In this session, we welcome contributions on novel frameworks for model development, evaluation and parametrization across spatio-temporal scales.
Potential contributions could (but are not limited to):
(1) advance seamless modeling of spatial patterns in hydrology and land models using new data products and earth observations;
(2) improve model structure by representing often neglected processes in hydrological models such as human impacts or vegetation dynamics;
(3) provide novel concepts for improving the characterization of internal and external model fluxes and their spatio-temporal dynamics;
(4) introduce new approaches for model calibration and evaluation, especially to improve process representation, and/or to improve model predictions under changing conditions;
Improvements in computer power have facilitated the automatic calibration of hydrological and ecohydrological models. Still, many internal hydrological processes remain poorly understood due to their inherent complexity, strong spatial heterogeneity, and highly interactive nature. This knowledge gap largely stems from the prevailing focus on catchment celerity responses (rainfall-runoff response) in hydrological studies, while the pathways and velocities of internal water fluxes remain largely unexplored. Consequently, many hydrological models function as grey boxes – capable of reproducing discharge dynamics yet often “for the wrong reasons.”
Stable water isotopes offer a powerful means to unbox the water cycle with improved process understanding across spatial scales. As conservative tracers, 2H and 18O are independent of most biogeochemical reactions and naturally integrate spatial heterogeneity, providing effective constraints on the spatial connectivity and velocities of hydrological flow paths. In this presentation, we synthesize our experience with isotope-enabled hydrological and ecohydrological modelling to demonstrate how such frameworks enhance process representation from plot to continental scales.
We will briefly introduce how we developed or refined isotope-aided ecohydrological models at plot, river, catchment, and continental scales. We then demonstrate how these models can be used to partition hydrological fluxes and to identify key flow pathways and their corresponding velocities. Specifically, we illustrate how stable isotopes can be used to (i) quantify depth-dependent root water uptake at the plot scale, (ii) resolve geometry-controlled channel recharge and leakage at the river scale, (iii) diagnose lateral hydrological connectivity among landscape units at the catchment scale, and (iv) characterize pathways and velocities of terrestrial water cycling at the continental scale. These process-based insights not only support more robust and sustainable water management strategies, but also advance our understanding of the co-evolutionary mechanisms linking water and nutrient cycles.
How to cite: Wu, S., Tetzlaff, D., and Soulsby, C.: Unboxing the Water Cycle across Spatial Scales with Isotope-Enabled Hydrological Modelling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9295, https://doi.org/10.5194/egusphere-egu26-9295, 2026.
Root zone storage (Sr,max) is a key parameter in hydrological and land-surface models that regulates water fluxes partitioning and the ability of vegetation to buffer against dry periods, and closely linked to the Budyko parameter ω describing long-term catchment precipitation partitioning (Ibrahim et al., 2025). Sr,max is defined as the maximum subsurface water volume accessible to plants to meet their transpiration demands. Because direct observations of rooting depth are scarce and limited to local scales, catchment-scale Sr,max is commonly calibrated or estimated using the memory-method. This method derives Sr,max from annual maximum water storage deficits calculated from water balance data, assuming a fixed extreme-value distribution (typically Gumbel), and a predefined return period of 20 years. Despite its broad application, uncertainties arising from these assumptions and their implications for hydrological modelling have rarely been quantified. Here, we systematically evaluate the uncertainty, robustness and practical applicability of memory-method Sr,max estimates across different hydroclimatic regions globally (≈ 5700 catchments). Annual maximum storage deficits (Sd) were derived following the original memory-method framework but instead of fitting Gumbel distribution to Sd , we used the Generalized Extreme Value (GEV) distribution to allow flexible tail behaviour. Analysis of the GEV shape parameter - which determines tail behaviour - within the Budyko framework reveals strong hydroclimatic control, with Pearson correlations of approximately -0.50 with both the aridity index and the evaporative index. Most water-limited catchments exhibit negative shape parameter indicative of bounded (reversed Weibull Type-III) extremes, whereas energy-limited catchments tend toward positive shape parameters associated with heavy tailed (Frechet type-II) behaviour.
Uncertainty in Sr,max estimates was quantified using bootstrap resampling and expressed as confidence bounds derived from 2-year and 80-year return periods. Uncertainty width was strongly climate dependent with the widest ranges (median ≈ 132mm) occurring in transitional climates (aridity index ≈ 0.5-2), while arid and humid regions exhibit comparatively narrow uncertainty envelops (median ≈ 72mm). To assess practical implications, Sr,max uncertainty bounds were propagated into hydrological model calibration (≈1950 catchments). Among the Pareto-optimal solutions, model performance metrics were very similar, indicating strong equifinality in Sr,max estimates. Median simulated Sr,max values show strong agreement with memory-method estimates, with a global Pearson corelation of 0.92 (RMSE ≈ 60mm) and corelations across Koppen-Geiger climate zones ranging from 0.91-0.98. When memory-method Sr,max was calculated using the GEV distribution, the strongest agreement with median simulated Sr,max occurred for return periods of 20-30 years (Pearson r ≈ 0.93) at the global scale, with particularly clear sensitivity in cold and temperate regions. Overall, our results demonstrate that the memory-method robustly captures spatial patterns of Sr,max and that the commonly used 20-year return period represents a physically meaningful and hydro-climatically consistent choice. Using memory-method-based Sr,max estimates, instead of calibrating it, can reduce model complexity and parameter uncertainty without compromising model performance, offering practical advantages for large-scale hydrological and land-surface modelling.
Reference:
Ibrahim, M., Van der Ent, R., Coenders, M., Markus Hrachowitz, M. & van Oorschot, F. 2025. Catchment precipitation partitioning in the Budyko framework is controlled by root zone storage capacity. Environmental Research Letters, (under review)
How to cite: Ibrahim, M., van der Ent, R., Coenders, M., and Hrachowitz, M.: Can improved root zone storage capacity estimates simplify hydrological modelling?, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13773, https://doi.org/10.5194/egusphere-egu26-13773, 2026.
Surface Soil Moisture (SSM) is a key variable in terrestrial hydrology, governing land–atmosphere exchanges of water and energy, influencing runoff generation, and mediating interactions with deeper soil layers and groundwater recharge. Accurate representation of SSM within land surface and hydrological models is critical for simulating these processes realistically. To achieve this, operational hydrometeorological systems assimilate satellite-derived SSM observations into models.
Looking ahead, next-generation hydrological modeling aims to develop “digital twins” of Earth’s water cycle, high-resolution (≤1 km), physically consistent systems that integrate advanced models with spaceborne observations through Data Assimilation (DA). However, implementing SSM DA at such fine spatial scales raises fundamental questions. For instance, the benefits and limitations of assimilating high-resolution (1 km) SSM products remain poorly understood. Furthermore, it is unclear how DA performance compares when using high-resolution but temporally sparse observations (e.g., every few days) versus coarser-resolution data available daily.
This study addresses these gaps by conducting two DA experiments: (i) assimilation of ASCAT-derived SSM at 25 km resolution with daily availability, and (ii) assimilation of Sentinel-1-derived SSM at 1 km resolution with a few-day revisit. Both experiments employ the World Wide Water Resources Assessment (W3RA) hydrological model, downscaled to operate at 1 km daily resolution. The Ebro River basin (Iberian Peninsula) serves as the testbed, chosen for its availability of 1 km precipitation forcing, in-situ discharge observations, and significant irrigation activity, which is an anthropogenic factor not explicitly represented in W3RA but potentially captured through SSM DA. Assimilation is implemented via a localized Ensemble Kalman Filter (EnKF).
Evaluation is carried out in three stages: (1) comparison of DA outputs against assimilated observations to assess assimilation skill; (2) analysis of spatial and temporal variability in SSM estimates to quantify DA method’s downscaling capability; and (3) validation of simulated river runoff against independent discharge measurements. Through this comparative framework, the study aims to elucidate the trade-offs, benefits, and challenges of high-resolution SSM DA for operational hydrological modeling.
How to cite: Retegui-Schiettekatte, L., Leopardi, F., Gaona, J., Brocca, L., Filippucci, P., Camici, S., Madsen, H., and Forootan, E.: Assimilation of SSM into hydrological models: comparing assimilation in higher spatial resolution (1km, few-daily) vs. higher temporal resolution (25km, daily), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19207, https://doi.org/10.5194/egusphere-egu26-19207, 2026.
Paddy rice is a major irrigated crop in Asia and plays a critical role in regional food security, particularly in countries such as India. It is also highly water-intensive, accounting for roughly 40% of global agricultural irrigation withdrawals. Flooded paddy systems exhibit a unique water balance characterized by continuous ponding, high evapotranspiration, seepage and percolation losses, return flows, and controlled drainage, leading to substantial seasonal water requirements. However, most basin-scale hydrological models are originally developed for upland crops and rainfall–runoff systems, and therefore have limited capability to represent irrigated, ponded systems.
