Abstract ID: 3.8597 | Accepted as Poster | Poster | TBA | TBA

Mountain snowfall is poorly observed, leading to large uncertainties and biases in weather models and therefore in our knowledge of water resources that are important to hundreds of millions of people. We present a novel set of observations from more than 30 sites in the Alps, Himalayas, Rockies, Finland, Greenland and Southern Ocean Islands that overcome key limitations of the conventional instruments that are routinely used to measure falling or accumulating snow water equivalent. We describe how our method uses frozen lakes as sensing surfaces, avoiding notable measurement biases of pluviometers and snow pillows and, most importantly, covering a very much larger surface area. The large scale of our lakes makes them comparable to the grid scale of operational weather models, which has allowed us to test their skill in representing mountain precipitation. We illustrate the instrument design as well as lessons learned from its deployment, and we present examples from these sites, with a particular focus on Austria’s Rofental INARCH catchment. Finally, this unique repository of snowfall measurements has also enabled us to produce a snowfall-optimised version of the atmosphere-only version of the UK Met Office Model (MetUM) at a spatial resolution of 1.5 km with the MetUM able to accurately simulate both the timing and amounts of snowfall observed.