Although a subsurface phenomenon, permafrost creates distinct features on the land surface which can be observed using remote sensing data. This is especially true for the East Siberian Arctic where ice-rich permafrost is abundant in the geological subsurface and only protected by a thin layer of organic soils. In the Lena River Delta, deeper seasonal thaw during increasingly frequent warm summers does not only result in irreversible loss of ground ice and subsequent land subsidence, but also in discharge of soil organic carbon that was previously fixed in permafrost. To characterize the dynamics of thawing permafrost and its impacts on landscapes, hydrology, and emission of methane and carbon dioxide we analyse optical remote sensing time series from various sensors. Local field measurements (meteorology, ground temperature, geodetic surveys) during several recent Russian-German Arctic expeditions complement our remote sensing studies and help differentiating factors causing relief and land cover changes. While previous studies concentrated on general inventory of thermokarst landforms, higher temporal resolution of contemporary image acquisitions provides unique information for the understanding of seasonal processes, such as ice-on and ice-off on thermokarst lakes, shore erosion on delta channels, water level changes and drainage events in lakes, and wettening/drying of thermokarst-affected areas. Ground truth data provides the basis for calibration and correction of 21 RapidEye scenes (level 1B) from 2014 using a bundle block adjustment procedure. Next steps will include extraction of seasonal variations of band metrics such as NDVI which we will compare to decadal Landsat time series of landcover change and multitemporal, photogrammetrically-derived digital elevation models in order to identify signatures and trends typical for permafrost thaw related processes on the surface. Our approach will allow assessment of rates and short-term changes in thermokarst dynamics and landscape evolution. In addition, the derived data will be valuable for permafrost-thaw model parameterization.
EPIC Alfred Wegener Institut