This session aims to explore advances in the study and modeling of hydroclimatic extremes, embracing a broad perspective that includes—but is not limited to—their spatial and temporal characteristics. Key challenges include the definition of multivariate and compound events; the quantification of uncertainties, of spatial and temporal dependence together with the introduction of flexible dependence structures; the identification and integration of physical drivers and processes across scales; the handling of high-dimensional data and the estimation of occurrence probabilities. Improved representation of spatial–temporal dependence, clustering, and uncertainty is also critical for robust hazard and risk assessments, with direct implications for infrastructure design, disaster preparedness, climate adaptation strategies, and risk management in the (re)insurance sector.
We welcome contributions that enhance our understanding of the mechanisms driving hydroclimatic extremes, propose innovative modeling frameworks, or offer new insights into the prediction, attribution, and risk assessment of these events across space and time. Studies addressing extremes from statistical, physical, or interdisciplinary perspectives are particularly encouraged.
Orals: Thu, 7 May, 14:00–08:50 | Room B
Flood risk management faces a fundamental challenge in robustly estimating flood quantiles in a changing climate and developing appropriate adaptation measures. Furthermore, sound risk estimates require spatially coherent and temporally consistent scenarios of extreme precipitation and flood events. In this contribution, we address both challenges by deploying a novel non-stationary climate-informed stochastic weather generator conditioned on dynamic and thermodynamic change signals from global climate models. We generate synthetic weather datasets for present and future climate states in Germany, which are subsequently used to estimate flood quantiles through continuous hydrologic simulations. The seasonality of extremes is analyzed and compared between present and future periods. The robustness of the weather generator-based estimates is exemplified for the flood frequency estimation in the Ahr basin hit by an extreme flood in July 2021 and benchmarked against temporal information expansion using historical floods.
How to cite: Vorogushyn, S., Nguyen, V. D., Han, L., Guan, X., and Merz, B.: Flood frequency hydrology with a non-stationary, climate-informed weather generator, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16267, https://doi.org/10.5194/egusphere-egu26-16267, 2026.
The multifractal Bernoulli-lognormal (BLN, traditionally known as beta-lognormal) cascade has been effectively used to model the intermittence and scale invariance found in precipitation intensities, particularly under extreme hydro-meteorological events that generate hydrologic and geomorphological hazards such as floods, landslides, and debris flows. However, the parametrization of its generator based on a single realization has been a challenge due to the inherent non-ergodic nature of the process, and it is relevant for understanding vulnerability, risk mitigation and societal response to weather-induced extremes. In this work, we compare two recently proposed advances in parametrisation: (i) an approximation for the distribution of the BLN breakdown coefficients (BDCs) and (ii) the explicit expression of the dressed-cascade autocorrelation function in terms of the moments of its generator. Based on these two statistics, we derive an equation system that directly links the parameters ($C_b$, $C_{ln}$) with the observable quantities: the BDCs' distributional moments and the decay rate of the autocorrelation. We use these two parametrisation methods on multiscale precipitation data obtained from Google Earth Engine, enabling the analysis of weather–precipitation relationships, socio-hydrological interactions, and their implications for preparedness, impact-based forecasting, and even insurance and reinsurance applications.
How to cite: Ruano, A., Villegas, I., Gaviria, E., Hernandez, C., and Carsteanu, A. A.: Comparative parametrization of the Bernoulli-lognormal cascade generator through binary breakdown coefficients and using its autocorrelation function, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-577, https://doi.org/10.5194/egusphere-egu26-577, 2026.
Rainfall frequency analysis is routinely required for hydrological applications such as in the derivation of intensity-duration-frequency (IDF) curves for engineering design and planning. Commonly the Generalized Extreme Value (GEV) is used for rainfall frequency analysis, but it encounters limitations in capturing rare events which have heavy tailed distributions. An alternative is to use the four-parameter Kappa distribution which is a generalization of commonly used three-parameter extreme value distributions. Here, the applicability of the four‐parameter Kappa distribution for modelling extreme daily rainfalls using a global data set of annual rainfall maxima is presented.
The second shape parameter (h) of the four‐parameter Kappa distribution was found to vary regionally. Consistent with theoretical expectations, the second shape parameter converged toward zero (i.e., toward the limiting GEV distribution) as the average number of rain days events per year increased. However, in arid regions h was greater than zero suggesting there is merit in using the four‐parameter Kappa distribution for modelling heavy tail behaviour, particularly in regions which experience a small number of rainfall events per year. Information on the uncertainty in h as a function of the number of wet days per year is provided to facilitate Bayesian inference for at-site analyses.
As the four‐parameter Kappa distribution can be difficult to estimate, parameter estimation can be improved by using a two-step fitting approach based on maximum likelihood estimation which separately models storm intensity and the arrival frequency. Leveraging additional information from a peak-over-threshold series in the fitting improves quantile estimation and reduces uncertainty compared to fitting using annual maxima. These results demonstrate that the four‐parameter Kappa distribution is suitable for both at-site and regional rainfall frequency analyses.
How to cite: Wasko, C., Strong, R., Borgstroem, O., O'Shea, D., and Nathan, R.: Using the 4-parameter Kappa distribution to model extreme rainfall, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3218, https://doi.org/10.5194/egusphere-egu26-3218, 2026.
Improving our estimates of extreme precipitation magnitudes with low exceedance probability under climate change scenarios is crucial for disaster preparedness. The task is particularly challenging for sub-daily extremes, as they are hardly resolved by current climate models and they are expected to change at faster rates than longer-duration extremes. A statistical approach to predict future sub-daily extremes using a physically-based dependence on temperature was proposed (TENAX). The approach establishes a functional dependence between the parameters of the statistical model and near-surface air temperature. A temperature model is then used to represent the probability of having a precipitation event at a given temperature. While an exponential relation between scale parameter and temperature can be physically obtained from the Clausius–Clapeyron relation, the dependence of the shape parameter (related to tail heaviness) on temperature is less trivial and may significantly affect the model’s accuracy. Here, we implement a Bayesian framework to investigate this issue and to include prior knowledge on the parameter in the statistical inference. We test both linear and exponential dependencies of the shape parameter on temperature, as well as different temperature models. Preliminary results on several stations in Germany, Japan, the UK, and the USA show consistency of the past return levels with the previous TENAX model (based on maximum likelihood estimation with only the scale parameter dependent on temperature), and with benchmark estimates from a non-asymptotic method (SMEV), in both its classic and time-dependent implementations.
How to cite: Darienzo, M., Canale, A., Thomas, E., Borga, M., and Marra, F.: Including prior information on temperature-dependent sub-daily extreme precipitation in a Bayesian framework, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9309, https://doi.org/10.5194/egusphere-egu26-9309, 2026.
Quantifying rainfall extremes and their temporal evolution is essential for hydrological risk analysis and infrastructure design, and is commonly based on intensity–duration–frequency (IDF) curves. In this study, we further develop the framework proposed by Bendjoudi et al. (1997), in which IDF curves are theoretically derived within the Universal Multifractals (UM) formalism. This approach is mathematically robust and parsimonious, and is grounded in the physically based concept of scale invariance inherited from the Navier–Stokes equations.
Relying on either a unique scaling regime or two scaling regimes with a break at roughly 14 days, and the existence of a multifractal phase transition associated with moment divergence, IDF curves can be derived where rainfall intensity follows a power-law relationship with both return period (positive exponent) and duration (inverse exponent). The values of the exponents and of a prefactor can be directly inferred from the UM characterization of the rainfall process.
The framework was tested using rain-gauge data from six stations in Northern Italy, with record lengths ranging from 30 to 38 years. The agreement between the theoretically predictions from the UM analysis and the observed values of the prefactor and the two exponents, according to the quality of the scaling, is discussed. Possible directions for further improvements of the framework will also be discussed.
Authors acknowledge the France-Taiwan Ra2DW project for financial support (grant number by the French National Research Agency – ANR-23-CE01-0019-01).
References:
Bendjoudi H., Hubert P., Schertzer D., Lovejoy S., 1997, Interprétation multifractale des courbes intensité-durée-fréquence des précipitations, Comptes Rendus de l'Académie des Sciences - Series IIA - Earth and Planetary Science, 325, 5, 323-326,https://doi.org/10.1016/S1251-8050(97)81379-1
How to cite: Gires, A., Dallan, E., Marra, F., Schertzer, D., and Tchiguirinskaia, I.: Universal Multifractals characterization of Intensity-Duration-Frequency curves, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14683, https://doi.org/10.5194/egusphere-egu26-14683, 2026.
Unprecedented rainfall events are characterized by extremely high magnitudes and very low probabilities. Such events are occurring more frequently under a warming climate, despite being poorly represented in historical records. While former studies investgated physical drivers of such extremes, statistical approaches to quantify their likelihood and impacts remain limited. The current study presents a serial-type stochastic rainfall generator (SRG) explicitly designed to simulate unprecedented rainfall by incorporating non-stationarity through resampling and perturbation of model parameters governing the power-law tails of rainfall distributions. The approach is first evaluated over Southeast Texas using daily rainfall simulations for the 2017 Hurricane Harvey event, based on rainfall accumulation data from eight weather stations. By adjusting two power-law tuning parameters to represent warming conditions, the SRG successfully generates Harvey-like rainfall extremes. Simulated rainfall magnitudes associated with 50-, 100-, 250-, and 500-year return periods substantially exceed historical estimates. Additionally, the inferred return period of Harvey-scale rainfall closely aligns with previous independent assessments. The framework is subsequently extended to the Indian region, where thirty-six climate-change-relevant precipitation scenarios are generated by perturbing SRG parameters. High-performance computing is used to simulate daily rainfall across the domain, from which rainfall return levels and depth–duration–frequency (DDF) curves are derived. Results indicate substantial increases in rainfall return levels across all frequencies when unprecedented events are considered, particularly in coastal, northeastern, and Himalayan regions. Consistent spatial patterns and low spatial uncertainty across climate zones demonstrate the robustness of the SRG despite its point-based formulation. The proposed framework provides a statistically grounded pathway for revising design storms and supporting climate-resilient flood risk management under non-stationary climate conditions.
