Improvement of the Ensemble Methods in the Dynamical–Statistical–Analog Ensemble Forecast Model for Landfalling Typhoon Precipitation

The Dynamical–Statistical–Analog Ensemble Forecast model for landfalling typhoon precipitation (the DSAEF_LTP model) identifies tropical cyclones (TCs) from history data that are similar to a target TC and then assembles the amounts distributions of those identified obtain TC. Two original ensemble methods in model, namely, mean maximum, tend under- over- forecast precipitation, respectively. In addition, these two unable at stations beyond their maxima. To overcome shortcomings improve performance following five new incorporated: optimal percentile, fuse, probability-matching mean, equal difference-weighted track Similarity Area Index-weighted mean. Then, experiments TCs over China 2018 conducted evaluate with methods. Results show overall percentile 90th percentile) method is superior, false alarm rate lower than As compared operational numerical weather prediction models, improved shows advantages predicting accumulated rainfall, especially rainfall 250 mm. When implementing experiments, above results, however, it found varies, depending on type tracks. That is, westbound significantly better northbound TCs. address this issue, different schemes used types. TCs, using method, respectively, much single all