Skill Test of the West-WRF and GFS Models Verified Using CalWater Dropsonde Observations

Atmospheric rivers (AR) play a crucial role in the horizontal transport of water vapor and moist static energy in the midlatitudes and in delivering water to a variety of continental climate zones. In California, up to 60% of the annual precipitation depends on the arrival of a small number of AR. Despite their importance, state-of-the art atmospheric circulation models are consistently poor in predicting AR location and timing. We will demonstrate that model predictions also contain large errors in the magnitude of AR horizontal vapor transport. In this study we aim to compare the prediction skill in horizontal water vapor transport from a modified version of the Weather Research and Forecast (West-WRF) and the Global Forecast System (GFS) models. We verify model skill using dropsonde observations taken from the CalWater 2014 - 2015 field campaigns and a ground-based network of co-located wind profiling radar and GPS receivers. We compare each model across a large number of lead times ranging from 12 hours to 8 days. Our preliminary results suggest that the Integrated Vapor Transport (IVT) and total vapor flux are more accurately predicted by the higher resolution West-WRF model. Furthermore, we find that GFS typically has a consistent 2-6 hour lag in the timing of peak water vapor flux compared to the West-WRF model. Physical explanations of the more accurate West-WRF horizontal vapor transport and the apparent delay in peak vapor flux timing are also examined.