Impacts and uncertainty of climate change on mountain hydrology and its extremes: An approach using high-resolution stochastic climate simulations

Abstract

This thesis delves into the hydrological responses of mountainous catchments in the Swiss Alps to climate change, with a particular emphasis on understanding and underscoring the importance of spatial variability, uncertainties in projections, and the impact of intensified extreme rainfall. Initially, the study explores the significant spatial variability in hydrological responses at the sub-catchment level, driven by changes in precipitation patterns, snowmelt, and evapotranspiration across different elevations. The analysis employs a high-resolution two-dimensional weather generator model alongside a distributed hydrological model, detecting notable shifts in streamflow patterns, particularly a winter increase and summer decrease projected towards the end of the 21st century under the RCP8.5 emission scenario. To address the inherent uncertainties in climate and hydrological projections, this thesis introduces a novel framework designed to quantify and partition the sources of uncertainty across different scales. The framework combines the use climate of model outputs, a high-resolution weather generator, and a distributed hydrological model to produce large ensembles of future climate and hydrological variables at high resolution. Applying this approach to two representative mountainous catchments, the analysis underscores the dominant role of precipitation's natural variability in engendering uncertainty, alongside the identification of robust change signals in specific hydrological components like snowmelt and liquid precipitation, especially during warm seasons and at higher elevations. Further, to enhance the realism in simulating the intensification of extreme rainfall, an improved version of the weather generator model is introduced. This revised model successfully mimics the Clausius-Clapeyron relation, reflecting the observed scaling of heavy rainfall with temperature. By employing this enhanced model to assess future rainfall impacts on hydrological responses, a more accurate depiction of the potential intensification of short-duration heavy rainfall in future climates is achieved, providing a more realistic assessment of future hydrological impacts even for short durations. Overall, this work provides an improved understanding of the complex interplay between climate change and hydrological processes in mountainous terrains, enhancing the predictive capacity regarding the hydrological behaviour of such regions in a warming climate. Through a detailed spatial analysis, rigorous uncertainty quantification, and an enhanced representation of extreme rainfall, this thesis contributes to deciphering the complex hydrological implications of climate change in mountainous environments.