Analysing forests naturalness in southern Norway using airborne laser scanning

Abstract

Biological and ecological important areas and species are in increasing focus now. Earlier protected areas have mostly excluded productive forest areas in Norway. The added focus on these areas combined with an increased interest in biodiversity has led to the production of this study. It was therefore examined to see if it was possible to categorize these degrees of naturalness and to see if it was possible to remotely sense these areas. The remote sensing technique if successful will decrease time and costs used identifying these areas in the field. In this study Airborne Laser Scanning data was used in conjunction with data collected in the field to see if it was possible to determine old natural forests and to see if it was possible to distinguish between the naturalness categories using ALS. It was decided to use Random Forest decision tree model to predict the factors from the ALS that were the most important in the predication of naturalness. Height variables were the variables that were deemed the most decisive according to the Random Forest Analysis. The aim of this thesis is to evaluate methods to map ONFs using ALS. More specifically the objective was to include methods to maps the ONFs gradient of definitions to six categories. Untouched (virgin), ancient, semi-natural, old-selective, old managed, and managed.