曹林兮,吴梦雯,罗亚丽,陈锋. 2024. 苏沪城市群区域出梅后期对流初生客观识别方法及其时空分布特征. 气象学报,82(3):275-288. DOI: 10.11676/qxxb2024.20230168
引用本文: 曹林兮,吴梦雯,罗亚丽,陈锋. 2024. 苏沪城市群区域出梅后期对流初生客观识别方法及其时空分布特征. 气象学报,82(3):275-288. DOI: 10.11676/qxxb2024.20230168
Cao Linxi, Wu Mengwen, Luo Yali, Chen Feng. 2024. An objective algorithm for identification of post-Meiyu convection initiation and its spatiotemporal distribution over the urban agglomeration of Jiangsu and Shanghai. Acta Meteorologica Sinica, 82(3):275-288. DOI: 10.11676/qxxb2024.20230168
Citation: Cao Linxi, Wu Mengwen, Luo Yali, Chen Feng. 2024. An objective algorithm for identification of post-Meiyu convection initiation and its spatiotemporal distribution over the urban agglomeration of Jiangsu and Shanghai. Acta Meteorologica Sinica, 82(3):275-288. DOI: 10.11676/qxxb2024.20230168

苏沪城市群区域出梅后期对流初生客观识别方法及其时空分布特征

An objective algorithm for identification of post-Meiyu convection initiation and its spatiotemporal distribution over the urban agglomeration of Jiangsu and Shanghai

  • 摘要: 研发了一套利用天气雷达观测资料对流初生(CI)客观识别算法。该算法定义“对流块”为雷达反射率因子达到35 dBz的连续强回波区域,如果某一对流块的前30 min、周边一定范围内没有任何对流块,且它维持超过20 min,则该对流块被识别为CI。将这套CI客观识别算法应用于苏-沪城市群区域出梅后期(从梅雨季结束到8月底,且排除热带气旋影响)CI识别,构建了该地区2019—2022年出梅后期高时、空分辨率(10 min、1 km)CI数据库,通过个例和统计研究,分析了CI识别结果对算法中两个参数的敏感性。判断对流回波是否为首次出现,当区域半径(R)设为10 km时,对流块最小面积设为4 km2(Test1)或2 km2(Test2)对CI识别结果的影响不大,两个试验都能识别出CI连续发生和相邻发生的复杂情况,Test1能够识别出Test2的83% CI事件,二者识别出的CI日数、逐日CI数、CI高频区分布、CI频次日变化等统计特征基本一致;当R设为30 km、对流块最小面积为4 km2(Test3)时,仅能识别出Test1的36% CI事件,其识别的CI数目不足,导致CI发生频次的空间分布和日变化的分析结果明显不同于Test1和Test2。识别结果表明,分析区域内有CI发生频次较高的3个子区域:沿长江的“上海—苏锡常—南京”城市带、从南通北部到黄海近海、分析区域西南侧的天目山系部分区域,CI频次日变化呈现午间(11—14时,北京时)主峰特征。

     

    Abstract: In this study, an objective algorithm to identify convection initiation (CI) based on weather radar observations is developed. In this algorithm, a convective cell is defined as a continuous area of strong radar echo (reflectivity factor reaching 35 dBz). The presence of a convective cell is identified as CI if two criteria are met: (1) no convective cells exist around the convective cell in a 30 min period prior to its presence; (2) this convective cell lasts for at least 20 min. This algorithm is applied to the urban agglomeration region over Jiangsu-and-Shanghai during the post-Meiyu period (from the end of the Meiyu season to 31 August, and excluding days influenced by tropical cyclones). Based on the results, A CI database with high spatiotemporal resolutions (10 min, 1 km) over the region for the 2019—2022 post-Meiyu periods is constructed. The sensitivity of the CI identification to two parameters in the algorithm is investigated by comparing three experiments in a case study and also based on statistics analysis. When the radius of the area (R), which is used to determine whether the convective echo is the first occurrence, is set to 10 km, whether the minimum area of the convective echo (Amin) is set to 4 km2 or 2 km2 (Test1, Test2) does not significantly influence the CI identification results. Both experiments can identify the complex situations with continuous and adjacent CI events, with Test1 being able to identify 83% of Test2's CI events. The statistical characteristics of CI events identified by the two experiments are essentially the same, including the number of CI days, daily CI numbers, the distribution of CI high-occurrence frequency areas, and the diurnal variation of CI occurrence frequency. When R is set to 30 km (with Amin of 4 km2, Test3), only 36% of the CI events identified in Test1 can be identified in Test3. The insufficient number of identified CI events results in significant differences in spatial distribution and diurnal variation of CI occurrence compared to those from Test1 and Test2. The results indicate that there are three subregions of high occurrence frequency of CI in the analysis region, i.e., the "Shanghai-Suzhou-Wuxi-Changzhou-Nanjing" city belt along the Yangtze river, from northern Nantong to the offshore of the Yellow sea, and part of the Tianmu mountain in the southwest of the analysis region. The diurnal variation of CI occurrence frequency is characterized by a major peak around noon (11:00—14:00 BT).

     

/

返回文章
返回