Thermo-insulation effect of a seasonal snow-cover on permafrost soil in Bayelva, Svalbard (1998 - 2017)

• Main Authors: Ebenhoch, Sabrina, Maturilli, Marion, Höfle, Bernhard, Roth, Kurt, Boike, Julia
• Format: Conference Object
• Language: unknown
• Published: University of Heidelberg 2018
• Subjects: Arctic; Bayelva; Svalbard; Svalbard Archipelago; Arctic Council; permafrost; Spitsbergen
• Online Access: https://epic.awi.de/id/eprint/48088/; https://epic.awi.de/id/eprint/48088/1/180226_SnowHydro_poster_SE.pdf; https://www.geog.uni-heidelberg.de/hydro/snow2018.html; https://hdl.handle.net/10013/epic.da0b83d2-f5fd-478f-8ca6-73b3bf39f152; https://hdl.handle.net/

Bayelva is a high-arctic research site on Spitsbergen Island in the Svalbard archipelago (78.551N; 11.571E) where climate, soil and snow components are recorded since 1998 by the Alfred-Wegener-Institute. This study site is underlain by permafrost with average temperatures around -2 °C and seasonally snow-covered from October to May. The thermal insulation by the snow-pack decouples the soil temperatures from air temperature. To gain better understanding of the thermal relationships, analysis of long-term measurements of air temperature, radiation, soil temperature, and snow characteristics were examined in this study. Mean annual air temperature has increased by 0.14 °C per year for the period from 1998 to 2016 which is higher than the global average temperature change in the same period. Radiation balance trend analyses are characterized by an increase in long-wave radiation during winter months. Results of our snow characteristic studies show that the last day in the year with snow cover has been moving to earlier dates with 0.5 days per year (from 1998 to 2016), extending the snow-free season, thus resulting in more time for soil warming. This corresponds well with warming trends of all soil temperature sensors as well as the thickening of the active layer (ALT), which is estimated to have reached a depth of 2 m in 2016 for the first time. Since it is difficult to measure the exact ALT, we used the Stefan model for an estimate. To account for different magnitudes of insulation, an effective snow depth (Sdep,eff) was calculated for each year. High Sdep,eff values represent high thermo-insulation, due to early snowfall and long snow depths over 0.4 m, and can be linked to years in which re-freezing of the whole active layer took until early March of the next year. Our results support the importance of snow physical characteristics for the permafrost thermal regime, as also detailed in the Arctic Monitoring and Assessment Programme - an Arctic Council Working Group.