一次粤西南暴雨过程的预报误差来源分析

卢楚翰; 林琳; 周菲凡

doi:10.3878/j.issn.1006-9895.2008.20130

一次粤西南暴雨过程的预报误差来源分析

doi: 10.3878/j.issn.1006-9895.2008.20130

卢楚翰^1,,
林琳^{1, 2},
周菲凡^{2, 3, ,}

1.
南京信息工程大学气象灾害教育部重点实验室/气候与环境变化国际合作联合实验室/气象灾害预报预警与评估协同创新中心，南京 210044
2.
中国科学院大气物理研究所云降水物理与强风暴实验室，北京 100029
3.
中国科学院大学，北京 100049

基金项目: 国家重点研发计划项目2018YFC1507405、2017YFC1501601

详细信息

作者简介:
卢楚翰，男，1981年出生，副教授，主要从事大气环流异常及气候变率方面研究。E-mail: luchuhan@nuist.edu.cn

通讯作者:
周菲凡，E-mail: zhouff04@163.com

中图分类号: P457
计量
- 文章访问数: 433
- HTML全文浏览量: 112
- PDF下载量: 148
- 被引次数: 0
出版历程
- 收稿日期: 2020-03-11
- 网络出版日期: 2020-08-15
- 刊出日期: 2020-11-20

Analysis of the Source of Forecast Errors for a Heavy Precipitation in the Southwest of Guangdong Province

Chuhan LU^1
,,
Lin LIN^{1, 2},
Feifan ZHOU^{2, 3
, ,}

1.
Key Laboratory of Meteorological Disaster, Ministry of Education/Joint International Research Laboratory of Climate and Environment Change/Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing 210044
2.
Laboratory of Cloud–Precipitation Physics and Severe Storms (LACS), Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029
3.
University of Chinese Academy of Sciences, Beijing 100049

Funds: National Key Research and Development Program of China (Grants 2018YFC1507405, 2017YFC1501601)

摘要

摘要: 本文基于WRF模式研究了2015年5月16～17日广东西南地区的一次暴雨过程的预报误差来源。首先比较了以NCEP_FNL为初始资料的WRF模式的模拟预报（记为WRF_FNL）和ECMWF（European Centre for Medium-Range Weather Forecasts）关于该次暴雨过程的确定性预报。结果表明，ECMWF具有较高的预报技巧，因此，认为ECMWF的模式和初始场都较为准确。进一步，以ECMWF的初值作为初始场，选用相同的物理参数化方案，再次用WRF模式进行预报（预报结果记为WRF_EC）。结果表明相对WRF_FNL，WRF_EC的预报结果有明显改善。这表明，初始场的改进对预报有较大的影响，初始误差是预报误差的重要来源。进一步，分析了初始误差的主要来源区域和来源变量。结果表明，南海北部湾至广西西南区域为本次暴雨预报初始误差的主要来源区域，而初始温度场和初始湿度场则为此次暴雨预报初始误差的主要来源变量。同时改进初始温度场和湿度场可以较大程度提高本次暴雨过程的预报技巧。
- 误差分析 /
- 华南暴雨 /
- 初始场 /
- 敏感区 /
- 敏感变量
Abstract: Based on the WRF model, in this study, the authors investigated the source of forecast errors for an extreme precipitation event in the southwest of Guangdong Province that occurred from 16 May to 17 May 2015. First, forecasts using the WRF model with the initials NCEP_FNL (hereinafter WRF_FNL) were compared with the deterministic forecasts generated by ECMWF (European Centre for Medium-Range Weather Forecasts), and the results showed that ECMWF had good forecast skills. Thus, the model and initials used by ECMWF are deemed to be accurate. Then, the authors generated a new forecast (hereinafter WRF_EC) using the WRF model with the ECMWF initials, keeping the same physical schemes as those in the previous WRF simulations. The results showed that WRF_EC has better forecast skills than WRF_FNL. This indicates that the improvement of initial states has a significant impact on forecasts, and thus the initial error is the main source of the forecast errors. The authors also analyzed the sensitive area and sensitive variables. The results showed that the initial errors mainly came from the North Gulf of the South China Sea and southwest of Guangxi Province. Both the initial temperature and the initial humidity played an important role in the heavy precipitation forecast. Thus, improving the accuracy of initial temperature and humidity could greatly improve the skills in forecasting this heavy precipitation phenomenon.
- Error analysis /
- Heavy precipitation in South China /
- Initial states /
- Sensitive areas /
- Sensitive variables

