A New Comprehensive Physical Index Based on Ingredients-based Methodology and Its Application in Forecasting Heavy Rainfalls

Heavy rainfall often leads to floods and related disasters, and it is also one of the difficulties for weather forecast. In recent years with the gradual development of numerical weather prediction models, the level of weather forecast is also rising. However, there is still a large deviation from the specific weather phenomena forecast especially strong convective precipitation forecasts. In China, the current short interval of six hours of heavy rainfall TS score is only 2.6%, far below of the wind field and temperature field prediction score. Hence, the accuracy rate of strong rainfall forecast still needs to be improved. Some studies found that analysis and data mining strong signals form the various early physical factors are very important to improve the level of heavy rainfall forecast. So by analysis and data mining the strong signal from the various early physical factors, based on the philosophy of Ingredients-based Methodology and combined with the advantage of Pattern-recognition Method and physics diagnosing, a new physical index THP is established, which integrated atmospheric dynamics, thermodynamics and moisture factors. The THP index not only reflects the thermodynamics, dynamics and moisture situation, but also integrated the main three factors of conducive occurrences of heavy rainfall. The THP index amplified strong signals, attenuated weak signals, and enhanced forecasting stability by calculated the intersection between the three physical factors. The THP index is then diagnosed and tested through two heavy precipitation processes in Henan province and Beijing in China by using NCEP/NCAR reanalysis data and ground observational data. The results showed that the forecasting performance and stability of the new index THP which based on Ingredients-based Methodology, is better than the method of using single physical parameter (θ_se flux divergence, relative horizontal helicity and PW) in forecasting analysis. Compared with some conventional physical indexes (K index, Quasi-geostrophic Q vector divergence, moisture vertical helicity), the high value distribution of THP is great consistent with precipitation areas in the following six hours. The precipitation phase lag behind THP phase variation, the variation of THP indicates the beginning and ending time of a heavy rainfall. Besides, the THP index is also some significance of precipitation intensity forecast which occurs in the pre-trough pattern. Therefore, the THP index can be used in the objective probability forecast and enrich the model results as a statistical post-processing of model output for heavy rainfall forecast. It provides a new reference to weather forecasters. Key words: Ingredients-based Methodology, heavy rainfall, THP index, upper trough