A wind profiler and GPS-based water vapor flux tool for precipitation forecasting in coastal mountains

The skill of quantitative precipitation forecasts (QPF) is often poor, including for extreme precipitation events enhanced by orography. This paper describes a new prototype operational tool that combines coastal wind profiles recorded by all-weather wind profiling radars with collocated integrated water vapor (IWV) measurements derived from global positioning system (GPS) receivers to estimate the bulk transport of water vapor, a process that contributes significantly to mountain precipitation enhancement. Based on earlier orographic precipitation work in California's coastal mountains, a controlling wind layer is defined that has maximum correlation between the horizontal component of the upslope wind in that layer and the rainfall measured downstream along elevated terrain. The altitude of the maximum correlation (~1 km above sea level) often corresponds to the altitude of the low-level jet that is typically present in the warm sector of approaching midlatitude cyclones and that often bears little resemblance to the wind measured at the surface, thus highlighting the need to obtain upper-air wind measurements for this particular application. The upslope wind in the controlling layer is then combined with the simultaneously measured IWV to calculate hourly, layer-mean, bulk water vapor fluxes. Although IWV is a column-integrated value, water vapor is typically concentrated in the lower troposphere. Hence, to first order, the bulk water vapor flux provides a close estimate of the low-level water vapor transport into the coastal mountains. An analysis of four winters of data demonstrate the close relationship between the magnitude of the bulk water vapor fluxes and mountain precipitation. These results are integrated into a prototype real-time diagnostic tool that has the potential to improve short-term QPF in coastal mountains. We will show initial bulk flux results across coastal California obtained during the landfall of an intense storm that heavily impacted the state in early January 2008. For comparison, the hourly bulk flux observations will be presented in tandem with numerical model results. The 48-h model forecast will be compared with the observations to help gauge how well the model is performing with respect to the orographic forcing and associated QPF of this storm.