Time series analysis of remotely sensed snow cover data: revealing permafrost thermal state and vegetation dynamics

Abstract

Snow cover plays a crucial role in climate change and is considered an essential climate variable by the Global Climate Observing System (GCOS). To accurately monitor daily snow cover extent, optical medium-resolution remote sensing systems like MODIS and VIIRS are employed. DLR's Global SnowPack (GSP) product, derived from daily MODIS/VIIRS snow cover data, addresses data gaps caused by clouds or polar night, providing gap-free daily datasets since February 2000. The extended time series allows the identification of trends in snow cover duration (SCD), which has implications for the thermal state of permafrost soils and vegetation dynamics. Snow acts as an insulating barrier against colder winter air temperatures, enabling the underlying permafrost layer to retain higher temperatures. The 23-year dataset of SCD data from GSP allows both the determination of a long-term average and the derivation of trends. We compared both the mean SCD and the trend with annual changes in parameters describing horizontal and vertical permafrost extent, as well as changes in land cover classifications (both provided by ESA CCI). Regarding "Greening of the Arctic" we found changes in land cover classes, but in the observed period since 2000 there was little dynamism, and this is only slightly reflected in the SCD. Obvious developments were found in the thermal state of the permafrost -mainly degradation, but increases were also noted, similar to the positive trends in snow cover duration. Changes in Active Layer Thickness (ALT) could be best explained with SCD changes. Overall, additional data are needed to make quantitative predictions about permafrost development using SCD.