How to cite: Das Bhowmik, R., Alexander, A. A., Rasool, T., and Dasika, N. K.: Stochastic Simulation of Unprecedented Rainfall Events under Climate Change: From Hurricane Harvey to Continental-Scale Risk Assessment, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-15379, https://doi.org/10.5194/egusphere-egu26-15379, 2026.
Many Clausius-Clapeyron (CC) scaling studies relate warming to changes in a single precipitation quantile, which can obscure how “extremes” are defined and overlook the fact that events of different rarity are expected to scale at different, yet physically related, rates. CC analyses are also commonly conducted separately for different event durations, limiting insight into whether distinct processes control precipitation variability across timescales. To overcome this limitation, we propose a framework in which the full precipitation-intensity probability distribution is allowed to vary with climate conditions, enabling multiple quantiles to respond differently and to be associated with different drivers.
We apply this approach to observations from 605 stations across the continental United States, exploring how the parameters of hourly and daily precipitation distributions vary with local thermodynamic covariates and indicators of large-scale atmospheric circulation. An additional set of 456 stations with dew point temperature data is used to further assess the role of atmospheric moisture. Stations are grouped by Köppen-Geiger climate zones to ensure robust and coherent relationships. Results show that at the hourly scale, changes in extremes are primarily explained by local temperature and atmospheric moisture availability, with distributional tail thickening under warmer and moister conditions leading to increasingly rapid intensification for rarer events. At the daily scale, controls shift toward non-local influences associated with large-scale circulation. By characterizing scaling behavior across the entire distribution, this framework provides a physically grounded view of how warming affects both typical precipitation and extremes, and highlights the limitations of CC-based approaches.
Our results suggest that the assessment of future extremes should fully account and resolve the physical processes, such as convection and orographic forcings, responsible for extreme rainfall generation rather than rely on simplistic CC-based methodologies.
How to cite: Andria, S., Borga, M., and Marani, M.: Redefining Clausius-Clapeyron Scaling to Disentangle Local Thermodynamic vs Large-scale Circulation Controls on Extreme Precipitation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19771, https://doi.org/10.5194/egusphere-egu26-19771, 2026.
Rainfall events are inherently spatially extended phenomena that can be described through multiple physical attributes. Nevertheless, return period estimates are still commonly derived from point-scale rainfall intensity series, whose extension to regional-scale hazard assessment and rainfall–runoff modeling relies on strong and often implicit assumptions.
This study presents an event-based framework for the multivariate analysis of extreme rainfall events and the estimation of their return periods at regional scale. Rainfall events are reconstructed over Tuscany (Italy) from high-resolution precipitation records collected from a dense rain gauge network over the period 1999–2024, using a spatio-temporal aggregation approach. Aggregated events are represented through a set of physically meaningful attributes describing their intensity, spatial extent, duration, and precipitation volume, allowing a coherent characterization at event scale.
Extreme-value behavior is modeled through a Peak Over Threshold approach applied to the selected event attributes. Multivariate dependence among extreme events is described using flexible dependence models, enabling the joint behavior of intensity- and extent-related characteristics to be captured without imposing restrictive assumptions. A large synthetic population of extreme events is then generated to support a probabilistic interpretation beyond the limits of the observed sample.
To define multivariate return periods in a consistent manner, events are analyzed within a reduced space of independent latent variables derived from the original attributes. This representation allows extreme events with different physical signatures to be compared within a unified probabilistic framework, while accounting for the multivariate nature of rainfall extremes.
The proposed approach provides a robust basis for the regional-scale assessment of extreme rainfall hazards and highlights key challenges related to the definition and interpretation of return periods for spatially extended events. The framework is designed to support more physically consistent comparisons of extreme rainfall events and to improve their integration into hydrological risk analyses.
How to cite: Di Bacco, M., Manzella, F., Mazzanti, B., and Castelli, F.: An Event-Based Framework for Multivariate Return Periods of Extreme Rainfall , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19999, https://doi.org/10.5194/egusphere-egu26-19999, 2026.
Extreme precipitation in complex Alpine terrain exhibits pronounced spatial and temporal variability, challenging the reliable estimation of design-relevant return levels. Rain-gauge networks provide accurate point measurements but are often sparsely distributed and located in accessible valley floors, with few instruments on steep slopes or exposed crests where wind-induced undercatch is substantial. This limits their ability to capture localised extremes and fine-scale spatial variability. Weather radar offers the necessary coverage and resolution, yet radar archives are typically short and subject to various uncertainties. The Simplified Metastatistical Extreme Value (SMEV) framework offers a solution by enabling robust inference of rare extremes from short and error-prone datasets.
We analyse summer (JJA) rainfall extremes in Switzerland to derive intensity–duration–area–frequency (IDAF) relationships across multiple spatial and temporal scales, using nine years (2016–2024) of 1-km²/5-min dual-polarisation radar data from the MeteoSwiss C-band network. Return levels for durations from 30 min to 24 h and areas from 1 to 500 km² are estimated for return periods of 2 to 100 years using the SMEV framework. The extension of the SMEV to the areal scale was first developed by Rosin et al. (2024) for the eastern Mediterranean. We adapt and apply it here to the complex, heterogeneous Alpine topography of Switzerland. To reduce sampling noise inherent to the short radar archive, we spatially smooth the Weibull shape parameter, preserving coherent physical gradients while suppressing pixel-scale artefacts. Radar-derived SMEV return levels show strong regional agreement with SMEV estimates from 60 long-term (≥30 yr) gauges.
Rainfall extremes across Switzerland exhibit strong dependence on both spatial and temporal aggregation, affected by orography and location. Short-duration, small-area extremes display sharp, topographically anchored maxima over the Jura, Pre-Alps, and southern Alpine slopes, and persistent minima across the Plateau and inner-Alpine valleys. With increasing duration and area, small-scale peaks are progressively smoothed and broad-scale maxima emerge. The southern Alps remain the most prominent hotspot across all scales. Derived IDAF relationships display pronounced spatial differences at sub-hour scales and increasing spatial coherence for 12–24 h events, with pronounced regional differences.
Case studies of recent significant flooding events demonstrate how hydrological impacts depend on the spatio-temporal characteristics of rainfall. For each event, return levels were computed across all duration–area combinations using the IDAF framework, enabling a direct assessment of how 'extreme' the event was at different hydrologically relevant scales. Events that are highly extreme at short durations and small areas trigger flash floods and debris flows, reflecting the rapid response of steep Alpine basins. Conversely, events most extreme at long durations and large spatial scales, even when short-duration intensities are unremarkable, cause more widespread river flooding, elevated lake levels, and prolonged saturation. These results highlight the importance of evaluating extremes across multiple scales, rather than relying solely on point-scale intensities.
Overall, our findings highlight the value of combining short-record high-resolution radar precipitation fields with the SMEV framework to obtain a scale-aware extreme-rainfall climatology. The resulting multi-scale return-level maps and IDAF relationships provide improved information for flood-hazard assessment and infrastructure design.
How to cite: Rosin, T., Marra, F., Gabella, M., Germann, U., Wolfsenberger, D., and Morin, E.: Summer Precipitation Intensity-Duration-Area-Frequency Patterns in Complex Terrain using Radar Data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-857, https://doi.org/10.5194/egusphere-egu26-857, 2026.
Extreme rainfall and its temporal evolution critically influence flood hazard, slope stability, and infrastructure resilience. Yet in Italy, where complex topography and diverse climates shape precipitation, studies of rainfall extremes have produced conflicting outcomes, with neighboring sites often showing opposite trends. Much of this inconsistency stems from differences in data length, baseline period selection, and orographic context.
This study builds upon and extends the recent national-scale analysis by Mazzoglio et al. (2025), which, for the first time, quantified trends in rainfall extremes across Italy for the 1960–2022 period. Using the Improved Italian - Rainfall Extreme Dataset (I2-RED), which compiles data of thousands of rain gauges, we apply distributed quantile regression to annual maximum precipitation for short (1 h) and long (24 h) durations. Trends are expressed as percentage variations per decade and evaluated over multiple baseline windows (1960–2022, 1970–2022, 1980–2022, and 1990–2022) to test the sensitivity of results to the observational timeframe. Elevation effects are assessed by stratifying rain-gauge samples into low- and high-altitude groups and by comparing the regression slopes obtained for each.
Results reveal that short-duration extremes exhibit widespread and coherent positive trends, while 24-hour events show more heterogeneous and regionally variable patterns. Shortening the analysis period strengthens the positive signal, indicating that the intensification of sub-daily rainfall is largely a recent phenomenon. The most pronounced increases occur at higher elevations, especially in the Alps and Apennines. By contrast, lowlands and coastal areas show weaker or negligible changes. The geographic segmentation further demonstrates that spatial patterns of change align closely with major Italian physiographic structures, highlighting the combined roles of orography and regional geography in shaping rainfall evolution.
These findings suggest that trends in rainfall extremes in Italy cannot be interpreted through a single national lens: both methodological choices (baseline period and rainfall duration) and environmental factors (topography and geography) fundamentally shape the detected signals. The combined sensitivity to time window and elevation highlights the importance of accounting for Italy’s physiographic diversity when assessing hydrological risk and designing climate-resilient infrastructure.
Reference
Mazzoglio P., Viglione A., Ganora D., Claps P. (2025). Mapping the uneven temporal changes in ordinary and extraordinary rainfall extremes in Italy. Journal of Hydrology: Regional Studies, 58, 102287. https://doi.org/10.1016/j.ejrh.2025.102287
How to cite: Mazzoglio, P., Lelli, G., Domeneghetti, A., and Ceola, S.: Reference period matters, so do altitude and geography: understanding trends in rainfall extremes across the Italian landscape, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1025, https://doi.org/10.5194/egusphere-egu26-1025, 2026.
