disturbance, and stronger water-retention capacity. Under Q2, very important and important zones accounted for the largest combined proportion (69.22%) among the three scenarios. These findings indicate that the ecological consequences of land-use change in coastal regions depend not only on the amount of ecological land retained but also on how land-use trajectories modify water redistribution, soil-moisture availability, groundwater support, and hydrological connectivity.

The hydrological and carbon cycles are two fundamental biogeochemical processes of the Earth system，and they co-evolve through water-carbon coupling at the watershed scale. As a natural unit in terms of water and material balance，the watershed provides an appropriate scale for understanding this coupling and forms an important basis for addressing climate change and watershed water security through nature-based solutions. This review clarifies the water-carbon coupling processes in natural watersheds，analyzes the underlying coupling mechanisms，and synthesizes observational techniques and modeling frameworks for watershed water-carbon interactions. A bidirectional coupling framework for the watershed water-carbon cycle is proposed. The influences of climatic conditions on watershed water-carbon coupling are further discussed. In addition，major knowledge gaps are identified in terms of bidirectional coupling processes，quantification of interface processes，data support，and the impacts of extreme events，and future research priorities are highlighted. This review is expected to provide a scientific basis for watershed carbon neutrality，water security，ecological security，and food security，as well as for climate change mitigation and adaptation based on nature-based solutions.

Accurate hydrological modeling is fundamental to reliable flood forecasting and water resources assessment. Conventionally, Hydrological models are commonly calibrated primarily against runoff observations. However, other critical variables such as evapotranspiration (ET) and soil moisture (SM) are frequently excluded from the calibration process, largely due to limited data availability. Consequently, the fidelity of runoff-calibrated models in simulating ET and SM accurately remains questionable. This limitation is particularly pronounced in arid and semi-arid regions, where only a small fraction of precipitation converts into runoff. To address this challenge, this study proposes a multi-objective calibration framework that jointly considers runoff, ET, and SM. The framework is evaluated in three representative arid and semi-arid catchments in northern China and implemented across five widely used hydrological models (WBM, XAJ, HBV, SWAT, and VIC) to test its general applicability. Results show that calibration using runoff as the sole objective can achieve satisfactory streamflow performance; however, it does not necessarily guarantee reliable simulations of ET and SM, with the non-target variables often exhibiting low accuracy. This discrepancy is mainly attributed to the fact that different model parameters exhibit distinct sensitivities to different hydrological variables, such that runoff-only calibration provides insufficient constraints on ET and SM, leading to relatively large predictive uncertainty. In contrast, the proposed multi-objective calibration framework improves the simulation accuracy of ET and SM while maintaining streamflow performance, and further reduces both parameter uncertainty and predictive uncertainty. This study provides a framework for improving multi-variable hydrological simulations and reduce model uncertainty, with implications for both theoretical research and practical water resource management.

This study compares the advantages and limitations of traditional CMIP6 data fusion methods and machine learning fusion methods when applied to drought identification in the Yangtze River Basin. We consider three traditional fusion methods and five machine learning fusion methods, and calculate drought indices over 3-, 6-, and 12-month periods based on precipitation data from meteorological stations in the Yangtze River Basin (1960-2014) and 15 CMIP6 model datasets. The drought identification index is used to evaluate the performance of the fusion methods. Results indicate that traditional statistical methods have significant limitations in the upper reaches of the basin, where the terrain is highly undulating, but perform better in the middle and lower reaches, which are relatively flat. Among the machine learning methods, neural networks tend to amplify the observational noise, whereas kernel-tuning methods better accommodate nonlinear relationships across different SPI time scales. The prediction performance of all methods decreases from the 12- to 3-month drought indices, but the extent of the decline varies. The Random Forest and Radial Basis Function methods give the smallest reduction in performance, while the Backpropagation and Backpropagation-Adaboost methods produce the largest drop in performance.

