The focus of this research is on atmospheric predictability at the upper ranges of deterministic forecasting (up to about two weeks). The study will use data assimilation based on the Local Ensemble Transform Kalman Filter method to examine the impact of observations on forecasts, and in particular to determine a priori the most needed locations for specific measurements ("targeted observations") to improve forecasts. This research extends previous work in perfect model scenarios (comparing models to models) to more realistic applications (forcast models versus real world). The study will also examine the predictability time limits for different circulation regimes, investigate seasonal effects on predictability, and assess the impacts of observations collected in the upcoming THORPEX Pacific Asian Regional Campaign (T-PARC) on forecasts. The study will use a recent operational version of the Global Forecast System model of the National Centers for Environmental Prediction. The fruits of the research have the possibility to improve forecasting by pinpointing where additional observations are most useful, thereby optimizing deployable resources. The use of an operational forecast model should ease the transition of knowledge from the basic research phase to operations. A graduate student will be supported and involved in the research.
