UP3.8 | From Global Large Ensembles to Regional Climate Models: leveraging models and AI to assess Climate Change Impacts on Extratropical Storms and Precipitation Extremes
From Global Large Ensembles to Regional Climate Models: leveraging models and AI to assess Climate Change Impacts on Extratropical Storms and Precipitation Extremes
Conveners: Giuseppe Zappa, Salvatore Pascale | Co-convener: Magdalena Mittermeier
Orals Wed1
| Wed, 10 Sep, 09:00–10:30 (CEST)
 
Room E1+E2
Posters P-Thu
| Attendance Thu, 11 Sep, 16:00–17:15 (CEST) | Display Wed, 10 Sep, 08:00–Fri, 12 Sep, 13:00
 
Grand Hall, P105–107
Orals Wed1 |
Wed, 09:00
Posters P-Thu
Thu, 16:00
Extratropical cyclones (ETCs), cut-off lows and mesoscale cyclones play a significant role in driving precipitation extremes and floods across midlatitude regions, with profound implications for hydrology, infrastructure, and climate risk management. While an increase in the intensity of precipitation extremes is generally expected because of thermodynamic effects of climate change, for adaptation purposes it is important to understand the frequency and magnitude of the ongoing changes and the possible amplitude of future worst case events. On one hand, large-scale internal climate variability modulates the occurrence of high-impact cyclones, while on the other hand resolving small-scale convective-scale processes is essential to reproduce precipitation extremes. From a modelling perspective this proves to be a challenge, since large initial-condition ensembles of convective-permitting simulations are currently not feasible.

This session will explore cutting-edge approaches to address this challenge, by gathering contributions showing the value of distilling and combining climate information from diverse datasets and modelling approaches, including: present and future simulations from single model initial-condition large ensemble (SMILEs), regional climate models including convective-permitting simulations (e.g. CORDEX) and regional large ensembles (regional SMILEs), UNSEEN approaches, statistical emulators to combine global and regional models, and the application of AI methods in this research field (e.g. downscaling large ensembles, AI-based precipitation modelling, detection of ETCs in climate models).

We particularly encourage contributions addressing the following topics:

- Novel approaches, e.g. dynamical-statistical emulators or AI-based methods, to combine information from global models with high-resolution regional climate simulations.
- Storylines of worst case extratropical storms due to the combination of climate change and internal variability.
- Projected changes in mid-latitude storm tracks, ETC frequency, intensity, and speed and their influence on extreme precipitation events.
- The added value from regional and convective-permitting simulations in quantifying future trends in extreme precipitation compared to global climate models.
- The role of climate change in intensifying recent European extreme precipitation events.
- Emerging insights from digital twins, e.g. Destination Earth.

The overall aim of the session is to foster discussion and interaction across global and regional modelling research communities to improve regional climate risk assessment.

Orals: Wed, 10 Sep, 09:00–10:30 | Room E1+E2

Chairpersons: Giuseppe Zappa, Magdalena Mittermeier, Salvatore Pascale
09:00–09:15
|
EMS2025-71
|
Online presentation
Machine Learning Emulation of Precipitation from km-scale Regional Climate Simulations using a Diffusion Model 
Henry Addison, Elizabeth Kendon, Suman Ravuri, Laurence Aitchison, and Peter Watson

Dynamical downscaling of climate simulations to local scales is valuable for understanding climate change impacts and planning adaptation measures, but is very computationally expensive. We present CPMGEM (Convection-Permitting Model Generative EMulator) [1]: a novel application of a generative machine learning model, a diffusion model, that skilfully emulates precipitation simulations by Met Office’s UK convection-permitting model (CPM). This achieves similar results to dynamical downscaling at a fraction of the computational cost. This emulator enables stochastic generation of high-resolution (8.8km) daily-mean precipitation samples, fine enough for use in applications such as flood modelling, conditioned on coarse (60km) weather states from a general circulation model (GCM).

We trained the emulator to produce output over England and Wales, using Met Office simulations from the United Kingdom Climate Projections (UKCP) Local product, covering 1980-2080. The output precipitation has a similarly realistic spatial structure and intensity distribution to the CPM simulations. Our generative emulator outputs a well-calibrated spread of predictions and reproduces the small-scale structure and frequency of extreme intensities better than a deterministic model. We also find evidence that the emulator captures the main features of the CPM-simulated 21st century climate change but exhibits some error in the magnitude.