The Soil and Water Assessment Tool (SWAT) is widely used for watershed-scale hydrological assessments but lacks explicit representation of flooded rice cultivation. Existing approaches in SWAT including the curve number (CN) method (treating paddy as upland) and pothole routines do not fully capture paddy-specific irrigation management or field-level water balance components. In contrast, field-scale crop models such as ORYZA and CERES-Rice provide more advanced representations of paddy water dynamics, but are not intended for watershed-scale analysis. To address this gap, this study develops a new process-based paddy module (SWAT-PADDY) by integrating soil water routing and irrigation management algorithms adapted from crop model frameworks into SWAT. The module accounts for key management practices, including transplanting, puddling, irrigation and drainage scheduling, and bunded field hydraulics. Soil water routing was reformulated to couple the ponded layer with the soil profile, enabling realistic simulation of infiltration, percolation, overflow, and return flows. The enhanced model was evaluated at ten paddy fields in South India over two cropping seasons using observed water levels, and key water balance components were assessed through cross-model comparisons with ORYZA, CERES-Rice, and a numerical soil water flow model (HYDRUS-1D) to assess consistency and process representation.
Results show that SWAT-PADDY realistically simulates ponded water levels and major water balance components, including evapotranspiration, infiltration, percolation, overflow, and soil water storage. The enhanced model demonstrated good statistical performance for observed water levels (NSE > 0.5) and achieved zero water balance closure error at field scale. Cross-model comparisons showed strong agreement with field-scale crop and numerical model simulations. The improved process representation broadens the applicability of SWAT for regions dominated by irrigated rice cultivation and enables basin-scale assessment of water use under diverse climatic, soil, and management conditions.
Keywords: Flooded rice systems, SWAT model development, Process-based modelling, hydrology, Water balance simulation
How to cite: Shyma, M. and Narasimhan, B.: Process-Based Modeling of Paddy Water Dynamics in SWAT Through Crop-Model Algorithm Integration, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14327, https://doi.org/10.5194/egusphere-egu26-14327, 2026.
Vegetation strongly influences evapotranspiration, the largest water flux from land to atmosphere, and a key land-surface process, and thus plays a central role for capturing soil moisture and groundwater dynamics. As drought duration and magnitude increase, reliable high-resolution simulations of these variables become increasingly important for informed water resource management and sustainable allocation decisions. Yet, many hydrological models still have difficulty reproducing spatially distributed variables, in part due to epistemic uncertainty arising from model structural choices. Epistemic uncertainty is further amplified by calibration practices that rely solely on streamflow, an integration variable. This common practice disregards internal states and, consequently, the spatial variability of hydrological processes. Because vegetation modulates the variability of evapotranspiration and soil moisture, it is particularly relevant for local-scale uncertainty analyses. Although numerous studies examine individual aspects of vegetation–water interactions and their parameterization, it remains unclear which vegetation processes are essential to capture small-scale spatial variability and which may be redundant or even exacerbate overfitting, thereby increasing uncertainty.
In this study, we examine how different model structures influence simulated soil moisture and related water storage variables using the mesoscale Hydrological Model (mHM) [1]. To this end, we incorporate Leaf Area Index–based evaporation control, alternative root water uptake schemes, and additional land-cover types, representing three types of forest as well as pastures, savanna, wetlands. We construct model structure variants representing different combinations of these processes, including configurations in which individual processes are disabled. We then evaluate the skill of each model variant to reproduce not only the streamflow but also catchment wide total water storage and spatial distribution of soil moisture at a resolution of 1 km, which allows vegetation heterogeneity and its impact on evapotranspiration and soil moisture dynamics to be explicitly represented. The investigation provides an improved understanding of model structural uncertainty associated with vegetation, and assesses whether additional calibration on spatially distributed variables can compensate for structural shortcomings. This is achieved by comparing the traditional streamflow calibration with the actual evapotranspiration calibration using the spatial pattern efficiency metric (ESP) defined by Dembélé et al. [2].
By identifying which vegetation processes meaningfully improve spatial predictions, this research supports the development of more robust hydrological models for drought assessment and sustainable water management under increasing hydro-climatic stress.
References:
[1] L. Samaniego, R. Kumar, S. Attinger, Water Resources Research 2010, 46.
[2] M. Dembélé, M. Hrachowitz, H. H. Savenije, G. Mariéthoz, B. Schaefli, Water resources research 2020, 56, e2019WR026085.
How to cite: Peter, C., Lüdke, V. S., Westermann, S. A., Boeing, F., Shrestha, P. K., Kelbling, M., Thober, S., Samaniego, L., and Hildebrandt, A.: Representing Vegetation in Hydrological Modeling: Between process detail and structural uncertainty, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17532, https://doi.org/10.5194/egusphere-egu26-17532, 2026.
Land surface models are often made increasingly complex to represent heterogeneity in soils and vegetation. However, the level of horizontal and vertical complexity actually required to capture drought processes in boreal forests remains unclear. This issue is critical for large-sample hydrological and climate applications, where detailed site-by-site calibration is rarely feasible. In this study, the Canadian Land Surface Scheme (CLASS v3.6) is applied in point mode and driven by ERA5-Land to quantify trade-offs between structural complexity and hydrological realism under a strictly calibration-free, rule-based parameterization.
The analysis is conducted at Forêt Montmorency, a humid, snow-dominated boreal catchment in Québec, Canada, instrumented with eddy-covariance towers, soil water content and temperature profiles, and long-term hydrometeorological observations. Vegetation and soil parameters are constrained by field data, LiDAR-based canopy metrics and CLASS defaults, without tuning to match fluxes. Two experiment sets are considered. In the first, the impact of progressively increasing the number of grouped response units (GRUs) and soil layers (from reduced 3–4 layer profiles to an 8-layer column) on model skill is assessed under identical ERA5-Land forcing. Second, the multi-decadal ERA5-Land record is used to isolate and evaluate model behavior during independently defined drought windows. Therefore, performance metrics specifically target water-limited conditions rather than aggregates over mixed wet and wet–dry periods.
Model behavior is evaluated for total evaporation and its components, soil moisture and temperature, diagnostically derived soil water potential (psi), and simple runoff and low-flow indicators at seasonal to annual scales, using Kling–Gupta efficiency, bias and error metrics. Drought periods are defined independently of the model from standardized climatic and soil-based indices (SPEI, SSMI, REW and psi thresholds), ensuring that the assessment targets genuinely water-limited conditions rather than artifacts of model structure.
Results indicate clear diminishing returns from added structural complexity. Increasing the number of GRUs and extending the soil column beyond a limited number of layers does not systematically improve evaporation skill and can degrade the coherence of shallow soil water content dynamics. However, a parsimonious configuration with a small set of dominant GRUs and a moderately deep soil profile is sufficient to reproduce seasonal energy partitioning and to capture the timing and relative severity of drought events within forcing-related uncertainty. These findings provide quantitative evidence that robust drought diagnostics in snow-affected boreal forests do not require highly complex land-surface setups and that carefully designed, rule-based configurations offer a pragmatic benchmark for regional hydrological and climate modelling studies.
How to cite: Razavi Ebrahimi, Y. and Anctil, F.: How much land-surface complexity is needed to simulate drought processes in boreal forests? A calibration-free CLASS assessment across an energy-water gradient, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17777, https://doi.org/10.5194/egusphere-egu26-17777, 2026.
Water resources are increasingly under threat across the UK. Climate change is driving more frequent and severe floods and droughts, while anthropogenic pressures such as abstractions and wastewater discharges are increasingly impacting streamflow. Accurate hydrological simulations are critical for water resources management, particularly in densely populated and water-stressed regions such as South-East England. However, many hydrological models omit or oversimplify key human activities such as surface and groundwater abstractions and discharges from wastewater treatment plants, limiting model performance in human-influenced catchments.
To address this challenge, we exploit a unique water resource management dataset, which includes decades-long records of monthly surface water and groundwater abstraction (1999–2023) and daily wastewater discharge time series (2005–2015) for thousands of locations across England. We first analyse this dataset to identify when and where river flows are most affected by abstractions and wastewater discharges, and characterise their intra-annual and interannual variability, providing evidence for integrating these data into hydrological models. We then implement water abstraction and wastewater discharge modules within the DECIPHeR-GW hydrological model, and quantify the resulting improvements in streamflow simulations. We also identify the conditions under which neglecting these processes leads to substantial model degradation. Scenario-based experiments are used to assess how water resource management data should be represented, for instance, the importance of temporal patterns, providing guidance for modelling human impacts in data-scarce regions.
How to cite: Zheng, Y., Coxon, G., Pianosi, F., Woods, R., Rahman, M., Devitt, L., and Howden, N.: When and where are abstractions and wastewater discharges crucial for streamflow modelling? Lessons learnt from large sample analysis and hydrological modelling in England, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8296, https://doi.org/10.5194/egusphere-egu26-8296, 2026.
A current challenge in hydrological modelling is to provide water resource managers with projections of future water resources under various climate change, land use and water management scenarios. This will help them to develop climate change adaptation strategies specific to their territories. The hydrological models used to simulate these projections can be complicated and time-consuming to implement at a local level. Therefore, the challenge lies in providing water managers with modelling tools that are simple enough to implement, yet realistic enough in their representation of processes to simulate the correct hydrological response.
