We investigate the evolution of a supercell storm within a quasi‐linear convective system (QLCS) that occurred in the Beijing area on 12 June 2022. Using high spatiotemporal resolution observations from a C‐band phased array radar (PAR), assimilated into a four‐dimensional variational data assimilation system, we primarily analyze dynamical processes contributing to the development of the supercell storm and its associated mesocyclone. Our study shows that just before the convective cell is triggered, a significant convergence zone develops to the west of the terrain, forming several meso‐γ vortices near the surface. During the merger of the convective cell and the QLCS from upper to lower levels, a strong downdraft generated by the QLCS enhances low‐level horizontal convergence, further producing a stretching effect on the vortices within the storm and significantly increasing vertical vorticity. With the formation of the mesocyclone in the mature stage of supercell storm, the height of the rotational center rises to 4.5 km, and the maximum rotational velocity reaches 20 m/s. Our results indicate that the surface convergence lines and the meso‐γ vortices along them strengthen low‐level convergence and generate strong updrafts, triggering the initial storm. These intense updrafts transform horizontal vorticity into vertical vorticity and transport it upward. Additionally, the process of convective merging leads to strengthen low‐level horizontal convergence, which forcibly stretches the mesovortex, enhancing vertical vorticity and allowing the convective storm to develop in a strong, organized manner and form the supercell storm.

Radar echo extrapolation is a critical forecasting tool in the field of meteorology, playing an especially vital role in nowcasting and weather modification operations. In recent years, spatiotemporal sequence prediction models based on deep learning have garnered significant attention and achieved notable progress in radar echo extrapolation. However, most of these extrapolation network architectures are built upon convolutional neural networks, using radar echo images as input. Typically, radar echo intensity values ranging from −5 to 70 dBZ with a resolution of 5 dBZ are converted into 0–255 grayscale images from pseudo-color representations, which inevitably results in the loss of important echo details. Furthermore, as the extrapolation time increases, the smoothing effect inherent to convolution operations leads to increasingly blurred predictions. To address the algorithmic limitations of deep learning-based echo extrapolation models, this study introduces three major improvements: (1) A Deep Convolutional Generative Adversarial Network (DCGAN) is integrated into the ConvLSTM-based extrapolation model to construct a DCGAN-enhanced architecture, significantly improving the quality of radar echo extrapolation; (2) Considering that the evolution of radar echoes is closely related to the surrounding meteorological environment, the study incorporates specific physical variable products from the initial zero-hour field of RMAPS-NOW (the Rapid-update Multiscale Analysis and Prediction System—NOWcasting subsystem), developed by the Institute of Urban Meteorology, China. These variables are encoded jointly with high-resolution (0.5 dB) radar mosaic data to form multiple radar cells as input. A multi-channel radar echo extrapolation network architecture (MR-DCGAN) is then designed based on the DCGAN framework; (3) Since radar echo decay becomes more prominent over longer extrapolation horizons, this study departs from previous approaches that use a single model to extrapolate 120 min. Instead, it customizes time-specific loss functions for spatiotemporal attenuation correction and independently trains 20 separate models to achieve the full 120 min extrapolation. The dataset consists of radar composite reflectivity mosaics over North China within the range of 116.10–117.50°E and 37.77–38.77°N, collected from June to September during 2018–2022. A total of 39,000 data samples were matched with the initial zero-hour fields from RMAPS-NOW, with 80% (31,200 samples) used for training and 20% (7800 samples) for testing. Based on the ConvLSTM and the proposed MR-DCGAN architecture, 20 extrapolation models were trained using four different input encoding strategies. The models were evaluated using the Critical Success Index (CSI), Probability of Detection (POD), and False Alarm Ratio (FAR). Compared to the baseline ConvLSTM-based extrapolation model without physical variables, the models trained with the MR-DCGAN architecture achieved, on average, 18.59%, 8.76%, and 11.28% higher CSI values, 19.46%, 19.21%, and 19.18% higher POD values, and 19.85%, 11.48%, and 9.88% lower FAR values under the 20 dBZ, 30 dBZ, and 35 dBZ reflectivity thresholds, respectively. Among all tested configurations, the model that incorporated three physical variables—relative humidity (rh), u-wind, and v-wind—demonstrated the best overall performance across various thresholds, with CSI and POD values improving by an average of 16.75% and 24.75%, respectively, and FAR reduced by 15.36%. Moreover, the SSIM of the MR-DCGAN models demonstrates a more gradual decline and maintains higher overall values, indicating superior capability in preserving echo structural features. Meanwhile, the comparative experiments demonstrate that the MR-DCGAN (u, v + rh) model outperforms the MR-ConvLSTM (u, v + rh) model in terms of evaluation metrics. In summary, the model trained with the MR-DCGAN architecture effectively enhances the accuracy of radar echo extrapolation.

