Eastern China is one of the most economically developed and densely populated areas in the world. Due to its special geographical location and climate, eastern China is affected by different weather systems, such as monsoons, shear lines, typhoons, and extratropical cyclones. In the near future, the rainfall rate becomes difficult to predict precisely due to these systems. Traditional physics-based methods such as numerical weather prediction (NWP) tend to perform poorly on nowcasting problems due to the spin-up issue. Moreover, various meteorological stations are distributed in this region, generating a large amount of observation data every day, which have great potential for application to data-driven methods. Thus, it is important to train a data-driven model from scratch that is suitable for the specific weather situation of eastern China. However, due to the high degrees of freedom and nonlinearity of machine learning algorithms, it is difficult to add physical constraints. Therefore, with the intention of using various kinds of data as a proxy for physical constraints, we collected three kinds of data (radar, satellite, and precipitation data) in the flood season from 2017 to 2018 in this area and preprocessed them into tensors (256×256) that cover eastern China with a domain of $\mathrm{12.8}\times \mathrm{12.8}{}^{\circ }$ . The developed multisource data model (MSDM) combines the optical flow, random forest, and convolutional neural network (CNN) algorithms. It treats the precipitation nowcasting task as an image-to-image problem, which takes radar and satellite data with an interval of 30 min as inputs and predicts radar echo intensity with a lead time of 30 min. To reduce the smoothing caused by convolutions, we use the optical flow algorithm to predict satellite data in the following 120 min. The predicted radar echoes from the MSDM together with satellite data from the optical flow algorithm are recursively implemented in the MSDM to achieve a 120 min lead time. The MSDM predictions are comparable to those of other baseline models with a high temporal resolution of 6 min. To solve blurry image problems, we applied a modified structural similarity (SSIM) index as a loss function. Furthermore, we use the random forest algorithm with predicted radar and satellite data to estimate the rainfall rate, and the results outperform those of the traditional, nonlinear radar reflectivity factor and rainfall rate (Z–R) relationships that use logarithmic functions. The experiments confirm that machine learning with multisource data provides more reasonable predictions and reveals a better nonlinear relationship between radar echo and precipitation rate. Apart from developing complicated machine learning algorithms, exploiting the potential of multisource data will yield more improvements.

Quantitative precipitation nowcasting (QPN) is a highly challenging task in weather forecasting. The ability to provide precise, immediate, and detailed QPN products is necessary for a variety of situations, including storm warnings, air travel, and large gatherings. To address this challenge, this paper proposes a new transformer LPT(Lightweight Physics-informed Transformer)-QPN for QPN tasks, utilizing vertical cumulative liquid water content (VIL) products. This model adopts novel transformer modules to model long-term evolution of precipitation and incorporates multi-head squared attention (MHSA) to model its highly non-linear relationships while reducing computational complexity. The results of experimental evaluations demonstrate the superiority of LPT-QPN when compared to existing state-of-the-art QPN models. In particular, the LPT-QPN model demonstrates greater accuracy for long lead time and in high-intensity areas, confirmed in both quantitative and qualitative evaluations. Additionally, through three customized fine-tuning schemes, we are able to further improve the predictability of the LPT-QPN model for specific precipitation events. By incorporating physical constraints of convection diffusion equation, our approach offers novel perspectives for future explorations that combine physical prior knowledge and deep learning techniques.

The atmospheric duct (AD) is an anomalous structure in which electromagnetic waves can make transhorizon propagation. ADs often occur in the formation, development and disappearance of tropical cyclones (TCs). In this work, the eXtreme Gradient Boosting (XGBoost) model is used to predict TC ducts and a relatively high accuracy of 81.3% is obtained. Shapely additional explanations (SHAP) values of the features including TC parameters and local meteorological parameters are employed to interpret XGBoost model predictions of the TC ducts existence. Furthermore, the importance ranking of the features is revealed, among which the distance between dropsondes and TC eyes is the most important. In addition, the detailed relationships between the AD existence and the features are presented. Hence, this work can not only improve the knowledge of the relationship between TC ducts and the features, but also be of great value to the ducts prediction.

The precipitation data from the Global Precipitation Climatology Project (GPCP) and CPC Merged Analysis of Precipitation (CMAP) were used to investigate the discrepancy of Centre and Eastern Pacific ITCZ (CEP-ITCZ) during two types of El Niño years. Two models of the heat source distribution during two types of El Niño events were constructed, and the causes of different CEP-ITCZ anomalies for two types of El Niño events were analyzed through the Gill model. The results show that the CEP-ITCZ precipitation is approximately 4.0° southward, and the intensity is enhanced by 3.6 mm/day during the mature period of Eastern Pacific El Niño (EP-El Niño), while during the mature period of Central Pacific El Niño (CP-El Niño), it is only 0.8° southward, and the intensity is enhanced by 3.2 mm/day. The meridional mode of the SST anomaly by means of EOF (Empirical Orthogonal Function) can indirectly affect the CEP-ITCZ by influencing the atmospheric Rossby wave response. In CP-El Niño years, the meridional mode of the SST anomaly is weak, and the atmospheric Rossby wave response enhances the northern and southern trade-wind zones at the same time. The anomaly of cross-equatorial flow is weak and the CEP-ITCZ moves southward a little. At the same time, the wind convergence zone is enhanced, and it is more conducive to the vertical transport of water vapor. In EP-El Niño years, the meridional mode of the SST anomaly is strong, and the atmospheric Rossby wave response strengthens the meridional wind on the northern side of the equator, leading to the southward shift of the CEP-ITCZ. At the same time, the wind convergence zone is weakened and widened, and to a certain extent, it suppresses the vertical transport increase of water vapor caused by the sea surface evaporation.

Impact of Local Grid Refinements of Spherical Centroidal Voronoi Tessellations for Global Atmospheric Models - Volume 21 Issue 5 - Yudi Liu, Taojin Yang