CACTUS (CustomizAble CaTchment model for water Use Scenarios) is an interactive tool that allows users to customise the characteristics of a simplified catchment, then run simulations using the distributed hydrological model J2000 and visualize the simulation results. In order to make the tool operational and quick to implement, several simplifications had to be made : (i) the shape of the catchment and river network is predefined and discretized into a fixed number of grid cells (185) and reaches (14), (ii) the input climate data are estimated for a specific localization based on a limited number of reference time series, (iii) the Penmann-Monteith formula to represent potential evapotranspiration has been replaced by a formula that depends only on latitude and temperature (Oudin formula), (iv) the model is not calibrated and the parameters are selected from standard values found in the literature, (v) climate change scenarios are produced by perturbing the climate variables of the present period (delta on the seasonal temperature, cumulative precipitation and number of rainy days). Thanks to these simplifications, the catchment can be configured in about ten to twenty minutes and a 40-year simulation can be run in a few seconds.
To evaluate the accuracy with which CACTUS can simulate catchment hydrology and its response to climate scenarios, simulation results obtained with this tool were compared with hydrological projections from an ensemble of 7 hydrological models and 4 climate projections (Sauquet et al., 2025), in 5 contrasted French catchments (1 high mountain basin, and 2 lower mountain basins, 1 agricultural basin in plains, 1 peri-urban basin). The comparison shows that CACTUS simulates hydrological regimes and changes in hydrological indicators (in terms of both signs and magnitudes) that fall within the uncertainty range of the benchmark 7-models ensemble. This demonstrates that quick-to-implement, simplified hydrological models can provide water managers with a valuable primary level information for initiating an exploration of adaptation strategies.
Sauquet, E., Evin, G., Siauve, S., Aissat, R., Arnaud, P., Bérel, M., Bonneau, J., Branger, F., Caballero, Y., Colléoni, F., Ducharne, A., Gailhard, J., Habets, F., Hendrickx, F., Héraut, L., Hingray, B., Huang, P., Jaouen, T., Jeantet, A., Lanini, S., Le Lay, M., Magand, C., Mimeau, L., Monteil, C., Munier, S., Perrin, C., Robelin, O., Rousset, F., Soubeyroux, J.-M., Strohmenger, L., Thirel, G., Tocquer, F., Tramblay, Y., Vergnes, J.-P., and Vidal, J.-P.: A large transient multi-scenario multi-model ensemble of future streamflow and groundwater projections in France, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2025-1788, 2025
How to cite: Mimeau, L., Héraut, L., Vidal, J.-P., and Branger, F.: Are more advanced hydrological models necessary to help water managers adapt to climate change?, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10150, https://doi.org/10.5194/egusphere-egu26-10150, 2026.
Hydrological model structures are often selected based on legacy considerations—such as habit, practicality, or experience—rather than whether they are fit for a specific modelling purpose. This is problematic, as model structure alone can substantially influence modelling results and hence outcomes for e.g. design flow assessment. Automatic Model Structure Identification (AMSI) offers a way to address this issue by framing model choice as an optimization problem. AMSI combines the modular modelling framework Raven with mixed-integer calibration algorithms (DDS/PA-DDS), allowing the simultaneous optimization of model structural choices and parameter values with respect to user-defined objectives.
Here, we apply AMSI to explore a hypothesis space of more than 13,500 conceptual model structures with zero to 12 parameters per model. We test 14 calibration routines, including six single-metric, four multi-metric, and four multi-objective formulations, designed to reflect different modelling purposes that target flood, drought, and water-resources management assessment. Model evaluation uses metrics and hydrological signatures associated with different aspects of the flow regime to assess model suitability across these different purposes. All experiments are conducted on a test catchment located on the Eastern Coast of the US.
Each calibration routine is performed 50 times, yielding a set of preferred model structures. These are analyzed regarding their individual processes and equations, as well as model performance across purpose-specific metric and flow signature groups. Results show that model structural preferences vary with modelling purpose, favouring different process descriptions for different intended applications of the model. Within the tested hypothesis space, identifying suitable model structures is easiest for water-resources management (average flow behaviour), followed by flood (peak flow) modelling, and most challenging for drought (low flow) modelling. Multi-metric and multi-objective calibrations provide more balanced representations than single-metric approaches, with multi-objective calibration revealing explicit trade-offs between structural choices and multi-metric calibration reducing structural equifinality.
How to cite: Spieler, D. and Stadnyk, T.: When Calibration Metrics Choose Your Model: Investigating Model Selection Choices for Different Modelling Purposes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15121, https://doi.org/10.5194/egusphere-egu26-15121, 2026.
The research and operational communities have developed many models to represent the complexity and diversity of hydrological processes and meet specific application needs. Previous studies have shown the limitations of a one-size-fits-all model structure (e.g. poor representation of local conditions, limited process representation, scalability issues). To address these barriers and improve model performance, multi-model approaches have been developed that select and/or combine outputs from an ensemble of models (e.g., catchment-specific selection based on performance scores, or weighting of ensemble members using methods such as Bayesian model averaging).
This study compares multi-model methods to improve streamflow simulation. Specifically, we evaluated five different approaches: a mosaic (i.e. per-catchment selection) based on performance, a mosaic based on performance-equivalence, a static combination in time and space (i.e. a fixed combination applied across all catchments), a static combination in time only (i.e. per-catchment combination) and a dynamic combination (i.e. evolving over time and space). To this end, an ensemble of 78 models was designed with the Framework for Understanding Structural Errors (FUSE) and applied to 559 catchments in the CAMELS dataset across the contiguous USA. The evaluation is based on a composite criterion to account, to some extent, for both high- and low-flow conditions. Sampling uncertainty (i.e. the variability in performance scores due to the evaluation period selected) was assessed using a bootstrap-jackknife method.
Results show that differences between multi-model approaches are small, even when complexity varies greatly (e.g., number of models per catchments, variability in space and time, computational time). Benefits compared with a one-size-fits-all model are not as large as expected, especially after considering sampling uncertainty. While perhaps surprising, this underscores the strength of the one-size-fits-all model selection used here, where model choice is guided by performance across a large ensemble of models and sample of catchments, and not arbitrarily or by convenience. These findings may also reflect limitations of common evaluation metrics, which may not fully capture the benefits of more complex approaches.
How to cite: Thébault, C., Knoben, W. J. M., Addor, N., and Clark, M. P.: Comparing multi-model approaches to simulate streamflow across a large sample of catchments, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5934, https://doi.org/10.5194/egusphere-egu26-5934, 2026.
Karst springs represent an important source of drinking water. Karst aquifers are highly complex and heterogeneous hydrogeological systems made of poorly known conduit networks which makes springs discharge difficult to forecast. Lumped hydrological models can only satisfactorily model karst systems if they consider both diffuse and localised infiltration, and matrix and conduit flow pathways. The French Geological Survey developed RAMEAU (River and Aquifer Model of the frEnch geological sUrvey), a flexible lumped hydrogeological model conceptually encompassing these processes in its model structure. The objective of this study is to demonstrate the versatility of the model to simulate discharge in a variety of karstic systems in France. This study evaluates the performance of the different model structures to simulate the discharge of a large sample of French springs. The models are calibrated on decades long time series of observed spring discharge, and the best model structure is selected independently for each spring. These models are then integrated in the Aqui-FR platform, a multi-model platform that provides hydrogeological seasonal forecasts over continental France.
How to cite: Hallouin, T., Vergnes, J.-P., Charlier, J.-B., and Audigane, P.: Testing multiple structures of a lumped hydrogeological model to simulate karst spring discharge in France and provide seasonal forecasts in the Aqui-FR platform, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7811, https://doi.org/10.5194/egusphere-egu26-7811, 2026.
The spatial resolution at which distributed hydrological models are implemented plays a critical role in their ability to represent dominant hydrological processes and to close the water balance consistently across scales (Blöschl & Sivapalan, 1995). Despite the increasing availability of satellite-based observations, their integration into multiscale hydrological modelling frameworks remains challenged by scale dependency, parameter transferability (Barrios & Francés, 2012; Medici et al., 2008), and computational constraints.
This study presents a multiscale performance assessment of the distributed hydrological model TETIS (Francés et al., 2007; GIMHA - Grupo de Investigación en Modelación Hidrológica y Ambiental Distribuida, 2021) by coupling in situ observations from 27 gauging stations with satellite-derived(García-García et al., 2026) state variables, including evapotranspiration (ET) and surface soil moisture (SSM), in the Tugela River basin (South Africa) (Droppers et al., 2024). The model is implemented at four spatial resolutions (250 m, 500 m, 1 km, and 5 km) to evaluate the sensitivity of key water balance components—ET, SM, and discharge (Q)—to spatial discretization.
A set of mono-objective (5 km) and multi-objective (1 km) calibration experiments is conducted using Q, ET, and SSM as target variables, supported by both satellite products and ground observations. Model performance is assessed using complementary efficiency metrics (correlation, variability ratio, bias ratio, KGE, and SPAEF), enabling a detailed analysis of scale-dependent behavior and spatial pattern consistency.
The results reveal systematic trends in model performance across spatial resolutions, highlighting scale-dependent sensitivities of individual water balance components. According to the KGE metric, model performance is consistently higher at finer resolutions and progressively degrades toward coarser ones, a behavior observed across all experiments regardless of the calibration scale. Furthermore, the integration of satellite data with ground observations leads to improved model performance across scales, as reflected by higher KGE values and a more balanced contribution of the correlation, variability, and bias components.
