Abstract:The spatial-temporal characteristics were well predicted for flood season in 2021, including typical features such as “normal to poor climatic conditions nationwide, concurrent drought and flood disasters in different regions and stages, more frequently extreme weather and climate events, main rainfall belt located in northern China”. We further successfully predicted that heavy floods might occur in parts of Haihe River Basin, Songhuajiang River Basin, and the upper and lower reaches of Yangtze River. The issued operational forecasts well captured the main features of summer climate except two shortcomings. Firstly, the intensity of abnormal heavy rainfall in northern China was underestimated, partly due to the predictability limits of extreme events. Secondly, the inconsistency of drought tendency existed between the forecast and observation over the eastern part of Northwest China, the middle reaches of the Yangtze River and the eastern part of South China. The mid-summer issued forecast in late June predicted there would be heavy precipitation in the middle and lower reaches of the Yangtze River and most of northern China, which was more consistent with the observation. The possible impacts of several key predictors on the East Asian summer monsoon (EASM) in 2021 were analyzed comprehensively in March including the cold phase of the Pacific decadal oscillation (PDO), decaying La Ni〖AKn~D〗a event, less snow cover over the Tibetan Plateau in pre-winter, and the positive phase of the North Atlantic tripole mode. These predictors and dynamic models all forecasted the EASM in 2021 would be stronger than normal and might lead to more rainfall in northern China in summer. However, both the precursory signals and dynamic models cannot predict the intraseasonal variations of EASM three months in advance. Finally, the external forcing signals and circulation in 2021 are compared with that in 2020. It should be note that the factors affecting precipitation in flood season are complicated and need further studying.