Forest Age Estimation In Northern Arkhangelsk Region Based On Machine Learning Pipeline On Sentinel-2 And Auxiliary Data

This article focuses on the development of a machine learning-based pipeline for forest age estimation. It examines the usage of Sentinel-2 data with auxiliary data, such as DEM measurement, in two ML pipelines for pixel-based and stand-based age estimation. The study area is located in the north of the European part of Russia, in the Arkhangelsk region. The forestry inventory data was collected in 2018 for several forestries in Arkhangelsk region, covering a total area of 126-641 ha. The article also considers two approaches to spatially distributed testing and training splits.