The Numerical Prediction Uncertainties of the Tibetan Plateau Impacting the Forecasts of Its Downsteam Region
The influence of the initial uncertainties on the NWP forecasts over the Tibetan Plateau (TP) and adjacent regions was studied based on the GFS forecasts/WRF model and OSSE experiments.The result shows that:(1) The large uncertainties of Tibetan Plateau (TP),because of the scarcely observations and complex terrain,can affect the NWP forecasts over the TP region and decrease the forecast skill of the downstream region of TP.After comparing the forecast errors over the TP and the downstream regions,it is found that the errors over the TP region are larger than the errors over the TP adjacent/Pacific Ocean and Rocky Mountain regions on the first 24 hours' integration:the error amplification rate over the TP adjacent region are larger than other regions evidently.(2) The forecast error rise faster over the downstream of TP region than the downstream of the Rocky Mountain (US).The WRF model forecasts results,with/without the large terrain (TP/Rocky Mountain),indicate that the forecast errors over the downstream region of the TP/Rocky Mountain are larger if the model domain contains the large terrain areas:this error divergence significantly after 2-days integration.(3) The observation data over TP region have effect on the forecasts of the downstream areas:additional "synthetic" observation data can decrease the initial uncertainties over the TP region and then improve the forecast skills over the downstream of TP region.Therefore,this study suggests that in order to reduce the initial uncertainties influence over the TP region,the additional observations even the "synthetic" observations over the TP region could assimilated to the NWP model.
NETL
NSTL
万方数据
维普
