Abstract
Hydrological dynamics in glacierized catchments of the Alps are shaped by temperature-driven processes, including snow and ice melt as well as precipitation, leading to diel streamflow cycles that vary in intensity within- and among-the seasons. During the summer melt period, the amplitude of these diel cycles increases due to diminished snow storage and the emergence of efficient subglacial drainage systems. Accurately modeling these sub-daily cycles remains difficult, due to a lack of high-resolution meteorological input data for melt simulations and due to challenges in parameterizing meltwater routing through dynamic glacial systems. This research develops an approach for downscaling daily streamflow timeseries to sub-daily timescales (daily flow duration curves) in alpine glacierized catchments influenced by snow and ice melt runoff. We adapt a maximum entropy framework (POME) to the specificities of glacial systems, that we calibrate on a 45-year data set of 15-min discharge records from seven glacier-fed catchments in the southwestern Swiss Alps. The calibrated method is then applied to the outputs of a semi-lumped hydrological model that simulates daily discharge and provides hydrological variables such as snow depth and ice melt to inform the downscaling, and the results are evaluated against observed discharge. Our results reveal that a sigmoid function effectively represents seasonally varying daily flow duration curves in glacierized catchments and highlight the influence of climate warming on sub-daily flow dynamics over recent decades. This downscaling method offers a robust tool for reconstructing sub-daily discharge in catchments with limited data, opening new perspectives for hydrological modeling at finer scales.