Convective storms frequently trigger flash floods, debris flows, and urban flooding, making robust sub-hourly precipitation statistics essential for risk assessment and infrastructure design. The TENAX model (TEmperature-dependent Non-Asymptotic statistical model for eXtreme return levels) offers a physically-based framework to estimate future extreme rainfall by linking precipitation intensity to near-surface air temperature. Its standard configuration adopts a 24-hour temperature window with zero offset preceding the rainfall event, which is mainly driven by compatibility with daily climate model outputs rather than empirical optimization. Yet, its sensitivity to alternative configurations remains largely unexplored. We hereby analyze 145 rain gauges from Arpae Emilia-Romagna (106) and ARPA Lombardia (39), from the Po plains to the Alpine forelands, spanning 2003–2024, each with at least 15 years of precipitation records at 10- to 15-minute resolution. Temperature data come from VHR-REA_IT reanalysis at 2.2 km resolution. We test twelve model configurations obtained by combining three alternative window durations (1, 12, and 24 h) with four temporal offsets (0, 1, 5, and 12 h). The analysis is performed both at the annual and at the seasonal levels, and model performance is assessed through repeated split-sample validation (50–50 random temporal splits), where the optimal configuration is selected by minimizing the mean squared error with respect to empirical return levels derived using Weibull plotting positions. Our annual analysis shows that the 24 h window with 12 h offset consistently outperforms the default configuration. In contrast, seasonal analyses reveal marked differences: summer extremes show a clear preference for short (1-h) temperature windows, consistent with convective storm dynamics, whereas autumn and winter exhibit higher variability with no single dominant configuration. Moreover, we identify a statistically significant relationship (p < 0.05) between the optimal temperature window configuration and station elevation, suggesting that elevation-dependent thermodynamic and convective processes modulate the temperature–precipitation link. The findings provide practical guidance for calibrating TENAX in data-rich regions and support more physically consistent applications to future climate projections.
How to cite: Lelli, G., Paschalis, A., Domeneghetti, A., and Ceola, S.: The role of antecedent temperature in controlling extreme rainfall statistics: multi-temporal and geomorphic patterns across Northern Italy, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-5288, https://doi.org/10.5194/egusphere-egu26-5288, 2026.
Climate change and the expected resulting changes in precipitation patterns call for robust and resilient climate adaptation solutions, including urban drainage systems that can handle the precipitation of the future. One of the most commonly used engineering tools when designing urban drainage systems is precipitation frequency estimates (PFEs) that allow for estimation of extreme precipitation rates, also called return levels, and the associated return periods. Oftentimes PFEs are based on point rain gauge measurements, either directly computed from a rain gauge precipitation time series or from a regionalized model that is based on a larger network of rain gauges, when representing extreme precipitation in ungauged areas. Weather radar precipitation measurements pose an alternative data source for computing PFEs as longer precipitation time series become available. Here, PFEs can be computed directly at the weather radar pixel scale (corresponding to the spatial resolution of the radar data) without the need for interpolation or other models of ungauged areas.
In this study, we aim to investigate how weather radar derived PFEs compare to rain gauge derived PFEs, especially at the short timescales that are necessary in urban drainage design. In addition to rain gauge radar pixel PFE comparisons, we aim to utilize the fully spatially distributed weather radar derived PFEs to analyze the spatial structure of model parameters over a study area in Denmark. Utilizing a 18-year long C-band weather radar record, PFEs are derived in the form of IDF curves at the pixel scale, along with the corresponding PFEs of rain gauges located within the study area. Timescales ranging from 1 minute to 2 days are considered. The weather radar and rain gauge data sets are analyzed using the median plotting formula for empirical return levels and extrapolated to longer return periods by constructing a partial duration series (PDS). The PDS is then modelled by the Generalized Pareto distribution, where model parameters are determined via maximum likelihood estimation. The resulting PFEs display clear scale differences, where weather radar derived PFEs are underestimated at short timescales. However, IDF curves converge at timescales around 200-300 minutes. The spatially distributed model parameters reveal novel insights with regards to spatial variation of extreme precipitation in the study area. Clear gradients are found in the number of yearly exceedances, the mean exceedance, and the shape parameter controlling the PFEs. Moreover, these parameters are also clearly dependent on the timescale considered, where higher timescales equal smoother parameter surfaces with higher spatial correlation. These results highlight the advantages of supplementing rain gauge data with weather radar data for supplementary information about spatial variation of extreme precipitation over a given area. They also underline methods for determining the specific timescales where users should be aware of scale differences, given the inherent different measurement techniques of rain gauges and weather radar.
How to cite: Vester, I. K., Nielsen, J. M., Nielsen, J. E., and Thorndahl, S.: Deriving Precipitation Frequency Estimates from High-Resolution Weather Radar Rainfall Data, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7545, https://doi.org/10.5194/egusphere-egu26-7545, 2026.
Extreme precipitation and its associated hydrometeorological hazards pose serious concerns to human well-being, society, and ecosystems. The frequency and intensities of such events are projected to increase in the future due to climate change. Despite substantial efforts to better understand these extremes and their underlying physical mechanisms, how these extremes will respond to increasing temperature remains an ongoing debate. In particular, the different response of different precipitation processes, such as convective and non-convective precipitation, to warming remains poorly understood. In this context, we explore sub-daily precipitation extremes at the quasi-global scale (60°N–60°S) from the high-resolution Integrated Multi-satellitE Retrievals for the Global Precipitation Measurement (IMERG). The dataset has 0.1° spatial and 30-minute temporal resolutions, enabling us to capture short-term convective events even at localized scales. We utilize lightning datasets to classify storms into convective and non-convective types, and we assess the scaling of extreme hourly precipitation intensities with temperature using a quantile regression approach. Our analyses will provide the first global-scale assessment of precipitation-temperature scaling rates across various storm types, providing new insights into sub-daily precipitation extremes. This will help us to better understand the underlying physical mechanisms of the extremes, and consequently to better prepare appropriate mitigation strategies.
How to cite: Pradhan, R. K. and Marra, F.: Extreme Precipitation Scaling with Temperature: A Storm-Type Perspective, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-7915, https://doi.org/10.5194/egusphere-egu26-7915, 2026.
Areal Reduction Factor (ARF) is a well-established hydrological concept used to convert point precipitation to areal precipitation. The aim of this work is to develop a dataset of ARFs all over Germany, which can be utilized to convert point precipitation extremes to areal precipitation extremes for different area sizes, using RADKLIM radar Product. Initially, point and areal precipitation quantiles, covering seven distinct area sizes up to 1225 km², are estimated at more than 10000 randomly selected RADKLIM pixels. Following the extreme value analysis, areal depth-duration-frequency (ADDF) curves are derived and pixels with the crossing problem - as defined in Goshtasbpour & Haberlandt (2025)- are filtered out. The remaining pixels are further analyzed as study locations. ARFs are then calculated at these study locations, for nine durations from 5 to 1440 minutes, and eight return periods from 1 to 50 years. ARFs typically increase with increasing duration and decrease with increasing area. To model the calculated ARFs as a function of area and duration, a well-performing four-parameter ARF expression from De Michele et al. (2001) is utilized. This model accurately represents the expected behavior of ARFs in relation to area and duration, and has been widely used in the literature. The application of the De Michele model simplifies the representation of ARFs at each study location and for each return period by representing them with only four estimated parameters, instead of 63 different ARF values considering all durations and area sizes. The estimated ARF fitting parameters show solid performance across most study locations, as indicated by the goodness-of-fit criteria: R², Percent Bias, and normalized Root Mean Square Error. Finally, the estimated parameters are interpolated in the space using various geostatistical techniques to provide countrywide raster based ARFs.
How to cite: Salehfard, G. and Haberlandt, U.: Estimation of areal reduction factors of extreme precipitation based on radar data , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12005, https://doi.org/10.5194/egusphere-egu26-12005, 2026.
The El Niño–Southern Oscillation (ENSO) is a dominant source of interannual climate variability, strongly influencing hydroclimatic extremes across the U.S. Great Plains (USGP). This study examines the seasonal and lagged impacts of ENSO phases—El Niño, La Niña, and Neutral—on precipitation-based extremes over the USGP for the period 1950–2023. ENSO phases were identified using the Oceanic Niño Index (ONI) with ±0.5 °C thresholds, and seasonal transitions (DJF, MAM, JJA, SON) were analyzed to characterize persistent, isolated, and whiplash ENSO extremes. High-resolution precipitation datasets from PRISM and NOAA Climate Divisions were integrated within a GIS framework to develop seasonal time series and conduct spatial analyses at the climate-division scale. Composite anomaly maps of precipitation percentiles were generated and spatially aggregated using zonal statistics, while Pearson and Spearman correlation analyses, including 3–12-month lags, quantified delayed and region-specific ENSO responses. Statistical significance of phase-wise differences was evaluated using ANOVA, Kruskal–Wallis, and Mann–Whitney U-tests. Results reveal pronounced seasonal asymmetry in ENSO impacts, with La Niña strongly associated with drought conditions in the southern plains and El Niño linked to enhanced wet anomalies across central and eastern regions. The identification of ENSO-sensitive zones improves regional climate predictability and provides actionable insights for anticipatory water-resources management. Overall, the study demonstrates the effectiveness of integrating geospatial analysis, long-term climatological datasets, and robust statistical methods to attribute hydroclimatic extremes to large-scale ocean–atmosphere variability.
How to cite: Talukdar, G. and Wadhawan, K.: Large-Scale Climate Drivers of Spatially and Temporally Compounding Hydroclimatic Extremes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18772, https://doi.org/10.5194/egusphere-egu26-18772, 2026.
