Radar reflectivity (RR) greater than 35 dBZ usually indicates the presence of severe convective weather, which affects a variety of human activities, including aviation. However, RR data are scarce, especially in regions with poor radar coverage or substantial terrain obstructions. Fortunately, the radiance data of space-based satellites with universal coverage can be converted into a proxy field of RR. In this study, a convolutional neural network-based data-driven model is developed to convert the radiance data (infrared bands 07, 09, 13, 16, and 16–13) of Himawari-8 into the radar combined reflectivity factor (CREF). A weighted loss function is designed to solve the data imbalance problem due to the sparse convective pixels in the sample. The developed model demonstrates an overall reconstruction capability and performs well in terms of classification scores with 35 dBZ as the threshold. A five-channel input is more efficient in reconstructing the CREF than the commonly used one-channel input. In a case study of a convective event over North China in the summer using the test dataset, U-Net reproduces the location, shape and strength of the convective storm well. The present RR reconstruction technology based on deep learning and Himawari-8 radiance data is shown to be an efficient tool for producing high-resolution RR products, which are especially needed for regions without or with poor radar coverage.

The prediction of summer precipitation patterns (PPs) over eastern China is an important and topical issue in China. Predictors that are selected based on historical information may not be suitable for the future due to non-stationary relationships between summer precipitations and corresponding predictors, and might induce the instability of prediction models, especially in cases with few predictors. This study aims to investigate how to learn as much information as possible from various and numerous predictors reflecting different climate conditions. An objective prediction method based on the multinomial logistic regression (MLR) model is proposed to facilitate the study. The predictors are objectively selected from a machine learning perspective. The effectiveness of the objective prediction model is assessed by considering the influence of collinearity and number of predictors. The prediction accuracy is found to be comparable to traditionally estimated predictability, ranging between 0.6 and 0.7. The objective prediction model is capable of learning the intrinsic structure of the predictors, and is significantly superior to the prediction model with randomly-selected predictors and the single best predictor. A robust prediction can be generally obtained by learning information from plenty of predictors, although the most effective model may be constructed with fewer predictors through proper methods of predictor selection. In addition, the effectiveness of objective prediction is found to generally improve as observation increases, highlighting its potential for improvement during application as time passes.

Missing data in weather radar image sequences may cause bias in quantitative precipitation estimation (QPE) and quantitative precipitation forecast (QPF) studies, and also the obtainment of corresponding high-quality QPE and QPF products. The traditional approaches that are used to reconstruct missing weather radar images replace missing frames with the nearest image or with interpolated images. However, the performance of these approaches is defective, and their accuracy is quite limited due to neglecting the intensification and disappearance of radar echoes. In this study, we propose a deep neuron network (DNN), which combines convolutional neural networks (CNNs) and bi-directional convolutional long short-term memory networks (CNN-BiConvLSTMs), to address this problem and establish a deep-learning benchmark. The model is trained to be capable of dealing with arbitrary missing patterns by using the proposed training schedule. Then the performances of the model are evaluated and compared with baseline models for different missing patterns. These baseline models include the nearest neighbor approach, linear interpolation, optical flow methods, and two DNN models three-dimensional CNN (3DCNN) and CNN-ConvLSTM. Experimental results show that the CNN-BiConvLSTM model outperforms all other baseline models. The influence of data quality on interpolation methods is further investigated, and the CNN-BiConvLSTM model is found to be basically uninfluenced by less qualified input weather radar images, which reflects the robustness of the model. Our results suggest good prospects for applying the CNN-BiConvLSTM model to improve the quality of weather radar datasets.