Hydrological time series provide essential information for identifying trends and variability in surface and subsurface water systems, understanding their response to extreme events such as floods and droughts, and improving predictive capabilities for sustainable water-resource management. Recent developments in statistical modelling and machine learning offer new opportunities to extract information from increasingly large and complex hydrological datasets and to improve forecasting across a range of temporal and spatial scales.

This session welcomes contributions on the analysis, forecasting, and interpretation of hydrological and hydroclimatic time series using statistical, machine learning, and hybrid approaches. Methods may include, but are not limited to regression, SARIMAX, LSTM, transformers, foundation models, knowledge-guided AI, and related approaches. We particularly encourage studies that evaluate model performance, interpretability, transferability, and robustness under changing or extreme conditions. Applications may address streamflow, flooding, groundwater, soil moisture, surface runoff, water quality, and other hydroclimatic and environmental variables.