Extreme precipitation events are critical phenomena that pose significant risks to societies. Understanding their spatial patterns and drivers is vital for both scientific and practical purposes, such as risk management strategies. This study investigates the spatial correlation of extreme precipitation events across Europe, their modulation by the North Atlantic Oscillation (NAO), and the detection of potential changes over recent years. Additionally, it evaluates whether these patterns and trends are accurately replicated by tools commonly employed in the insurance industry.

Correlations between extreme precipitation events across European regions are investigated. In addition, the NAO, which is a dominant mode of atmospheric variability in the North Atlantic, is widely considered to be a significant modulator of these spatial patterns. Positive phases of the NAO are associated with intensified extreme precipitation in northern and western Europe, while negative phases shift these patterns towards southern Europe. By coupling precipitation data with NAO indices, we demonstrate how changes in NAO phases alter the spatial coherence and intensity of extreme events. Furthermore, a critical aspect of this study is comparing these patterns and trends with the tools and methods used in the insurance industry.

This study contributes to a better understanding of extreme precipitation dynamics in Europe, offering insights for practical applications in risk management. By highlighting gaps in current approaches, it underscores the need for integrating advanced climate diagnostics into risk assessment frameworks.

How to cite: Tsaknias, D.: Spatial patterns of extreme precipitation in Europe, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-15034, https://doi.org/10.5194/egusphere-egu25-15034, 2025.

The assessment of extreme precipitation statistics is essential for managing flood hazards and developing effective climate change adaptation strategies. These design values are typically estimated through the frequency analysis of precipitation data, with limited understanding of their generative atmospheric phenomena. We aim to go beyond the statistical extrapolation of observed extremes with extreme value distributions towards enhancing their physical comprehension: this may be beneficial for improving our estimates of extreme precipitation probability and our predictions of future changes. Our analysis is based on a network of ∼300 rain gauges and temperature stations in a complex-orography region of the Alps. We estimate the magnitude of extreme precipitation from sub-hourly to daily durations for return periods up to 100 years (1% annual exceedance probability). We employ a non-asymptotic extreme value approach based on the concept of storms (independent meteorological objects) and ordinary events (duration maxima within each storm). We focus on the ordinary events exceeding high percentiles (e.g., 85^th, 90^th, 95^th) at some duration, and we extract several characteristics of the corresponding storms, such as the event peak and average intensity, total lifetime, seasonality, temporal profile, peakedness, temperature, etc. We then assess their relationships with the parameters of our non-asymptotic extreme value model.

Our preliminary results show that variations in the model parameters depend on topography and event duration. Heavier tails in the extreme precipitation distribution emerge at sub-hourly durations in mountainous regions and for parts of the lowlands, but at longer durations in the pre-Alps. The scale parameter is generally higher in the lowlands and the pre-Alps. As a result, extreme precipitation intensity for short duration is generally higher in the lowlands than in the mountains (“reverse orographic effect”), with higher intensities in the pre-Alps at longer durations. Storm characteristics also vary with topography, precipitation duration, and event extremeness. In summer, front-loaded storms are prevalent at short durations, where heavier tails are observed. In the pre-Alps, storms are characterized by the highest extremes at long durations, have a more symmetric temporal profile, are most common in autumn, and have a longer total lifetime compared to the rest of the region.

Further investigation is needed to clarify the relationship between storm characteristics and statistical properties. This work enhances understanding of the key processes shaping precipitation extremes and provides insights for improving predictive models, ultimately aiding in risk assessment and climate resilience planning.

This study is 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., Marra, F., Papacharalampous, G., Fowler, H. J., and Borga, M.: Can the climatology of heavy storm characteristics explain extreme precipitation statistics?, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5069, https://doi.org/10.5194/egusphere-egu25-5069, 2025.

Extreme precipitation events have become increasingly frequent and intense in recent decades, resulting in severe flooding and substantial socio-economic losses. These events are typically associated with intense weather systems that vary across numerous meteorological factors and exhibit significant temporal and spatial variability. A comprehensive understanding of the underlying processes and the identification of key meteorological factors driving extreme precipitation are critical for enhancing the accuracy of extreme rainstorm predictions and flood warnings.

