Search

Skilled seasonal forecasting will effectively reduce the economic losses caused by droughts and floods. Because of the powerful data mining capability of deep learning networks, it is increasingly applied in studies of seasonal rainfall prediction. However, there remain two prominent issues in the modeling process: the lack of enough training samples and the effect of a small number of extreme values on the model optimization. To tackle these deficiencies, we combine strategies such as principal component analysis, reduction of model hidden layers, and early‐stopping with Attention U‐Net to construct a rainfall classification forecasting model. These steps reduced the model outfitting and improved the model generalization. The results show that the prediction accuracy of this network with leads of 1–3 months is obviously better than that of the numerical model. Further analysis also supports that the spatial features of precipitation predicted by the network are very close to the observations. Plain Language Summary: Accurate summer precipitation forecasts can effectively reduce economic losses and casualties caused by heavy precipitation and flooding in China. Nowadays, seasonal forecasting still mainly relies on numerical models based on physical knowledge. However, the predictors of rainfall are so broad and complex that some physical processes are not yet fully understood. As a result, the improvement of the model skills encountered a bottleneck. Recently, deep learning (DL) network can achieve complex function approximations of data by automatically extracting and learning data features. Yet the existing DL seasonal prediction models suffer from over‐fitting and interference from noise and extreme values. Here, we use Attention U‐Net to predict rainfall levels and combine strategies such as principal component analysis, reduced model depth, and early‐stopping. This network not only solves the two drawbacks mentioned above. It also shows the superiority in the prediction results. First, the rainfall level prediction of this network is more accurate than the numerical model at 1–3 months ahead. Second, the network captures the spatial features with excellent performance. The predicted distributions are largely consistent with the true values. Key Points: The Attention U‐Net model outperforms the numerical model in predicting summer precipitation levels with a forecast lead of 1 monthConverting a regression task into a classification task can eliminate the bad effects of noises and extremesThe principal component analysis and the early‐stopping strategy facilitate the model's generalization [ABSTRACT FROM AUTHOR]

Taking the semiannual cycle of surface air temperature (SAT) into account in fitting the finer structure of the seasonal cycle, this study reveals the specific role of the semiannual cycle in modulating variations of the season onsets and lengths for the first time. The time-varying amplitudes and phases of the annual and semiannual SAT cycle for the period 1964 to 2017 in China are extracted from daily observations based on harmonic analysis. The results show that the phase differences between the two harmonics remain essentially unchanged, while the amplitude ratio of the semiannual to annual harmonics tends to increase after 1988. This enhanced semiannual cycle together with the global warming and annual cycle changes jointly led to a much longer summer with an average speed of 7.45 days/10 years and shorter spring, autumn, and winter with speeds of − 2.43, − 1.63, and − 4.26 days/10 years, respectively. Such obvious asymmetry cannot be found in fitting seasonal cycles by only annual harmonics. The average absolute percentages of the linear trends of season onsets and lengths caused by the intensification of the semiannual cycle are 52.63% and 78.66%. It indicates that season onsets and length not only depend on the phase of annual and semiannual harmonic, but also highly related to the strength of semiannual cycle. The time evolutions of semiannual cycle strength in China are found to be significantly correlated to the meridional temperature advection. The anomalous stronger southward wind in December can result in colder temperature and hence intensify the semiannual component. [ABSTRACT FROM AUTHOR]