Projected changes in the seasonality and interannual variability of lowland and mountain runoff and potential consequences for global water use

Mountains play a critical role in water provision and use in the lowlands, with the timing and magnitude of runoff being key determinants of water availability. Climate change is projected to alter runoff dynamics globally. This will affect the interplay between mountain and lowland runoff and, thus, lowland water availability. While many studies have focused on runoff changes in individual river basins, occasionally contrasting lowlands and mountains, we lack information about how the magnitude and timing of mountain and lowland runoff differ across river basins worldwide and how these characteristics may change in the future. Moreover, studies examining future changes in the relevance of mountain runoff for future lowland water use beyond decadal averages are rare.

Therefore, in this study we examined differences in runoff magnitude, seasonality and interannual variability between lowlands and mountains in all large river basins globally, both in the past and under future projections. A key focus was to determine whether future changes might lead to increasing similarity between lowland and mountain runoff characteristics. We then investigated future changes in the seasonality and interannual variability of lowland surface water abstractions (LSWA) and the share stemming from mountain runoff. This second part of the study builds upon previous work (Hanus et al., 2024a) by exploring future changes in the relevance of mountain runoff for meeting lowland water demand including interannual variability.

To improve the representation of mountain runoff, we used global hydrological simulations from the Community Water Model (CWatM, Burek et al., 2020) coupled with the Open Global Glacier Model (OGGM; Maussion et al., 2019) (Hanus et al., 2024b). Future simulations were conducted until the end of the century using a low-emission (SSP1-2.6) and high-emission (SSP5-8.5) scenario.

Our results show that mountains have a lower interannual variability, a later seasonality timing and higher specific runoff magnitude compared to lowlands. In contrast, the strength of seasonality is higher or lower in the mountains depending on the region. The projected directions for the change of these runoff signatures are agreed upon in most river basins between mountains and lowlands. Only in Central Europe are all of the analysed runoff signatures projected to become more similar between mountains and lowlands.

Focusing on interannual variability, our analysis shows that the contribution of mountain runoff to lowland surface water abstractions varies substantially between years. For example, for the Po River basin, the long-term average mountain runoff contribution to LSWA in July is 58%, whereas it can reach 76% in one year. Notably, the sign in runoff anomaly agrees in most years among the lowlands and mountains in the river basins, with minimal future changes. Still, anomaly strength is mostly higher in the lowlands. The reliance of lowland water use on mountain runoff is generally the largest in years with a negative lowland runoff anomaly, even if the mountain runoff anomaly is also below average.

Overall, our study demonstrated that mountain runoff is an important water source in many world regions, especially in specific years with low lowland runoff.