Identifiering och validering av potentiella översvämningsriskområden med hjälp av GIS-baserad multikriterieanalys

Sammanfattning: Floods are natural disasters that often have significant socio-economic consequences. Urban areas with uncontrolled urban development, rapid population growth, an unregulated municipal system and an unplanned change in land use belong to the very sensitive areas where floods cause devastating economic and social losses. This study aims to present a reliable GIS multi-criteria methodology for risk zone mapping of flood-prone areas in urban areas. This study also intends to study and investigate the relationship between criteria and field of study. This was carried out with a literature study to investigate and determine the criteria to be used. In this study, the focus is on fluvial flooding. The study area is the centre of Karlstad (17.5 km²) in Värmland County. The study area is interesting because the municipality of Karlstad is trying to increase the number of inhabitants in the city and invest manyresources to build attractive areas close to water, even though those areas were affected by flooding from 2000 to 2001. All data processing was performed in ArcGIS Pro 3.1.0 and FME 2022.2.2 software. All hydrological analyses such as flow direction, accumulation and DEM sinks were performed in System for Automated Geoscientific Analyses (SAGA) 8.0.1 software.  The criteria used in this study are land use, DEM, slope, curvature, flow accumulation and distance from water. The result was then validated with MSB's flood model, previous studies and reports, on the Vänern flood 2000/2001 and insurance claims, because there are no previous flood risk maps in the study area to directly validate the maps generated in this study. In addition, 3D models were generated to facilitate visualization and simulation of the event. The proposed method is based on a combined application of geographic information systems (GIS) and multi-criteria analysis (MKA). AHP was used as a weighting method with the help of five experts from Karlstad University. Flood risk maps were constructed and classified into five classes with flood potential from very low to very high. Respectively 0.5 % (very low), 10 % (low), 28.9 % (moderate), 43 % (high), and 17.6 % (very high). The result of this study shows that 60.6 % of the area is associated with areas of high to very high flood risk. This study is a valuable practical tool for flood risk analysis.