High Conservation Value Forests, Sweden (HCVFSw)

HCVFSw has been developed through a machine learning algorithm (Random Forest Classifier) as a part of a research project financed by the Swedish Environmental Protection Agency. It is based on a large set of biophysical and socio-economic variables, including known occurrences of high conservation values forests (HCVF) and provides a wall-to-wall prediction of the relative likelihood that single hectares (100x100 m), dominated by forest (≥50%), constitute HCVF. The predictions are continuous from 100% to 0% (1 to 0). HCVFSw can therefore be regarded as a first step towards identifying areas that are in need of protection, conservation management and restoration as well for increased environmental concern during forestry operations. HCVFSw is primarily aimed for geographical planning and not a description of actual natural values in singel forest stands. HCVFSw may also be used to identify areas where forestry can continue without major conflicts with existing natural values