﻿ 基于欧洲中心细网格预报资料和观测资料的浙江省春夏降水性质分类指标
 气候与环境研究  2018, Vol. 23 Issue (5): 543-550 PDF

Precipitation Classification Index Based on ECMWF Fine Grid Forecast Data and Observation Data for Spring and Summer in Zhejiang Province
WANG Jiyin, CHEN Yini, SUN Zhang, WANG Liying, YU Pei
Zhejiang Province Meteorological Observatory, Hangzhou 310021
Abstract: The authors used the plot data and T-logP data of Micaps (Meteorological Information Comprehensive Analysis and Processioning System) to analyze thunderstorm precipitation and non-thunderstorm precipitation in Hangzhou, Quzhou, and Taizhou. The prediction factors in this study that can tell the differences between the environments of thunderstorm precipitation and non-thunderstorm precipitation are convective available potential energy, temperature differences between 850 hPa and 500 hPa, K index, and 2-m height temperature. An index has been formulated from analysis of parameters derived from proximity soundings and decision trees. The quality of this index is examined with precipitation activities for the period from 2004 to 2013 and 2016 using ECMWF fine grid forecast data and observation data. The results show that the threat score (TS) is more than 0.53, the hit rate is 71% and the false rate of thunderstorm forecast (43%) is higher than that of non-thunderstorm precipitation (10%) for most areas and seasons. This method basically can also predict the areas of different types of precipitation from examining two thunderstorm processes which occurred in the spring and summer of 2016, respectively. In summary, the advantage of this method is reflected in the fact that it not only can make fine precipitation classification forecast but also can create expressive prediction for medium and long-term forecast.
Keywords: Precipitation thunderstorm     Non-thunderstorm precipitation     Decision tree     Precipitation classification index     ECMWF fine grid forecast data and observation data
﻿
1 引言

2 资料和方法 2.1 资料

2.2 分析方法

2.2.1 临近探空分析方法

2.2.2 决策树CART算法

 $G(A) = 1 - \sum\limits_{i = 1}^C {{P_i}^2},$ (1)

3 预报指标的建立与检验 3.1 预报指标的建立

 图 1 决策树降水分类模型 Fig. 1 Decision tree classification model of precipitation

 图 2 降水性质分类流程图 Fig. 2 Flow chart of precipitation classification
3.2 结果检验 3.2.1 历史样本回报

3.2.2 预报检验

（1）从不同城市来看：雷雨命中率杭州、衢州、台州3市分别为100%、97%与67%，平均为89%；阵雨的命中率分别为59%、56%与69%，平均为，61%；命中率高则漏报率低，两者和为1；雷雨空报率3市接近分别为46%、41%、40%，阵雨空报率为0、3%、25%；TS评分基本超过0.50。

（2）从不同月份来看：雷雨命中率3~8月分别为：75%、85%、74%、100%、100%与93%，春季为78%，夏季为98%；阵雨命中率分别为：93%、65%、59%、39%、38%与25%，春季为70%，夏季为37%；同样，命中率高则漏报率低，两者和为1；空报率雷雨高阵雨低，阵雨几乎无空报；TS评分有月季变化，但雷雨夏季高，阵雨春季高。

（3）总体平均来看：TS评分都超过0.5，雷雨和阵雨差不多，分别为0.53和0.57；命中率雷雨（89%）高于阵雨（61%），阵雨雷雨两者综合的准确率为71%；空报率也是雷雨（43%）大于阵雨（10%）。杭州、衢州降水性质预报效果好于台州；夏季的雷雨预报效果好于春季，夏季的阵雨预报效果比春季差。

 图 3 2016年4月2、3日（a、c）降水量实况与（b、d）ECMWF预报的降水量分布（填色）及降水性质（红色标记为雷雨）：（a、b）2016年4月2日；（c、d）2016年4月3日 Fig. 3 Distributions of (a, c) real-time precipitation, (b, d) precipitation forecasts and classification (thunderstorm is denoted by red mark) during the period of 2-3 April 2016: (a, b) 2 April 2016; (c, d) 3 April 2016