Potential applications include producing high-resolution precipitation predictions for large-ensemble climate simulations and predictions using inputs from different GCMs and scenarios to better sample uncertainty. We will also discuss extending the emulator to predict sub-daily precipitation by downscaling temporally as well as spatially.

[1] Addison, H., Kendon, E. J., Ravuri, S., Aitchison, L. and Watson, P.A.G., 2024. Machine learning emulation of precipitation from km-scale regional climate simulations using a diffusion model. arXiv preprint arXiv:2407.14158

How to cite: Addison, H., Kendon, E., Ravuri, S., Aitchison, L., and Watson, P.: Machine Learning Emulation of Precipitation from km-scale Regional Climate Simulations using a Diffusion Model, EMS Annual Meeting 2025, Ljubljana, Slovenia, 7–12 Sep 2025, EMS2025-71, https://doi.org/10.5194/ems2025-71, 2025.

Show EMS2025-71 recording (16min) recording
09:15–09:30
|
EMS2025-528
|
Onsite presentation
Storylines of unprecedented flood risk using reforecast ensemble pooling 
Elena Bianco, Agostino Manzato, Giuseppe Zappa, Paolo Davini, Antonio Giordani, and Paolo Ruggieri

Assessing the frequency and impact of large-scale flood events is crucial for improving risk preparedness and emergency management. This effort, however, places emphasis on the need for large data samples to derive statistically robust probabilistic risk estimates. Recent studies have demonstrated the potential of using ensemble climate simulations to drastically extend the sample size of extreme events with respect to the relatively short historical record. This approach has become known as UNSEEN (UNprecedented Simulated Extremes using ENsembles) and has been successfully applied to different types of hazards. In this study, we use ensemble simulations from Copernicus C3S – specifically, seasonal reforecasts from the European Flood Awareness System (EFAS) and ECMWF SEAS5 – to generate a catalogue of flood events that are unprecedented in the historical record, yet plausible in the current climate. We focus on two target regions with recent episodes of severe flooding: The Panaro river basin in northern Italy and the Turia river basin in eastern Spain. We then employ the probabilistic event-based impact model CLIMADA (Aznar-Siguan et al. 2024) to rank the UNSEEN flood events in both basins by their estimated socio-economic impact and develop storylines of unprecedented flood risk. We find that high-impact UNSEEN events generally display recurrent synoptic patterns, similar to those of historical extremes but with higher intensity. Our results suggest that seasonal reforecasts by C3S simulate realistic extremes with respect to historical data following minimal bias correction. This framework can be leveraged to explore risk storylines for stress-testing, and to support the development of mitigation and adaptation strategies to manage disasters.

How to cite: Bianco, E., Manzato, A., Zappa, G., Davini, P., Giordani, A., and Ruggieri, P.: Storylines of unprecedented flood risk using reforecast ensemble pooling , EMS Annual Meeting 2025, Ljubljana, Slovenia, 7–12 Sep 2025, EMS2025-528, https://doi.org/10.5194/ems2025-528, 2025.

Show EMS2025-528 recording (14min) recording
09:30–09:45
|
EMS2025-599
|
Onsite presentation
Exploring extreme floods in two Italian watersheds through unseen ensemble scenarios  
Agostino Manzato, Elena Bianco, Giuseppe Zappa, Paolo Davini, and Paolo Ruggieri

Panàro and Reno are two rivers (165 and 212km long, respectively) having adjacent mid-sized watersheds (covering areas of 1775 and 4828km2, respectively) in N-Italy. Given their spatial proximity they are expected to behave similarly, particularly in response to meteorological forcing leading to extreme floods. Three different river discharge (RD) datasets are analyzed: 1) the historical daily-mean RD observed timeseries by ARPAE in the stations of Bomporto (Panàro) and Casalecchio (Reno); 2) the EFAS5 historical simulations of daily-mean RD, computed using the LISFLOOD hydrological model forced with the EMO1 dataset (1992-2023); and 3) the EFAS seasonal-reforecast daily-mean RD dataset, computed by LISFLOOD forced by the 25-member SEAS5 seasonal-reforecast ensemble (2000-2023).  