Understanding the evolving dynamics of rainfall extremes is critical for assessing hydroclimatic risks in the Indian Ganga Basin (IGB), one of the world’s most densely populated and monsoon-dependent river systems. This study presents a comprehensive, century-long (1901–2023) assessment of rain spell dynamics across the IGB using a multivariate and probabilistic framework. Rain spells are characterized through the joint consideration of duration, intensity, and rainfall volume, enabling a clear distinction between short-duration (1–3 days) and long-duration (>3 days) events. For each category, a joint probability-based severity index is developed to quantify rainfall extremeness in an integrated manner. The analysis reveals a pronounced basin-scale reorganization of rainfall regimes over the last century. Historically, the IGB was dominated by spatially coherent and persistent long-duration rainfall events. However, recent decades show a marked shift toward increasingly frequent, intense, and spatially fragmented short-duration spells. Since the 1990s, short-duration rainfall events have exhibited rising persistence, increased recurrence rates, and enhanced severity across most parts of the basin. In contrast, long-duration wet spells display declining spatial continuity, reduced stability, and weakening basin-wide coherence. Notably, the entire basin now experiences an elevated occurrence of short, high-intensity events, indicating a fundamental transformation in monsoon rainfall behaviour. These evolving patterns significantly amplify hydrological hazards, including flash floods, rapid surface runoff, soil erosion, and landslides. Concurrently, the decline in sustained rainfall limits groundwater recharge, reduces soil moisture replenishment, and poses challenges for agricultural productivity and water security. The novelty of this study lies in its integration of multivariate rain spell characteristics within a joint probability framework to assess the long-term evolution of rainfall regimes. The findings provide robust evidence of hydroclimatic reorganization across the IGB and establish a probabilistic foundation to inform water resource management, disaster risk reduction, and climate adaptation strategies under a changing monsoon system.
How to cite: Maurya, A. K. and Swarnkar, S.: Spatiotemporal Dynamics of Rain Spell Persistence across the Indian Ganga Basin (IGB), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19991, https://doi.org/10.5194/egusphere-egu26-19991, 2026.
Flood Frequency Analysis (FFA) is a fundamental tool for flood‐risk assessment and hydraulic design and constitutes the statistical basis of the flood hazard scenarios defined under the European Floods Directive and its national implementations. Classical FFA is typically applied under assumptions of temporal independence and spatial representativeness at individual gauging stations and within the framework of the design storm paradigm. However, these assumptions are increasingly challenged during more extreme compound hydroclimatic events, where rainfall and runoff responses occur synchronously across multiple connected catchments and in successive phases in time. The October 2024 flood in southern Valencia Metropolitan Area (Spain) offers a unique opportunity to revisit FFA under such conditions. Over the course of that day, a spatially extensive and temporally clustered rainfall event affected the five catchments draining in this area, producing an exceptional hydrological response. The accumulated hydrograph had a peak of 7,500 m3/s, but with significant delays between the individual hydrograpghs. The event was characterized by two distinct rainfall phases, with an initial episode in the morning modifying the antecedent hydrological conditions of the catchments, followed by an extreme afternoon-evening phase that induced a strongly non-linear runoff response. Several tributaries responded almost simultaneously, resulting in spatial compounding of peak discharges and unprecedented flow magnitudes at the basin scale. Such a response challenges the assumptions underpinning classical FFA and highlights the need for alternative frameworks capable of representing compound hydrological behavior.
Rather than relying solely on point-based discharge records, this study proposes an integrated approach that combines regional extreme rainfall analysis, stochastic weather generation, and distributed hydrological modelling to estimate discharge quantiles beyond the limitations imposed by short instrumental records and thee design storm hypothesis.
The results indicate that applying the proposed integrated framework leads to a substantial downward revision of discharge quantiles associated with fixed return periods when compared to classical point-based FFA. Flood frequency estimates derived exclusively from local discharge records are strongly influenced by limited sample sizes and by the extrapolation of the upper tail, which can result in unrealistically high discharge quantiles. By combining regional precipitation analysis, stochastic weather generation, and distributed hydrological modelling, the proposed approach better constrains the range and frequency of rainfall-runoff conditions capable of producing extreme flows. As a consequence, discharge magnitudes previously associated with very long return periods are shown to occur more frequently, implying lower discharge values for a given return period and a higher effective frequency of potentially damaging flows.
Overall, this study demonstrates that the proposed framework provides a more consistent and physically grounded basis for estimating flood quantiles under spatially and temporally compounding hydroclimatic conditions, and offers a robust foundation for the derivation of flood hazard maps within the context of current European and national flood-risk management frameworks.
How to cite: Francés, F., Beneyto, C., and Aranda, J. Á.: Flood Frequency Analysis revisited under spatially and temporally Compound Flood Extremes: evidence from southern Valencia Metropolitan Area, Spain, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-17214, https://doi.org/10.5194/egusphere-egu26-17214, 2026.
Hydroclimate extremes such as floods and droughts are associated with increasing socio-economic losses worldwide, reflecting their diverse spatial and temporal characteristics and growing exposure. Reliable forecasting across seasonal to interannual timescales is therefore critical for mitigating their impacts and informing risk management. Although machine learning approaches have demonstrated considerable potential, they often depend on large volumes of high-quality data and on distributional transformations of predictors, while neglecting mismatches in temporal scale and spectral structure between predictors and hydrological responses. These mismatches can mask physically meaningful signals, particularly for extremes influenced by scale-dependent climate variability.
Here we address this limitation by introducing the Wavelet System Prediction (WASP), a frequency-domain method designed to enhance hydroclimate predictors through spectral transformation. WASP employs discrete wavelet transforms to decompose predictors and responses into scale-specific components and systematically adjusts the spectral variance of predictors to align with that of the response under an assumed stationary predictor–response relationship. This approach explicitly accounts for temporal dependence and scale interactions, enabling the extraction and amplification of predictive signals that are weak or hidden in the raw predictor space.
We apply WASP to two contrasting hydroclimate extremes and spatial contexts: seasonal flood forecasting across multiple European catchments and interannual drought forecasting at the continental scale over Australia. In both applications, the proposed method substantially improves forecast skill compared to conventional methods. These results highlight the value of scale-aware, frequency-based transformations for advancing statistical modelling of hydroclimate extremes, contributing to improved hazard assessment and climate risk management.
How to cite: Jiang, Z. and Sharma, A.: Spectral transformation of hydroclimate predictors enhances flood and drought forecasting, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3661, https://doi.org/10.5194/egusphere-egu26-3661, 2026.
Climate change is an obvious driver of changes in water regimes in Polish rivers. However, human interventions in catchments strongly magnify its negative effects. One of the most visible consequences is the increasing intermittence of rivers that only a few years ago flowed throughout the year. In Poland, river intermittence is a relatively new phenomenon. Smaller rivers now disappear for significant parts of the year due to prolonged hydrological droughts.
Within the project “Intermittent rivers of Central Europe: Identifying threats to protection goals and biodiversity for efficient nature conservation and climate-proof environmental management”, we analysed all available Polish records of daily discharges and identified 22 gauging stations where extremely low or zero flow occurred at least once during the observation period.
We observed strong temporal unevenness in the occurrence of low-flow events, suggesting that gradual climatic change alone may not fully explain the development of river intermittence. Indeed, when compared with land-cover changes derived from successive CORINE Land Cover maps, some stations revealed sudden increases or decreases in the frequency of low-water events. Although this pattern was not observed for all analysed intermittent rivers, it may provide further evidence that unsustainable water management practices in catchments amplify the effects of climate change.
How to cite: Kochanek, K., Mammadova, A., Grodzka-Łukaszewska, M., Sinicyn, G., and Grygoruk, M.: Patterns of low-flow and zero-flow events in Polish rivers: climate signal or catchment impact?, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10332, https://doi.org/10.5194/egusphere-egu26-10332, 2026.
Drought-to-flood transitions, where low-flow conditions are rapidly followed by high flows, are increasingly framed as compound hydrological hazards. However, it remains unclear whether such transitions are genuinely rare events or simply reflect how they are defined. Most existing studies apply uniform magnitude thresholds and fixed time windows across diverse catchments, implicitly assuming comparable extremeness. Here, we challenge this assumption by reframing transitions in probabilistic terms, quantifying the conditional likelihood of large streamflow swings across a range of severities, durations, and seasonal contexts.
Using daily streamflow records from 4,299 European catchments, we perform three conditional probability experiments to assess how transition likelihood depends on threshold choice, low-flow duration, and timing within the hydrological year. We identify pronounced and spatially coherent patterns in transition probability. Very rapid transitions (e.g. within 14 days) are common in the Alps, coastal Scandinavia, and the United Kingdom, while catchments with strong hydrological memory exhibit consistently low probabilities, even over long time windows (up to 365 days). Transition probability generally decreases with increasing low-flow duration, except in snow-influenced catchments, where seasonal processes can increase the likelihood of transitions when only longer-duration low flow periods are considered. Examined continuously, low-flow events also exert a persistent influence on subsequent streamflow distributions, particularly when they occur in phase with the climatological dry season.
Our results show that transition definitions commonly used in the literature correspond to frequent events in some regions and extremely rare events in others. This demonstrates that the extremeness of drought-to-flood transitions cannot be inferred from magnitude and timing alone, but must be evaluated relative to their conditional or joint probability of occurrence. We argue that compound hydrological transitions should be defined consistently with other extremes, using probability-based or impact-relevant criteria rather than uniform thresholds.
How to cite: Anderson, B., Schillinger, M., Muñoz-Castro, E., Tarasova, L., Berghuijs, W., and Brunner, M.: How rare are rapid transitions in streamflow? A conditional probability approach, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11369, https://doi.org/10.5194/egusphere-egu26-11369, 2026.
Projected increases in temperature and alterations in precipitation patterns across two major river systems in South Africa necessitate the implementation of adaptation strategies to address water scarcity and flood hazards. The Integrated Vaal River System (IVRS), the primary freshwater supply system for Johannesburg, is increasingly challenged by extreme drought conditions, and the coastal rivers, including the Umgeni, Mlazi, and Mbokodweni rivers, east of the Lesotho highlands in the Greater Durban region, face significant flood risks. To develop adaptation measures, compound climate extremes, such as coincident or sequential meteorological droughts and heatwaves, as well as meteorological droughts followed by extreme precipitation, are of particular interest.
In the present study, the focus is on the changes in frequency and spatial distribution of coincident and sequential compound climate extremes across both river systems. Using the bias-adjusted CORDEX-CORE Africa climate RCP8.5 projection ensemble at a 0.22° spatial resolution, generated by three regional climate models that dynamically downscaled three distinct Earth system models, enables a comprehensive assessment of model uncertainties. Initial results indicate that the occurrence of coincident meteorological droughts and heatwaves increases along a south-to-north gradient, with longer durations over the IVRS toward the end of the century. This research is conducted in the WaRisCo project, which is a part of the “Water Security in Africa – WASA” programme.
