Annual glacier mass balance estimation through ASTER DTMs and snowlines extracted from Landsat and Sentinel-2 images

Direct glaciological measurements are an important dataset of glacier mass balances but remain concentrated on a small number of glaciers. On hydrological years 2019/20 and 2020/21, 318 annual mass-balance observations were compiled based on 169 glaciers worldwide (Zemp et al., 2023). On the other hand, the current climate crisis now requires a description of cryosphere evolution at a larger scale by quantifying annual snow and ice losses on a larger number of glaciers. A relevant attempt to fill this gap has been provided by Huggonet et al. (2021) where a global dataset of mass balances at a glacier scale have been generated from 2000 to 2020. While being an extremely valuable glacier mass balance dataset, it is limited to provide mass balance estimation with a time scale longer than 5 years, i.e. annual mass balances cannot be considered reliable. On the other side, the equilibrium line altitude (ELA) method (Rabatel et al., 2016) have been proven to be an effective approach to reconstruct annual glacier mass balance time series as soon as annual estimation of ELA from satellite multispectral images (e.g. Landsat, Sentinel-2) and at least two digital terrain models (DTMs) acquired at different years are available. Typically, highly accurate DTMs (e.g. airborne LiDAR or photogrammetric DTMs), which are only available on a regional scale base, have been employed within the ELA method. The main objective of this work is to test the ELA method using as input: 1) Landsat and Sentinel-2 estimation of ELA and 2) ASTER DTMs (Hugonnet et al., 2021). In this way, annual mass balances can be retrieved using satellite data only. We initially tested this approach over the glaciers in Trentino and South Tyrol where seven glaciers have been monitored through glaciological measurements and different airborne DTMs have been acquired during the last 20 years. Our results show that the use of the ELA method with high resolution airborne DTMs can produce mass balance estimations characterized by an error around 0.3 m w.e. with respect to ground measurements. This error value is in line with estimations conducted with the same method in other regions (e.g. Rabatel et al., 2016) and it is in the error range of ground based measurements. The use of ASTER-based 5 years DTM differences as input of the ELA method can produce estimations with a similar error range. Therefore, the combination of ASTER DTM and ELA extracted from Landsat or Sentinel-2 images may be an interesting approach to produce accurate annual mass balance estimations for many glaciers in the world.   References : Hugonnet, R. et al. (2021). Accelerated global glacier mass loss in the early twenty-first century. Nature 592, 726–731 Rabatel, A. et al. (2016). Spatio-temporal changes in glacier-wide mass balance quantified by optical remote sensing on 30 glaciers in the French Alps for the period 1983–2014. Journal of Glaciology, 62(236), 1153-1166. Zemp, M. et al. (2023). Global Glacier Change Bulletin No. 5 (2020-2021). WGMS.