Cluster analysis of historical harvesting sites in Sweden : spatiotemporal, density-based and non-density-based cluster analysis of forestry-related variables in historical harvesting data

Sammanfattning

Human activity has had a profound effect on the environment and has increased the global average temperature through emissions of greenhouse gases. Although the member states of the United Nations signed Agenda 2030 towards sustainability, harvesting operations in Sweden are still heavily dependent on fossil fuels. Solutions to this are shifting to renewable fuels and electrifying work processes in harvesting operations. These solutions are however challenged by a lack of knowledge on the spatial distribution of worksites, the conditions under which work is done and requirements for the machines. The availability of data from wide scale and everyday data collection of forestry operations enables cluster analysis as a viable option for identifying patterns and help understand the requirements for shifting to renewable energy sources in forestry. This study aims to identify clusters and patterns of harvesting worksite activity in combination with varying site conditions and landscape infrastructure through clustering analysis of spatial and spatiotemporal data on harvesting operations, soil and weather data. This study identifies spatiotemporal clusters in historical harvest data through unsupervised cluster analysis, clusters harvesting operations through forestry-related variables and identifies common patterns in harvesting operations. These results can be used to better understand the spatial and temporal distribution of harvesting operations and can be used to further this area of research. This study also identifies the problem of a lack of accurate and wide scale data for cluster analyses, which should be addressed by policymakers to help better understand and mitigate the present challenges in forestry.