Overall, this work contributes to the ongoing discussion on scale dependency in hydrology and directly relates to several open questions identified by Blöschl et al. (2019), particularly those addressing the consequences of spatial heterogeneity in hydrological fluxes, the existence of hydrological laws across catchment scales, and the effective use of innovative observation technologies to characterize hydrological states and fluxes across resolutions. By integrating satellite-derived state variables into a multiscale distributed modelling framework, this study establishes a methodological baseline for future research on calibration transferability, multiscale equifinality, and synthetic basin experimentation.
How to cite: Cortés-Torres, N., Güiza-Villa, N., Salazar-Galán, S., and Frances, F.: Evaluating multiscale performance of the TETIS distributed hydrological model using satellite and in situ observations in the Tugela basin, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12226, https://doi.org/10.5194/egusphere-egu26-12226, 2026.
Over the past few years, the increasing global population and climate change have intensified pressure on existing waterbodies, with many regions experiencing both water shortages and prolonged periods of water stress. As a result, the demand for reliable, detailed information on water availability and storage has grown rapidly. Despite this, many of the hydrological models used today either do not explicitly represent storage or operate at scales too coarse for practical water management.
A clear example of this can be seen in Sweden. Historically largely spared from water scarcity, the country has in recent years experienced recurrent shortages and increasing water stress, particularly in the southern regions. Traditional water storage assessments have relied on S-HYPE, the national adaptation of the widely used HYPE model (Hydrological Predictions for the Environment). Like HYPE, S-HYPE is a semi-distributed catchment model used for flood and drought forecasting, water quality assessment, and evaluating hydromorphological and climate change impacts. While S-HYPE can estimate total storage at the catchment scale, the current setup does not support assessments of individual waterbodies, severely limiting the model’s usefulness in providing in-depth information about local storage changes.
To address this, we explored a modified version of the Australian Water Resources Assessment Landscape model (AWRA-L). A case study was conducted on three lakes in the Lagan River catchment in southern Sweden to evaluate the model’s performance and applicability. Initial results showed generally good performance, with an average NSE of 0.68 and a KGE’ of 0.64. However, systematic differences between simulated and observed storage were noted. Preliminary analysis indicated that surface runoff is a major contributor to these residuals, while the influence of individual model parameters remains unclear. It is also uncertain whether the model fully captures all relevant processes under varying climatic conditions, particularly during cold periods.
This study aims to improve the model by combining physics-informed parameter optimization with detailed residual diagnostics. First, a randomized one-at-a-time sensitivity analysis was conducted to assess the overall contribution of the various input variables used to calculate the surface runoff. Parameter optimization was then performed using physics-informed rating curves constrained to physically plausible ranges, and optimized inputs were used to recalculate surface runoff. Model performance was evaluated against previous simulations, with residuals analyzed for systematic biases and potential missing processes using statistical and machine learning methods. Finally, temporal and seasonal patterns, autocorrelation, and correlations with auxiliary variables such as air temperature were analysed to identify model deficiencies and areas of improvement.
How to cite: Bjerkén, A., Ahmadi, K., and Klante, C.: Advancing water storage model development through physics-informed machine learning and residual diagnostics: a case study of the Lagan River catchment, southern Sweden, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17170, https://doi.org/10.5194/egusphere-egu26-17170, 2026.
Abstract: The water resources of the Nile Basin are under mounting pressures due to population growth, climate change, and growing transboundary tensions. These pressures intensify upstream–downstream trade-offs, making adaptive, data-driven planning essential to meet demands. wflow_sbm is a well-suited tool for understanding and modelling large-basin hydrological processes, yet it lacks an automated multi-objective calibration workflow. To overcome this limitation, we develop Optiverse, an AI-driven, multi-objective Python framework to calibrate wflow_sbm. Optiverse is designed as a modular, general-purpose package for multi-objective optimization of simulator workflows. Building on the Python package Platypus, Optiverse implements multi-objective evolutionary algorithms to search for Pareto-optimal solutions using NSGA-II and NSGA-III. This study presents a case study of calibrating a wflow_sbm model of the Blue Nile Basin for the period 1991-2020. The calibration was run on high-performance computing resources to meet the computational demands of iterative calibration, enabling reliable convergence to Pareto-optimal solutions. Early results indicate promising improvements across multiple calibration objectives for wflow_sbm, considering multiple calibration locations within the same optimization formulation. This framework provides a practical pathway for AI-enabled calibration of wflow_sbm and, for the Nile, provides a practical tool for decision support under increasing variability and risk.
Keywords: Distributed hydrology; wflow_sbm; optiverse; AI optimization; NSGA-II; calibration; evolutionary algorithms
How to cite: Alaminie, A. and Basheer, M.: AI-Enabled multi-Objective calibration of a wflow_sbm hydrological model of the Nile Basin, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12916, https://doi.org/10.5194/egusphere-egu26-12916, 2026.
Representing hydrological and hydraulic processes consistently across spatial scales remains a major challenge for large-scale flood modelling. Besides using simplified routing schemes that struggle to accurately represent in-channel river flow dynamics, most large-scale hydrological models adopt a Cartesian-grid discretization scheme due to their compatibility with widely available gridded datasets. However, this grid-based structure poorly captures river geometry and oversimplifies natural drainage boundaries, leading to scale-dependent biases in runoff production and streamflow simulations, commonly referred to as the “catchment size problem”. In contrast, hydrodynamic models implement more physically-based routing schemes with fine-scale geometric representation of river channels, but face important parametrization challenges at larger scales. To address these challenges, we introduce an integrated hydrological–hydrodynamic (H&H) modelling framework that enables a seamless coupling between coarse-resolution gridded hydrological modelling and fine-scale vector-based river routing, leveraging a sub-grid representation of the river network derived from high-resolution topography. Notably, sub-grid information is propagated into the hydrological model by replacing regular grid-cell areas with realistic drainage areas derived from sub-grid topography, thereby addressing the aforementioned “catchment size problem”. The integrated H&H framework is implemented within the SMASH modelling platform (Spatially distributed Modelling and ASsimilation for Hydrology, https://smash.recover.inrae.fr/). For this application, we coupled the grid-based conceptual hydrological model GR4 to a vector-based hydrodynamic model solving a simplification of the 1D shallow water equations without convective acceleration terms. We evaluated this framework over the Garonne River catchment (France, ~50,000 km²) using the MERIT digital elevation model (resampled at 100 m) and the reference national river network BD TOPAGE®. We conducted H&H simulations across three spatial resolutions: 1 km, 5 km, and 10 km, where we considered the 1 km configuration as a baseline and kept the same hydrological and hydrodynamic parametrization across resolutions (no recalibration): semi-distributed hydrological parameters, uniform channel friction, and simplified rectangular channel geometry where widths and depths are estimated from geomorphological relationships, and bathymetry is subsequently derived from geomorphological depths and elevation. Results show that the sub-grid river representation maintains a consistently high spatial accuracy across spatial scales, with mean separation distance from the reference hydrography of around 25 m and minimal omission of the mapped network (<6%). This geometric accuracy consistency is further supported by H&H simulations showing robust preservation of flow timing across scales. Furthermore, H&H simulations demonstrate improved consistency across spatial scales when leveraging sub-grid drainage areas, compared to the conventional grid-based delineation method that shows scale-dependent volume bias. This bias reflects the impact of drainage area misrepresentation as resolution is coarsened, which results in biased precipitation volumes and propagates into runoff production. Overall, these results highlight the potential of the proposed integrated H&H framework to enable scalable hydrological–hydrodynamic modeling at large scales and provide a flexible foundation for leveraging increasingly available multi-source water surface observations, such as satellite altimetry (e.g., SWOT, ICESat-2), to infer key H&H model parameters and enhance modeling accuracy in data-sparse regions.
How to cite: Berkaoui, M. A., Saadi, M., Colleoni, F., Huynh, N. N. T., Akhtari, A., Larnier, K., Garambois, P.-A., and Roux, H.: Towards Consistent Multi-Scale Flood Modelling Using an Integrated Hydrological–Hydrodynamic Framework, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9617, https://doi.org/10.5194/egusphere-egu26-9617, 2026.
