The NWP models' evaluation of landfall processes of the strong typhoon Fitow

Xiaofeng WANG,Xiaolin XU,Xuchao YANG,Yuncheng ZHANG
DOI: https://doi.org/10.13878/j.cnki.dqkxxb.20170124001
2017-01-01
Abstract:Based on the observation data and the ECMWF reanalysis data,the forecast products made by STI-THRAPS (Shanghai Tropical cyclone High resolution Analysis and Prediction System) and other four models,ECMWF/EC(European Centre For Medium Range Weather Forecasts),GFS (The Global Forecast System),JMA (Japan Meteorological Agency) and T639(TL639L60 Global Model),are verified and assessed by statistic method for typhoon Fitow.The results show that the STI-THRAPS has significant advantages for the prediction of torrential rain and severe rain.The score of STI-THRAPS is better than other models.There is little difference between each two models of the prediction from light rain to heavy rain.GFS and STI-THRAPS have made a good prediction for the wind area.The special related degree of wind filed between STI-THRAPS and observation is the most optimal.Besides that,STI-THRAPS has significant advantages in false alarm ratio and false negative ratio.The STITHRAPS model is the best at the prediction of the track after 24 h,but the center pressure and the maximum wind speed are higher than observation.
What problem does this paper attempt to address?