Evaluation of the CAMS-CSM Model in predicting a Northeast Cold Vortex Event at the early Onset of 2020 Mei-yu season

Abstract

Intimately related to the Asian summer monsoon, the Mei-yu rainfall is also strongly influenced by atmospheric circulation in the middle to high latitudes, especially the Northeast Cold Vortex (NECV), and thus making prediction of the Mei-yu rainfall a challenging issue. The prediction skills of the CAMS-CSM (full name as the Chinese Academy of Meteorological Sciences Climate System Model) for a NECV event at the early onset of 2020 Mei-yu season and the associated Mei-yu rainfall are evaluated. Hindcast experiments with two different horizontal resolutions are employed to assess the climate model in predicting the synoptic system. The ERA5 reanalysis data and the CN05.1 precipitation data are adopted for comparison. Results indicate that both the middle-resolution (T106, ∼1°) and high-resolution (T255, ∼0.47°) version of the CAMS-CSM model are able to capture this NECV event, while showing some biases on its intensity, duration and location. Overpredictions in the strength and duration of the NECV are found in the T106 version, which could be attributed to a stronger dry invasion from higher levels during the development stage, an intensified Okhotsk blocking high, and a weakened upper-level jet at the decaying phase. As for the T255 version, a stronger NECV is predicted further north than the reanalysis, and sources of the biases are identified, including northward displacements of the high-level dry invasion and the westerly jet. Further investigation suggests that the T106 version performs better than the T255 version in predicting the Mei-yu rainfall because of a more accurate prediction on the location of the NECV. Positive biases in the cold advection and Mei-yu front are identified in the T106 version, contributing to a stronger Mei-yu rainfall than the observation. The Mei-yu rainfall is poorly predicted in T255 version due to prediction errors in the location of the NECV-induced cold advection. Therefore, improving the prediction skill of the NECV, not only its intensity but also location, is of vital significance to achieve a better prediction of the Mei-yu rainfall.