The evolution of lightning generation and extinction is a nonlinear and complex process, and the nowcasting results based on extrapolation and numerical models largely differ from the real situation. In this study, a multiple-input and multiple-output lightning nowcasting model, namely Convolutional Long- and Short-Term Memory Lightning Forecast Net (CLSTM-LFN), is constructed to improve the lightning nowcasting results from 0 to 3 h based on video prediction methods in deep learning. The input variables to CLSTM-LFN include historical lightning occurrence frequency and physical variables significantly related to lightning occurrence from numerical model products, which are merged with each other to provide effective information for lightning nowcasting in time and space. The results of batch forecasting tests show that CLSTM-LFN can achieve effective forecasts of 0 to 3 h lightning occurrence areas, and the nowcasting results are better than those of the traditional lightning parameterization scheme and only inputting a single data source. After analyzing the importance of input variables, the results show that the role of numerical model products increases significantly with increasing forecast time, and the relative importance of convective available potential energy is significantly larger than that of other physical variables.

China New Generation Weather Radar (CINRAD) operated in volume coverage pattern 21 (VCP21) mode may miss important information about precipitation for storms that evolving rapidly and moving fast due to its volume scanning interval (about 6 minutes). However, despite the rapid changes and fast movement of storms, according to the continuity equation, the pattern and location of storms are continuous in temporal and spatial. An attempt is made in this research to use interpolation to supplement the missing information about storm evolution and improve the radar quantitative precipitation estimation (QPE) quality. Three interpolation methods are used to supplement storm evolution information between scan intervals for two fast-moving storm events. In these events, weather radar using VCP-21 mode cannot capture storm evolution well. The results show that all of the three interpolation methods do enhance the accuracy of the radar QPE. However, the linear interpolation method only describes the linear variation of storms, and the optical flow method can consider the movement of the storm and has a significant improvement in capturing the rainfall area, yet it cannot delineate the variation in rainfall shape and intensity. The Bi-directional optical flow (BIO) can describe not only the storm movement, but also the change of intensity and pattern of storms. BIO can greatly enhance the accuracy of radar QPE among the three methods. This study suggests that the interpolation of fast-moving storms between two neighboring radar volume scans needs to take into account both intensity variation and precipitation movement. The application of BIO method can significantly improve the accuracy of radar QPE, namely the rainfall intensity and pattern.

Accurate and refined nowcasting of short-term intense precipitation can be utilized to provide timely warnings, and mitigate the damage caused by hydrological and meteorological disasters. Radar echo extrapolation can provide highly efficient forecasts of precipitation with high spatial and temporal resolution during the first 3 hours and has been widely used in operational nowcasting systems in aviation, meteorology, hydrology and other fields. However, these extrapolation methods still have drawbacks and need to be improved to provide more accurate nowcasting. In this article, an improved nowcasting method, based on continuity of tracking of radar echo with correlations (COTREC), named ensembled COTREC (EnCOTREC) was proposed. A comparison was then made between the COTREC method and the EnCOTREC method for several types of precipitation processes. By comparing the EnCOTREC method and the COTREC method in squall line, convective and stratiform precipitation processes, the results show that the EnCOTREC method is better at describing precipitation motion than the COTREC method and thus performs better in precipitation nowcasting. Especially for processes with a large precipitation area and uniform precipitation intensity, the accuracy improvement of EnCOTREC is more pronounced. Through time series analysis, the results of the EnCOTREC method are more stable than those of the COTREC method, the EnCOTREC methods can perform better than COTREC with an overall improvement of 5%.

Biases in high-spatiotemporal-resolution radar quantitative precipitation estimation (QPE) can substantially compromise the reliability of meteorological and hydrological disaster warning. Existing QPE correction methods commonly assume predefined spatial smoothness, whereas localized biases can vary sharply over short spatial distances. We therefore propose an implicit neural representation (INR) framework for continuous radar QPE bias-field reconstruction (QPE-BiasINR). QPE-BiasINR introduces QPE-background modulation, which conditions bias estimation on the local precipitation context to account for precipitation-dependent spatial variations in radar QPE errors. Moreover, a multiscale spatial consistency (MSC) constraint is incorporated into the loss function to explicitly strengthen spatial consistency. QPE-BiasINR was evaluated using hourly multi-radar mosaic QPE and gauge observations within the coverage of four radars in the Daqing River Basin,China, during June–September 2023. Test set performance was benchmarked against the uncorrected radar QPE, mean-field bias correction, and Kriging. QPE-BiasINR achieved the highest correlation coefficient (0.84) and the lowest root mean square error (3.29 mm) and mean absolute error (1.68 mm). Ablation experiments on QPE-BiasINR demonstrated that QPE-background modulation improved accuracy, while the MSC constraint reduced localized overcorrection. These results indicate that QPE-BiasINR can improve high-resolution radar QPE, with potential value for hydrological modeling and operational flood-risk monitoring.