The UNSEEN approach (UNprecedented Simulated Extremes using ENsembles) uses many simulations to identify “plausible” extremes. Building upon this approach, an ensemble of 100 “surrogate” RD timeseries is built by concatenating 3-month blocks from the original 25-SEAS5-EFAS members, but starting from four different initial dates (beginning of April, May, June, and July). To ensure statistical independence, the first four months of each timeseries are discarded.  
Watershed are compared by inspecting  quantiles, interannual trends, and annual cycles. It is found that -without any bias correction- the 100 surrogate timeseries at the Bomporto can produce extremes RD higher than those seen in the historical EFAS dataset: conversely, this is not observed for the Casalecchio. The surrogate series show a mostly bell-shaped distribution centered around zero for the slope of the interannual trend, while both the historical EFAS timeseries and the ARPAE observations show a positive trend in recent years. Finally, with respect to the annual cycle, autumnal extremes appear to be underestimated in the surrogate time series, particularly at Casalecchio. 

Possible interpretations for such behaviors are explored by examining the meteorological conditions associated with the most extreme floods, derived from ERA5 for the historical EFAS and from the corresponding simulations of SEAS5 for the surrogate. We find that the meteorological conditions leading to floods in Bomporto and Casalecchio stations (only 33 km apart) are different. For instance, the most extreme events at Bomporto are typically associated with deep trough over the Mediterranean or a cyclone over Central-Italy, while those at Casalecchio seems to be linked to a more zonal flow and a stronger Atlantic depression. Additionally, the frequency of synoptic configurations associated with extreme RD is studied in both the historical and surrogate datasets, to assess whether the surrogate simulations reveal unseen patterns not present in the historical record. We suggest the UNSEEN approach can be a powerful tool to compare the amplitude of climate signals to internal climate variability and highlight the differences in the characteristics of floods in nearby catchments.

How to cite: Manzato, A., Bianco, E., Zappa, G., Davini, P., and Ruggieri, P.: Exploring extreme floods in two Italian watersheds through unseen ensemble scenarios , EMS Annual Meeting 2025, Ljubljana, Slovenia, 7–12 Sep 2025, EMS2025-599, https://doi.org/10.5194/ems2025-599, 2025.

Show EMS2025-599 recording (14min) recording
09:45–10:00
|
EMS2025-187
|
Onsite presentation
¿Can regional climate models reduce global model large-scale biases? A climate ensemble approach for downscaling cut-off lows using the HCLIM model 
Ignacio Prieto Rico, Juan Carlos Sánchez Perrino, and Esteban Rodríguez Guisado

Dynamical downscaling with regional climate models (RCMs) may inherit biases from the boundary conditions of a reanalysis or a general circulation model (GCM), leading to flawed results in the higher-resolution output. Typical examples of large-scale biases in CMIP5/6 models are the “too equatorial and too zonal” North Atlantic storm track (Schemm, 2023) or the bias in location and frequency of cut-off lows (COLs) (Pinheiro et al., 2022). These large-scale deviations may be even more noticeable when the geographical domain of interest is small, such as islands (Adinolfi et al., 2025). Even when using reanalysis as boundary conditions, some events may be lost after downscaling (Lavin-Gullon et al., 2021).

Although the primary objective of an RCM is to develop the smaller scales, if the RCM integration domain is large enough it can resemble the behaviour of a global model and the finer resolution potentially may correct some of the large-scale bias (Diaconescu & Laprise, 2013)

This study focuses on the optimal configuration for simulating COL events in convection-permitting scales using ensembles. The geographical area of interest is Southern Europe -including the Mediterranean and the eastern Atlantic coast-, one of the three areas of higher COLs occurrence worldwide (Nieto et al., 2005). Frequently the presence of a COL is related to high precipitation, accounting for about 80% of the extreme rainfall in the Valencia region of Spain (Nieto, 2021).

A convection-permitting resolution experiment is proposed, using the habitual double nested approach in RCMs, but defining an intermediate domain size appropiate to capture the possible weather regimes in an ensemble. The ensemble consists of several runs with different initial conditions, each lagged by one day. Results are shown for some COL specific events over Canary Islands and Iberian Peninsula using the regional model HCLIM (Belušić et al., 2020), demonstrating that in certain cases this approach can better reproduce an event that is not properly captured by the large-scale model.

Although a climate ensemble approach requires more computational resources, it can be appropriate for shorter runs, as in certain pseudo global warming approaches or event-based downscaling. It can also be used for longer simulations by taking advantage of hybrid techniques: physical models for the intermediate runs (that require relatively low HPC resources) and ML/AI emulators for the convection-permitting resolution.