HTML全文

图 1 2015年5月16日00时（北京时，下同）至17日00时华南地区降水量（单位：mm）分布：（a）OBS；（b）TIGGE_EC；（c）WRF_FNL；（d）WRF_EC。图a中西侧、北侧、南侧的黑色虚线框区分别为敏感性试验中的区域一、二、三

Figure 1. Distribution of precipitation (units: mm) in South China from 0000 BJT (Beijing time) 16 May to 0000 BJT 17 May 2015: (a) OBS (hourly precipitation grid data set (version 1.0) fused by China automatic station and CMORPH precipitation products); (b) TIGGE_EC [TIGGE (THORPEX Interactive Grand Global Ensemble) ensemble forecast data from ECMWF (European Centre for Medium-Range Weather Forecasts)]; (c) WRF_FNL [forecasts using the WRF model with the initials NCEP_FNL (Final Global Data Assimilation System) data]; (d) WRF_EC (forecasts using the WRF model with the initials ECMWF data). In Fig. a, black boxes on the west, north, and south sides are areas 1, 2, and 3 in sensitive experiments, respectively

下载: 全尺寸图片幻灯片

图 2 2015年5月16日00时（a、c、e）FNL、（b、d、f）EC两初始场的（a、b）500 hPa、（c、d）850 hPa、（e、f）地面的温度场（阴影，单位：°C）和风场（箭头，单位：m s⁻¹）

Figure 2. Temperature (shadings, units: °C) and wind (arrows, units: m s⁻¹) at (a, b) 500 hPa, (c, d) 850 hPa, (e, f) surface obtained from the two initial fields (a, c, e) FNL (initial forecast field of WRF_FNL), (b, d, f) EC (initial forecast field of WRF_EC) at 0000 BJT 16 May

下载: 全尺寸图片幻灯片

图 3 2015年5月16日00时（a）FNL、（b）EC两初始场的850 hPa相对湿度场（阴影）、风场（箭头，单位：m s⁻¹）

Figure 3. Relative humidity field (shadings) and wind (arrows, units: m s⁻¹) at 850 hPa obtained from the two initial fields (a) FNL, (b) EC at 0000 BJT 16 May

下载: 全尺寸图片幻灯片

图 4 2015年5月16日00时（a、c）FNL、（b、d）EC两初始场的（a、b）500 hPa位势高度场（单位：dagpm）、（c、d）海平面气压场（阴影，单位：hPa）、风场（箭头，单位：m s⁻¹）。图a、b中的粗实线表示槽线

Figure 4. (a, b) 500-hPa gopotential height (units: dagpm), (c, d) sea level pressure (shadings, units: hPa), and surface wind (units: m s⁻¹) obtained from the two initial fields (a, c) FNL, (b, d) EC at 0000 BJT 16 May 2015. Black bold lines in Figs. a and b represent trough lines

下载: 全尺寸图片幻灯片

图 5 2015年5月16日01时（a）FNL、（b）EC两初始场对应的850 hPa垂直速度场（单位：m s⁻¹）

Figure 5. 850-hPa vertical velocity (units: m s⁻¹) obtained from the two initial fields (a) FNL, (b) EC at 0100 BJT 16 May

下载: 全尺寸图片幻灯片

图 6 2015年5月16日00时EC、FNL两初始场的（a）500 hPa、（b）850 hPa、（c）地面温度（阴影，单位：°C）的差值场、EC风场（箭头，单位：m s⁻¹），（d）850 hPa相对湿度（阴影）的差值场叠加同层次EC风场（箭头，单位：m s⁻¹），（e）2015年5月16日01时EC、FNL两初始场850 hPa垂直速度的差值场（单位：m s⁻¹）

