登陆热带气旋引发的广东沿海地区降水业务预报研究

图1. (a) ECMWF模式预测热带气旋Kujira在汕尾引发的日降水量。(b) 降水预报范围与降水观测对比：箱型图为历史同类型热带气旋引发汕尾站降水范围，其中红色阴影代表ECMWF预报汕尾地区降水范围，蓝色阴影代表观测到的汕尾地区降水范围。(c) 汕尾地区的实际日降水分布。

        Following previous studies of the rainfall forecast in Shenzhen owing to landfalling tropical cyclones(TCs), a nonparametric statistical scheme based on the classification of the landfalling TCs is applied to analyze and forecast the rainfall induced by landfalling TCs in the coastal area of Guangdong province, China. All the TCs landfalling with the distance less than 700 kilometers to the 8 coastal stations in Guangdong province during 1951—2017 are categorized according to their landfalling position and intensity. The daily rainfall records of all the 8 meteorological stations are obtained and analyzed. The maximum daily rainfall and the maximum 3 days’ accumulated rainfall at the 8 coastal stations induced by each category of TCs during the TC landfall period(a couple of days before and after TC landfalling time) from 1951 to 2013 are computed by the percentile estimation and illustrated by boxplots. These boxplots can be used to estimate the rainfall induced by landfalling TC of the same category in the future. The statistical boxplot scheme is further coupled with the model outputs from the European Centre for Medium-Range Weather Forecasts(ECMWF) to predict the rainfall induced by landfalling TCs along the coastal area. The TCs landfalling in south China from 2014 to 2017 and the corresponding rainfall at the 8 stations area are used to evaluate the performance of these boxplots and coupled boxplots schemes. Results show that the statistical boxplots scheme and coupled boxplots scheme can perform better than ECMWF model in the operational rainfall forecast along the coastal area in south China.

        在前次研究“An operational statistical scheme for tropical cyclone induced rainfall forecast”(《热带气象学报》(英文版)2015年第2期文章)的基础上，本研究使用根据登陆热带气旋分类别的非参数统计方法分析和预报登陆热带气旋引发的广东沿海地区的降水。作者将1951–2017年间登陆地点距离广东省8个沿海气象站700公里以内的所有热带气旋按其登陆位置和强度进行分类，并对所有8个气象站的日降水数据进行分析。对1951–2013年间由每一类别热带气旋在登陆期间（热带气旋登陆日前后2–3天）引发的8个沿海气象站的最大日降水量和最大累计3日降水量进行百分位数估算，并用统计箱型图表示。这些箱型图可以用于预报未来同类别登陆热带气旋引发的沿海区域降水。统计箱型图方法进一步与欧洲中期天气预报中心（ECMWF）的数值模式进行耦合，预报由登陆热带气旋引发的广东沿海地区降水。本研究用2014–2017年间登陆广东的热带气旋和8个沿海气象站的降水量情况来评估这些箱型图方法及耦合箱型图方法的预报性能。结果显示，在预报广东沿海地区由热带气旋引发的降水时，统计箱型图方法和耦合箱型图方法的预报能力优于ECMWF数值模式。

Highlights (文章亮点)：

        Typhoon induced precipitation is a difficult problem in weather forecast. This paper attempts to combine the boxplots method based on historical data analysis with the numerical model interpretation method to give a forecast model of typhoon induced precipitation. In this paper, among the 674 typhoons that made landfall in Guangdong from 1951 to 2017, 649 typhoons that made landfall in Guangdong from 1951 to 2013 and the historical observation data of precipitation at 8 stations along the coast were selected to establish the prediction model, and the prediction performance of the model was tested with the data of 25 typhoons that made landfall and their induced precipitation from 2014 to 2017. The results show that the prediction model performs well in the evaluation, and the technical method is reliable and easy to implement. This research is enlightening to meteorological professionals.

        台风引发降水是气象预报的难题，本文尝试把基于历史数据分析的Boxplots方法和数值模式释用方法结合起来，给出台风引发降水的预报模型。本文在1951–2017年在广东登陆的674个台风过程中，选取1951–2013年在广东登陆的649个台风过程及沿海8个台站降水历史观测资料建立预报模型，用2014–2017年间登陆的25个台风及其引发降水资料对模型进行预报性能检验。结果表明，预报模型在评估中表现较好，技术方法合理，易于实施，对实际业务有启发。

For full text, please visit (全文链接): http://jtm.itmm.org.cn/en/article/doi/10.16555/j.1006-8775.2020.001

Citation (引用): LI Qing-lan, LIU Bing-rong, WAN Qi-lin, et al. Operational forecast of rainfall induced by landfalling tropical cyclones along Guangdong coast [J]. J Trop Meteor, 2020, 26(1): 1-13, https://doi.org/10.16555/j.1006-8775.2020.001.