How to cite: Prieto Rico, I., Sánchez Perrino, J. C., and Rodríguez Guisado, E.: ¿Can regional climate models reduce global model large-scale biases? A climate ensemble approach for downscaling cut-off lows using the HCLIM model, EMS Annual Meeting 2025, Ljubljana, Slovenia, 7–12 Sep 2025, EMS2025-187, https://doi.org/10.5194/ems2025-187, 2025.

Show EMS2025-187 recording (13min) recording
10:00–10:15
|
EMS2025-29
|
Onsite presentation
Unravelling drivers of the future Mediterranean precipitation paradox during cyclones 
Marco Chericoni, Giorgia Fosser, Alessandro Anav, Marco Gaetani, and Emmanouil Flaounas

In recent years, the Mediterranean region has experienced intense cyclones with heavy precipitation, resulting in severe flooding with multiple fatalities and significant damage to infrastructures. Additionally, the basin is increasingly recognized as a climate change hotspot. For these reasons, it is fundamental to understand how Mediterranean cyclones respond to climate change, identifying the key processes driving these changes.

Our study is the first combining CMIP6 models with a high-resolution atmosphere-ocean coupled regional climate model (AORCM) over the Mediterranean to investigate how changes in moisture advection and surface diabatic processes influence precipitation during intense cyclones under three SSP scenarios (SSP5-8.5, SSP2-4.5, SSP1-2.6). The AORCM is the only available with three SSP scenarios for the Mediterranean basin and it is essential for identifying the physical mechanisms driving cyclone-related precipitation changes. Despite a strong reduction in the number of intense cyclones and a subsequent decline in seasonal precipitation, our results underscore a significant increase in cyclone-related extreme precipitation, especially under the SSP5-8.5 scenario. The AORCM shows that this intensification is driven by enhanced moisture exchange from the ocean to the atmosphere and increased mid-level moisture transport toward the coastal regions of Southern France, Italy, and the Balkans.

These results offer valuable information for regional climate impact assessments in the Mediterranean basin and provide novel insights into the physical processes driving precipitation changes. The analysis underscores the potential risk of more damaging inland flooding in a warming climate, particularly in southern Spain and France (including their islands), the Italian Peninsula, the Balkan region and the Levantine Coast, posing severe issues for such densely populated areas. Besides, our results highlight the importance of combing different models within coordinated frameworks to disentangle the influences of large-scale forcings and regional climate processes on the future Mediterranean climate under varying radiative forcing levels. This approach is crucial for improving confidence in climate projections.

How to cite: Chericoni, M., Fosser, G., Anav, A., Gaetani, M., and Flaounas, E.: Unravelling drivers of the future Mediterranean precipitation paradox during cyclones, EMS Annual Meeting 2025, Ljubljana, Slovenia, 7–12 Sep 2025, EMS2025-29, https://doi.org/10.5194/ems2025-29, 2025.

Show EMS2025-29 recording (13min) recording
10:15–10:30
|
EMS2025-453
|
Online presentation
Understanding future projections for a high-impact Mediterranean storm track 
Paolo Ghinassi, Giuseppe Zappa, Federico Grazzini, Cristina Iacomino, Salvatore Pascale, Alice Portal, and Claudia Simolo

Mediterranean cyclones often cause severe damage due to extreme winds and intense precipitation. Two notable examples are the November 1966 storm that flooded Florence and Storm Adrian in October 2018. These cyclones caused significant damage in central-northern Italy and share common characteristics in their mature stages, including a south-to-north trajectory, rapid intensification, and the presence of a deep upper-level precursor trough.

In this study, we identify cyclones with similar dynamics using reanalysis data (ERA5) and an ensemble of EC-Earth3 coupled climate simulations of present and future climate conditions. Then we apply a storlyine approach, whereby the probability of an extreme precipitation event over Italy is decomposed into the product of three terms: the probability of the occurrence of the large-scale trough precursor, the probability of cyclogenesis resembling the Adrian/1966 track given the large-scale trough, and the probability of extreme precipitation events in Italy, given an Adrian-like cyclone track.  We first compare reanalysis data with model outputs to assess the representation of such cyclones in the EC-Earth3 historical ensemble. Next, we extend our analysis to EC-Earth future projections in the SSP5-8.5 scenario. We evaluate the probability of the three terms composing the storyline in both the historical and future simulations, and then we estimate the risk ratio, which quantifies how each component has contributed to the changes in the probability of this class of precipitation extremes. 