Climate-change impact assessments using agro-hydrological models often assume that farmers stick to their historical planting and harvest calendars under future climate conditions. This fixed-calendar assumption conflicts with observed and expected shifts in crop phenology and management, leading to errors in simulated crop timing, yields, and water use. Here, we develop and test a generic workflow to replace static calendars with dynamic, state-triggered crop management in SWAT+ using decision tables. Focusing on winter wheat and corn silage in a rainfed catchment in eastern Germany, we first run a conventional, date-based SWAT+ setup and analyze the management log to determine the heat units and weather conditions under which farmers actually plant, harvest, till, and fertilize. Empirical distributions of base-0 potential heat units, days since planting/harvest, and recent precipitation are converted into compact decision-table guards. These guards, mainly expressed as PHU windows and dry-day constraints, are tuned so that dynamic management reproduces historical planting and harvest dates within about one week, while maintaining baseline yield performance. Next, we run the decision-table model with three contrasting late-century EURO-CORDEX climates (cool–dry, cool–wet, warm–wet), without changing crop parameters or management intensities. Under cool scenarios, winter wheat maintains a long growing season of approximately 313–314 days, but under a warm–wet climate, later autumn planting is followed by an earlier midsummer harvest, shortening the season to roughly 287–290 days. Corn silage shows marked advances in planting (up to about three weeks earlier in the warm–wet scenario) while harvest remains anchored near early September, lengthening the growing period from ≈150 to ≈170 days across scenarios. Yields for corn silage stay high and fairly stable, whereas winter wheat yields show modest scenario-dependent changes in the average and more variation between years. The study demonstrates that SWAT+ decision tables can encode historically accurate, climate-responsive management directly from observed practices and apply it consistently under future climates. The proposed workflow provides a transferable model for representing adaptive cropping calendars, reducing structural bias, and enhancing the credibility of climate-change impact studies on crops and agricultural water management.
How to cite: Eini, M., Müller, B., and Strauch, M.: From Fixed Calendars to Dynamic Triggers: Climate-Responsive Field Management with SWAT+ Decision Tables, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-252, https://doi.org/10.5194/egusphere-egu26-252, 2026.
Vegetation dynamics play a critical role in canopy interception and transpiration, yet its representation in hydrological models is often simplified or even entirely omitted. Rising air temperatures have been shown to shift the timing and extend the duration of the vegetation period, directly affecting evapotranspiration. Incorporating vegetation dynamics into hydrological models is therefore essential, particularly in studies assessing the impacts of climate change. In this study, we employ satellite‑derived Normalized Difference Vegetation Index (NDVI) data to parameterize vegetation processes within a distributed HBV-type rainfall–runoff model. For each land cover class in the Upper Danube basin, NDVI regimes over a 25‑year period were derived by averaging values from 1,000 randomly selected points. Seasonal vegetation dynamics were then characterized by fitting trapezoid functions to the annual NDVI regimes, yielding estimates of the onset and end of the growing season.
The analysis of vegetation characteristics revealed that certain land cover classes (particularly deciduous forest, agricultural land and pastures) exhibit notable changes including increases in mean annual NDVI values and earlier onset of the growing season. Moreover, the timing of the active growing season was found to correlate with air temperature indices, such as the number of days above or below certain thresholds. These relationships were used to calibrate temperature thresholds and consecutive day counts to estimate the start and end of the vegetation period. The methodology was implemented in the Upper Danube basin as a case study, providing a foundation for further evaluation of its impact on hydrological simulations. By explicitly linking vegetation dynamics to temperature indices, the approach enables hydrological models to operate independently of direct NDVI observations, which are unavailable in climate change impact studies, while also accounting for elevation effects, as cooler temperatures at higher altitudes naturally delay vegetation onset.
How to cite: Valent, P. and Parajka, J.: Linking NDVI-derived vegetation dynamics with air temperature to model interception and transpiration processes of a conceptual hydrological model, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1612, https://doi.org/10.5194/egusphere-egu26-1612, 2026.
Accurate representation of vegetation dynamics is critical for hydrological modeling and climate change impact assessments. Leaf Area Index (LAI) influences ecohydrological processes, including evapotranspiration, interception, and soil moisture. The conventional Soil and Water Assessment Tool Carbon (SWAT-C) model has a simplified vegetation growth module, which limits the accuracy of the model in decision making. The present study evaluates how improved simulation of LAI affects ecohydrological responses of a watershed under historical and future climate scenarios. The study employed setting up of two models: original SWAT-C as a baseline model, and a modified version of the SWAT-C model with an improved plant growth module to simulate LAI more realistically in forested areas. Both models were calibrated using streamflow, evapotranspiration, and sediment yield using A Multi Algorithm Genetically Adaptive Multiobjective (AMALGAM) optimizer. The climate projections from different bias corrected global circulation models were applied to understand the sensitivity of ecohydrological simulations of streamflow, evapotranspiration, sediment yield, net primary productivity (NPP), biomass, net ecosystem exchange (NEE), soil organic carbon (SOC), and lateral carbon fluxes to vegetation-driven changes in LAI. The differences in the future climate scenarios highlights the influence of vegetation feedbacks on projected hydrological responses and carbon dynamics. This approach provides better insight into vegetation–water–carbon interactions and can support improved strategies for managing water and ecosystem resources under changing climate conditions.
How to cite: Vema, V. K. and Aashi, A.: Influence of Vegetation Dynamics on Hydrological and Carbon Responses under Future Climate Scenarios Using SWAT-C, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3703, https://doi.org/10.5194/egusphere-egu26-3703, 2026.
Traditional raster-based distributed hydrological models often face challenges in representing the geometric complexity of landscape features, leading to fragmented boundaries and oversimplified topological relationships. To address these limitations, this study proposes a novel Integrated Vector-Raster Hydrological Model (VeRHyM). By employing a flexible spatial discretization scheme, the VeRHyM utilizes vector polygons to represent irregular land features (e.g., glaciers, reservoirs, agricultural fields) and vector polylines for river networks, while maintaining raster data for continuous field variables (e.g., precipitation, topography) and describing internal heterogeneity within individual land features.
Enabled by its hybrid data structure, the proposed model offers several key advantages over conventional grid-based approaches: (1) Geometric Integrity: It preserves the precise boundaries of land features, preventing the fragmentation of physical objects into disjointed pixels; (2) Topological Accuracy: It provides a more rigorous description of river networks, water conveyance structures, and the spatial connectivity between different land features; and (3) Multi-scale Coupling: It facilitates the seamless coupling of hydrological processes across varying spatial scales, from individual glaciers to the entire watershed. This also enables precise coupling between distinct physical models (e.g., glacier runoff and crop water stress) at their native spatial scales.
We applied the VeRHyM to the Urumqi River Basin in Tianshan, China, a typical complex watershed characterized by diverse landscapes ranging from high-altitude glaciers and alpine vegetation to arid piedmont zones containing oases, cropland, and urban settlements. The model performance was rigorously validated using multi-source data: River discharge was calibrated against observations from the Tianshan No. 1 Glacier station and the mountain outlet hydrological station; simulated evapotranspiration was compared with remote sensing products; and human water consumption estimates were verified against regional statistical records. Results demonstrate that the VeRHyM captures the spatiotemporal variability of the water cycle effectively in this complex terrain. The successful application suggests that the vector-raster integration strategy significantly improves the representation of heterogeneous landscapes and provides a robust tool for integrated water resources management in arid regions.
How to cite: Liu, X.: A Novel Integrated Vector-Raster Model for Multi-Process Hydrological Simulation in Heterogeneous Landscapes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6697, https://doi.org/10.5194/egusphere-egu26-6697, 2026.
The GeoLaB project (https://geolab.helmholtz.de/en/) aims to establish an underground laboratory in the crystalline rocks of the Odenwald, Germany, to conduct controlled experiments on thermal-hydraulic-mechanical-(bio)chemical processes relevant to the development and operation of enhanced geothermal systems. The facility will provide access to representative reservoir rocks of the Upper Rhine Graben, specifically the Tromm granite located along its northeastern margin.
One of the key questions to be addressed during the exploratory phase of GeoLaB concerns the occurrence and flow rates of groundwater in the area. The geological setting is complex and includes plutonic, metamorphic, and sedimentary rocks, all affected by different degrees of fracturing and intersected by regional faults. These units, particularly the hard rocks, are difficult to investigate directly, which results in a general lack of subsurface information in the study area. In addition, the alluvial and colluvial deposits besides the soils developed under grasslands and forests are expected to exert significant control on groundwater flow, together with the weathered zones and associated saprolites. This combination of factors makes the hydrogeological assessment particularly challenging.
To numerically explore the influence of the different hydrogeological units on groundwater dynamics, we employ integrated hydrologic modelling using ParFlow–CLM. The focus is placed on testing alternative assumptions regarding the hydraulic properties and geometries of the shallow deposits and weathered horizons, as well as different hydraulic conductivities for the underlying hard rock units, all of which are expected to exert strong control on groundwater flow. Although groundwater is the central objective, the use of ParFlow–CLM also provides insights into surface water resources and the complete water balance. As a first step, we simulate a suite of steady state models to assess groundwater table depths under different conceptualizations. Subsequently, we run transient simulations for the most plausible scenarios and compare the results with available observations of discharge, groundwater level and soil moisture profiles. Preliminary results highlight the sensitivity of groundwater table depths to the conceptualization of the shallow and weathered units and illustrate the potential of this approach to constrain hydrogeologic conditions in the GeoLaB area.
Overall, this work provides a first hydrogeological assessment for the GeoLaB site and outlines a modelling framework that can be progressively refined to support future explorations and the design of the laboratory.
How to cite: Muñoz-Vega, E., Horovitz, M., and Schulz, S.: Investigating hydrogeological controls at the GeoLaB site using ParFlow-CLM , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-13038, https://doi.org/10.5194/egusphere-egu26-13038, 2026.
