Advances in dynamic-statistical analog ensemble forecasting and its application to precipitation prediction of landfalling typhoons: A renewed understanding
-
摘要: 动力统计结合是提高天气、气候预报水平的重要途径之一，关键问题是如何将数值模式与历史资料进行有效结合；相似预报这一传统方法与动力统计的结合是未来提高天气、气候预报水平的一个重要方向，尽管其原理目前仍停留在相似假设基础上且缺乏坚实的物理基础。文中从准确模式的初值问题出发，提出准确模式初值扰动概念，进而发展了动力统计相似集合预报（Dynamical Statistical Analog Ensemble Forecast，DSAEF）理论。DSAEF理论不仅回答了为什么可以进行相似预报，同时还指出了如何进行相似预报，即其原理是利用准确模式来做预报，并采用集合预报的方式实现预报。基于 DSAEF 理论，建立了登陆台风降水动力统计相似集合预报DSAEF_LTP （Landfalling Typhoon Precipitation，LTP）模型，该模型包括4个步骤：台风路径预报、广义初值构建、初值相似性判别和台风降水集合，其中广义初值由影响台风降水的物理因子构成。DSAEF_LTP模型具有可持续发展特性—可通过引入新因子或改善模型参数来改进模型的性能；目前该模型发布了广义初值包含台风路径、登陆季节和台风强度3个物理因子的1.0版和在此基础上改进了“相似区域”和“集合方案”的1.1版。该模型的性能提升很快，已完成的最新版本（1.1版）3次大样本预报试验均显示，与ECMWF、CMA-GFS、NCEP-GFS和SMS-WARMS （上海区域模式）对比，对≥100 mm和≥250 mm台风过程降水预报的TS评分，DSAEF_LTP模型（V1.1）排名第1。今后，围绕广义初值不断改善，研究引入更多影响登陆台风降水的物理因子，DSAEF_LTP 模型的发展前景广阔。
-
关键词:
- 相似预报 /
- 原理 /
- 动力统计相似集合预报理论 /
- 登陆台风降水 /
- DSAEF_LTP 模型
Abstract: Combining dynamical and statistical methods is one of the important ways to improve weather and climate prediction. A key issue is how to effectively combine numerical model results with historical data. Combining the above two methods with an analog method is an important direction for future improvement of weather and climate prediction, although the analog method is still limited to similarity assumptions and lacks solid physical basis. Based on the initial condition problem of a perfect model, this work proposes the concept of initial condition perturbation of the perfect model and develops a Dynamical-Statistical-Analog Ensemble Forecast (DSAEF) theory. The DSAEF theory shows not only why the analogue-based forecast can be conducted, but also how it can be conducted. That is, the perfect model is used to produce forecasts and an ensemble prediction scheme is used to achieve the forecast accuracy. Based on the DSAEF theory, the DSAEF_LTP (Landfalling Typhoon Precipitation, LTP) model has been developed. This model includes the following four steps: (Ⅰ) forecast typhoon track, (Ⅱ) Construct generalized Initial Value (GIV), (Ⅲ) identify analogs from historical observations, and (Ⅳ) produce an ensemble forecast of typhoon precipitation. The GIV is constructed by physical variables that affect LTP. The DSAEF_ LTP model has the characteristic of sustainable development, which can be improved by introducing new variables or refining the existing parameters of the model. At present, the model versions 1.0 and 1.1 have been released. In version 1.0, GIVs include three physical variables, i.e., typhoon track, landfall season and typhoon intensity. In the version 1.1, two extra improved parameters of "similarity region" and "ensemble scheme" are added. The most recent version 1.1 has shown significant improvements in the performance for