Intercomparison of Forecast Skill in Warm Season across CMA Regional Ensemble Forecasting Systems in Zhejiang Complex Topographical Areas
Article
Figures
Metrics
Preview PDF
Reference
Related
Cited by
Materials
Abstract:
In this study, we conduct a statistical verification and analysis on near-surface temperature, wind speed and precipitation forecasts during the warm season in Zhejiang using CMA regional ensemble prediction systems at 3-km and 10-km horizontal resolutions (CMA-REPS 3 km and CMA-REPS 10 km). The study results, which provide a scientific basis for objectively evaluating CMA multi-horizontal-resolution regional ensemble prediction systems across various terrains and for subsequent improvements in ensemble forecasting methods, can be summarized as follows. (1) Both CMA-REPS effectively capture the diurnal variations of near-surface elements. However, the forecasts exhibit a negative bias in temperature and overestimation of wind speed, while overestimating precipitation from afternoon to early morning. (2) Compared to CMA-REPS 10 km, CMA-REPS 3 km reduces forecast errors for temperature and wind speed, with a maximum decrease of 23.07% in continuous ranked probability score(CRPS). It also enhances the detectability of precipitation forecast, improving the area of relative operating characteristic(AROC) by up to 19%. Meanwhile, CMA-REPS 3 km demonstrates superior fraction skill score(FSS) for heavy rainfall and provides more accurate forecast in diurnal variations of near-surface elements. However, the spread of early forecast is smaller. (3) In complex terrains, CMA-REPS 3 km shows significant improvements in the probability forecast errors of temperature and wind speed. Additionally, CMA-REPS 3 km exhibits the better spread-skill relationship for wind speed in hilly and mountainous, as well as for temperature in plain and basin. Not only that, CMA-REPS 3 km has better capability for the spread of heavy precipitation and the development of rain bands in steep terrain areas. As such, CMA-REPS 3 km demonstrates superior near-surface forecasting capability over complex terrains, particularly for 2 m temperature, 10 m wind speed and precipitation.