Verification of winter precipitation forecasts and analyses

English Abstract:

Numerical Weather Prediction (NWP) models lack verification for mountainous regions during the winter season, although they are providing input for hydrological modelling and for decision makers. Winter precipitation from two NWP models (GEM-LAM and GEM15) and from a precipitation analysis system (CaPA) was verified at approximately 100 stations in the mountains of western Canadian and northwestern US. Instead of rain gauges which have a systematic undercatch for solid precipitation and a rather unknown performance in complex terrain, ultrasonic snow depth sensors and snow pillows were used to observe daily precipitation amounts. Another advantage of these sensor networks are that they are available in relevant elevations in Canada. A detailed objective validation scheme highlights many aspects of forecast quality. Overall, the models underestimated precipitation amounts, although low precipitation categories were overestimated. This is oppositely to former verifications of these models, which were mainly performed in the summer and in flat terrain and/or with rain gauges. The finer resolution model GEM-LAM performed best in all analysed aspects of model performance, while the precipitation analysis system performed worst. An analysis of the economic value of large precipitation categories showed that only mitigation measures with low cost/loss ratios (i.e. measures that can be performed often) will benefit from these NWP models. This means that measures with large associated costs (relative to anticipated losses when the measure is not performed) should not or not primarily depend on forecasted precipitation.