The Lightning Mapping Imager (LMI) onboard the Fengyun-4A (FY-4A) satellite is the first independently developed satellite-borne lightning imager in China. It enables continuous lightning detection in China and surrounding areas, regardless of weather conditions. The FY-4A LMI uses a Charge-Coupled Device (CCD) array for lightning detection, and the accuracy of lightning positioning is influenced by cloud top height (CTH). In this study, we proposed an ellipsoid CTH parallax correction (ECPC) model for lightning positioning applicable to FY-4A LMI. The model utilizes CTH data from the Advanced Geosynchronous Radiation Imager (AGRI) on FY-4A to correct the lightning positioning data. According to the model, when the CTH is 12 km, the maximum deviation in lightning positioning caused by CTH in Beijing is approximately 0.1177° in the east–west direction and 0.0530° in the north–south direction, corresponding to a horizontal deviation of 13.1558 km, which exceeds the size of a single ground detection unit of the geostationary satellite lightning imager. Therefore, it is necessary to be corrected. A comparison with data from the Beijing Broadband Lightning Network (BLNET) and radar data shows that the corrected LMI data exhibit spatial distribution that is closer to the simultaneous BLNET lightning positioning data. The coordinate differences between the two datasets are significantly reduced, indicating higher consistency with radar data. The correction algorithm decreases the LMI lightning location deviation caused by CTH, thereby improving the accuracy and reliability of satellite lightning positioning data. The proposed ECPC model can be used for the real-time correction of lightning data when CTH is obtained at the same time, and it can be also used for the post-correction of space-based lightning detection with other cloud top height data.

精细尺度降水的临近预报对于提升现代城市内涝和山洪地质灾害预警能力具有重要意义.深度学习作为一种新兴方法,在挖掘数据内部特征及物理规律方面更具优势,近年来在天气雷达图像领域的应用已初见成效.为进一步提升精细尺度降水的临近预报能力,基于深度学习网络模型RainNet,研究建立了两种滚动预报方式,开展了京津冀地区1 km分辨率精细尺度降水滚动式临近预报试验和对比分析.试验结果表明:与传统基于交叉相关的外推预报相比,深度学习网络模型RainNet总体可以明显改进降水1 h临近预报的绝对误差和相关系数;两个RainNet相结合的滚动预报方式对1.04 mm/(10 min)及以下阈值降水,在10-50 min预报性能一致优于传统的交叉相关外推预报.深度学习模型对降水消亡过程的时、空演变趋势刻画更好,尤其更适用于降水消亡过程的临近预报.采用两个RainNet模型相结合的滚动式预报方式优于单一模型滚动预报方式.

基于自主获取的双金属球电场探空仪穿云观测数据,结合地面大气电场、天气雷达、闪电定位等综合观测资料,对华北平原地区一次中尺度对流系统(Mesoscale Convective System,MCS)层云区域内的探空电场和电荷结构进行研究.探空系统上升和下降阶段均处于该MCS层云区域中,此时雷暴正处于成熟阶段.完整的上升阶段探空数据表明,该MCS层云区域内存在6个正、负极性交替的电荷区,主正电荷区的高度范围为8.2～9.5 km,对应温度层-20～-14℃;主负电荷区高度范围为7.4～8.2 km(-14～-10℃);紧靠下方为一个薄正电荷区;最上方为一负极性电荷屏蔽区;0℃层附近有一对正、负极性电荷区.探空下降阶段(间隔约1h)获取的层云区域电荷分布与上升阶段大致对应,但电荷层所处高度以及厚度和电荷密度均有所差异.

A double-metal-sphere three-dimensional electric field sonde was developed successfully, and the thunderstorm electric-meteorological integrated sounding system was constructed in combination with weather radiosonde to measure the comprehensive sounding data of electric field, temperature and relative humidity. In summer of 2019, the field experiment was carried out in the North China Plain. The vertical electric filed profile and the corresponding charge structure inside the thunderstorm in this region are presented for the first time, through analyzing the sounding data, and synchronous surface electric field, radar echo and the three-dimensional wind field given by Variational Doppler Radar Analysis System (VDRAS). The sounding system was released during the dissipation stage of a mesoscale convective system (MCS) on August 7, 2019 and passed through the weak echo region of the thunderstorm. The sounding results show that there were five charge regions in the thunderstorm and the charge polarity altered in the vertical direction. The upper positive charge region was at 4. 4 similar to 5. 6 km (near 0 degrees C), and the complex middle negative charge region was at 3. 6 similar to 4. 4 km. The height of lower positive charge region was 1. 0 similar to 3. 6 km. Besides, there was a negative charge region below 1 km, and a weak negative shielding charge region was near the top of the thunderstorm. Both middle negative charge region and lower positive charge region were composed of several charge layers with different thicknesses and charge densities. Furthermore, the electric field sounding system experienced the stages of rising, falling and rising again during 3. 6 similar to 4. 4 km, which were in the range of the middle negative region. The sounding data indicates that the dynamic field inside the cloud was complicated, and the detailed charge structures of the three stages in the negative charge region were similar but different, which reflected the real charge structure inside the thunderstorm was very complicated and inhomogeneous spatio-temporally.