Farm dams are a controversial solution for ensuring the supply of water for summer agricultural crops. Their potential impact on catchment hydrological balance and river ecology is still an open question. In France, research work is mostly focused on small catchments, where it is possible to collect information. However, it is necessary to study the cumulative impact of farm dams at a larger scale and in contrasting (geographical, agricultural, geological, and climatic) contexts to guide the development of national regulations. To this end, large-scale distributed hydrological modelling integrating representations of farm dams and agricultural uses of water is a solution. This requires spatialised data on agricultural practices (e.g. types of crops, irrigated areas), farm dams (e.g. their locations, sizes, connections to the river network), and their uses (e.g. agricultural, industrial, recreational). These data are not always available, and if they exist, additional challenges are posed by the partial nature of the data, lack of documentation, and varying or overly coarse resolutions. The work undertaken here proposes a methodology to exploit these imperfect databases and to optimise the representation of the territories heterogeneity for hydrological modelling. We collected recent public data (last 20 years) for two major French river basins, the Rhône and the Loire (~100,000 km² each). Agricultural statistics (crops and irrigation) from national census, have an overly coarse spatial resolution and therefore need to be spatially redistributed. National inventory of water bodies (locations, surface) built using satellite imagery, contains limited data on the volumes and uses of water bodies. Annual water abstractions database (locations, volumes, uses, water origins), where the use of farm dams are heterogeneously documented. The first step reconstructs crop and irrigation surfaces at a fine spatial resolution over the domain of study in order to allocate variables in the model units, using a dedicated optimisation algorithms. The second step synthetises and allocates an equivalent farm dam to the model units, crossing informations from the national inventory embedded by local databases. The third step connects irrigated areas to water abstraction origin (farm dams, rivers or groundwater) using statistics rebuilt for both catchment. The spatial reconstruction and allocation of irrigation (i.e. farm dams) is validated by comparing the model units with the original data. In addition, the integration of farm dams and agriculture water use is validated by exploring the hydrological variables simulated with the J2000 hydrological model, in comparison with abstraction volumes. This modelling approach will enable the assessment of the impact of farm dams and their uses on catchment hydrology. It will also evaluate the capacity of farm dams to meet irrigation demands at different spatial scales, while accounting for the uncertainties associated with the imperfect nature of the databases.
How to cite: Pellerin, N., Brown, N., Mimeau, L., Vidal, J.-P., and Branger, F.: Dealing with imperfect data to integrate farm dams and agricultural water uses in hydrological modelling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12060, https://doi.org/10.5194/egusphere-egu26-12060, 2026.
Climate change is intensifying hydrological extremes, making droughts and floods more frequent and severe in agricultural landscapes. These changes pose growing challenges for water availability, crop productivity and sustainable land management. Effective adaptation requires a better understanding of how hydrological fluxes and storages respond to changing climatic conditions and human interventions. Interactions between soil moisture, groundwater, surface water and the atmosphere play a key role in determining water availability and system resilience in the context of increasing extreme events. However, many commonly used hydrological models simplify these interactions, which limits their ability to adequately assess the effects of water management measures under climate extremes.
This research aims to investigate how agricultural water management measures influence hydrological fluxes and storages by integrating physically based hydrological modelling with observational data. The focus is on agriculture-dominated catchments where management interventions such as controlled drainage, adjustable weirs and retention measures are applied. A spatially distributed, integrated hydrological model will be applied to simulate coupled surface and subsurface processes, including soil moisture dynamics, groundwater fluctuations, evapotranspiration and surface water flow, enabling a holistic assessment of hydrological system behaviour under climate extremes.
Model calibration and validation will be conducted using multi-source observational data, including in situ measurements of soil moisture at multiple depths, groundwater levels, surface water observations and satellite-derived products. This approach allows the evaluation of the model’s ability to reproduce internal hydrological states in addition to discharge. The validated model will then be used to assess the impacts of different water management strategies under variable climate conditions. The analysis initially focuses on one or two small agricultural catchments with dense observational coverage and in a later stage the modelling framework can be upscaled to larger catchments to explore the implications of water management strategies beyond the local scale.
By improving the representation and evaluation of subsurface–surface–atmosphere interactions, this work aims to support the development of more robust and resilient agricultural water management strategies under a changing climate.
How to cite: Signora, F., Tügel, F., Booij, M., van der Tol, C., and Rientjes, T.: Assessing effects of adapted agricultural water management on hydrological processes using integrated hydrological modelling and field observations, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14034, https://doi.org/10.5194/egusphere-egu26-14034, 2026.
The struggle to reliably close the water balance at the catchment scale has been a major issue for conceptual hydrological models. Inter-catchment groundwater flows (IGF), human influences, forcing biases and streamflow uncertainties all unite to provide high-end puzzles even to the most enduring hydrologist. In the absence of a better assessment of each of these non-closure sources, many mesoscale conceptual models either choose to accept the mismatch between observed and simulated streamflow, or to force the water balance closure through calibration. Here, we argue that, when working at the mesoscale, ridding ourselves of the actual non-closure complexity through calibration can (or should) be avoided: rather, we propose to grapple with water balance anomalies by explicitly addressing all of their potential causes, ahead of any subsequent parameter estimation.
Specifically, we take advantage of the recent proliferation of natural streamflow datasets, enriched with a number or regional and national contributions to evaluate water-balance anomalies as computed with the CERRA-Land climate reanalyses over the period 1984-2024. Thereon, we develop a methodology based on the spatial analysis of the distribution of catchments’ distances to a Budyko-type curve. Our belief is that the spatial coherence of this water balance anomaly signal can be used to disentangle the different causes of the water balance non-closure and, in particular, to discriminate between those which have local determinants (such as groundwater flows) and those involving regional factors (such as forcing biases). Finally, we further break down the contribution of the forcing biases based on a priori knowledge about precipitation (P) and potential evaporation (E0) biases. We present our results under the form of a pan-European map displaying all catchment-specific non-closure sources, with a three-way scale (e.g. ternary plot) measuring the respective weights of P, E0 and IGF.
How to cite: Rosales, M., Sarrazin, F., Hendrickx, F., and Andréassian, V.: On attempting to close the water balance when all fluxes are questionable: what can we learn from the spatial coherence of the water balance anomaly signal ?, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9813, https://doi.org/10.5194/egusphere-egu26-9813, 2026.
Specific Catchment Area (SCA) and Total Catchment Area (TCA) are two widely used topographic attributes in the study of hydrological, geomorphological and biological processes at the watershed scale. They are typically estimated starting from a Digital Terrain Model using Flow Direction (FD) algorithms. In the ideal case of constant and uniform rainfall excess, SCA and TCA are directly proportional to the steady-state specific discharge and discharge of surface runoff, respectively. This study investigates an alternative approach that computes SCA and TCA fields using a rain-on-grid Shallow Water Equations (SWE) model (PARFLOOD-Rain). Given a DTM representing a terrain, a constant and spatially uniform rainfall rate is applied, and the simulation is run until a steady-state regime is reached everywhere. At each pixel, the ratio between the steady-state discharge and the imposed rainfall rate yields the TCA value, while SCA is obtained by dividing the steady-state specific discharge by the rainfall rate. In the first part of the study, SCA and TCA fields generated by PARFLOOD-Rain are compared against outputs from six commonly used FD algorithms (namely D8, Rho8, D-infinity, MFD-Quinn, MFD-md and FD8) and from the recent IDS algorithm proposed by Prescott et al. (2025). All outputs are validated against analytical solutions on four synthetic surfaces (inclined plane, saddle, convergent and divergent surfaces). All the methods, including PARFLOOD-Rain, are further validated – on another synthetic surface – against a numerical solution of the differential equation proposed by Gallant & Hutchinson (2011), which defines SCA along a flow line. On all test surfaces, PARFLOOD-Rain predicts SCA and TCA with errors one to two orders of magnitude smaller than those of the FD methods, and its accuracy improves with grid refinement – unlike the FD algorithms – which show no such convergence behavior. In the second part of the study, PARFLOOD-Rain is used to estimate SCA and TCA in a real catchment, and its results are used as a reference solution to validate the FD algorithms in a complex, channelized terrain. A small catchment downstream of Blanca Peak, Colorado (USA) is selected as a case study. The analysis highlights that, while SCA and TCA’s sensitivity to the rainfall intensity is negligible on smooth synthetic surfaces, it is a major controlling factor in irregular, natural terrains featuring channels and carvings. Whereas it is trivial that FD methods could never fully describe the complex hydrodynamics captured by SWE-based approaches, the results of the study suggest that more sophisticated FD algorithms, like IDS, can offer potential advantages over traditional FD methods in the prediction of TCA and SCA.
How to cite: Carta, S., Prost, F., Aureli, F., and Mignosa, P.: Validation of Specific and Total Catchment Area estimated via Flow Direction Algorithms through a 2D Shallow Water Equations Numerical Solver, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18718, https://doi.org/10.5194/egusphere-egu26-18718, 2026.