Figure 6. Differences of temperature (between initial fields EC and FNL, shadings, units: °C) at (a) 500 hPa, (b) 850 hPa, (c) ground and wind (from initial field EC, arrows, units: m s⁻¹), (d) 850-hPa relative humidity differences (between initial fields EC and FNL, shadings) and wind (from initial fields EC, arrows, units: m s⁻¹) at 0000 BJT 16 May, and (e) 850-hPa vertical velocity differences (between initial fields EC and FNL, units: m s⁻¹) at 0100 BJT 16 May

下载: 全尺寸图片幻灯片

图 7 2015年5月16日00时EC、FNL两初始场的整层大气扰动湿能量的差值场（单位：10⁶ J）

Figure 7. Differences (between initial fields EC and FNL, units: 10⁶ J) of the wet energy field disturbance from the entire atmosphere at 0000 BJT 16 May

下载: 全尺寸图片幻灯片

图 8 24小时（2015年5月16日00时至17日00时）累积预报降水量分布（单位：mm）：（a）WRF_FNL；EC替换FNL中（b）区域一、（c）区域二、（d）区域三（如图1中所示）内基础要素场

Figure 8. 24-h (from 0000 BJT 16 May to 0000 BJT 17 May 2015) accumulative forecast precipitation (units: mm) distribution in South China: (a) WRF_FNL; EC replaces the basic element fields of FNL in (b) area 1, (c) area 2, (d) area 3 (as shown in Fig. 1)

下载: 全尺寸图片幻灯片

图 9 2015年5月16日00时至5月17日替换FNL敏感区（区域一）内不同变量后24 h预报累积降水量分布（单位：mm）：（a）温度；（b）湿度；（c）风场；（d）气压场；（e）同时替换温度和湿度

Figure 9. Forecast precipitation (units: mm) distribution of precipitation (units: mm) in South China of the replacing sensitive variable experiment in the sensitive area (area 1) of FNL from 0000 BJT (Beijing time) 16 May to 0000 BJT 17 May 2015: (a) Temperature; (b) humidity; (c) wind field; (d) pressure field; (e) temperature and humidity

下载: 全尺寸图片幻灯片

表 1 2015年5月16日00时至17日00时华南地区WRF_FNL、WRF_EC、TIGGE_EC 24 h累积预报降水量分别与OBS的相关系数及TS评分

Table 1. Correlation coefficients and TS scores between OBS and WRF_FNL, WRF_EC, TIGGE_EC 24-h accumulative precipitation forecast in South China from 0000 BJT (Beijing time) 16 May to 0000 BJT 17 May 2015

相关系数 TS评分

WRF_FNL 0.333 0.0889
WRF_EC 0.363 0.1393
TIGGE_EC 0.478 0.1639

下载: 导出CSV

表 2 敏感区试验中24小时（2015年5月16日00时至17日00时）预报累积降水量与OBS的相关系数及TS评分

Table 2. Correlation coefficients and TS scores between OBS and the experimental forecast precipitation results in sensitive-area experiments from 0000 BJT 16 May to 0000 BJT 17 May 2015

相关系数 TS评分

区域一 0.512 0.1287
区域二 0.241 0.0639
区域三 0.323 0.0948

下载: 导出CSV

表 3 敏感变量试验中24小时（2015年5月16日00时至17日00时）累积降水量预报结果与OBS的相关系数及TS评分

Table 3. Correlation coefficients and TS scores between OBS and the experimental forecast precipitation results in sensitive-variable experiments from 0000 BJT 16 May to 0000 BJT 17 May 2015

替换物理量相关系数 TS评分

温度场 0.182 0.0520
湿度场 0.112 0.0937
风场 0.094 0.0322
气压场 0.099 0.0329
温度场和湿度场 0.487 0.1713

下载: 导出CSV

参考文献(27)

