The Urban Heat Island Effect in Karlstad : An Analysis of Long-Term Trends in Land Use, Vegetation, and Temperature Variations

This study has analyzed the surface urban heat island effect (SUHI) in the urban area of Karlstad during the period 2000-2018 using satellite data, geographic information systems (GIS) and the remote sensing platform Google Earth Engine (GEE). The aim has been to investigate the relationships between urban land consumption, vegetation (NDVI), land surface temperature (LST) and the spatial development of SUHI. Four research questions have formed the basis of this study, concerning changes in land use, the relationship between vegetation and land surface temperature, and the identification of risk areas for future heat islands. The results show a clear increase in urban land area in Karlstad, with a total increase of 6.41% during the studied period. At the same time, the analyses indicate that vegetation in urban areas has decreased, which in several cases coincides with rising land surface temperatures and intensified SUHI effects. Particularly vulnerable areas include industrial areas such as Örsholmen and the Välsviken shopping area, where SUHI values have been high throughout the period. NDVI and LST generally show a negative relationship, although the statistical analysis suggests that NDVI only explains a very small part of the temperature variations, which suggests that other factors - such as building materials and the city's structure - play a larger role. The SUHI effect has been strengthened in line with urban expansion, especially between 2000 and 2006. Cartographic visualizations also show that Karlstad Central Hospital exhibits high SUHI values over the entire period, making it a priority area from a health perspective. The study underlines the importance of integrating climate perspectives into urban planning and suggests that future research should cover more years, a larger geographical scope and include the effect of adaptation measures such as green infrastructure. GEE proved to be an effective tool for both data analysis and visualization in environmentally related urban studies.