Statistical-dynamical analog ensemble system for real time quantitative precipitation forecasts (QPFs) at local scale in the north-west Himalaya (NWH), India

This research work focuses on the development of an analog ensemble system with 10-members for real time quantitative precipitation forecasts (QPFs) at local scale (specific site) in the north-west Himalaya (NWH), India, by using the observed precipitation amounts in analog days (ANOB) and precipitation amounts simulated with the help of the weather research and forecasting model (WRF model) for day1 [d1, lead time 24 h (24 h)] in analog days (ANWRF). Developed analog ensemble system is validated for QPFs in the hindcast mode to explore on the possibility of developing statistical-dynamical analog ensemble system for real time QPFs. The performances of the WRF model for d1, ANOB, ANWRF and climatological forecast model for d1, d2 (lead time 48 h) and d3 (lead time 72 h) are evaluated and compared. The ANOB, ANWRF and WRF model exhibit similar patterns for QPFs but differ in their QPF biases. The ANOB and ANWRF exhibit marginally poor performances relative to the WRF model for the QPFs for d1. However, these models exhibit better skills relative to the climatological forecast model for the QPFs for d1, d2 and d3. The results suggest fair possibility of developing statistical-dynamical analog ensemble system which can be driven by the local scale meteorological observations for the QPFs at local scale. Moreover, it is fairly possible to use precipitation amount simulated with the help of the WRF model [or the other model(s)] for short lead time (d1) in analog days to produce real time QPFs for longer forecast lead times (d2 and d3).