This study utilized cumulative distribution function (CDF) analysis based on ERA5 hourly reanalysis data and employed the eXtreme Gradient Boosting (XGBoost) algorithm to identify the key meteorological factors contributing to 24-hour extreme precipitation across three distinct climatic zones in China. Additionally, forecasting models were developed to predict these events. The results highlighted the efficacy of this methodology and demonstrated its ability to achieve the following key advancements:

• Mapping data into the CDF space effectively addressed the challenges posed by the spatial heterogeneity in the value ranges of meteorological factors in regional system analyses, thereby significantly enhancing the spatial scalability of the predictive model.
• The integration of SHAP (SHapley Additive exPlanations) value interpretation with XGBoost successfully identified the critical meteorological factors influencing extreme precipitation events. This facilitated the construction of classification and regression models to predict both the occurrence and the return periods of these events.
• The application of SHAP values enhanced the interpretability of the "black-box" XGBoost model by incorporating physical insights and elucidating the interactions between different factors, thus providing valuable information for the construction and refinement of the final model.

In summary, this study presents a novel and interpretable machine learning framework for analyzing and predicting extreme precipitation events based on the CDF analysis. By effectively addressing spatial heterogeneity and enhancing model interpretability, the proposed methodology offers significant advancements in the prediction of extreme rainfall and associated flood risks, contributing to improved disaster preparedness and mitigation efforts.

How to cite: Wu, X., Jiang, Z., and Sharma, A.: Predicting extreme precipitation events using machine learning techniques based on cumulative distribution function (CDF) analysis of meteorological factors, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-13959, https://doi.org/10.5194/egusphere-egu25-13959, 2025.

Flooding represents a growing concern for the (re)insurance industry, with precipitation extremes as a key driver of flood risk. Some of the most destructive flood events in 2024 were driven by extreme rainfall occurrences, although with important differences in spatial and temporal scales (e.g. Dubai floods, Ex-Hurricane Debby floods in Canada, Central Europe Floods, Hurricane Helene flooding in Georgia and North Carolina, Valencia floods).

Ongoing climate trends introduce additional uncertainty in the estimates of intensity, frequency, and distribution of rainfall extremes, complicating their quantification and risk assessment. Understanding and modelling these extremes is critical for improving flood risk management and financial preparedness.

This study investigates rainfall extremes in the United States across various temporal scales, focusing on their role in different types of flood risks. We compare multiple statistical models to estimate extreme precipitation values, including approaches that incorporate climate trends. By analysing spatial and temporal patterns of extremes, we evaluate how well these models capture underlying processes and improve predictive accuracy.

Our findings suggest that integrating additional information about climate trends and hydrometeorological processes enhances the accuracy of extreme rainfall estimates, moving in the right direction, although given the rare nature of these extremes looking at historical data alone leaves space for future unexpected outcomes. These results provide valuable insights for improving catastrophe models and stress-testing (re)insurance portfolios.

How to cite: Nicotina, L., Jewson, S., Petrie, R., Cox, T., and Ball, P.: Rainfall Extremes in a Changing Climate: Implications for Flood Risk and (Re)Insurance, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-18938, https://doi.org/10.5194/egusphere-egu25-18938, 2025.

Hydrological extremes pose significant challenges to flood risk assessment and mitigation, particularly under non-stationary climatic and hydrological conditions. Hydraulic structures, such as large reservoirs, modify flood distributions by attenuating peak flows, with their effectiveness varying over return periods. This variability introduces non-stationarity in flood frequencies and has a significant impact on the tail of the distribution. As a result, data-driven approaches to flood frequency estimation can lead to under- or overestimation of flood quantiles, especially when limited observations are available. To address these challenges, we propose an analytical framework capable of defining the full probability distribution of floods at a control section. This method explicitly incorporates key physical processes, including the influence of reservoir volume, non-linear spillway behavior and threshold discharge on inflow hydrographs. The accuracy of the estimations is demonstrated by comparisons with numerical simulations of reservoir routing using the continuity equation in a real case study. Our results highlight the critical role of integrating physical processes into flood modelling to capture tail behavior, and show how statistical approaches applied to small samples of flood peak observations can instead lead to significant biases. The proposed analytical solution provides a robust and parsimonious tool for estimating the impact of reservoirs on floods, with applications in both risk assessment and infrastructure planning.