The spatial scale and delineation of Representative Elementary Watersheds (REWs) are fundamental determinants of fidelity in physically based hydrological modeling. The REW framework facilitates a semi-distributed representation of dominant processes—such as surface runoff, subsurface flow, and channel routing—by averaging conservation equations of mass and momentum over discretized sub-units. Leveraging the Tsinghua Hydrological REW model (THREW), this study investigates the sensitivity of hydrological simulations to REW spatial scale across a diverse set of over 80 catchments. By systematically varying drainage area thresholds for REW delineation, we observed distinct scale-dependent behaviors: for larger basins, higher spatial resolution generally enhances model accuracy with relatively low sensitivity to the specific delineation threshold. Conversely, in smaller catchments, excessive discretization often degrades performance and exhibits heightened sensitivity to threshold selection. In the context of daily time-step simulations, we found that for smaller catchments, the detriments of increased data noise and parameter uncertainty often outweigh the marginal gains derived from resolving spatial heterogeneity. In contrast, the explicit characterization of this heterogeneity is critical for optimizing model performance in larger basins.
How to cite: Xu, B. and Tong, R.: Assessing the impact of the spatial scale of Representative Elementary Watershed (REW) delineation for hydrological modeling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9215, https://doi.org/10.5194/egusphere-egu26-9215, 2026.
The groundwater system is dynamic in time, in particular in high productivity aquifers such as the heterogeneous and fractured Chalk aquifer. Unlike river catchments, which are generally topographically controlled and therefore stable, groundwater catchments vary seasonally. The location and magnitude of intercatchment groundwater flow (IGF) can therefore also vary seasonally. This can pose a significant challenge for hydrological conceptual models. Building on work previously conducted by the authors on perceptualising the spatial variability of IGF and incorporating this into the DECIPHeR conceptual rainfall-runoff model, we have followed the same data-led approach in an investigation of the temporal variability of IGF. An evidence-based perceptual model of the River Kennet, UK (a tributary of the River Thames) was first developed and then used to inform the design of model edits that capture seasonal IGF in-line with the perceptual understanding. From review of recorded data, it was found that a strong sinusoidal climate signal propagates through to the groundwater table, river flow and catchment water balance annual profiles, but that this signal is highly variable between years. The temporal variability observed in the test sub-catchments was applied to an IGF flux within the DECIPHeR model, and the results compared to both the baseline model and the spatial IGF model developed in previous work. Four model structures were developed and tested, show-casing an increasing level of hydrogeological information and seasonal analysis. Model calibration at the annual scale was no better than the spatially-variable IGF model, but there was a marked improvement in the representation of the monthly flow profile when an additional IGF flux sinusoidal amplitude parameter was introduced. The timings of the autumnal increase in river flow, plus the slow spring recession, are now able to be replicated. The findings prompted a discussion on the challenges that remain when representing intercatchment groundwater flow in conceptual hydrological models. Most notably, these include a lack of representation of long-term catchment storage limiting a model’s ability to replicate the inter-annually variable groundwater catchment areas that are so characteristic of Chalk catchments.
How to cite: Oldham, L., Coxon, G., Howden, N., and Jackson, C.: From perceptualisation to modelling: Improving the representation of temporally variable intercatchment groundwater flow in hydrological models, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6570, https://doi.org/10.5194/egusphere-egu26-6570, 2026.
Climate warming is rapidly modifying glacier and snowmelt processes, with impacts that propagate across spatial and temporal scales, from high-elevation cryospheric dynamics to basin-scale streamflow. Capturing these cross-scale interactions requires hydrological models that explicitly represent cryospheric processes, such as glacier dynamics and snow redistribution, while also accounting for uncertainty arising from data scarcity and model structure. To address these challenges, we present a one-way coupling that integrates the Global Glacier Evolution Model (GloGEM) with Raven, a flexible hydrological modeling framework designed to emulate multiple model structures for systematic uncertainty assessment. Glacier simulations are first improved by incorporating snowline observations as an additional calibration constraint to the geodetic mass balance, reducing parameter equifinality and enhancing glacier runoff estimates. The resulting glacier runoff is then provided to the hydrological model, leading to a more robust representation of snowmelt processes and their contribution to streamflow. By evaluating multiple hydrological model structures within Raven, we further quantify uncertainties in simulated melt contributions arising from structural model choices. The framework is applied to multiple data-scarce Indus headwaters as representative high-mountain catchments, simulating melt contributions and streamflow across scales. By explicitly coupling cryosphere and hydrological models, this framework aims to improve projections of future water availability in data-scarce regions and provides a transferable approach for integrated cryosphere-hydrology modeling in complex mountain regions.
How to cite: Berg, J., Horton, P., Kauzlaric, M., von der Esch, A., and Schaefli, B.: Coupling a Global Glacier Model and Hydrological Model: Constraining Cryospheric Contributions in the Data-Scarce Indus Basin, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16142, https://doi.org/10.5194/egusphere-egu26-16142, 2026.
Analysing parameter sensitivity and identifiability are important steps in hydrological modelling as they help to detect the most relevant parameters and suitable parameter ranges. Within such studies we can categorize parameters is diverse ways – one of them being capacity vs. flux parameters. While capacity parameters regulate the magnitude or storage of hydrological components, flux parameters determine the timing of water flow. We demonstrate in two steps that separating capacity and flux parameters is beneficial for model analyses.
First, we conducted a temporally resolved parameter sensitivity analysis, targeting the rate of change of eight hydrological components in addition to the absolute time series. Using the rate of change as target variable improved the representation of dynamic flux parameters, while capacity parameters were more precisely represented by the absolute time series of the hydrological components.
In a second step, we conducted a parameter identifiability analysis across six contrasting German catchments using sixteen diverse performance metrics and hydrological signatures. Our analysis with four hydrological models (Raven-GR4J, HBV, SWAT+ and mHM) reveals that capacity parameters can be precisely identified using several performance criteria, in particular those related to mid and low flows. In contrast, accurate identification of flux parameters requires specific performance criteria such as hydrological signatures related to the process timing.
Separating parameters into capacity and flux parameters improves the detection of the sensitivity signal and enables a more precise identification of parameter values. The reduced uncertainty in estimating the dominant parameters is a valuable step towards efficient model calibration.
How to cite: Guse, B., Herzog, A., Houska, T., Spieler, D., Staudinger, M., Wagner, P., Pool, S., Loritz, R., Kiesel, J., Pfannerstill, M., Duethmann, D., Wagener, T., Gupta, H., and Fohrer, N.: Capacity and flux model parameters should be addressed separately in parameter sensitivity and identifiability analyses, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16951, https://doi.org/10.5194/egusphere-egu26-16951, 2026.
The Horizon Europe project FOCCUS (https://foccus-project.eu) aims to enhance Copernicus Marine Service's coastal dimension by developing innovative products as well as facilitating seamless ocean monitoring and forecasting. The scope of the project includes improvement of the estimation of water and matter runoff to the European coastal regions using large-scale hydrological models. Effort is directed on improving the pan-European HYPE (E-HYPE) hydrological model (see 1) both through a traditional calibration and a hybrid modelling approach through an AI-based enhancement.
The calibration framework was revised and updated, including changes in the application of common criteria (NSE, KGE, RE, R²). The previous E-HYPE calibration focused on improving domain-average model performance. Instead of relying on domain-average model performance, the proportion of calibration stations meeting acceptable performance thresholds was used to increase the number of well-calibrated stations. This approach reduces the influence of stations with highly unreliable data that may otherwise bias criteria-based parameter selection. The model validity was also assessed for key physical processes including snow accumulation, reservoir siltation, and sedimentation-resuspension dynamics. The final step involved manual inspection of time series and performance distributions for streamflow, nutrient and sediment concentrations, as well as snow water dynamics. Validation was conducted at the spatial extent across gauged catchments (not used for calibration), and at major coastal outlets. The updated E-HYPE model shows improved overall performance compared to its previous benchmark version, especially in streamflow, sediment concentration and evapotranspiration.
Finally, a hybrid modelling approach was applied, which included an AI-based post-processing to improve the streamflow predictions at coastal outlets (see 2). This effort involves transferring the knowledge learned from the upstream gauged locations, providing improved predictive performance at ungauged locations (not included in the training stations). Overall, the final dataset includes daily streamflow, sediment and nutrient concentration at 5,302 European coastal outlets for the period 2000-2024 and will be soon publicly available on Zenodo.
References:
1. Brendel, C., Capell, R., & Bartosova, A. (2023). To tame a land: Limiting factors in model performance for the multi-objective calibration of a pan-European, semi-distributed hydrological model for discharge and sediments. Journal of Hydrology: Regional Studies, 50, 101544.
2. Du, Y., & Pechlivanidis, I. G. (2025). Hybrid approaches enhance hydrological model usability for local streamflow prediction. Communications Earth & Environment, 6(1), 334.
How to cite: Terskii, P., Du, Y., Brendel, C., Pechlivanidis, I., and Bartosova, A.: From continent to coast: advances in HYPE hydrological model calibration and AI-based model enhancement, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10607, https://doi.org/10.5194/egusphere-egu26-10607, 2026.
