Severe hailstorms are becoming more frequent in Central Europe, showing increasing interannual variability. The Pre-Alpine and Alpine region is significantly affected due to its complex terrain, which initiates convection and can intensify many hail-favoring processes. In particular, large hail events are often very local phenomena and are becoming increasingly intense. Ground-based observations from weather radars are the most reliable for detecting hail; however, they are challenging in the Alpine region due to interference at mountain ranges.Passive Microwave satellite observations offer a valuable alternative for detecting hail: a hail probability can be directly derived from Passive Microwave channels with a high spatial coverage. However, this data is only available at certain times during satellite overpasses, thus capturing only snapshots of a few of these events. Visible, near-infrared, and infrared data from MSG give the highest spatial and temporal coverage. Though not directly sensitive to hail, its high spatiotemporal resolution can identify early stages of severe storm development leading to large hail formation by peculiar characteristics in their spatiotemporal evolution.Whereas various ML approaches already classify spatial cloud patterns from satellite measurements, the temporal component of cloud development remains less explored. This work aims at classifying the evolution of typical cloud patterns leading to severe storms over the Alpine region. In particular, learning about formation of hail storms and the conditions in which they develop.We initially adopted a supervised deep-learning framework for classifying spatiotemporal cloud evolution patterns as hail and non-hail, obtaining better performance than a simple logistic regression. We trained the network using MSG timeseries displaying cloud evolution patterns and labeled the presence of hail using a hail probability product based on passive microwave radiometry. Then, we exploited the same architecture in a self-supervised approach, using the labeled dataset as fine-tuning to capture hail development characteristics better. We characterize such classes of cloud developments in a spatiotemporal embedding, exploiting its characteristics and investigate the physical properties of the classes with ancillary datasets.

Climate change is intensifying and increasing the frequency of storms in the Alpine region. However, accurately capturing these events remains a challenge for current weather models due to the complexity of atmospheric processes over mountainous terrain. Realistic rainfall predictions require precise cloud representation, especially for deep convective systems that cause extreme precipitation.We present a framework for classifying cloud structures observed by the Meteosat Second Generation (MSG) geostationary satellite. Leveraging its good spatio-temporal resolution and extensive historical data, our approach exploits brightness temperature in the 10.8 µm infrared channel alone and in combination with the 6.2 µm water vapor channel. This synergy enables a better representation of diurnal cycles and cloud top heights. Our classification framework employs a self-supervised deep learning (DL) model to generate a feature space where cloud structures group together based on their semantic similarity.We characterize the identified cloud classes using a range of physical parameters, including cloud properties, precipitation amounts, lightning activity, and morphological indices. Additionally, their diurnal and seasonal variability are analyzed to determine whether some cloud types are most likely to occur. Once the classes are physically described, cloud development are tracked in the feature space associated with extreme convective rainfall and hailstorms in the Alps, as recorded in the European Severe Storms Laboratory (ESSL) database. We study the transition of convective systems to extreme precipitation across space and assess the associated environmental conditions.Ultimately, this framework can enhance the evaluation of numerical weather prediction models by analyzing how simulated cloud evolution aligns with observed transitions in extreme events. Furthermore, it can be used to improve nowcasting and early warning systems for extreme precipitation by leveraging observation-based transition probabilities derived from past severe weather events.

The cloud systems of the North Atlantic trades (NAT) have been a topic of curiosity due to significant uncertainty in their physical characteristics, physical process understanding across various spatial and temporal scales, and their impact on the regional climate system. Initial research has provided the causal link for cloud systems having distinct organizational aspects (formerly described as Sugar, Gravel, Fish, and Flower) with the net radiative flux over the region. Questions have been raised about how the changing climate will influence the frequency of occurrence of these cloud regimes and how the net radiative impact will change the regional climate system. However, cloud systems represent a continuous spectrum where not-so-visually distinct systems also occur. Existing clustering mechanisms sort organizations into separate classes. Yet, in reality, the organization often does not align with those pre-defined classes but transitions amongst them or simply occurs in a continuous spectrum. Using two complementary neural networks in self-supervision (without human interference), we investigate the representation learning of cloud systems both in a continuous space describing a cloud system spectrum and in a discreet space aiming to identify distinct cloud systems. 50,000 GOES-ABI cloud optical depth NAT images from 2017 – 2021 covering the EUREC4A study area are randomly cropped to 256 x 256 pixels and sorted/labeled by the machine. We study the climatological representation of EUREC4A’s cloud patterns in the continuous embedding space. We follow the Maria S. Merian ship track inside the feature space matching the ship-based atmospheric remote sensing and ERA5 profiles with the K-nearest satellite images. This analysis examines the consistency of the environmental conditions for cloud systems identified as close to each other in the continuous feature space. We investigate the relationship between the net cloud radiative effect and the radiative flux characteristics in the continuous space, finding a strong functional relationship with the cloud system’s pattern and distributions. In the discrete space, we aim to identify the optimal number of classes that could represent the continuous space. We also aim to understand if these discrete classes correspond to the categories identified in the physical and visual space. Moreover, to better understand the decision of the neural network for a particular cloud pattern, we visualize the network’s focus in the activation space. We find that different self-attention heads of the neural network learn to focus on different semantics of the cloud system distribution. Finally, we found that different regularizations applied during the training of the network directly impact the representation learning of the cloud system, and we show how to use such regularizations to improve the understanding of cloud systems.