Soil could represent a potentially notable source of carbon for achieving global carbon neutrality. However, how the land surface soil organic carbon (SOC) stock, which is more sensitive to climate change than other carbon stocks, will change naturally under the influence of global warming remains unknown. In this work, the global land surface SOC trends from 1981 to 2019 were explored, and the driving factors were identified. A random forest model (a type of machine learning method) was proposed to predict future global surface SOC trends integrated with climate scenarios of the Coupled Model Intercomparison Project Phase 6 (CMIP6) models. The results revealed that the global surface SOC content will increase, while the temperature and precipitation are the main climate drivers at the global scale, and vegetation cover is a crucial local factor influencing the increase in SOC. However, under the 1.5 degrees C global warming scenario, the land SOC sink will increase by 13.0 petagram carbon (PgC) at most compared with that under the SSP2-4.5 scenario, which accounts for only 19% of the total carbon emission capacity at the current 1.1 to 1.5 degrees C global warming level. Moreover, this value is far from the Paris Agreement target of four out of one thousand for the annual increase in the soil carbon stock 40 cm below the surface over the next 20 years (2.72 PgC a-1). This illustrates that overreliance on natural carbon sinks is a high-risk strategy. These findings highlight the urgency of implementing mitigation and removal strategies to reduce greenhouse gas emissions. (c) 2025 THE AUTHORS. Published by Elsevier LTD on behalf of Chinese Academy of Engineering and Higher Education Press Limited Company. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Under evolving environmental conditions, significant alterations in streamflow have been observed, necessitating a systematic attribution analysis to inform watershed governance and water resource management. Previous studies predominantly focus on the individual impacts of climate change and human activities on runoff variation, often neglecting their synergistic effects, which are frequently merged into the human-induced contribution. Leveraging the nonlinear dynamics of hydrological systems, this study introduces a segmented hydrological calibration and modelling approach to quantify the nonlinear synergistic influence of environmental changes on runoff. The framework is applied to disentangle the contributions of climate change, human activities, and their interplay in altering runoff patterns across the Wei River Basin. Key findings include: (1) From 1955 to 2022, observed runoff exhibited a phased decline with reductions of 31.8 % during 1971-1995 and 43.8 % during 1996-2022 relative to the baseline period (1955-1970). (2) During 1971-1995, climate change and human activities accounted for runoff declines of 4.29 mm and 22.49 mm, respectively, while their synergy contributed an additional 1.16 mm, representing 15.3 %, 80.5 %, and 4.2 % of the total reduction. (3) From 1996 to 2022, runoff decreased by 38.48 mm compared to the baseline, with climate change (3.62 mm), human activities (32.27 mm), and their synergistic effect (2.59 mm) responsible for 9.4 %, 83.9 %, and 6.7 % of the decline, respectively. (4) Human activities emerged as the dominant driver of runoff reduction, with an escalating influence over time, whereas the role of climate change gradually decreased. Critically, the synergy between these factors amplified the overall decline. These findings underscore the imperative of incorporating synergistic climate-human effects into future watershed management strategies to enhance the precision of water resource planning.

Soil temperature is a key state variable in terrestrial hydrological systems, and characterizing its spatiotemporal evolution is essential for understanding hydrological responses to climate change. However, deep soil temperature exhibits lagged and nonlinear responses to surface climate forcing, and long-term continuous observations are limited, complicating accurate modeling and future projection. This study proposes a depth-resolved, data-driven framework for soil temperature modeling based on 30 years (1985–2014) of in-situ observations from five stations in the Huaibei Plain, explicitly incorporating feature selection and depth-dependent lag effects into model construction. Based on this framework, multiple deep learning models (CNN, LSTM, and CNN–LSTM) are implemented and compared to simulate soil temperature at depths of 10, 40, 100, and 200 cm, and the best-performing model is selected for climate scenario projections, followed by sensitivity analysis and SHAP-based explainability to quantify the contributions of meteorological drivers. Results indicate that the hybrid CNN–LSTM model outperforms the standalone network models across all soil depths. Future soil temperature exhibits a clear warming trend, with warming rates exceeding those observed during the historical period. Near-surface layers (10 and 40 cm) show the strongest response to climate change, accompanied by greater interannual variability compared to deeper soils. Soil warming further accelerates during the mid- to late 21st century (2051–2100) and reaches its maximum under the high–radiative forcing SSP5–8.5 scenario, which is also associated with the largest inter-model uncertainty. These findings show that the proposed framework effectively characterizes depth-dependent soil temperature responses and provides depth-resolved information for future hydrological and land-surface assessments.

The Tibetan Plateau (TP) is one of the most climate-sensitive regions on Earth, where soil respiration (Rs) and ecosystem respiration (Re) are expected to respond strongly to climate change. However, due to sparse observations and complex ecological processes, the spatial characteristics of respiration fluxes across the Plateau and their responses to future climate changes remain poorly understood. In this study, we developed a knowledge-guided multi-task deep neural network (KG-MTDNN) model that leverages prior ecological information and shared representations to improve the ecological plausibility and model performance of respiration simulations under data-sparse conditions. Based on this model, we conducted a comprehensive assessment of both the current status and future trajectories of Rs and Re across the TP. Our results reveal a southeast-to-northwest decreasing spatial pattern of respiration, along with a persistent increasing trend under future climate scenarios. Seasonally, the majority of carbon emissions occur during the growing season, whereas spatially, non-permafrost areas contribute most of the total emissions. In terms of Re components, Rs accounts for the dominant portion (similar to 70%) of Re, and this dominance is expected to further strengthen under ongoing climate change. Notably, according to our estimates, climate-induced enhancement of Re under high radiative forcing scenario will offset nearly 70% of the TP's current carbon sink. Our study highlights the potential risk of significantly increased carbon emissions from the Earth's Third Pole under future climate change, and improves understanding of climate-carbon cycle feedback mechanisms in cold-region ecosystems.

