Assessing skill and use of CMIP6 decadal re-forecasts in global glacier mass balance modelling

Within the earth system, glaciers serve as important indicators of climate change, being principally governed by temperature and precipitation. Additionally, they provide essential freshwater storage on various scales, ranging from long-term in firn and ice, to short-term storage in snow cover. By preventing precipitation from immediately turning into runoff, glaciers fulfill a buffering role within their basins, providing downstream runoff during melt season. With changes in glacier mass balance in response to changes in climate, a glacier's buffering capacity is altered simultaneously. In order to predict the evolution of runoff on temporal scales relevant to water resource management (5-15 years), it is essential to observe and simulate glacier mass balance on the same scale. The current research presents a global modelling approach using the Open Global Glacier Model (OGGM), forced with a multi-model, multi-member retrospective ensemble of monthly temperature and precipitation re-forecasts (hindcasts) from the Decadal Climate Prediction Project (DCPP), part of the Coupled Model Intercomparison Project, phase 6 (CMIP6). The decadal hindcasts are initialized each year in the period 1960-2010 and are bias corrected for model drift, while retaining the period's warming trend, using a lead-time based correction. The hindcasts are then downscaled to the glacier scale and used to compute the climatic mass balance with OGGM, with fixed glacier geometries. The method is validated using 274 reference glaciers, which have a >5 year observational record. It is then applied globally, to all land-terminating glaciers in the Randolph Glacier Inventory (RGI), outside the Greenland Ice Sheet and Antarctica. The results indicate merit in using decadal re-forecasts to model glacier mass balance, paving the way for reliable decadal scale runoff predictions on regional and global scales.