[1]	卞建春, 杨培才. 2003. 关于大气过程可预报性问题的一些讨论 [J]. 高原气象, 22(4): 315−323. doi: 10.3321/j.issn:1000-0534.2003.04.001 Bian Jianchun, Yang Peicai. 2003. Some remarks on the predictability of atmospheric processes [J]. Plateau Meteor. (in Chinese), 22(4): 315−323. doi: 10.3321/j.issn:1000-0534.2003.04.001
[2]	陈静, 薛纪善, 颜宏. 2003. 华南中尺度暴雨数值预报的不确定性与集合预报试验 [J]. 气象学报, 61(4): 432−446. doi: 10.11676/qxxb2003.042 Chen Jing, Xue Jishan, Yan Hong. 2003. The uncertainty of mesoscale numerical prediction of South China heavy rain and the ensemble simulations [J]. Acta Meteor. Sinica (in Chinese), 61(4): 432−446. doi: 10.11676/qxxb2003.042
[3]	陈涛, 孙军, 谌芸, 等. 2019. 广州“5·7”局地突发特大暴雨过程的数值可预报性分析 [J]. 气象, 45(9): 1199−1212. doi: 10.7519/j.issn.1000-0526.2019.09.002 Chen Tao, Sun Jun, Chen Yun, et al. 2019. Study on the numerical predictivity of localized severe mesoscale rainstorm in Guangzhou on 7 May 2017 [J]. Meteor. Mon. (in Chinese), 45(9): 1199−1212. doi: 10.7519/j.issn.1000-0526.2019.09.002
[4]	杜钧, 李俊. 2014. 集合预报方法在暴雨研究和预报中的应用 [J]. 气象科技进展, 4(5): 6−20. doi: 10.3969/j.issn.2095-1973.2014.05.001 Du Jun, Li Jun. 2014. Application of ensemble methodology to heavy-rain research and prediction [J]. Advances in Meteorological Science and Technology, 4(5): 6−20. doi: 10.3969/j.issn.2095-1973.2014.05.001
[5]	傅慎明, 赵思雄, 孙建华, 等. 2010. 一类低涡切变型华南前汛期致洪暴雨的分析研究 [J]. 大气科学, 34(2): 235−252. doi: 10.3878/j.issn.1006-9895.2010.02.01 Fu Shenming, Zhao Sixiong, Sun Jianhua, et al. 2010. One kind of vortex causing heavy rainfall during pre-rainy season in South China [J]. Chinese Journal of Atmospheric Sciences (in Chinese), 34(2): 235−252. doi: 10.3878/j.issn.1006-9895.2010.02.01
[6]	管勇, 刘玉明, 胡丽华, 等. 2010. 广东两次特大暴雨成因的诊断对比 [J]. 气象科技, 38(5): 565−571. doi: 10.3969/j.issn.1671-6345.2010.05.006 Guan Yong, Liu Yuming, Hu Lihua, et al. 2010. Diagnostic analysis of two torrential rain events in Guangdong [J]. Meteorological Science and Technology, 38(5): 565−571. doi: 10.3969/j.issn.1671-6345.2010.05.006
[7]	胡亮, 何金海, 高守亭. 2007. 华南持续性暴雨的大尺度降水条件分析 [J]. 南京气象学院学报, 30(3): 345−351. doi: 10.3969/j.issn.1674-7097.2007.03.008 Hu Liang, He Jinhai, Gao Shouting. 2007. An analysis of large-scale condition for persistent heavy rain in South China [J]. Journal of Nanjing Institute of Meteorology, 30(3): 345−351. doi: 10.3969/j.issn.1674-7097.2007.03.008
[8]	Huang Ling, Luo Yali. 2017. Evaluation of quantitative precipitation forecasts by TIGGE ensembles for South China during the presummer rainy season [J]. J. Geophys. Res. Atmos., 122(16): 8494−8516. doi: 10.1002/2017JD026512
[9]	李春晖, 吴志伟, 蒙伟光, 等. 2017. 影响华南后汛期季风持续性暴雨和热带气旋持续性暴雨的大尺度环流背景分析 [J]. 热带气象学报, 33(1): 11−20. doi: 10.16032/j.issn.1004-4965.2017.01.002 Li Chunhui, Wu Zhiwei, Meng Weiguang, et al. 2017. Compared analysis characteristics of impact of large scale circulation on the persistent heavy rains in post-flood season in South China associated with monsoon and tropical cyclone [J]. Journal of Tropical Meteorology (in Chinese), 33(1): 11−20. doi: 10.16032/j.issn.1004-4965.2017.01.002