Infiltration is a vital component of the water cycle, governing the runoff generation and concentration processes. A more accurate description of the infiltration process is essential to improve hydrological modeling and facilitate effective water management strategies. In cold regions, soil water infiltration process is affected by freeze-thaw cycle and more crucial due to its sensitivity to global warming. Current infiltration models are developed for stable, temperate conditions and constant soil properties, which contrast sharply with the thermal and hydraulic conditions in freeze-thaw environments. Based on the widely used infiltration Horton equation, this study proposed a Horton-RCCC (Research Center for Climate Change) formula specifically designed for the freeze-thaw environment in cold regions, which are achieved by several steps as follows. A large-size undisturbed soil sampling and restoration method to enable laboratory simulation of soil freeze-thaw processes. The soil freeze-thaw state is described using the ratio of current freeze-thaw depth to maximum freeze-thaw depth. The calculation equations for soil water initial infiltration rate and stable infiltration rate are developed, shifting from the traditional approach of using a single parameter set for entire watersheds and improving distributed characterization of soil infiltration capacity under varying underlying surface conditions. The Horton-RCCC formula is calibrated and validated in permafrost and seasonally frozen ground regions. In complete thawing stage, the Horton-RCCC formula demonstrates comparable accuracy with that of Horton equation on a condition of zero slope, and shows its advantage to calculation infiltration under sloped conditions. In freeze-thaw period, the Horton-RCCC formula effectively addresses the dynamics of underlying surface infiltration conditions, and shows a significant improvement in accuracy. This formula breaks away from the conventional research paradigm of directly applying infiltration formulas for temperate conditions, could stimulate improvements in hydrological modeling, water resource management and climate change adaptation in cold regions.

In stage simulation for river-type reservoirs, realistic representation of lateral inflows from ungauged intervening areas is essential for maintaining water balance. However, under strong backwater effects and complex operational boundary conditions, hydrodynamic models are prone to equifinality, and whether explicit coupling of lateral inflow can substantially improve stage simulation remains insufficiently understood. To address this issue, this study focused on the Three Gorges Reservoir Area and developed a high-resolution coupled hydrologic-hydrodynamic framework, together with attribution analysis based on Shapley Additive Explanations (SHAP), to evaluate the actual benefits of explicitly representing incremental lateral inflow under four typical hydrodynamic regimes, namely low stage, high stage, fluctuating stage, and operation-varying conditions including drawdown and impoundment. Results show that the simulation gain from lateral inflow coupling is not universal, but strongly regime dependent, and is fundamentally controlled by whether the lateral inflow signal can be effectively expressed under a given hydrodynamic state. Under quasi-steady regimes with strong downstream control, including high stage and low stage conditions, local perturbations induced by lateral inflow are readily masked by backwater buffering, parameter compensation, friction desensitization, and shallow-flow conveyance bias, resulting in only limited improvement from explicit coupling. In contrast, under unsteady regimes characterized by rapid stage fluctuations or impoundment, the dominant role of the downstream boundary weakens, and stage evolution becomes jointly shaped by upstream propagation, downstream adjustment, and local inflow, allowing explicit lateral inflow coupling to more effectively reduce overall and peak-stage errors. These findings suggest that lateral inflow should not be treated as a static modeling option in reservoir-affected rivers, but should instead be evaluated dynamically in relation to the hydrodynamic control background of each event.

Floods are among Earth's most devastating natural disasters, with cascading societal and ecological impacts. Flood timing shifts amplify risks by disrupting preparedness, yet their global patterns remain largely unquantified. Using multi-model ensembles, we provide a global-scale assessment of flood timing changes under incremental warming (1.5 °C-4.0 °C). Here we show that anthropogenic forcing advances global flood timing by 0.43 ± 0.25 days per 0.5 °C of warming, with regional divergence: early-flood regions shift even earlier, while late-flood regions experience further delay. At 1.5 °C, 50.73 ± 4.23% of global land area faces flood timing shifts greater than 7 days, escalating to 52.85 ± 3.20% at 2.0 °C. Countries including China, India, and the United States are projected to experience greater population exposure. These findings highlight the need to change how flood risk is conceptualized and managed. The traditional focus on flood magnitude and frequency must expand to incorporate timing as a fundamental variable in climate adaptation planning.