forecasting LTP. Three large-sample forecast experiments using version 1.1 show that, compared ECMWF, CMA-GFS, NCEP-GFS and SMS-WARMS (Shanghai regional model), version 1.1 performs best in forecasting accumulated rainfall greater than 250 mm and 100 mm. Of course, there is a considerable room for improvement in the DSAEF_LTP model forecast performance, which can be achieved by incorporating more physical variables that affect LTP into the GIV of the model. This implies that the DSAEF_LTP model will have broad prospects for future development. -
图 1 利用准确模式做预报之动力统计相似集合预报 （理论） 示意
Figure 1. Schematic diagram of the Dynamical Statistical Analog Ensemble Forecast （theory） using the perfect model to produce forecasts
图 2 动力统计相似集合预报 DSAEF 模型示意
Figure 2. Schematic diagram of the Dynamical Statistical Analog Ensemble Forecast （DSAEF） model
图 3 影响登陆热带气旋降水的物理因子示意
Figure 3. Schematic diagram of physical factors affecting precipitation induced by landfalling tropical cyclones
图 4 对于某起报点完整的目标台风路径示意
Figure 4. Schematic diagram of the complete target typhoon track for a certain initial point
图 5 登陆台风降水的动力统计相似集合预报 （DSAEF_LTP） 模型流程
Figure 5. Flowchart of the Dynamical Statistical Analog Ensemble Forecast for Landfalling Tropical cyclone Precipitation （DSAEF_LTP） model
图 6 DSAEF_LTP-2及DSAEF_LTP-3A模型对2015—2016年华南台风过程降水预报试验与数值模式TS评分对比
Figure 6. Comparison of Threat Scores of accumulative precipitation forecast experiments with the DSAEF_LTP-2 and DSAEF_LTP-3A models and numerical models for typhoons during 2015—2016 in South China
图 7 不同版本DSAEF_LTP模型对台风“利奇马”（2019）过程降水模拟试验TS评分对比
Figure 7. Comparison of Threat Scores of accumulative precipitation simulation using different versions of the DSAEF_LTP model for Typhoon Lekima （2019）
图 8 不同改进集合方案DSAEF_LTP模型与数值模式对2018年台风过程降水模拟试验TS评分对比
Figure 8. Comparison of Threat Scores of accumulative precipitation forecast by the DSAEF_LTP model with different improved ensemble schemes and numerical models for typhoons in 2018
图 9 不同方法对2017—2018年8个影响华南台风过程降水预报的平均TS评分对比 （DSAEF_LTP-1、-2和-3为不同过渡版本，DSAEF_LTP-4为1.1版；数值模式为3个全球模式 （ECMWF、NCEP-GFS和CMA-GFS） 和上海区域模式 （SMS-WARMS））
Figure 9. Comparison of Threat Scores of accumulative precipitation forecast using different methods for eight typhoons influencing South China during 2017—2018 （DSAEF_LTP-1，-2 and -3 are different versions while DSAEF_LTP-4 is V1.1； meanwhile，numerical models are three global models （ECMWF，NCEP-GFS and CMA-GFS） and the Shanghai regional model （SMS-WARMS））
表 1 DSAEF_LTP 模型参数表
Table 1. List of the DSAEF_LTP model parameters
参数名称 取值方式 取值个数 起报时刻 陆面出现热带气旋降水前两天12：00 UTC、00：00 UTC 2×2=4 相似区域 热带气旋路径面积相似指数 （TSAI）中的一个参数，首先选取起报时刻和最大预报时效之预报时刻的热带气旋位置作为相似区域（矩形框）的两个对角点；起报时刻对应的顶点可以变为提前12、24、36或48 h的热带气旋观测位置，而另一个顶点可以变为时效缩短6或12 h的热带气旋预报位置 3×5 = 15 纬度极值点分割度阈值 TSAI的一个参数，可取值0.1、0.2和0.3 3 重叠度阈值 TSAI的一个参数，可取值0.9、0.8、···和0.4 6 季节相似 可取值1—12月、5—11月和7—9月 3 最佳相似热带气旋个数 可取值1—16个 16 集合预报方案 可取值平均值或最大值 2 方案总数 4×15×3×6×3×16×2 103、680 