[10]	Li Shuwen, Yang Shuai, Liu Haiwen. 2018. Sensitivity of warm-sector heavy precipitation to the impact of anthropogenic heating in South China [J]. Atmospheric and Oceanic Science Letters, 11(3): 236−245. doi: 10.1080/16742834.2018.1469952
[11]	Li Zhenghui, Luo Yali, Du Yu, et al. 2020. Statistical characteristics of pre-summer rainfall over South China and associated synoptic conditions [J]. Journal of the Meteorological Society of Japan, 98(1): 213−233. doi: 10.2151/jmsj.2020-012
[12]	林晓霞, 冯业荣, 张诚忠, 等. 2017. 华南一次暴雨过程热力和动力特征的诊断分析 [J]. 热带气象学报, 33(6): 975−984. doi: 10.16032/j.issn.1004-4965.2017.06.018 Lin Xiaoxia, Feng Yerong, Zhang Chengzhong, et al. 2017. Diagnostic analysis of thermal and dynamic characteristics of a rainstorm process in southern China [J]. Journal of Tropical Meteorology (in Chinese), 33(6): 975−984. doi: 10.16032/j.issn.1004-4965.2017.06.018
[13]	刘瑞鑫, 孙建华, 陈鲍发. 2019. 华南暖区暴雨事件的筛选与分类研究 [J]. 大气科学, 43(1): 119−130. doi: 10.3878/j.issn.1006-9895.1803.17245 Liu Ruixin, Sun Jianhua, Chen Baofa. 2019. Selection and classification of warm-sector heavy rainfall events over South China [J]. Chinese Journal of Atmospheric Sciences (in Chinese), 43(1): 119−130. doi: 10.3878/j.issn.1006-9895.1803.17245
[14]	鲁蓉, 孙建华, 傅慎明. 2018. 近海水汽初值和对流影响一次华南前汛期沿海强降水对流系统发展过程的机理研究 [J]. 大气科学, 42(1): 1−15. doi: 10.3878/j.issn.1006-9895.1705.16261 Lu Rong, Sun Jianhua, Fu Shenming. 2018. Influence of offshore initial moisture field and convection on the development of coastal convection in a heavy rainfall event over South China during the pre-summer rainy season [J]. Chinese Journal of Atmospheric Sciences (in Chinese), 42(1): 1−15. doi: 10.3878/j.issn.1006-9895.1705.16261
[15]	Luo Yali, Li Liye, Johnson R H, et al. 2019. Science and prediction of monsoon heavy rainfall [J]. Science Bulletin, 64(21): 1557−1561. doi: 10.1016/j.scib.2019.09.005
[16]	罗雨, 张立凤. 2010. 一次梅雨暴雨预报中的误差演变及可预报性分析 [J]. 气象学报, 68(3): 411−420. doi: 10.11676/qxxb2010.040 Luo Yu, Zhang Lifeng. 2010. A case study of the error growth evolution in a Meiyu front heavy precipitation forecast and an analysis of the predictability [J]. Acta Meteor. Sinica (in Chinese), 68(3): 411−420. doi: 10.11676/qxxb2010.040
[17]	史学丽, 丁一汇. 2000. 1994年中国华南大范围暴雨过程的形成与夏季风活动的研究 [J]. 气象学报, 58(6): 666−678. doi: 10.11676/qxxb2000.068 Shi Xueli, Ding Yihui. 2000. A study on extensive heavy rain processes in South China and the summer monsoon activity in 1994 [J]. Acta Meteorologica Sinica (in Chinese), 58(6): 666−678. doi: 10.11676/qxxb2000.068
