The monsoonal rainy season in eastern China has distinct regional and temporal characteristics (Ding, 1994; Ding and Chan, 2005; Wang and Ding, 2008). Every year, with the northward advance and southward retreat of the East Asian summer monsoon, southern China, the Yangtze River Basin, and northern China successively enter a period of concentrated precipitation (Wang and Ding, 2008). With the first northward jump of the Western Pacific subtropical high, the East Asian summer monsoon pushes the rain belt toward the Yangtze and Huai River basins in June and July to form plum rains (also called Meiyu) (Qian et al., 2009). Extreme precipitation events such as floods and droughts during the Meiyu rainy season have severely affected the economic development and the people’s safety in eastern China. In recent years, the precipitation amount during the Meiyu rainy season has shown an increasing trend in some areas (Wu and Zhan, 2020; Wu et al., 2021). In 2020, the Yangtze and Huai River basins recorded the largest amount of the Meiyu rainy season precipitation since 1961. The direct economic losses and deaths far exceeded other meteorological disasters (Zhang et al.,2020; Chen et al., 2020). With the second northward jump of the subtropical high, the 2nd and 3rd rainy seasons in North China begin in July. The average duration of the rainy season can reach 32 days and generally accounts for half of the total summer precipitation in the region (Zhao et al., 2017; Chen et al., 2019). The sub-seasonal precipitation variation in the rainy season in North China is often accompanied by an extensive period of rainstorms and floods or drought disasters, which cause great economic loss and casualties. Thus, an accurate precipitation forecast in Meiyu and rainy season in North China is vital for disaster prevention and mitigation.
The direct cause of precipitation anomaly during the Meiyu rainy season is the anomalous atmospheric circulation of the East Asian summer monsoon. The East Asian summer monsoon is mainly affected by the South China Sea monsoon trough, the Meiyu front, the Western Pacific subtropical high, the Qinghai-Tibet high and the northern cold air (Tao and Chen, 1987). The influence of the northern cold air can be explained as the static Rossby wave of the high-altitude subtropical jet, which has a greater impact on the climate of the Meiyu rainy season (Tao and Wei ,2006). When the Okhotsk Sea high forms and stabilizes, the anomalous waves in the middle and high latitudes of Asia and East Asia often cause precipitation anomalies in East Asia during the Meiyu rainy season (Zhang and Tao, 1998). In addition, external forcing such as the ENSO cycle, thermal conditions in the Western Pacific, snow cover and polar ice are closely related to the interannual variation of the plum rains (Tao et al., 1988; Huang and Sun, 1994; Liu and Cao, 1994; Zhang and Tao, 1998; Wu and Qian, 2000; Wei and Song, 2005; Feng et al., 2016). The precipitation in the rainy season in North China has a good relationship with the intensity of the East Asian summer monsoon (Zhu, 1934; Yang et al., 2012; Hao et al.,2016). The circulation anomaly in middle and high latitudes, the position of high-altitude jets, the east-west position of the South Asian high, and the Arctic Oscillation also have an impact on precipitation in the rainy season in North China (Wei et al., 2003; Huang et al., 2006; Wang and He,2015; Xu et al., 2015; Lu, 2002; Du et al., 2009; Gong et al., 2002), while external forcing such as ENSO, different modes of the Indian Ocean, and land surface conditions of the Qinghai-Tibet Plateau are all related to the anomalous precipitation in the rainy season in North China (Lin and Yu, 1993; Yang and Ding, 2007; Zhang and Tao, 2001; Yu and Chen, 2012).
The development of dynamical model prediction systems contributed to important advances in sub-seasonal to seasonal climate prediction (Saha et al., 2006, 2014; Vitart et al., 2017; Ren et al., 2021). Some models have shown reasonable skill in predicting interannual variability of the Asian monsoon intensity (Jiang et al., 2013; Zhu and Shukla, 2013; Liu et al., 2014, 2015). However, reproducing regional precipitation over China remains a great challenge for climate models (Liu et al., 2013, 2019; Gong et al.,2017; Liang et al., 2019). The low prediction skill is partly associated with the fact that the regional climate processes cannot be well simulated or predicted in climate models. The sub-seasonal prediction skill of the summer precipitation over eastern China is low on average in the Beijing Climate Center (BCC) Atmospheric General Circulation Model (BCC_AGCM2.2), which is linked to the ability of the model in simulating the western Pacific subtropical high (WPSH). Moreover, as a major contributor to predictability, the spatial pattern and sub-seasonal evolution of the Pacifc-Japan region are not well predicted (Liu et al., 2020). In addition, some physical mechanisms in the model show discrepancies, and the impact of some key circulation systems on precipitation has noticeable deviations. In most comprehensive average forecasts, even for short-term forecasts, the relationship between some dynamic monsoon indices and precipitation patterns is overestimated in the BCC Climate System Model (BCC CSM1.1) (Liu et al., 2015).
The circulation systems that affect Meiyu and rainy season in North China are more complex. The two rainy seasons are affected by low- and mid- and high-latitude circulation systems (Chen et al., 2019). The model’s circulation forecasting skills at various latitudes are different. In other words, the circulation in low-latitudes is more predictable than in mid- and high-latitudes (Wu et al., 2017; Ren et al., 2017). The two rainy seasons belong to different stages of the East Asian summer monsoon on China. Circulation system configurations at low- and high- latitudes are different (Chen et al., 2019). The predictability of the two rainy seasons is likely to be different. Therefore, discussing the prediction of key circulation systems in the two rainy seasons and presenting their relationship with precipitation based on models is of great significance for understanding the climatic factors and physical processes that affect the precipitation during the extended periods of the two rainy seasons.
In this study, the difference in the precipitation prediction skills between Meiyu and rainy season in North China are analyzed and compared. The ability of the model (The S2S dataset produced by ECMWF) to predict the key circulation systems that affect precipitation in the rainy seasons and its relationship with precipitation is assessed. The possible reasons for the difference in the prediction skills of precipitation are discussed. Section 2 describes the data and methods. Section 3 presents the contrasting precipitation prediction between Meiyu and rainy season in North China. Section 4 assesses the ability of the model to simulate ENSO and the general circulation associated with precipitation in Meiyu and rainy season in North China. The sources of prediction biases between Meiyu and rainy season in North China are discussed in Section 5. Summary and discussion are given in Section 6.