冬季降水相态及其转变时间的精细化客观预报对提高气象预报和服务质量具有重要的现实意义.利用京津冀地区国家级自动气象站观测资料及网格化快速更新精细集成产品,统计分析了京津冀地区复杂地形下各类降水相态温度和湿球温度平均气候概率的分布差异及不同降水相态时网格化快速更新精细集成产品中可能影响降水相态判断的特征信息.然后将地面观测天气现象资料、复杂地形下降水相态气候特征及高分辨率模式输出产品作为特征向量,分别基于梯度提升(XGBoost)、支持向量机(SVM)、深度神经网络(DNN)3种机器学习方法建立了降水相态的高分辨率客观分类模型,并对同样条件下3种机器学习方法对雨、雨夹雪和雪3种京津冀主要降水相态的预报效果进行了对比检验,进一步提升了雨夹雪复杂降水相态的客观分类预报技巧.

基于雷达资料快速更新四维变分同化(RR4DVar)技术和三维数值云模式发展的快速更新雷达四维变分分析系统(VDRAS),通过在系统中加入地面自动气象站观测资料的同化方法,对发生在北京地区的10个强对流过程开展了地面资料同化的高分辨率模拟分析和检验评估,并与已经业务使用的地面资料融合方法进行对比.研究结果发现,地面观测资料同化使边界层1 km高度以下的分析场改善最为明显,风速和风向的均方根误差分别平均降低0.1 m/s和7.2°,温度的均方根误差降低0.2℃.模式最低层100 m高度的风速均方根误差降低0.5 m/s,风速的误差随高度上升逐渐增大.模式最低层风向的均方根误差降低15.5°,温度的均方根误差降低0.4℃,且1.5 km高度以下的温度偏差都减小.区域内地面10 m高风速的均方根误差平均降低0.2 m/s,风向的均方根误差降低10.8°,地面2 m气温的偏差也降低.随着预报时效的延长,地面温度和风场的误差不断增大,但地面资料同化方法在一定程度上可以提高1 h内地面气象要素的预报效果.对2019年5月17日北京地区局地强对流新生和增强过程的详细分析表明,地面自动气象站观测资料的同化方法相对于融合,可以通过更细致准确地分析低层大气的热动力特征,改善低层气象要素的预报效果.在此基础上,通过探究对流单体的局地触发机理发现,海风锋辐合线与城市的相互作用一定程度上影响了对流的局地新生和发展,该同化方法可以进一步提高北京地区局地突发强对流的临近数值预报能力.

将优化冠层参数的城市冠层模式耦合到快速更新循环系统RMAPS-ST中,探究城市冠层效应对地面气象要素预报的影响.设计耦合城市冠层模式(UCM)和未耦合城市冠层模式(NOUCM)两组对比试验,对华北区域2017年1月中旬和7月中旬的地面气象要素(2 m气温、10 m风速和2 m比湿)进行短期预报评估.结果表明:与NOUCM试验相比,UCM试验能够显著提高RMAPS-ST系统对地面气象要素的预报性能,使华北区域城市地面气象要素的预报更接近于观测.考虑冠层效应能够有效减小华北城市站点的2 m气温预报偏差,1月中旬和7月中旬24 h气温预报准确率分别提高42％和15％.冠层效应的引入增加了城市摩擦系数,改善了10 m风速预报偏差明显偏大现象,预报准确率均提高40％以上.对于2 m比湿,两组试验1月中旬的预报偏差均较小,冠层的引入对预报结果无显著影响;两组试验对7月中旬的比湿预报全天偏干,冠层效应能显著改善城区比湿偏干的情况,预报准确率提高17％.

Terrestrial gamma-ray flashes (TGFs) are impulsive (~1 ms) but intense sources of gamma-rays associated with lightning activity and thunderstorms. Till now, few observation/study for those TGFs occurred in China are carried out. From June 2013, we started to install high sensitivity magnetic sensors in China and collected the lightning waveforms and other data related with TGF that detected by Fermi. Here we mainly introduce one special TGF occurred in South China which had three-station magnetic recordings. The results show that the TGF-related sferic is an energetic in-cloud pulse (EIP), with high current of 204.5 kA and the altitude at about 10 km. The characteristics of TGF parent thunderstorm are also discussed with comprehensive data of weather radar, Black Body Temperature (TBB), lightning location and weather sounding.