Tree Cover Mapping Based on Sentinel-2 Images Demonstrate High Thematic Accuracy in Europe

摘要: Abstract The spatial and temporal distribution of trees has a large impact on human health the environment through contributions to important climate mechanisms as well commercial, recreational social activities in society. A range tree mapping methodologies been presented literature, but cover estimates still differ widely between individual datasets, comparisons thematic accuracy resulting maps are rather scarce. Copernicus Sentinel-2 satellites, which were launched 2015 2017, have combination high resolution. Given that this is new satellite, substantial amount research development algorithms assessment said be done years come. To contribute process, map produced unsupervised classification was created for six tiles. agreement corresponding national forest inventory, function band chosen, analysed assessed two out results show highest present inventory found bands 2, 3, 6 12. relative difference 8% 79% compared previous estimates, characterised by scatter. Lastly, it shown overall up 90%, with user’s ranging from 34.85% 92.10%, producer’s 23.80% 97.60% various classes. This demonstrates can Sentinel-2. In future increased even more use averaging procedure, will enable an accurate estimate European cover.