Understanding the dynamics of cetacean distribution in ecologically vulnerable regions is essential to interpret the impact of environmental changes on species ecology and ecosystem functioning. Species distribution models (SDMs) are helpful tools linking species occurrences to environmental variables in order to predict a species’ potential distribution. Studies on baleen whale distribution are comparably rare in polar regions, mainly due to financial or logistic constraints and habitat suitability models are scarce. Using SDMs, this master thesis aims at identifying areas of suitable habitats for fin whales (Balaenoptera physalus) in the Nordic Seas during summer. A further aim is to identify important environmental variables that potentially drive the species’ distribution. Opportunistic data were collected during ten RV Polarstern cruises from 2007 to 2018 during summer months (May to September) along with complementary opportunistic data from open source databases. Environmental covariates were chosen based on ecological relevance to the species, comprising both static and dynamic variables. MaxEnt software was used to model fin whale distribution, with presence-only data as a function of carefully chosen environmental covariates. This master thesis is one of the first studies to use SDMs to model suitable habitats of fin whales in the Arctic Ocean and revealed a link of the occurrence of fin whales to specific environmental variables. Most contributing variables were distance to shore and distance to sea ice edge, suggesting both static and dynamic variables to have an impact on habitat suitability in the Arctic Ocean. Four other environmental variables, namely bathymetry, slope, variability of sea surface temperature and mean salinity at 100 m depth were shown to also have an impact. Areas of high suitability were pronounced around the southwestern and -eastern side of Svalbard, as well as on the northern tip of Norway and southern East Greenland. These results generally demonstrate the effective use of SDMs to predict species distribution in highly remote areas, constituting a cost-effective method for targeting future surveys and prioritizing the limited conservation resources. Results can be applied in a variety of purposes, such as designing marine protected areas, guiding seismic surveys and support the further use of opportunistic data in research.