Unravelling the sources of uncertainty in glacier runoff in the Patagonian Andes (40–56° S) 

Glaciers are retreating globally and are projected to continue to lose mass in the coming decades, directly affecting downstream ecosystems through changes in glacier runoff. Estimating the future evolution of glacier runoff involves several sources of uncertainty in the modelling chain, which have not been comprehensively assessed on a regional scale. In this study, we used the Open Global Glacier Model (OGGM) to estimate the evolution of each glacier (area > 1 km2) in the Patagonian Andes (40–56° S), which together represent 82% of the glacier area of the Andes. We used different glacier inventories (n = 2), ice thickness datasets (n = 2), historical climate datasets (n = 4), general circulation models (GCMs; n = 10), emission scenarios (SSPs; n = 4), and bias correction methods (BCMs; n = 3) to generate 1,920 possible scenarios over the period 1980–2099. For each scenario and catchment, glacier runoff and melt time series were characterised by ten glacio-hydrological metrics. We used the permutation feature importance of random forest regression models to assess the relative importance of each source on the metrics of each catchment. Considering all scenarios, 30% ± 13% of the glacier area has already peaked in terms of glacier melt (year 2020), and 18% ± 7% of the glacier area will lose more than 80% of its volume this century. In terms of glacier melt metrics, future sources of uncertainty (GCMs, SSPs and BCMs) were the main source for only 18% ± 21% of the total glacier area. In contrast, the reference climate was the main source in 78% ± 21% of the glacier area, highlighting the importance of the choices we made in the calibration procedure. The results provide a basis for prioritising future efforts to reduce glacio-hydrological modelling gaps in poorly instrumented regions, such as the Patagonian Andes.