| Summary: | Tropical cyclones (TCs) are one of the most severe weather hazards in the tropics, causing many human deaths and substantial property loss. Therefore, it is always urgent to significantly reduce systematic errors that may hinder the accurate forecast of TC activity. To evaluate the ability of the Euro-Mediterranean Center for Climate Change (CMCC) forecast system to represent TC activity and to verify the feasibility of TCs forecast, we use a new version of the CMCC Seasonal Prediction System (SPS3.5), which configures with high spatial resolution (0.5 degrees) in the atmospheric component and a large (40) number of ensemble members. We compared SPS3.5 hindcasts from 1998 to 2016 July-September with observed TC tracks derived from the International Best Track Archive for Climate Stewardship (IBTrACS). To identify storms and track their trajectories in SPS3.5, we use the Geophysical Fluid Dynamics Laboratory (GFDL) Tropical Cyclone tracking algorithm and set strict thresholds to ensure the authenticity of the captured TCs. We find that the SPS3.5 system captures the spatial distributions in the North Hemisphere (R~0.8) well, although it underestimates TCs' number magnitude, which relates to the model resolution and thresholds. TCs spatial distribution is best captured over the Pacific, where the largest peak season (August) is covered by our dataset, while lower-skill is shown over the North Atlantic Ocean and the North Indian Ocean. For TC variability, the model performs well only in the Western North Pacific (R up to 0.6). We also investigate the relationship between TCs' intensity and interannual variability with ENSO.
|
|---|