The Importance of Sea Surface Information in Forecasts of Precipitation from Landfalling Atmospheric Rivers

Numerical weather prediction models are most often prescribed low resolution (~ 0.5 degree) and time-invariant temperature at the sea surface boundary when creating real-time forecasts. When the forecast includes an atmospheric river (AR) event, this may lead to biases in the simulated boundary layer and low level jet. While AR most commonly exhibit neutral stability, near-saturatation and moderate vertical wind shear from the surface to near 3 km MSL. These same features are prone to modification by sea surface fronts and temporal eddies. In this study, we examine the impact of daily and subdaily, high resolution, satellite-derived sea surface temperature on forecasts of landfalling atmospheric rivers using the Weather Research and Forecast (WRF) model.  We investigate the improvement in accuracy of forecasted precipitation, water vapor transport, and marine boundary layer structure in AR of varying strength using in-situ data. Additionally, we comment on the role that the development of energy spectral density at appropriate scales has on the simulated atmospheric water vapor budget, including the resultant precipitation.