[18]	Wu Mengwen, Luo Yali, Chen Fei, et al. 2019. Observed link of extreme hourly precipitation changes to urbanization over coastal South China [J]. J. Appl. Meteor. Climatol., 58(8): 1799−1819. doi: 10.1175/JAMC-D-18-0284.1
[19]	吴亚丽, 蒙伟光, 陈德辉, 等. 2018. 一次华南暖区暴雨过程可预报性的初值影响研究 [J]. 气象学报, 76(3): 323−342. doi: 10.11676/qxxb2018.001 Wu Yali, Meng Weiguang, Chen Dehui, et al. 2018. A study of the impact of initial conditions on the predictability of a warm-sector torrential rain over South China [J]. Acta Meteor. Sin. (in Chinese), 76(3): 323−342. doi: 10.11676/qxxb2018.001
[20]	张诚忠, 万齐林, 黄燕燕, 等. 2008. 华南暖区降水数值预报的初值同化试验 [J]. 热带气象学报, 24(6): 576−589. doi: 10.3969/j.issn.1004-4965.2008.06.002 Zhang Chengzhong, Wan Qilin, Huang Yanyan, et al. 2008. Numerical experiments on the impact of initial conditions upon precipitation in warm region over south of China [J]. J. Trop. Meteor. (in Chinese), 24(6): 576−589. doi: 10.3969/j.issn.1004-4965.2008.06.002
[21]	Zhang Murong, Meng Zhiyong. 2018. Impact of synoptic-scale factors on rainfall forecast in different stages of a persistent heavy rainfall event in South China [J]. J. Geophys. Res. Atmos., 123(7): 3574−3593. doi: 10.1002/2017JD028155
[22]	Zhang Renhe, Ni Yongqi, Liu Liping, et al. 2011. South China Heavy Rainfall Experiments (SCHeREX) [J]. J. Meteor. Soc. Japan, 89A: 153−166. doi: 10.2151/jmsj.2011-A10
[23]	周菲凡. 2009. 条件非线性最优扰动方法在台风目标观测中的应用研究 [D]. 中国科学院大气物理研究所博士学位论文, 126pp. Zhou Feifan. 2009. Application of conditional nonlinear optimal disturbance method in typhoon target observation [D]. Ph. D. dissertation (in Chinese), Institute of Atmospheric Physics, Chinese Academy of Sciences, 126pp.
[24]	Zhou Feifan, Yamaguchi M, Qin Xiaohao. 2016. Possible sources of forecast errors generated by the global/regional assimilation and prediction system for landfalling tropical cyclones. Part Ⅰ: Initial uncertainties [J]. Advances in Atmospheric Sciences, 33(7): 841−851. doi: 10.1007/s00376-016-5238-4
[25]	Zhou Feifan, Duan Wansuo, Zhang He, et al. 2018. Possible sources of forecast errors generated by the global/regional assimilation and prediction system for landfalling tropical cyclones. Part Ⅱ: Model uncertainty [J]. Advances in Atmospheric Sciences, 35(10): 1277−1290. doi: 10.1007/s00376-018-7095-9
[26]	钟雄炎, 张东, 管勇. 2017. 2015年5月16~17日粤西南沿海暴雨过程分析 [J]. 广东气象, 39(1): 14−18. doi: 10.3969/j.issn.1007-6190.2017.01.004 Zhong Xiongyan, Zhang Dong, Guan Yong. 2017. Analysis of a heavy rain process in the coastal area of southwestern Guangdong May 16-17, 2015 [J]. Guangdong Meteorology, 39(1): 14−18. doi: 10.3969/j.issn.1007-6190.2017.01.004
[27]	Zhong Suixin, Yang Shuai, Guo Chunya, et al. 2019. Capabilities and limitations of GRAPES simulations of extreme precipitation in the warm sector over a complex orography [J]. J. Trop. Meteor, 25(2): 180−191.