How to cite: Weber, T., Biskop, S., Ismail, M. F., Ernst, Y., and Engelbrecht, F.: Analysis of projected compound climate extremes across two major river systemsin South Africa, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12036, https://doi.org/10.5194/egusphere-egu26-12036, 2026.
Derived flood frequency analysis (DFFA) allows the estimation of design floods with hydrological modelling for both poorly observed basins and for catchments under nonstationary conditions. For mesoscale catchments long records of sub-daily precipitation are required. As these are usually not easily available, stochastic weather data can be used as an alternative. Objective of this research is to find the optimal calibration strategies of a hydrological model for DFFA using stochastic weather data as input by comparing various calibration alternatives. The optimal calibration of the hydrological model should a) consider long records regarding robust estimation of the extremes b) select the most informative parts from these records and c) utilise the stochastic input data.
Hourly climate variables are disaggregated from long daily records using a k-nearest neighbour approach. For hydrological modelling the semi-distributed conceptual HBV model is used. The model is calibrated alternatively on observed flow data and on various flow statistics considering different temporal discretisations and time periods. The main validation of the hydrological model is based on long term flood statistics. The calibration approaches are tested for several mesoscale catchments of the Mulde River basin in Germany. The results will reveal the advantages and disadvantages of the different calibration strategies and if there is an optimal approach.
How to cite: Haberlandt, U. and Brandt, A.: Optimal calibration of hydrological models for derived flood frequency analyses using stochastic rainfall - revisited, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12536, https://doi.org/10.5194/egusphere-egu26-12536, 2026.
Floods are the most common natural hazard in the Republic of Uzbekistan and cause loss of life for humans and livestock, damage infrastructure, destroy or impair habitats, and disrupt the economy. To inform infrastructure design and water management scenarios, statistical flood frequency analyses are performed. Ideally, statistical flood frequency analyses are based on instantaneous annual peak streamflows because peak streamflows are causative to maximum flood inundation surfaces. When instantaneous annual peak streamflows are not available, such as the historical hydrologic data portfolio in Uzbekistan, the largest daily mean streamflow becomes the surrogate for the annual peak. These 1-day annual maxima are usually an underestimation of the true peak streamflow for the year, particularly in a region where flood hydrograph durations are short and flashy. This problem in hydrologic risk analysis is exemplified in the Kashkadarya Region of Uzbekistan where long-term (50+ years) daily mean streamflow data exist, but digitized streamflow data is limited to 1991 to present at ten streamgages. Given that instantaneous peaks are not available for the Uzbek streamgages, a correction factor was calculated based on 3,466 station-years of daily mean streamflow and peak streamflows at 185 streamgages in, New Mexico, USA. New Mexico was chosen because it is a comparatively data-rich region with somewhat analogous topography and precipitation to Kashkadarya, Uzbekistan. The analysis showed that on average, instantaneous annual peaks were 38% higher than annual daily maxima. A regional statistical model was made using basin characteristics as explanatory variables to estimate an adjustment factor to increase flood streamflows based on the annual daily maxima. The modeled adjustment factor was then applied to annual exceedance probability streamflows from a flood frequency analysis performed at the ten streamgages in the Kashkadarya Region. The frequency analysis was performed using generalized extreme value probability distribution on daily streamflows from 1992 to 2020.
How to cite: Crowley-Ornelas, E., Asquith, W. H., Khudoyberdiev, A., Barnhart, T., and Umirzakov, G.: Flood frequency analysis and a regional peak streamflow correction factor model for Kashkadarya Region, Uzbekistan, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14732, https://doi.org/10.5194/egusphere-egu26-14732, 2026.
Extreme climate hazards can occur in isolation or interact as concurrent, compound or transitional events, amplifying their impact on key socioeconomic sectors such as agriculture, tourism and health. In Europe and the Mediterranean basin region, these interactions pose significant risks over the densely populated regions which are highly vulnerable to the combined occurrence of climate hazards. Hence, the study of this aggravating issue in the context of global warming emphasizing the analysis of concurrent, compound, sequential and transitional climate extreme events is crucial to better comprehending their relationships and improving early warning systems and adaptive strategies in emerging climate hotspots.
In this study daily high-resolution datasets from different sources, (ROCIO-IBEB, EMO1, CERRA, EFFIS, ICV, ERA-5 and MED-REP-L4), have been used to identify atmospheric and marine heatwaves, droughts, wildfires, extreme precipitation events and extreme wind, as well as to detect emerging hotspots.
Our findings over specific Mediterranean climate change hotspot such as the Valencia Region in eastern Spain reveal a rising frequency of concurrent hazards, with droughts emerging as a key driver of both summer wildfires and extreme autumn precipitation. Besides, our results also indicate an increasing influence of Mediterranean Sea warming on both maximum 2-meter air temperature over land and extreme autumn precipitation highlighting the relevance of the welldocumented Mediterranean SST increase on climate extremes. Besides, relationships among key climate variables have been studied using different methodologies, such as lagged correlations and normalized information flows, to estimate climate factors influences on climate extremes.
The extension of the analysis to Europe and the Mediterranean basin yielded results that were consistent with those of the regional analysis. It has been determined that the proportion of hazards and drivers that compound forest fires is similar between in and out identified hotspots. Furthermore, AHW-drought and drought-AHW transitions have been analyzed, with heightened intensity observed in the latter. Evidence suggests that drought-EPE transitions occur most severely in regions where droughts and EPEs are most intense as a singular event, too. Regarding MHW analyses in the northeastern Atlantic Ocean and Euro-Mediterranean seas, the results reveal the presence of large high-intensity MHW hotspots over northern seas, especially in the Artic Sea, in contrast with the localized Mediterranean hotspots.
The present study seeks to determine whether areas susceptible to dry-heat-wet hazards are concomitantly exposed to forest fires and floods. Furthermore, an ongoing analysis of flooding risk will provide additional information on a local scale, which is crucial for identifying interactions among climate hazards, and for evaluating potential risks and vulnerability over these areas.
How to cite: Rico-Bordera, B., Benetó, P., and Khodayar, S.: Emerging Links Between Droughts, Heatwaves and Extreme Precipitation in Europe and the Mediterranean basin, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-18039, https://doi.org/10.5194/egusphere-egu26-18039, 2026.
As torrential flood-inducing heavy rainfall intensifies under climate change, new indicators for quantifying short-term precipitation concentration are essential. This study introduces the Modified Inter-Amount Time (M-IAT), which measures the duration required to reach critical precipitation thresholds, and develops the Standardized Torrential Flood Index (STFI) using Generalized Extreme Value (GEV) and Generalized Pareto Distribution (GPD) models. Analysis of 65 ASOS stations (1990–2024) shows that as critical rainfall values (CV) increase, the GPD model evaluates extreme temporal concentration more conservatively than the GEV model. Validation against 39 historical flood events revealed that the GPD-STFI median reached 3.72 (99.99th percentile) during actual damage occurrences, effectively identifying extreme risks. Conversely, the GEV-STFI established stable long-term and structural risk baselines for different regions. The STFI facilitates a paradigm shift from precipitation-centered forecasting to dynamic, hydrological response-time-centered warnings. This study presents an integrated risk management strategy by combining design-oriented GEV models with operation-oriented GPD models, providing a robust framework for flood mitigation.
How to cite: Yoon, S., Kwak, M., and Lee, B.: Development and Application of a Time-Based Standardized Torrential Flood Index via Modified Inter-Amount Time (M-IAT), EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20262, https://doi.org/10.5194/egusphere-egu26-20262, 2026.
Recent theory and model projections indicate that climate change should intensify and reorganize global precipitation patterns; however, observational confirmation has been hindered by the proliferation and interdependence of gridded products. This study revisits the changing precipitation characteristics using an artifact-controlled ensemble of gauge-, satellite-, and reanalysis-based datasets at 0.25° daily and monthly resolution for the 1995–2024 period. Concentrated along the tropics, a drying pattern has emerged, while annual maxima daily precipitation has increased simultaneously. In other words, our results indicate that a growing share of annual precipitation is delivered by upper-percentile daily events, even as the annual mean precipitation decreases. The co-occurrence of drying and intensification patterns suggests that extreme events are efficiently depleting atmospheric moisture, leading to longer dry spells and reduced total precipitation. The results highlight regions shifting toward a more intense and abrupt hydrological regime, with higher flood and drought risks despite declining mean precipitation.
How to cite: Vargas Godoy, M. R., Markonis, Y., Papalexiou, S. M., and Jenicek, M.: It Never Rains but It Pours, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-650, https://doi.org/10.5194/egusphere-egu26-650, 2026.
Intensity-duration-frequency (IDF) relations describe the major statistical characteristics of extreme precipitation events (return level, return period, time scale). These IDF relations help to visualize either how extreme (in terms of probability/frequency/return period) a specific event is or which intensity is expected for a given probability. We model the distribution of annual precipitation maxima in an extreme-value-statistics setting for the study region Berlin, Germany. To increase model efficiency, we include the accumulation duration and model a duration-dependent GEV. The durations range from 5 minutes to days and are modeled in one single model in order to prevent quantile-crossing. Latitude and longitude are considered as covariates for the GEV parameters.
A major challenge is the need for long precipitation records in order to reliably estimate return levels of long return periods. Especially for short durations (minutes to hours), long records are rare. Therefore, we pool 3 data sources: radar-based Radklim (5-minute) and spatially-interpolated HYRAS (daily) and station-based measurements (minutely). This way, data from sources with daily resolution can borrow information from sources with minutely resolution at nearby locations. This is possible because we assume a functional relationship between short and long durations. Also we assume similar characteristics between nearby stations. This requires a spatial model since different data sources are not collocated. IDF relations will be estimated for any given point in space by using all available multi-source data in a radius of a few kilometers. Two different models are compared to do that: (1) A parametric model is using latitude and longitude as covariates. (2) We plan to create and show a non-parametric Bayesian Hierarchical Model (BHM), including a Gaussian process which models the spatial dependence between locations. The quality of estimated IDF relations will be assessed in terms of a cross-validated quantile score.
