<
doi:  10.12013/qxyjzyj2022-022
新SWAN江西本地化平台产品应用和检验评估

Application, inspection and evaluation of the localization platform product of new swan in Jiangxi
摘要点击 128  全文点击 97  投稿时间:2021-12-06  修订日期:2022-03-30
查看HTML全文  查看全文  查看/发表评论  下载PDF阅读器
基金:  江西省气象局2021年面上项目“SWAN3.0在江西的本地化技术研究和省市县三级推广应用”; 江西省气象局2019年青年人才培养项目“江西温度小时变化特征及其影响要素研究”.
中文关键词:  SWAN,短临预报,TS评分,强对流识别
英文关键词:  SWAN  short-term forecast  TS score  strong convection identification
        
作者中文名作者英文名单位
李海俊Li Haijun江西省气象台
单九生Shan Jiusheng江西省气象台
张 帆Zhang Fan赣州市气象局
引用:李海俊,单九生,张 帆.2022,新SWAN江西本地化平台产品应用和检验评估[J].气象与减灾研究,45(2):151-160
中文摘要:
      新SWAN3.0示范版通过融合“天擎”和LINUX改造等方式使系统框架更高效,在强对流识别算法、产品数量、功能模块等方面较以前版本均有了较大提升。文中使用空间领域TS评分法,并结合典型个例,对平台输出的临近预报效果和强对流识别产品进行了评估,结果表明:光流外推法和融合外推法结果趋势一致,预报时效越近预报能力越强;预报稳定性方面,大于35 dBz回波整体好于55 dBz以上的强回波中心;光流法预报效果好于融合法;平台能有效识别冰雹、下击暴流等,能为短临监测预警预报业务提供支撑。
Abstract:
      The new SWAN3.0 demonstration version makes the system framework more efficient by integrating “Tianqing” and LINUX transformation. Compared with the previous version, the strong convection identification algorithm, the number of products, and the functional modules have been greatly improved. The nowcasting effect and strong convection identification products output by the platform were evaluated by TS scoring method in the space domain and combined with typical cases. The results showed that the results of the optical flow extrapolation method and the fusion extrapolation method presented the same trend. The closer the forecast time was, the stronger the forecast ability was. The forecast stability of the echoes greater than 35 dBz was generally better than the strong echo centers above 55 dBz. The forecast effect of the optical flow method was better than the fusion method. The platform can effectively identify hail, downbursts, etc. and provide support for short term monitoring, early warning and forecasting services.
主办单位:江西省气象学会 单位地址:南昌市高新开发区艾溪湖二路323号
联系电话: (0791)82713175传真:0791-82713175 邮编:330096 Email:qxyjzyj@163.com
本系统由北京勤云科技发展有限公司设计