Heavy rain events in the Arctic archipelago Svalbard can have a high impact on nature and settlements, and are predicted to become more frequent and intense in the coming decades. This thesis work focuses on heavy rain events ( > 10 mm/24 hr) in summer and autumn season of 2016, 2017 and 2018. Heavy rain events in this period were found to be most common during southwesterly synoptic air flow over the archipelago. By comparing observations of heavy rain events with precipitation forecasts from the operational weather models AROME Arctic and ECMWF IFS HRES, the ability of AROME Arctic to predict such heavy rain events is evaluated. Both models are found to underestimate heavy rain events for the West-Coast stations Hornsund, Isfjord Radio and Ny-Ålesund, while for Longyearbyen there is a slight overestimation. By evaluating precipitation metrics from a model grid point box, spanning a larger area around the stations, the mean error and mean absolute error are substantially lowered. To complement the measurements from official MET stations measuring precipitation in Svalbard, six additional automatic weather stations with precipitation instruments were built and set up in the remote Tempelfjorden area, and around Longyearbyen, in the summer and autumn of 2018. The results from the field campaign show that it is possible to make reasonable measurements of heavy rain events in Svalbard from a relatively low-cost field campaign. In the observation period, the field campaign rain gauges measured one heavy rain event (19. August 2018 - 13 mm) with a maximum hourly intensity of 5 mm/hr. Additionally, a positive precipitation gradient of ~10 % per 100 m elevation was calculated based on field campaign observations. In Tempelfjorden, there seems to be a relation between the direction of large scale synoptic air flow, and when the model is underestimating or overestimating precipitation. Through the field period a persistent undercatchement was found for the official precipitation instrument in Adventdalen, the Geonor-T200B, when compared with field campaign measurements.