How to cite: Fauer, F. S. and Rust, H. W.: Tackling Sparse High‑Resolution Data in Extreme‑Value Statistics: A Spatial Multi‑source Approach, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-3109, https://doi.org/10.5194/egusphere-egu26-3109, 2026.
In recent decades, climate change has intensified extreme rainfall events and expanded their spatial extent, highlighting the need for area-based design rainfall estimation in regional flood control planning. Conventional Depth–Area–Frequency (DAF) curves rely on point rainfall observations combined with empirical Area Reduction Factors (ARFs), which limits their ability to represent actual area-averaged rainfall and spatially connected rainfall structures. This study develops a probabilistic DAF framework that explicitly accounts for spatial adjacency and area-averaged rainfall characteristics. Using 30 years of rainfall observations from the Automated Synoptic Observing System (ASOS) across South Korea, spatially connected area combinations were constructed through adjacency analysis, and representative area sets were selected using the Latin Hypercube Sampling technique. Area-averaged annual maximum rainfall was then derived for each area scale, and multiple probability distributions were applied to characterize extreme rainfall behavior. Goodness-of-fit evaluations indicate that the Generalized Extreme Value (GEV) distribution most appropriately describes area-based extreme rainfall across different spatial scales. Based on the selected GEV distribution, probabilistic DAF curves corresponding to various return periods were derived. The proposed framework eliminates reliance on empirical ARFs and provides a physically consistent and probabilistically rigorous approach for estimating design rainfall, thereby improving the reliability of regional and national-scale flood control and hydrologic design applications.
How to cite: Kim, J., Shin, J.-Y., Lee, G., and Kim, S.: Derivation of Probabilistic Depth–Area–Frequency Curves Based on Spatial Adjacency Using the Generalized Extreme Value Distribution in South Korea , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6136, https://doi.org/10.5194/egusphere-egu26-6136, 2026.
Under global warming, high-density coastal cities face the dual challenge of intensifying precipitation extremes and increasing atmospheric evaporative demand. While rainfall trends in Hong Kong have been widely monitored, determining whether the region is becoming wetter or drier requires a comprehensive assessment of the water-energy balance beyond simple precipitation totals. This study investigates the spatiotemporal characteristics of hydro-climatic changes in Hong Kong over the past 40 years (1985–2024), utilizing a hybrid data fusion approach. We integrate hourly precipitation records from 84 Geotechnical Engineering Office (GEO) stations with temperature data from the Hong Kong Observatory (HKO) and reanalysis products from ERA5-Land. To address the coarse resolution of reanalysis data in complex terrains, a topography-based bias-correction and downscaling scheme is applied to generate high-precision, 1-km resolution fields of both Evaporation (E) and Potential Evapotranspiration (PET).
The analysis evaluates hydro-climatic indices across wet (April to September) and dry (October to March) seasons to capture the changing patterns of the urban water cycle. Precipitation metrics include accumulated rainfall, total wet/dry days, and Consecutive Dry Days (CDD), while thermal stress is assessed through daily maximum temperatures, the aggregate count of hot days (>30°C), and the duration of consecutive hot days. Beyond statistical trend analysis, the study adopts the Budyko framework to physically characterize the shift in hydro-climatic regimes. We analyze the joint trajectories of the Aridity Index (PET/P) and the Evaporative Index (E/P) within the Budyko space. This framework is applied spatially across four distinct subregions—Hong Kong Island, Kowloon, New Territories, and Lantau—to reveal how varying degrees of urbanization and vegetation cover alter the partitioning of available water and energy.
By exploring these metrics, this study elucidates the potential decoupling between water supply and atmospheric demand. The research aims to identify transitions towards compound extremes, such as the alternation between intense rainfall pulses and prolonged, hotter dry spells. These insights provide a physical basis for understanding the changing flashiness of the local climate, offering critical guidance for adaptive water resource management in the Guangdong-Hong Kong-Macao Greater Bay Area.
How to cite: Tang, X., Wang, D., and Lu, Y.: Wetter or Drier? Spatiotemporal Evolution of Hydro-climatic Extremes in Hong Kong via High-Resolution Data Fusion and the Budyko Framework, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-6272, https://doi.org/10.5194/egusphere-egu26-6272, 2026.
Tailings storage facilities (TSFs) present a persistent stability challenge to the global mining industry, particularly given the large quantity of inactive, closed or abandoned deposits. Due to the potential for catastrophic failures, hydrometeorological hazards, such as extreme precipitation, are one of the main threats to these structures. This stems from the stress that infrastructure can undergo when dealing with these extreme events.
Current design standards require that TSF be capable of handling the Probable Maximum Precipitation (PMP) and the associated Probable Maximum Flood (PMF). However, the reliability of these design values, traditionally derived from stationary statistical records, is increasingly uncertain in the context of global warming.
Here, we assessed the hydrological failure hazard of four TSFs across significantly diverse climatic zones -ranging from arid to cold-humid climates- in Chile, a leading country in global copper production with nearly 800 TSF associated with these activities, most of which are inactive or abandoned. To do so, we first estimated PMP values over the historical period 1960-2014 using physically based hydrometeorological methods, including moisture and wind maximization, and contrasted these values with statistically obtained estimations typically used in consultancy. Secondly, to assess long-term safety, projected PMP values for the 21st century were calculated using data from four GCMs following SSP2-4.5, SSP3-7.0, and SSP5-8.5 climate projections with the same hydro-meteorological approach. Changes in the values of PMP throughout the century were analyzed through overlapping 30-year rolling windows over the period 2015-2100.
Preliminary results for the historical period reveal marked methodological discrepancies between physically based hydrometeorological and statistical methods. For example, while moisture maximization yields estimated values closely aligned with statistical baselines, the incorporation of wind maximization drives PMP values significantly higher, surpassing other methods by up to 78%. Furthermore, no convergence of trends is observed among the four sites in the near future (2015-2044). However, consistent upward trajectory in PMP becomes evident by the century’s end. This is most pronounced under high-emission scenarios, where estimates for the 2075–2100 period rise by 24% to 81% relative to the historical baseline.
Ultimately, these findings highlight that relying solely on historical statistics may significantly underestimate failure risks due to hydroclimatic extreme events. Ongoing efforts are focused on better understanding how changes in PMP propagate into PMF and how methodological decisions influence hydrological design.
How to cite: Baquedano, P., Gómez, T., Muñoz-Castro, E., and Vargas, X.: Assessing the risk of failure of tailings storage facilities due to changes in hydroclimatic stressors in a warming world - Case study of Chile, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-8124, https://doi.org/10.5194/egusphere-egu26-8124, 2026.
Extreme precipitation events (EPEs) are expected to increase in frequency and intensity under global warming and can trigger various impacts such as floods and landslides, which can undermine the socio-economic stability by raising the risk of loss of human lives, infrastructure failure and agricultural losses. Consequently, the study of hydroclimatic extremes has grown substantially in the recent decades, supported by high-resolution data and multivariate event-based analytical frameworks that improve understanding and resilience to climate-related risks. The rainfall induced impacts often show a compound nature because the underlying processes are scale-dependent and can overlap and intensify each another. Therefore it is important to consider extremeness across spatio-temporal scales when assessing EPEs. However, the the complex topography and diverse climatic conditions in the Indian Peninsular region pose a key challenge in assessing and characterizing the EPEs. In this context, a comprehensive ranked catalog of EPEs is developed from the 73 year long data set, based on their extremity across spatio-temporal scales. To increase the robustness of the underlying statistical analysis and to make an optimal use of the data, a combination of the peak-over-threshold (POT) method and the cross-scale weather extremity index (xWEI) is introduced to quantify the spatiotemporal extremity. In addition, the study exemplifies the applicability of POT method and compares the resulting extremeness with the conventional annual maxima approach. The catalog identifies EPEs that are jointly extreme across spatial and temporal scales and distinguishes short-lived localized storms from persistent, widespread events, thereby enabling a systematic characterization of EPE typologies. By linking each EPEs xWEI value to the season and meteorological divisions, the catalog offers a consistent basis for comparing historical events, and advances process-based understanding of regional hazard regimes. In summary, the resulting catalog can be a valuable tool in improving the robustness of quantitative risk assessments and enhancing the reliability of climate change attribution analyses.
How to cite: Ganapathiraju, S. A., Voit, P., Marwan, N., and Rathinasamy, M.: Ranked Multiscale Catalog of Precipitation Extremes using Cross Scale Extremity for the Indian Peninsular Region, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10790, https://doi.org/10.5194/egusphere-egu26-10790, 2026.
The southern African region is heavily impacted by climate change, which significantly alters water availability. The intensity and frequency of hydrological extremes, such as droughts and floods, have greatly increased in the past decades and will likely persist into the future due to projected rises in extreme precipitation and rising temperatures. In this regard, water resource management remains a major challenge in this region. Climate change impacts make it even more critical by transforming water security risks into substantial water insecurity and management challenges, especially for one of the key river systems in South Africa, the highly complex Integrated Vaal River System (IVRS). The IVRS involves inter-basin and transboundary water transfers (i.e., Lesotho Highlands) and is considered a lifeline for Gauteng Province’s water supply. The system faces the risk of a day-zero drought when water levels drop to around 20% or lower in the Vaal Dam, causing taps to run dry.
This study offers insights and prospects on how integrating advanced hydrological models with km-scale (i.e., 4km) high-resolution projected climate change data can help better understand and quantify the role of hydrological extremes in the IVRS.
Initial calibration at different gauging stations shows Kling-Gupta Efficiency (KGE) ranges between 0.60 and 0.70, and the Talsim hydrological model effectively captured seasonal flow and storage dynamics in the Vaal Dam. The storage volumes within the Vaal dam show approximately 8% deviation from observations when operational rules are excluded. The absence of operational rules is identified as the main limitation in current simulation runs. The future work will focus on integrating operational rules and long-term storage changes within the IVRS.
This research is part of the WaRisCo (Water Risks and Resilience in Urban-Rural Areas in Southern Africa - Co-Production of Hydro-Climate Services for Adaptive and Sustainable Disaster Risk Management) project, which is funded within the “Water Security in Africa – WASA” programme.