How to cite: Cipollini, S., Volpi, E., Vorogushyn, S., and Fiori, A.: Estimation of flood peak distributions considering reservoir effects on tail behavior, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-20629, https://doi.org/10.5194/egusphere-egu25-20629, 2025.

Global climate change poses critical challenges to food security and market stability as extreme weather events become increasingly frequent and severe. The combined effects of compound extreme events and global trade dynamics on food security, however, remain insufficiently explored. Here, we employed a copula-based statistical approach, integrating international trade data to estimate maize yield failures under compound drought and heat events (CDHEs) and to assess how global trade dependence and supply diversity impact food security under such stressors. Our findings reveal a 70.1% probability of global maize yield failure as CDHE intensity increases, with key breadbasket regions, including Northeast China, Europe, North America, Latin America, and South Africa, particularly vulnerable. Both drought and heat events contribute similarly to global maize yield risk; however, regional desynchronies, such as distinct effects in China and Brazil, highlight differing vulnerabilities. Furthermore, countries heavily dependent on imports from regions with high yield failure risk, such as Vietnam and Colombia, face an increased probability of maize yield failure exceeding 40%. Conversely, supply diversity offers a modest buffering effect, mitigating some adverse impacts of CDHEs, albeit with notable uncertainties. Our findings underscore the compounded vulnerability of maize yields to CDHEs, intensified by trade dependencies, while highlighting the potential for supply diversification to enhance resilience. Urgent adaptations, transformative strategies, and policy interventions are critical to mitigate cascading risks within the global food system, bolster resilience to climate change, and ensure food security.

How to cite: Liu, S., Shi, T., Zhang, W., Li, T., Wang, Z., and Ma, X.: Balancing risks: How global trade dependence exacerbates and supply diversity mitigates yield failures under compound drought and heat events, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-732, https://doi.org/10.5194/egusphere-egu25-732, 2025.

Chile is one of the most vulnerable countries to the impacts of climate change. This suggests challenges to mitigate their impacts and adapt existing infrastructure. Despite a consensus that future climate change will lead to an increase in hydrometeorological extremes and the importance of including this factor in the hydrological design of hydraulic infrastructure, clear national guidelines on how to achieve and implement this in practice are still lacking. To address this gap, this study aims to align national hydrological design methodologies with international best practices by offering recommendations for addressing extreme precipitation and surface runoff generation.

Key considerations include the temporal and spatial scales of precipitation and temperature, and methodologies for flow estimation in gauged and ungauged basins. Dynamic modeling, statistical methods, and synthetic unit hydrograph approaches are explored, with applied examples highlighting the integration of climate change in estimating peak flows, extreme precipitation, and intensity-duration-frequency (IDF) curves.

Our results show that dynamic hydrological modeling yields projections with lower associated uncertainty by accurately capturing historical patterns. Dynamic models account for interactions such as antecedent soil moisture and snowline shifts during extreme events. For northern Chile, spatially distributed or semi-distributed models are recommended to capture the heterogeneity of extreme events. In contrast, statistical and synthetic hydrograph methods present limitations due to their reliance on historical precipitation-runoff relationships and lack of spatial heterogeneity.

Finally, the study underscores the need for flexible, transdisciplinary approaches to address future climate challenges, advocating for hydrological system modeling and a deeper understanding of processes driving extreme hydrometeorological responses.

How to cite: Vargas, X., Muñoz-Castro, E., Jorquera, J., Muñoz-Castro, O., Ricchetti, F., and Gómez, T.: Hydrological design of hydraulic infrastructure in a changing climate – Insights for practitioners in Chile, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-18901, https://doi.org/10.5194/egusphere-egu25-18901, 2025.