An upscaling approach of 3D hydrological processes for Land surface models (LSM) has recently been developed and tested on the Little Washita experimental watershed (Ok., USA). We show here how the resulting upscaled model, which depends on two hydrologic variables, may be encapsulated into a vertical soil column model. This upscaling approach allows to establish the relationships between the physical and geometrical parameters of a watershed and the empirical parameters of the LSM ORCHIDEE. A comparison of both models on the Little Washita is discussed. Then, we test this new model on five watersheds located in Oklahoma (USA) and picked from the CAMELS database. All the geometrical and physical parameters come also from the database except the van Genuchten infiltration parameters obtained by a calibration algorithm. The five watersheds cover a wide range of areas and hydroclimatic conditions of the south great plain region. The results show that the monthly inter-annual means and annual means KGE values are comparable to those obtained with SACMA which is the reference model of CAMELS. As a conclusion, it is worth to emphasize that the quality of our results depends essentially on the calibration of the infiltration parameters. This work will be extended to other watersheds of the CAMELS database.
How to cite: Mouche, E. and Picourlat, F.: Application of an upscaled hydrological model to five Oklahoma watersheds of the CAMELS database., EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7610, https://doi.org/10.5194/egusphere-egu26-7610, 2026.
Climate change is one of the anthropogenic factors inducing alterations in hydrological processes, known as hydrological changes. To assess these changes at catchment scale and evaluate their impacts on the hydrology-related hazards, spatially-distributed and physically-based hydrological models represent reliable and suitable tools. However, significant challenges remain in achieving robust parameterization and calibration, mainly due to data limitations, spatial and temporal scale mismatches, and the intrinsic complexity of hydrological processes.
In this study we adopt the Triangulated Irregular Network-based real-time integrated basin simulator (tRIBS) hydrological model to build a proper laboratory tool enabling a comprehensive high-resolution distributed analysis on the response to scenarios of changes of multiple hydrological aspects, such as runoff components partitioning, evapotranspiration fluxes andsoil moisture dynamics. The case study is the Cedarchis basin, a part of the municipality of Arta Terme in the Friuli-Venezia Giulia (FVG) region, located in the north-eastern Italy. The basin covers 125 km2, as part of the main Tagliamento river basin, and the Chiarsò stream flows through it. Stream flow is monitored through an ultrasonic stream gauge installed at the outlet and managed by the Civil Protection Department of the FVG region. Additional stream measurements are available along the Chiarsò stream due to a small private hydropower plant. Rainfall data are available for the Paularo station, located approximately at the center of the basin. The calibration was conducted using data from the 2023-2024 period, which includes the proper spin-up time. To update and enhance the rating curve, required to transform hydrometric readings into discharge, we carried out multiple field measurements of stream flow managed by the water resources management service (Servizio Gestione Risorse Idriche) of the region. This time series was subjected to procedures of corrections, such as filling, extrapolation and filtering, with two moving windows, to reduce instrumental noise. The Nash-Sutcliffe efficiency coefficient (NSE) was used to perform the calibration and validation. Finally, to evaluate the response of the basin to climate change, synthetic precipitation series generated using an advanced weather generator for the 2050 and 2100 horizons under the RCP 4.5 and RCP 8.5 scenarios are used.
This research received funding from European Union NextGenerationEU – National Recovery and Resilience Plan (PNRR), Mission 4, Component 2, Investiment 1.1 -PRIN 2022 – 2022ZC2522 - CUP G53D23001400006.
How to cite: Flaugnacco, D., Bassi, A., Thomas, J., and Arnone, E.: Exploring hydrological changes across a mountain basin in Friuli-Venezia Giulia through a distributed hydrological model, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9884, https://doi.org/10.5194/egusphere-egu26-9884, 2026.
To understand water scarcity, it is vital to have reliable water demand data at suitable spatial and temporal resolutions. While more hydrological models are including human influences, and at increasingly fine resolutions, this improvement in process representation is not matched by improved data for driving or validating the models. Water demand data is normally very difficult to access and, when available, is usually at a coarse spatial resolution (often a country level). Downscaling methods for irrigation demand are well developed but domestic and industrial demands are generally naively downscaled using population as a proxy.
This work explores the potential for Machine Learning models in spatial downscaling of industrial demands at a range of resolutions. Various ensemble-tree type models are presented, trained on a recently published high-resolution water abstraction dataset from England, and using easily accessible spatial datasets as explanatory variables. The results are compared to a population-proxy downscaling and demonstrate minor improvements but without achieving the desired level of skill for application in water resource assessments.
Further avenues for exploration are proposed, with the aim of achieving a transferable downscaling method for water demands which can be trained in data-rich regions and applied to data-scarce areas. A successful approach would enhance water resource modelling through improved driving data, and improve our understanding of water scarcity, to support decision making in water allocation under increasingly water-scarce conditions.
How to cite: Baron, H., Kulranjan, R., Barr, A., Christie, M., and Rickards, N.: Novel methods to spatially downscale water demands, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19167, https://doi.org/10.5194/egusphere-egu26-19167, 2026.
Between 2023 and 2024, Emilia-Romagna (Italy) was affected by highly anomalous, persistent, and spatially homogeneous frontal precipitation events that triggered disastrous floods. Analysis of small and very small Apennine catchments shows that maximum hourly areal rainfall was significant but never extreme, whereas aggregated rainfall over 6–48 hours reached exceptional values, with return periods exceeding 200 years based on records dating back to the early 1900s. The resulting flood discharges were extremely intense, challenging the traditional design storm approach: synthetic storms with durations equal to the hydrological response time provided very limited guidance on the frequency of the resulting flood, even for catchments with relatively low permeability. Continuous simulation, instead, demonstrated strong potential for flood estimation.
A parsimonious lumped hourly rainfall–runoff model, GR5H, was calibrated under conditions of scarce discharge data using hydrological signatures—Flow Duration Curve and cumulative probability of annual maxima (both Gumbel and GEV distribution were considered, conducting two parallel analysis)—rather than conventional performance metrics. Over 30-year long simulated hourly series were validated through temporal validation and further analysed through log-normal frequency fitting of AMS. Despite limited observations and complex antecedent conditions, the approach provided plausible estimates of extreme flood peaks, highlighting its effectiveness for small, data-scarce basins and its relevance for improving flood risk assessment under evolving hydro-climatic patterns.
How to cite: Smerilli, G. and Castellarin, A.: Questioning the Design Storm Approach: Empirical Evidence for Small Basins from Emilia-Romagna’s Recent Catastrophic Floods, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1248, https://doi.org/10.5194/egusphere-egu26-1248, 2026.
Machine-learning models, particularly Long Short-Term Memory (LSTM) networks, often outperform process-based rainfall–runoff models, yet the specific process limitations driving this performance gap remain underexplored. We hypothesize that a major contributor is the use of simplified channel routing formulations that insufficiently represent temporal variability in flow velocity. Here, we couple the HBV rainfall–runoff model with the recently proposed Iterative Routing Model (IRM), a parsimonious and non-linear channel routing framework that explicitly allows flow velocity to vary with discharge. We evaluate the coupled HBV–IRM model over 64 CAMELS catchments across the United States. The hybrid model attains a median NSE of 0.72, improving on the original HBV (0.65) and approaching the performance of global LSTM benchmarks (0.74). The results indicate that improving process representation in channel routing can substantially reduce the performance gap between process-based and data-driven models, while retaining process understanding and physical interpretability.
How to cite: Sarkar, E., Kadu, A., Katari, V., and Biswal, B.: Do Process-Based Models Really Fall Short? Rethinking Channel Routing to Bridge the Gap with Machine Learning Models, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-2852, https://doi.org/10.5194/egusphere-egu26-2852, 2026.
Posters virtual: Wed, 6 May, 14:00–18:00 | vPoster spot A
EGU26-9647 | Posters virtual | VPS9
Signature-Based Evaluation of Hydrological Processes Using the SWAT Model in the Bharathapuzha River Basin, IndiaWed, 06 May, 14:21–14:24 (CEST) vPoster spot A
Understanding catchment behaviour is essential for effective water resources planning and sustainable watershed management. Traditional model evaluation approaches based solely on time-series performance metrics often fail to capture the full spectrum of hydrological functioning. This study employs a hydrological signature–based evaluation framework to assess the capability of the Soil and Water Assessment Tool (SWAT) model in reproducing the dominant hydrological processes in the Bharathapuzha River Basin, a monsoon-dominated river system in southern India. The SWAT model was implemented using spatial datasets of topography, land use, and soil characteristics, together with long-term hydro-meteorological inputs, and calibrated and validated against observed daily streamflow. Beyond conventional performance indices, key hydrological signatures including flow duration curves, runoff ratio, baseflow index, seasonal flow patterns, and characteristics of low- and high-flow events were extracted from both observed and simulated datasets. Comparison of observed and simulated signatures provided a process-oriented evaluation of model behaviour, offering key insights into how well runoff generation, seasonal variability, and hydrological extremes are represented. These perspectives are not readily evident from traditional model performance metrics alone. This study demonstrates the value of hydrological signatures as diagnostic tools for enhancing model realism and improving confidence in hydrological simulations for climate impact assessment and water resources management in monsoon-driven catchments.
How to cite: Krishnan Sreelatha, G., Anupama Rajith, A., Sreekumar, A., Girish, G., and Reghunath, G.: Signature-Based Evaluation of Hydrological Processes Using the SWAT Model in the Bharathapuzha River Basin, India, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9647, https://doi.org/10.5194/egusphere-egu26-9647, 2026.
