Improving precipitation ensemble forecasts of typhoon heavy rainfall over East China with a modified probability-matching technique

In this paper, four schemes involving the probability-matching technique were studied to obtain ensemble-based quantitative precipitation forecasts (QPFs) associated with Typhoon Lekima over East China. With the use of this technique, synthetic ensembles were created by blending low- and high-resolution rainfall forecasts. To effectively derive high-resolution ensemble forecasts, the neighborhood method was applied to mesoscale deterministic forecasts. Four schemes were explored based on the probability-matching technique. Two schemes resulted in ensemble forecasts, and the other two schemes yielded deterministic forecasts. By analyzing quantitative precipitation forecasts (QPFs) and ensemble forecasts, modified probability-matching-based schemes were determined to substantially reduce or eliminate the intrinsic model rainfall bias and to provide better QPF guidance. These encouraging results suggest that the modified probability-matching technique is a useful tool for QPFs of typhoon heavy rainfall over East China using dual-resolution ensemble forecasts.