Pardeep Pall, Lena M. Tallaksen, Frode Stordal

Rain-on-snow (ROS) events are complex multivariate hydrometeorological phenomena requiring a combination of rain and snowpack. Impacts include floods and landslides, and rain may freeze within the snowpack or on bare ground, potentially affecting vegetation, wildlife, and permafrost. ROS events occur mainly in high-latitude and mountainous areas, where sparse observational networks hinder accurate quantification – as does a scale mismatch between coarse (50-100 km) resolution re-analysis products and localised events. Variability in the rain-snow temperature threshold, and temperature sensitivity of snowmelt, adds additional uncertainty. Here the high-resolution (1 km) seNorge hydrometeorological dataset, capturing complex topography and drainage networks, is utilized to produce the first climatology of ROS events for mainland Norway. For daily data spanning 1957-2016, suitable rain and snowpack thresholds for defining ROS events are applied to construct ROS climatologies for 1961-1990 and 1981-2010, and investigate trends. Differing ROS characteristics are found, reflecting Norway’s diverse climates. Compared to 1961-1990, events in the 1981-2010 period decrease most in the southwest in winter, southeast in spring, and north in summer (consistent with less snow cover in a warming climate), and increase most in the southwest, central mountains, and north in winter-spring (consistent with increased precipitation and/or more snow falling as rain in a warming climate). Events also broadly correlate with the North Atlantic and Arctic Oscillations, with the latter dominating slightly. Finally, the extreme May 2013 ROS event in central Oppland county, causing devastating flooding, is examined to better understand the interplay between rain, snow, and temperature, with a view to future studies.