Severe hailstorms are becoming more frequent in Central Europe showing increasing interannual variability. The Pre-Alpine and Alpine region seems to be especially affected due to its complex terrain, that initiates convection and can intensify many hail favoring processes. This results in increasingly strong large hail events, which are often very local phenomena. Ground-based observations from weather radars are most reliable for detecting hail, however, prove to be challenging in the Alpine region due to interference at mountain ranges.Passive Microwave satellite observations offer a useful alternative for detecting hail: a probability for hail can directly be derived from Passive Microwave channels with a high spatial coverage. However, this data is only available at certain times during satellite overpasses, thus, capturing only a few of these events. The highest temporal coverage is given by visible, near-infrared and infrared data from MSG. Though not directly sensitive to hail its high spatiotemporal resolution can identify early stages of severe storm developments.Recently, self-supervised machine learning approaches have been used to classify spatial cloud patterns from satellite measurements from MSG over the Atlantic and Germany. The model learns to sort similar cloud organization patterns into the same classes.In this work, we aim at adapting this model to also include the temporal component to then classify the evolution of typical cloud patterns leading to severe hailstorms over the Alpine region. The framework will later be used to characterize changes in spatiotemporal evolution of large hail bearing systems and associated environmental conditions across a multi-year dataset. First steps are presented here including the investigation of the optimal training dataset using the available data sources.

Extreme precipitation and hail events over complex terrain are strongly dependent by the multiscale evolution of convective systems, yet their development remains challenging to observe and model. In this study, we apply a self-supervised learning (SSL) framework to Meteosat infrared 10.8 μm imagery over the Alpine region to derive physically meaningful representations of convective cloud organization across the full daily cycle. The learned representation organizes convective cloud scenes into three primary regimes: early convection (EC), deep convection (DC), and overcasted anvil systems (OA), each exhibiting distinct cloud-top properties, spatial structures, and temporal variability consistent with established convective development stages. Extreme precipitation and hail events preferentially occur within these regimes, occupying the upper tail of cloud-top height, optical thickness, and cloud coverage distributions within the latent space, rather than forming separate clusters. This indicates that extremes correspond to rare, intense realizations of continuous convective variability. Two dominant storm evolution pathways are identified: a rapid EC→DC transition and a persistent DC mode. The former is characterized by fast cloud growth and intensification, while the latter corresponds to long-lived, mature convective systems with sustained cloud properties. Overall, the results demonstrate that infrared-based SSL representations capture physically consistent convective stages and their transitions, providing a novel observational framework for characterizing extreme convective events and supporting the evaluation of convection-resolving numerical prediction models.

Extreme precipitation and hail events over complex terrain are strongly dependent on the multiscale evolution of convective systems, yet their development remains challenging to observe and model. In this study, we apply a self-supervised learning (SSL) framework to Meteosat infrared 10.8 μm imagery over the Alpine region to derive physically meaningful representations of convective cloud organization across the full daily cycle. The learned representation organizes convective cloud scenes into three primary regimes: early convection (EC), deep convection (DC), and overcast anvil systems (OA), each exhibiting distinct cloud-top properties, spatial structures, and day–night distributions consistent with established convective development stages. Within the developed SSL framework, reported extreme precipitation and hail events are preferentially associated with these regimes, occupying the upper tail of cloud-top height, optical thickness, and cloud coverage distributions within the latent space. This indicates that extremes correspond to rare, intense realizations of continuous convective variability. Two dominant pathways of cloud structure evolution are identified: a rapid EC→DC transition and a persistent DC mode. The former is characterized by fast cloud growth and intensification, while the latter corresponds to mature convective systems with sustained cloud properties. Overall, the results demonstrate that infrared-based SSL representations capture physically consistent convective stages and their transitions, providing a novel observational framework for characterizing extreme convective events and supporting the evaluation of convection-resolving numerical weather prediction models.