Preliminary results show that EC-Earth3 has a satisfactory representation of the North Atlantic and Mediterranean storm-tracks in Autumn, although it seems unable to simulate storms with a trajectory and intensity matching storm Adrian. Our aim is now to identify the probability of such events in the EC-Earth3 ensemble, and to decompose the probability in the product of the three terms. Based on previous literature, a possibile outcome is that while the probability of the large scale trough occurrence, and of a developing cyclone decreases under climate change - due to large scale circulation changes - the probability of high-impact precipitation increases.  Our analysis will enable to understand the relative importance of these processes in triggering precipitation extremes via this Mediterranean storm-track. Future applications of this approach to different climate models will further enable to explore uncertainties and to develop physical climate storylines based on plausible future changes in large-scale atmospheric dynamics, Mediterranean cyclone development and moisture environment.

How to cite: Ghinassi, P., Zappa, G., Grazzini, F., Iacomino, C., Pascale, S., Portal, A., and Simolo, C.: Understanding future projections for a high-impact Mediterranean storm track, EMS Annual Meeting 2025, Ljubljana, Slovenia, 7–12 Sep 2025, EMS2025-453, https://doi.org/10.5194/ems2025-453, 2025.

Show EMS2025-453 recording (12min) recording

Posters: Thu, 11 Sep, 16:00–17:15 | Grand Hall

Display time: Wed, 10 Sep, 08:00–Fri, 12 Sep, 13:00
Chairpersons: Salvatore Pascale, Magdalena Mittermeier, Giuseppe Zappa
P105
|
EMS2025-103
Can large-ensembles downscaled using generative AI be used for climate attribution? 
Daniel Cotterill, Georgie Logan, Mark Mccarthy, Andrew Ciavarella, Henry Addison, Peter Watson, and Tomas Wetherell

Attributing rainfall extremes in the context of climate change, requires climate model data run at sufficiently high resolution both in the present and pre-industrial climates. Nevertheless, climate runs for the latter do not exist at sufficiently high resolution to adequately represent rainfall extremes involving convection or over regions with complex topography due to the coarseness of the model’s spatial resolution. This limits our ability to make robust attribution statements on many types of rainfall events and their consequent flood impacts, as high resolution spatial-patterns of rainfall are also required to produce realistic flood inundation mapping for flood modelling. Furthermore, large-ensemble climate model simulations that are valuable for attribution are very expensive to run at convection permitting resolution. In this work we downscale our large ensemble of attribution runs (HadGEM3-A) providing over 5000 years of data in both present and pre-industrial climates to convection-permitting resolution for England and Wales using a generative AI approach. We use the diffusion model CPMGEM from Addison et al. (2024), trained on UK Climate model projections to map from coarse to convection-permitting resolution over England and Wales, to downscale the attribution runs. We test the ability of the diffusion model to transfer to generating high-resolution precipitation for the attribution system, in both present and pre-industrial climates. Exploratory testing and validation of the results may be able to provide answers on whether data from such ML techniques are suitable for attribution studies. If they are, the capacity to attribute rainfall extremes and flooding will increase greatly. We then intend to redo a couple of past attribution studies that used the coarser data, with the newly downscaled data from the ML model to compare results.

How to cite: Cotterill, D., Logan, G., Mccarthy, M., Ciavarella, A., Addison, H., Watson, P., and Wetherell, T.: Can large-ensembles downscaled using generative AI be used for climate attribution?, EMS Annual Meeting 2025, Ljubljana, Slovenia, 7–12 Sep 2025, EMS2025-103, https://doi.org/10.5194/ems2025-103, 2025.

