Improvement of Forecasted Sea Surface Wind Initialized with Cycling 3DVAR-WRF

Abstract

A cycling three-dimensional variational data assimilation (3DVAR) method based on the WRF is developed to assimilate surface and upper-air meteorological observations. A cycling 3DVAR scheme with 6-h assimilation window is designed and employed to generate the initial conditions for this sea surface wind forecasting. The result shows that the forecasted sea surface wind improved compared to the result without cycling 3DVAR. Although the impacts of assimilating the observations on near-surface wind forecasts are limited, the fine structure of local weather systems illustrated by the WRF-cycling 3DVAR system suggests that assimilating the observations has a positive effect on the near-surface wind forecast under conditions that the observations in the initial condition are properly amplified. Assimilating conventional data is an effective approach for improving the forecast of the near-surface wind.oyed to generate the initial conditions for this sea surface wind forecasting. The result shows that the forecasted sea surface wind improved compared to the result without cycling 3DVAR. Although the impacts of assimilating the observations on near-surface wind forecasts are limited, the fine structure of local weather systems illustrated by the WRF-cycling 3DVAR system suggests that assimilating the observations has a positive effect on the near-surface wind forecast under conditions that the observations in the initial condition are properly amplified. Assimilating conventional data is an effective approach for improving the forecast of the near-surface wind.