Extreme mesoscale weather in the Arctic region often forms in the vicinity of the ice edge. A polar low was observed in the region between Svalbard and the ice cap, 8. january 2010. This was the first polar low to have been spotted north of Spitsbergen, but also the first one to form north of 80°N.A series of fine mesh (1 km) experiments using the Weather Research and Forecasting (WRF) model are performed in order to examine the structure of the cyclone and the airflow within it and to determine the physical processes important for its development. The HIRLAM12 model was used for comparison.Observations show that the ice edge was simulated to close east of Svalbard. Due to the strong stability around the ice edge, and the fact that the ice edge deviated from the observed one in the satellite images, the polar low was underestimated. A control run was performed where the ice edge was moved further north and east. The position of the low became more exact, and due to the reduced static stability and the increased surface heating, the low became more intense. The polar low was similar to previous case studies in that it had a clear, calm, and warm eye structure, with the convection organized in a comma-cloud band. The highest wind speed (20 ms-1) was found in the eye wall.The high wind speed during the deepening stage resulted in high surface sensible and latent heat fluxes of about 500 and 250 Wm-2, respectively. Several sensitivity experiments were designed to analyse the physical properties of the polar low, and to test the possibilities of triggering a polar low through certain modifications of the surface conditions. Physical processes such as condensational heating and/or latent/sensible heat fluxes were turned on/off throughout the simulation, making it suitable to study the direct effects of the physical processes. The experiments suggest that the deepening stage of the polar low was dominated by baroclinic growth and that upper-level potential vorticity forcing contributed throughout its life cycle. After the deepening stage, the baroclinicity was still important in maintaining the disturbance. It is argued that its development is enhanced by large lapse rates that exist in the region of formation, since baroclinic theory indicates that growth rates are large when the static stability is low. The sensible heat transfer from the ocean surface was not an immediate cause of the deepening, but it plays an important role in the overall development by decreasing the stability of the environment, thereby allowing the aforementioned mechanisms to operate more effectively.