P106
|
EMS2025-259
A classification of high-risk atmospheric circulation patterns for Italian precipitation extremes 
Cristina Iacomino, Salvatore Pascale, Giuseppe Zappa, Marcello Iotti, Federico Grazzini, Paolo Ghinassi, and Alice Portal

Precipitation extremes are a significant natural hazard that has caused considerable destruction in Italy over the past decade. However, our understanding of the effects of climate change on these extremes remains incomplete, with unclear trends in the intensity and frequency of precipitation extremes. As a first step to address this issue, here we develop a comprehensive classification of the Weather Types (WTs) associated with 1985-2019 Extreme Precipitation Events (EPEs) in the 156 operational Warning Areas (WAs) used by the Italian Department of Civil Protection by applying  Self-Organizing Maps to sea level pressure and 500 hPa geopotential height. We identify six different WTs influencing EPEs in Italy associated with different large-scale dynamical drivers:  Atlantic cyclone over France/northern Tyrrhenian Sea (WT1), Mediterranean cyclone over Central Italy (WT2), Western Mediterranean cyclone associated with upper level trough over Iberia (WT3), Westerly zonal flow (WT4), Western Mediterranean cyclone associated with upper level trough over North Italy and ridge over Southern Italy (WT5), and Mediterranean cyclone over Southern Italy (WT6). The relevance of these WTs for different WAs is evaluated through composites of moisture transport, the probability of EPEs to be associated with each WT and their seasonality. Trend analysis for the annual frequency of these circulation patterns shows no significant trends, with the except of WT3 which has experienced a significant increase in autumn. These results add to the existing knowledge of drivers of extreme precipitation events in Italy, providing an understanding of underlying large-scale atmospheric circulation and a tool to investigate the role of anthropogenic climate change in  climate model simulations.

How to cite: Iacomino, C., Pascale, S., Zappa, G., Iotti, M., Grazzini, F., Ghinassi, P., and Portal, A.: A classification of high-risk atmospheric circulation patterns for Italian precipitation extremes, EMS Annual Meeting 2025, Ljubljana, Slovenia, 7–12 Sep 2025, EMS2025-259, https://doi.org/10.5194/ems2025-259, 2025.

P107
|
EMS2025-665
Climate attribution of Italian cyclone-driven precipitation extremes using a regional large ensemble 
Alice Portal, Cristina Iacomino, Paolo Ghinassi, Salvatore Pascale, Federico Grazzini, Marcello Iotti, and Giuseppe Zappa

Climate attribution science deals with understanding the role of climate change in impactful weather and climate events. It consists in comparing two sets of weather instances which are selected as analogues of the chosen event, one set in the ‘past climate’ and one in the ‘present climate’. The comparison allows to demonstrate the influence of climate change in the hazards produced by the present analogues against those by the past analogues (or future analogues against present analogues, if the interest is for the changes in future events). These analyses can lack in robustness because of the limited size of the datasets where the weather analogues are searched for – often the case in observation-derived datasets such as reanalyses, or because of the uniqueness of the event – e.g., when the small-scale details are important for the development of the hazards. In the Mediterranean region attribution of cyclone-driven hazards is made difficult by the fact that systems which are synoptically similar to each other may lead to a pletora of surface impacts. This is because the impacts depend on the small-scale interactions between the cyclone-driven circulation and the complex Mediterranean surface topography. Morover, the small scales are not always well captured by global atmospheric models such as those used to run reanalysis.

In this work we focus on a selection of high-impact, cyclone-driven precipitation extremes that have affected Italy during the recent climatological period. We search for analogues of these events in the regional large-ensemble climate projections run with CRCM5 model as part of the ClimEx project. The use of a large 50-member ensemble and the high spatial resolution (12 km) are crucial for addressing both sampling limitations and the representation of small-scale processes, which in the Mediterranean region contribute to precipitation extremes. The extensive ensemble size allows for a meaningful investigation of present and future changes in cyclone-driven extremes over Italy. Furthermore, the ensemble is sufficiently large to perform convergence tests, which help determine the minimum sample size required to obtain reliable attribution results for each selected case. By comparing these convergence behaviours across multiple extreme events, we explore the consistency and reliability of climate attribution statistics in the Mediterranean context. The results of this study therefore provide new insights into both the influence of anthropogenic climate change on severe precipitation events in Italy and the methodological soundness of attribution studies in complex regional settings like the Mediterranean.

How to cite: Portal, A., Iacomino, C., Ghinassi, P., Pascale, S., Grazzini, F., Iotti, M., and Zappa, G.: Climate attribution of Italian cyclone-driven precipitation extremes using a regional large ensemble, EMS Annual Meeting 2025, Ljubljana, Slovenia, 7–12 Sep 2025, EMS2025-665, https://doi.org/10.5194/ems2025-665, 2025.