How to cite: Ismail, M. F., Biskop, S., Lohr, H., Weber, T., Engelbrecht, F., Maviza, A., Schaudt, D., Kralisch, S., Frisius, T., Malherbe, J., Moseki, C., and Ernst, Y.: Critical role of hydrological extreme events in future water security and management of the Integrated Vaal River System, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-10900, https://doi.org/10.5194/egusphere-egu26-10900, 2026.
Sub-daily precipitation extremes are a primary trigger of flash floods and debris flows in the Great Alpine Region, yet their future evolution is still uncertain, especially in relation to changes at the catchment scale. Recent work using convection-permitting ensembles has demonstrated added value and different change signals relative to the Regional Climate Models (RCMs) that drive them, but most analyses remain focused on grid-point indices. This study addresses this gap by focusing on areal rather than local precipitation. It provides a unified comparison of Convection Permitting Models (CPMs) and RCM projections of areal extremes, together with a temperature-scaling framework to provide a physical interpretation of the projected changes.
We use the CORDEX-FPS kilometer-scale CPMs and their driving regional climate models to assess changes in areal extreme precipitation between a historical (1996–2005) and far-future (2090–2099) period under the RCP8.5 emission scenario. We quantify projected changes in extreme precipitation across durations from sub-daily to daily and across spatial scales up to 5000 km². We directly compare the change signals from CPMs against those from their driving RCMs. To understand the physical mechanisms behind these changes, we analyse precipitation-temperature scaling relationships, diagnosing where they follow thermodynamic expectations (Clausius-Clapeyron-like scaling) versus where they deviate from those, pointing to more dynamical controls across spatial scales.
How to cite: Akbary, R., Dallan, E., Astagneau, P., Wood, R., Marra, F., Brunner, M., and Borga, M.: Future changes in sub-daily extreme areal precipitation and their temperature scaling in the Great Alpine Region, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-11355, https://doi.org/10.5194/egusphere-egu26-11355, 2026.
Early warning of drought in the southern Peruvian Andes could enable anticipatory action to reduce the economic, social and environmental impacts. However, accurate and timely drought prediction is inhibited by a complex hydroclimate across time and space, owing to mountainous topography and the influence of multiple interacting climate drivers. Current understanding of the mechanistic link between oceanic and atmospheric variability, and drought, is limited, and possible intra-annual variation in the driving mechanisms of drought remains unconstrained. In this study we explore the effects of large-scale climate variability on the dominant modes of atmospheric circulation over South America, and the subsequent influences on precipitation- and temperature-driven drought. We find that meteorological drought during the onset of the wet season occurs during La Niña, which inhibits the development of the Bolivian High. In contrast, during the peak and termination of the wet season, El Niño causes drought via a weakening and northeast shift of the Bolivian High. Propagation to soil moisture and vegetation drought occurs quickly and is broadly driven by these same driving mechanisms, although temperature variability becomes more influential than precipitation variability. Propagation is modulated locally by land cover heterogeneity; higher elevation grasslands are particularly vulnerable. Hydrological drought develops over longer timescales due to buffering by catchment-scale processes. We conclude that actionable early warning of drought in the southern Peruvian Andes must be localised in time and space to account for this complexity in drought driving mechanisms.
How to cite: Atkins, O., Milla, P., Lavado-Casimiro, W., Espinoza, J. C., and Buytaert, W.: Intra-annual variation in the driving mechanisms of drought in the southern Peruvian Andes, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-12498, https://doi.org/10.5194/egusphere-egu26-12498, 2026.
Event-based hydrologic modeling is typically governed by a fundamental trade-off: lumped models are straightforward to implement but neglect spatial variability, whereas fully distributed models require extensive parameterization, limiting their applicability. This study proposes a semi-distributed modeling framework coupled with data-driven parameter estimation, requiring minimal calibration. The studied basin is divided in sub-catchments, within which runoff generation is modeled using the Soil Conservation Service (SCS) Curve Number (CN) method. Basin-specific CN relationships are developed for November–April and May–October, and used to rescale subbasin CNII values, preserving spatial heterogeneity. The effective precipitation is transformed to direct-runoff using the SCS Unit Hydrograph. This approach avoids over-parameterization while maintaining spatial detail and consistent performance at ungauged locations. In a case study over the Housatonic River Basin, the model reproduces observed storm peak discharges without calibration and performs consistently across gauges. Systematic and random error components, as well as CN uncertainty, are quantified to assess their effects on the simulated peak discharges. The findings show that the proposed modeling framework is well-suited for basin-scale applications, including integration into infrastructure risk assessment models.
How to cite: Farmakis, C., Langousis, A., Anagnostou, E. N., and Emmanouil, S.: A Parsimonious Semi-Distributed Framework for Event-Based Runoff Modeling, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-14969, https://doi.org/10.5194/egusphere-egu26-14969, 2026.
The stationarity of rainfall extremes is increasingly challenged by a changing climate, necessitating a deeper understanding of both remote and regional atmospheric drivers. While traditional risk assessments for India often rely on global climate indices like the El Niño–Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD), these one-dimensional approaches often struggle with covariate multicollinearity and fail to capture interacting physical processes. This study explores the Principal Component Analysis (PCA) with wavelet coherence to evaluate the influence of nine climate indices on extreme monthly rainfall across peninsular India (1901–2021). By transforming correlated predictors into orthogonal joint modes, we found that while the primary modes of global climate variability account for nearly half of the total variance, their direct coherence with localized rainfall extremes remains weak and intermittent. In contrast, principal components dominated by regional thermodynamic indicators (specifically Integrated Vapor Transport (IVT) and local temperature anomalies) demonstrated the most persistent and statistically significant coherence, affecting over 80% of the study area. Furthermore, cross-correlation analysis revealed that while ENSO exhibits a 2–3 month lag, regional variables exert a contemporaneous influence on extreme events. Our findings suggest that the governance of rainfall extremes is shifting toward regional-scale processes. Consequently, we argue that for the development of non-stationary extreme value models, local covariates should be prioritized over remote teleconnections. In practical applications, high-resolution products from regional climate models, offer a more physically representative and contemporaneous basis for capturing the drivers of extreme events. This shift in covariate selection has critical implications for improving the accuracy of hydrological hazard assessments and infrastructure design in a non-stationary world.
How to cite: Dixit, S. and Pandey, K.: Assessing the Shifting Drivers of Rainfall Extremes in Peninsular India: From Remote Teleconnections to Regional Thermodynamics, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16072, https://doi.org/10.5194/egusphere-egu26-16072, 2026.
Multi-purpose reservoirs in Alpine regions must balance competing demands for flood protection, hydropower generation, and water supply. This requires robust flood risk assessment frameworks to support decision-making under uncertainty. To this end, the aim of this work is to develop an innovative copula-based approach to evaluate flood risk mitigation strategies for Alpine reservoirs by simulating compound events of flood peaks and volumes through Monte Carlo generation.
Bivariate copulas are fitted to observed flood peak discharges and corresponding event volumes extracted from streamflow data, and subsequently employed to generate Monte Carlo synthetic flood events for risk assessment. This enables estimation of conditional probabilities of flood volumes given fixed peak discharges, the key variable controlling available storage capacity and thus the reservoir's ability to mitigate subsequent flood events. The simulated scenarios allow systematic exploration of reservoir responses across diverse flood conditions, evaluating how different initial water levels and water release patterns affect downstream flood risk.
A key innovation of our framework is the operation-based definition of flood events rather than statistical percentiles: we use the maximum turbine discharge capacity as the minimum peak threshold, which varies across reservoirs based on their operational characteristics. This directly links the statistical analysis to management constraints. A minimum inter-event duration, determined through sensitivity analysis, distinguishes between multi-peaked events (where volume accumulates from successive peaks) and truly independent flood occurrences.
The framework provides a quantitative basis for optimizing risk-based trade-offs among multiple water uses, explicitly accounting for how stored volumes affect both flood protection and competing demands, enabling reservoir operators and local authorities to quantify flood risk under alternative water allocation scenarios.
How to cite: Avesani, D., Di Marco, N., Zambon, F., Majone, B., and Rizzi, G.: Flood risk assessment under different multi-purpose reservoir allocation strategies: an operational driven copula approach , EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16793, https://doi.org/10.5194/egusphere-egu26-16793, 2026.
Global warming has intensified the frequency and intensity of precipitation anomalies, resulting in extreme drought and wetness that severely affect ecosystems and society. While most existing studies often examine spatial or temporal aspects separately, few have treated these extremes as spatiotemporally contiguous events. Here, we analyze the distinct characteristics of spatiotemporally contiguous extreme drought and wetness events across China during 2001-2024, employing a three-dimensional perspective. The results show that since the 21st century, both extreme drought and wetness events have increased in duration. However, the spatial extent and intensity of drought events have decreased, whereas those of wetness events have expanded significantly. During the growing season, drought events tend to suppress vegetation growth in arid regions yet promote it in humid areas, whereas wetness events exhibit an opposite pattern. Moreover, drought events have detrimental impacts on forests, croplands, and grasslands, while wetness events benefit croplands and grasslands but continue to adversely impact forests. Our findings emphasize the necessity of studying extreme events from a three-dimensional spatiotemporal perspective.
How to cite: Su, R. and Li, Y.: Spatiotemporally Contiguous Extreme Drought and Wetness Events in China and their Impacts on Vegetation, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19283, https://doi.org/10.5194/egusphere-egu26-19283, 2026.
Several extreme value frameworks are available for modelling rainfall extremes. These include classical asymptotic approaches, such as the Generalised Extreme Value (GEV) distribution applied to annual maximum series, as well as more recently proposed non-asymptotic methods, namely the Metastatistical Extreme Value (MEV) and the Simplified MEV (SMEV) distributions applied to ordinary events. While the non-asymptotic frameworks have been evaluated at daily and hourly timescales, they have not yet been systematically evaluated at sub-hourly timescales across climatic regimes. As a result, it remains unclear whether relative differences in predictive performance observed at longer timescales extend to sub-hourly durations.
