A two‐dimensional stochastic‐dynamical quantitative precipitation forecasting model

Abstract

Presented is a two‐dimensional model for quantitative precipitation forecasting at spatial and temporal scales relevant to hydrologic forecasting. The model uses as input operational forecasts of surface air temperature, pressure and dew‐point temperature, and wind speed and direction at midtropospheric levels. Such forecasts are expected to be available at the spatial scales of the large‐scale numerical weather prediction models. The two‐dimensional precipitation model produces as an output mean areal precipitation at scales down to 100 km2 in space and 1 hour in time. Thermodynamics and microphysics are used to determine the source and sink terms in the conservation equation for the mass of condensed liquid water equivalent utilized by the model. A state estimator has been designed to update precipitation model states from observations of precipitation in real time and to produce estimates of the variance of the precipitation forecasts. Preliminary tests of the model with hourly data from Oklahoma appear encouraging. The model developed offers a link in the operational environment between the prediction scales of the operational weather prediction models and the hydrologic prediction scales.