A climate-informed statistical framework to indirectly estimate trends in future seasonal high flows in snow-dominated watersheds using short-term climate variability indices

Abstract

The intensification of the hydrological cycle under climate change has brought changes in the temporal variability of flood-generating mechanisms and extreme hydrological events. To better anticipate these changes, modelling approaches integrating climate models, emissions scenarios, and hydrological models have been widely employed. However, their application remains challenging because of inherent uncertainties, in particular from hydrological models. This study aims to use a climate-informed statistical framework to indirectly estimate the temporal variability of seasonal high flows indices (HFI) using a set of short-term climate variability indices (SCI) characterizing likely causative mechanisms over different aggregated look-back periods. An ensemble of climate models, two future scenarios, and 31 SCIs were used to estimate future HFIs trends from 1997 to 2100 using as a proof of concept two snow-dominated watersheds in Southern Quebec, Canada. A statistical framework was used including linear and monotonic partial correlations along with significant trend tests. The results indicated that future temporal variability of HFIs could be anticipated using highly correlated SCIs as proxies. At least 50% of the HFI temporal variability was explained by a single SCI, such as cumulative total precipitation or climatic demands over 1 to 2 weeks, or drought indices like the Effective Drought Index (EDI) over 180 days. Furthermore, significant trends in highly correlated SCIs were consistent with significant trends observed in HFIs. These findings offer valuable insights for future analysis of HFI temporal variability, particularly in more comprehensive water management analyses aimed at informing regional mitigation and adaptation strategies.