We compare the predictive performance of the GEV, MEV, and SMEV distributions using sub-hourly rain gauge observations from 2,810 stations across six European countries. We conduct a cross-validation experiment in which at-site distribution parameters are estimated from a training subset and used to predict the return level associated with the most extreme event in an independent test subset. Performance is quantified as the root mean square error between predicted return levels and observed extreme events, computed over 1,000 iterations per rain gauge and duration.
Results show systematic differences in relative predictive performance across durations and regions, with SMEV being favoured at short durations (up to 3 hours) for the majority of rain gauges, MEV at longer durations, and GEV being competitive for a non-negligible fraction of rain gauges. Overall, no framework consistently outperforms the others across countries and durations, indicating that superior predictive performance of any one extreme value framework cannot be assumed across space or timescales.
How to cite: Hansen, S. S., Löwe, R., Sørup, H. J. D., and Mikkelsen, P. S.: Extreme value frameworks for sub-hourly rainfall: comparison of predictive performance across Europe, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-19494, https://doi.org/10.5194/egusphere-egu26-19494, 2026.
Quantifying extreme precipitation is fundamental for effective flood risk management and climate change adaptation. This study seeks to advance the physical interpretation of extreme precipitation statistics by explicitly connecting the properties of statistical distributions to the characteristics of the underlying physical processes. High-temporal resolution observations from approximately 400 rain gauges and temperature stations distributed across the Alpine region are analyzed. Extreme precipitation depths are estimated for durations ranging from sub-hourly to daily, and for return periods of up to 100 years, using a non-asymptotic framework based on the duration maxima of independent meteorological events (storms). Key storm characteristics, such as peak and mean intensity, storm duration, temporal variability, temporal profile metrics, antecedent temperature, are derived and examined in relation to extreme precipitation statistics.
Preliminary findings reveal a strong dependence of extreme precipitation estimates on both topography and accumulation duration. At short timescales, extremes are more intense in lowland regions than in mountainous areas, indicating a reverse orographic effect, whereas the pre-Alpine zone exhibits larger extremes at longer durations. These spatial patterns are consistent with variations in the parameters governing storm intensity and tail behavior of the precipitation distributions. Distribution parameters exhibit weak to strong correlations with storm characteristics, varying across accumulation durations. At sub-hourly scales, the intensity and tail-heaviness parameters display opposite correlations with the same storm properties (that is, an antagonistic effect on return level estimates). Although at these durations the heavy storms are predominantly convective across the whole domain, our results indicate that local storm features play a key role in shaping the extreme precipitation distribution.
By exploring the links between storm structure and extreme precipitation statistics, this work contributes to a more robust characterization and improved prediction of precipitation extremes.
This study was carried out within the RETURN Extended Partnership and received funding from the European Union Next-GenerationEU (National Recovery and Resilience Plan – NRRP, Mission 4, Component 2, Investment 1.3 – D.D. 1243 2/8/2022, PE0000005).
How to cite: Dallan, E.: Storm-scale characteristics governing extreme precipitation statistics in an Alpine region, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-20115, https://doi.org/10.5194/egusphere-egu26-20115, 2026.
Climate change is reshaping the statistical characteristics of precipitation, leading to an increased probability of precipitation events that exceed the range of historically observed extremes. Such ultra-extreme precipitation events are exceedingly rare with limited representation in observational record. However, they pose substantial risks to urban systems and hydrologic infrastructure. The scarcity of observations makes it challenging to robustly quantify their frequency and intensity, constraining the scientific basis for climate risk assessment and long-term adaptation planning. To address these challenges, we propose a statistical framework to characterize ultra-extreme precipitation by integrating observational records and climate model projections. The probability of ultra-extreme precipitation events is estimated at each station by counting the number of occurrences with a standardized deviation from the station mean that exceeds a specified threshold. These exceedances are divided by the total number of observations to derive the regional probability of exceedance. In order to evaluate changes under future climate, daily precipitation from Coupled Model Intercomparison Project Phase 6 (CMIP6) models is statistically downscaled to individual station using observation-based quantile mapping. This ensures consistency between modeled and observed precipitation distributions. The framework is applied to approximately 200 global major cities with populations exceeding one million and Gross Domestic Product (GDP) over 100 billion USD. Using this framework, we evaluate changes in ultra-extreme precipitation characteristics between historical and future climate conditions. We expect this framework to facilitate the analysis of spatial and temporal patterns of ultra-extreme precipitation and their potential changes in future. The framework further supports the interpretation of rare but high-impact precipitation events and provides insights for urban flood risk management. Therefore, this study contributes to the development of hydrologic infrastructure design and adaptation strategies that are robust to increasing precipitation extremes under climate change.
Acknowledgment
This work was supported by Korea Environment Industry & Technology Institute(KEITI) through R&D Program for Innovative Flood Protection Technologies against Climate Crisis Program(or Project), funded by Korea Ministry of Environment(MOE)(RS-2023-00218873).
How to cite: Jeong, Y., Hamlet, A., and Byun, K.: Global Characteristics of Ultra-Extreme Precipitation in Major Cities, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-21628, https://doi.org/10.5194/egusphere-egu26-21628, 2026.
The RAINSIM stochastic daily rainfield generator is based on weather pattern sub-sampling and meta-gaussian models (Ayar et al., 2020). RAINSIM is coupled with an air temperature generator to feed a a distributed hydrological model, allowing to simulate large hydrological chronicles for extreme estimation (both floods and low flows). A large-scale application of RAINSIM to the whole French continental territory (about 500 000 km²) is presented here.
The parameters of the statistical models (both at-site distributions and temporal and spatial covariances) are infered from observed precipitation data at stations, sub-sampled into subsets by seasons and weather types. Before the rainfield generation, sequences of weather types are generated by a Markov model. Here the seasonal transition matrixes are conditionned to observed large-scale climatic indexes such as NAO and WeMO. This conditionning allows a better representation of the year-to-year and decadal variabilities.
The presented application challenges a key assumption of RAINSIM: the stationarity of the spatial covariance. At the French scale, the diversity of climatology and of the spatial structures of rain fields are significant, thus questioning this hypothesis. To tackle this, an approach based on the deformation of the geographical space (Monestiez et al., 2007) has been tested, thanks to its implementation in the deform R-package (Youngman, 2023). The deformations are computed independently for each subset, illustrating that the spatial covariance structure of the rain fields depends on the weather, and to a lesser extend to the season. Comparisons to observed data with suitable metrics are presented to score this use of covariance-oriented deformations of space.
Perspectives and first developments for application in projected climate are also evoked.
References:
Ayar, P. V., Blanchet, J., Paquet, E., & Penot, D. (2020). Space-time simulation of precipitation based on weather pattern sub-sampling and meta-Gaussian model. Journal of Hydrology, 581, 124451.
Monestiez, P., Meiring, W., Sampson, P. D., & Guttorp, P. (2007). Modelling Non‐Stationary Spatial Covariance Structure from Space—Time Monitoring Data. In Ciba Foundation Symposium 210‐Precision Agriculture: Spatial and Temporal Variability of Environmental Quality: Precision Agriculture: Spatial and Temporal Variability of Environmental Quality: Ciba Foundation Symposium 210 (pp. 38-51). Chichester, UK: John Wiley & Sons, Ltd..
Youngman, B. D. (2023). deform: An R Package for Nonstationary Spatial Gaussian Process Models by Deformations and Dimension Expansion. arXiv preprint arXiv:2311.05272.
How to cite: Paquet, E.: A meta-Gaussian stochastic rainfield generator for France, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9156, https://doi.org/10.5194/egusphere-egu26-9156, 2026.
Posters virtual: Tue, 5 May, 14:00–18:00 | vPoster spot A
EGU26-1058 | ECS | Posters virtual | VPS8
A Statistical Methodology for Regional Scale Future Projection of the Seasonal Frequency of Sub-daily Extreme Rainfall EventsTue, 05 May, 14:42–14:45 (CEST) vPoster spot A
Sub-daily extreme rainfall (SDER) events frequently lead to natural disasters, including flash floods, urban floods, landslides, and soil erosion. It is essential to make a reliable prediction of its frequency at the local/regional spatial scale for the future period, in order to devise improved disaster mitigation and adaptation strategies. It has been observed that current CMIP6 GCMs have limitations in simulating short-duration (sub-daily) heavy rainfall events, and large biases are often observed in the control run simulations compared to historical observations at various locations worldwide. Hence, there is a lack of confidence in considering the crucial future projections obtained from those GCMs as reliable. In this study, we present a novel statistical methodology for predicting the seasonal frequency of SDER for 99 river sub-basins (RSBs) in India, encompassing tropical, temperate, arid, and polar climates across various topographies. The methodology identifies the scaling relationship between the SDER frequency and the associated potential atmospheric variables/drivers for each RSB. Results indicated that the seasonal frequency of SDER scales with (i) near-surface air temperature (SAT), and (ii) moisture content in the air, which is measured by near-surface dew-point temperature (DPT). The scaling relationship exhibits an increasing (scaling) phase followed by a decreasing (reverse scaling) phase as the (dew point) temperature increases. The range of SAT and DPT in the scaling relationship varies with RSB and climate. The SAT and DPT values at peak frequency are high for mountainous areas and lower for non-mountainous areas. The effectiveness of those scaling relationships in predicting SDER frequency at the seasonal scale was assessed/validated for the recent past (1981-2020). The method performed fairly well for RSBs with non-mountainous topography and moderately well for RSBs with mountainous topography across climate zones, except for years with an abnormally high or low SDER frequency. A finer spatial-resolution scaling relationship is deemed necessary for mountainous topographies where SDER exhibits a rather local nature. In addition, the time trends in simulated and observed frequencies closely matched. The proposed methodology is applied to predict the future seasonal frequency of SDER in the RSBs for different SSP climate scenarios till the end of the twenty-first century. The performance of various GCMs in projecting the seasonal frequency of SDER is also evaluated.
How to cite: Varshney, A. and Srinivas, V. V.: A Statistical Methodology for Regional Scale Future Projection of the Seasonal Frequency of Sub-daily Extreme Rainfall Events, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-1058, https://doi.org/10.5194/egusphere-egu26-1058, 2026.