 图 4 2016年5月15日（a）降水量实况与（b）ECMWF预报的降水量分布（填色）及降水性质（红色标记为雷雨） Fig. 4 Distributions of (a) real-time precipitation, (b) precipitation forecasts and classification (thunderstorm is denoted by red mark) on 15 May 2016
4 结论与讨论

（1）雷雨发生的条件为有降水并且有雷暴，根据雷暴发生三要素（静力不稳定、水汽和抬升触发机制）计算确定的几个基本、独立预报参数为CAPE、KT8-5T0。判别流程为是否有降水，如果有降水且T8-5大于23.95 ℃，则判定为雷雨，其余进行进一步判断；如果CAPE小于142.5 J kg-1，则判定为阵雨，其余进行下一步判断；如果T0小于29.45 ℃，则判定为雷雨，否则为阵雨。

（2）从数学统计分类角度结合气象预报因子，初步建立了一个适用于浙江省的区分降水性质的指标，且实际业务化的时候计算量小，预报周期长，短期内预报时效高、精细化。本文所建立的指标在除去有降水预报误差个例后，命中率综合达到70%以上，TS评分超过0.53，空报率雷雨与阵雨分别为43%和10%，在不同地区和季节稍有区别（命中率高，且本省尚无类似指标进行对比）。总体来看杭州、衢州降水性质预报效果更好，杭州、衢州雷雨预报效果好于台州，阵雨略差于台州；夏季的雷雨预报效果好于春季，夏季的阵雨预报效果好于春季。从浙江省两次大范围过程来看，都能很好预报和区分出阵雨和雷雨的落区以及演变情况。

（3）降水分类指标在地域上（东西部，是否沿海）、季节上有差异，不可排除偶尔性，但本文所做分类指标在力求全面、合理、计算量低的前提下还是有一定局限性，有待结合其它资料（卫星、雷达）或地形、环境类等因子，进一步完善指标；另外，降水分类指标在很大程度上依赖降水预报的准确性，降水的漏报和空报都会直接导致误差，故未来希望能对数据进行订正，更好地服务于浙江省的降水性质预报，为气象业务人员提供参考和服务。

