The forest ecosystem displays a great variation in structure and habitat which is important for many species. To manage nature and species diversity in the best possible way, information about the forest ecosystem is therefore essential. The increasing amount of available remotely sensed data from aerial photography and airborne laser scanning through national mapping programs and satellite data, e.g. through the Landsat or the European Copernicus programs, provides an increased potential for establishing information on forest ecosystems in a cost-effective way. In this project, we link the work on mapping and monitoring the forest ecosystem to the ecological base map (Meld. St. 14 2015-2016) and use the term forest ecological base map for the map layers produced. An ecological base map is not a specific map, but will in the long term consist of a collection of map layers that will be relevant as part of a forest ecological base map. The project nevertheless uses this term because central map layers are established for the description of the condition and pressures of forest ecosystems. The report describes the work of using remotely sensed data to describe the condition and pressures on the forest ecosystems. Available map data (AR5) is used together with Sentinel-2 data from the Copernicus program to delineate the forest ecosystem. Furthermore, airborne laser scanning data from the national laser scanning program in Norway (NDH) are used together with the reference data from the Norwegian National Forest Inventory (NFI) to establish maps of selected forest attributes i.e. volume, biomass, carbon, tree height, tree density, crown coverage, and variation in stem diameter. The volume and tree height forest attributes are well determined, while there is somewhat lower accuracy for the other attributes. Furthermore, probability maps and area estimates for natural forests and for the presence of Sitka spruce (non-native species) are established. The estimates, also showing the evolution of natural forests and non-native species over time, have been computed at the county level from the NFI data. The change in area from the mid-90s until today is small for non-native species. Natural forest mainly show an increase in area, except when natural forests are defined as what was harvest class V in the NFI in the mid-90s and which is still harvested class V. A main challenge in producing probability maps for Sitka spruce and natural forest is the small proportion of reference data for those classes. Therefore, additional reference data has been collected in Hordaland, Oppland, and Møre og Romsdal. The collection of extra reference data improved the models in all cases, but worked better for natural forests than for Sitka spruce in terms of cost-efficiency. For the natural forest definitions, the classification worked best for the NFI and "biological old forest" definitions. However, the accuracy cannot be characterized as more than moderate for any of the models. This varies with the definitions and county some models are very poor while others work better. A major trend is a large proportion of false positives.An increased amount of reference data is central to improving the models. In addition, change detection based on time-series of satellite images from the Landsat archive has been established and developed as a map layer in the forest ecological base map. Furthermore, the IPCC area category was classified for the detected changes. The possibilities of establishing a near real-time detection of changes have also been evaluated. The project has established satellite mosaics from Sentinel-2 and Landsat 8 for the years 2015-2018 for the entire country. Airborne laser scanning data is processed for the counties Østfold, Akershus, Oslo, Hedmark, Oppland, Buskerud, Vestfold, Telemark, Agder, Hordaland, and Møre og Romsdal. The map layers of the forest ecological base map were also established for these counties.