Comprehensive mapping of individual living and dead tree species using leaf-on and leaf-off ALS and CIR data in a complex temperate forest

Abstract Tree species information is crucial both for understanding forest composition and supporting sustainable forest management, but also for monitoring biodiversity and assessing ecosystem services. Remote sensing data has been widely used to map the spatial distribution of tree species across large areas. However, there is currently a lack of studies demonstrating the potential of airborne laser scanning data collected during different seasons to identify multiple individual tree species/genera, including dead individuals. The main objective of this study was to map the ecologically valuable forest area constituting the Polish part of the Białowieża Forest using leaf-on and leaf-off airborne laser scanning (ALS) data and color-infrared imagery. Eleven living species/genera (alder, ash, aspen, birch, hornbeam, lime, maple, oak, pine, spruce and other deciduous) and four dead classes (dead deciduous, dead pine, dead spruce and snag) were classified at the individual tree level. Applying the Random Forests algorithm and a set of 30 predictor variables, 15 classes were classified with an overall accuracy of 82 per cent. The mapping of nearly 20 million individual trees revealed that in 2015, the most common tree species in the upper part of the Białowieża Forest stands was spruce (20.1 per cent), followed by alder (19.0 per cent) and pine (18.1 per cent). Among dead trees, dead deciduous trees (2.2 per cent) and dead spruce (1.7 per cent) were the most common. Our results can serve as a first cornerstone for carrying out further in-depth analyses of forest biodiversity using remote sensing data in this exceptional forest area.