 [] Breiman L, Friedman J H, Stone C J, et al. 1984. Classification and Regression Trees[M]. Monterey, CA: Chapman and Hall/CRC. [] 陈洪滨, 朱彦良. 2012. 雷暴探测研究的进展[J]. 大气科学, 36(2): 411–422. Chen Hongbin, Zhu Yanliang. 2012. Review on the observation investigation of thunderstorms[J]. Chinese Journal of Atmospheric Science (in Chinese), 36(2): 411–422. DOI:10.3878/j.issn.1006-9895.2011.11064 [] 陈明轩, 俞小鼎, 谭晓光, 等. 2004. 对流天气临近预报技术的发展与研究进展[J]. 应用气象学报, 15(6): 754–766. Chen Mingxuan, Yu Xiaoding, Tan Xiaoguang, et al. 2004. A brief review on the development of nowcasting for convective storms[J]. Journal of Applied Meteorological Science (in Chinese), 15(6): 754–766. DOI:10.3969/j.issn.1001-7313.2004.06.015 [] Doswell Ⅲ C A. 1987. The distinction between large-scale and mesoscale contribution to severe convection:A case study example[J]. Wea. Forecasting, 2(1): 3–16. DOI:10.1175/1520-0434(1987)002<0003:TDBLSA>2.0.CO;2 [] Doswell Ⅲ C A. 2001. Severe convective storms-an overview[M]//Doswell C A. Severe Convective Storms. Boston: American Meteorological Society, 1-26. [] 巩崇水, 曾淑玲, 王嘉媛, 等. 2013. 近30年中国雷暴天气气候特征分析[J]. 高原气象, 32(5): 1442–1449. Gong Chongshui, Zeng Shuling, Wang Jiayuan, et al. 2013. Analyses on climatic characteristics of thunderstorm in China in recent 30 Years[J]. Plateau Meteorology (in Chinese), 32(5): 1442–1449. DOI:10.7522/j.issn.1000-0534.2012.00134 [] 姜文瑞, 王玉英, 郝小琪, 等. 2012. 决策树方法在气温预测中的应用[J]. 计算机应用与软件, 29(8): 141–144. Jiang Wenrui, Wang Yuying, Hao Xiaoqi, et al. 2012. Application of decision tree in temperature prediction[J]. Computer Applications and Software (in Chinese), 29(8): 141–144. DOI:10.3969/j.issn.1000-386X.2012.08.037 [] 雷蕾, 孙继松, 魏东. 2011. 利用探空资料判别北京地区夏季强对流的天气类别[J]. 气象, 37(2): 136–141. Lei Lei, Sun Jisong, Wei Dong. 2011. Distinguishing the category of the summer convective weather by sounding data in Beijing[J]. Meteorological Monthly (in Chinese), 37(2): 136–141. DOI:10.7519/j.issn.1000-0526.2011.2.002 [] 刘璇, 唐慧强, 许遐祯, 等. 2009. 决策树算法在农业气象灾害统计中的应用[J]. 农机化研究, 31(7): 200–203. Liu Xuan, Tang Huiqiang, Xu Xiazhen, et al. 2009. The application of decision-tree algorithm in the statistic of meteorological hazards of agriculture[J]. Journal of Agricultural Mechanization Research (in Chinese), 31(7): 200–203. DOI:10.3969/j.issn.1003-188X.2009.07.059 [] Ostby F P. 1992. Operations of the national severe storms forecast center[J]. Wea. Forecasting, 7(4): 546–563. DOI:10.1175/1520-0434(1992)007<0546:OOTNSS>2.0.CO;2 [] Quinlan J R. 1986. Introduction of decision trees[J]. Machine Le-arning(1): 81–106. [] 孙凌, 周筠珺, 杨静. 2009. 雷暴预警预报的研究进展[J]. 高原山地气象研究, 29(2): 75–80. Sun Ling, Zhou Yunjun, Yang Jing. 2009. Advances in early-warning and forecasting of thunderstorms[J]. Plateau and Mountain Meteorology Research (in Chinese), 29(2): 75–80. DOI:10.3969/j.issn.1674-2184.2009.02.015 [] 中国气象局监测网络司. 1999. 地面气象电码手册[M]. 北京: 气象出版社: 100pp. Monitoring Network Department of China Meterological Administration. 1999. The Manual of Surface Meteorological Code (in Chinese)[M]. Beijing: China Meteorological Press: 100pp. [] 王霁吟, 陈宝君, 宋金杰, 等. 2015. 基于再分析资料的我国龙卷发生环境和通用龙卷指标[J]. 气候与环境研究, 20(4): 411–420. Wang Jiyin, Chen Baojun, Song Jinjie, et al. 2015. Atmospheric conditions of tornado genesis and universal tornadic index based on reanalysis data[J]. Climatic and Environmental Research (in Chinese), 20(4): 411–420. DOI:10.3878/j.issn.1006-9585.2014.14127 [] 向俊莲, 王丽珍. 2001. PUBLIC在云南气象预报中的应用[J]. 云南大学学报(自然科学版), 23(1): 16–29. Xiang Junlian, Wang Lizhen. 2001. The application of PUBLIC in predicting the climatic phenomenon in Yunnan Province[J]. Journal of Yunnan University (in Chinese), 23(1): 16–29. DOI:10.3321/j.issn:0258-7971.2001.01.005 [] 俞小鼎, 周小刚, 王秀明. 2012. 雷暴与强对流临近天气预报技术进展[J]. 气象学报, 70(3): 311–337. Yu Xiaoding, Zhou Xiaogang, Wang Xiuming. 2012. The advances in the nowcasting techniques on thunderstorms and severe convection[J]. Acta Meteorologica Sinca (in Chinese), 70(3): 311–337. DOI:10.11676/qxxb2012.030 [] 曾淑玲, 巩崇水, 赵中军, 等. 2012. 动力-统计方法在24小时雷暴预报的应用[J]. 气象, 38(12): 1508–1514. Zeng Shuling, Gong Chongshui, Zhao Zhongjun, et al. 2012. The application of dynamical-statistical method to 24-h thunderstorm forecast[J]. Meteorological Monthly (in Chinese), 38(12): 1508–1514.