Landcover information is of relevance for a range of applications in Arctic environments including up-scaling of in situ measurements, permafrost monitoring and modelling and climate change impact assessment. But not only natural landcover types, also the identification of human impacted areas is required. A wide range of satellite records are available, but often there are coverage and resolution limitations. Due to the spatial heterogeneity typical for the Arctic, an as high as possible spatial resolution is needed. Sentinel-1 as well as Sentinel-2 offer 10m nominal resolution for specific bands and modes and are freely available across the entire Arctic. The multi-spectral information from Sentinel-2 is specifically of value for discrimination of tundra types. The Synthetic Aperture Radar mission Sentinel-1 provides added value through representation of land surface structure features. A combination of both allows significantly improved characterization of landcover over larger areas. Here, we review recent progress in monitoring the land surface close to Arctic coasts (focus region of HORIZON2020 Nunataryuk), specifically the distribution of human impacted areas beyond of what is represented in databases such as OpenStreetMap or global settlement maps. The added value of machine learning techniques will be discussed and results based on Sentinel-1 und Sentinel-2 presented. Further on, permafrost change information provided through the ESA CCI+ Permafrost project is combined with the novel maps. The CCI+ Permafrost datasets cover 1997-2018 for ground temperature, active layer thickness and permafrost fraction with 1 km gridding (partially derived from MODIS LST, https://climate.esa.int/en/projects/permafrost/data/) and therefore allow detailed assessment of areas with human presence over permafrost.