Hydroclimatic extremes such as droughts, floods, heatwaves and compound events emerge from complex interactions between atmospheric, hydrological and human processes across multiple spatial and temporal scales. The growing availability of Earth observation, in-situ and low-cost monitoring, citizen science observations, reanalysis and modelled data, together with advances in data-driven methods, provides new opportunities to detect, characterize and understand these events and their evolution.
This session invites contributions that use data-driven, computational and hybrid approaches to advance the understanding of hydroclimatic extremes. We particularly welcome studies that integrate multiple data sources, investigate spatial and temporal patterns, identify drivers and dependencies, improve prediction, or connect hazard dynamics with impacts and risk.
Relevant approaches include, but are not limited to, machine learning and artificial intelligence, remote sensing and Earth observation, data fusion, spatiotemporal analysis, event detection and tracking, statistical and causal methods, process-informed machine learning, hybrid modelling, and the integration of observations with hydrological and climate models.
We welcome applications addressing droughts, floods, heatwaves, low flows, compound and cascading events, and transitions between extremes, from catchment to regional and global scales. Contributions addressing uncertainty, interpretability, and transferability are particularly welcome, as are studies focusing on the Global South and other data-scarce regions, where limited observations create distinct challenges for characterizing and predicting extremes.
The session aims to bring together researchers working across hydrology, climate science, Earth observation, data science and risk assessment to discuss how emerging data and methods can improve our understanding of hydroclimatic extremes and their impacts.