下载: 导出CSV
-
[1] 陈德辉,薛纪善. 2004. 数值天气预报业务模式现状与展望. 气象学报,62(5):623-633 doi: 10.11676/qxxb2004.061Chen D H,Xue J S. 2004. An overview on recent progresses of the operational numerical weather prediction models. Acta Meteor Sinica,62(5):623-633 (in Chinese) doi: 10.11676/qxxb2004.061 [2] 丑纪范. 1974. 天气数值预报中使用过去资料的问题. 中国科学,(6):635-644Chou J F. 1974. Problems of using historical data in numerical weather forecasting. Scientia Sinica,(6):635-644 (in Chinese) [3] 丑纪范. 1986. 为什么要动力-统计相结合?—兼论如何结合. 高原气象,5(4):367-372Chou J F. 1986. Why the combination of dynamic and statistics?—How to combine. Plateau Meteor,5(4):367-372 (in Chinese) [4] 丑纪范,任宏利. 2006. 数值天气预报—另类途径的必要性和可行性. 应用气象学报,17(2):240-244 doi: 10.11898/1001-7313.20060216Chou J F,Ren H L. 2006. Numerical weather prediction-necessity and feasibility of an alternative methodology. J Appl Meteor Sci,17(2):240-244 (in Chinese) doi: 10.11898/1001-7313.20060216 [5] 顾震潮. 1958a. 作为初值问题的天气形势数值预报与由地面天气历史演变作预报的等值性. 气象学报,29(2):93-98Gu Z C. 1958a. On the equivalency of formulations of weather forecasting as an initial value problem and as an "evolution" problem. Acta Meteor Sinica,29(2):93-98 (in Chinese) [6] 顾震潮. 1958b. 天气数值预报中过去资料的使用问题. 气象学报,29(3):176-184Gu Z C. 1958b. On the utilization of past data in numerical weather forecasting. Acta Meteor Sinica,29(3):176-184 (in Chinese) [7] 黄建平. 1991. 环流异常相似性演变的动力学机制. 北京大学学报(自然科学版),27(1):99-108Huang J P. 1991. The dynamical mechanism of analogous evolution for circulation anomaly. Acta Sci Nat Univ Pekin,27(1):99-108 (in Chinese) [8] 黄伟,余晖,梁旭东. 2009. GRAPES-TCM对登陆热带气旋降水的预报及其性能评估. 气象学报,67(5):892-901Huang W,Yu H,Liang X D. 2009. Evaluation of GRAPES-TCM rainfall forecast for China landfalling tropical cyclone in 2006. Acta Meteor Sinica,67(5):892-901 (in Chinese) [9] 贾作. 2020. 基于DSAEF_LTP模型的登陆热带气旋日降水预报研究[D]. 北京: 中国气象科学研究院. Jia Z. 2020. Study on prediction of daily precipitation associated with landfalling tropical cyclones based on the DSAEF_LTP model [D]. Beijing: Chinese Academy of Meteorological Sciences （in Chinese） [10] 姜丽黎,余晖. 2019. 基于动力相似方法的台风极端降水概率预报研究. 热带气象学报,35(3):353-364Jiang L L,Yu H. 2019. A research on the prediction of typhoon extreme precipitation based on dynamic similitude methods. J Trop Meteor,35(3):353-364 (in Chinese) [11] 李博,赵思雄. 2009. 用SMAT建立台风暴雨预报模型的试验研究. 气象,35(6):3-12Li B,Zhao S X. 2009. Development of forecasting model of typhoon type rainstorm by using SMAT. Meteor Mon,35(6):3-12 (in Chinese) [12] 黎慧琦,张大林. 2021. 中小尺度对流系统的高分辨率数值模拟近况和未来挑战. 气象科技进展,11(3):75-91Li H Q,Zhang D L. 2021. High-resolution modeling of convective storms:Progress and future challenges. Adv Meteor Sci Technol,11(3):75-91 (in Chinese) [13] 林志强,德庆,边巴扎西等. 2015. 孟加拉湾热带风暴西藏降水的动力相似预报方法. 科技通报,31(9):25-28,35 doi: 10.13774/j.cnki.kjtb.2015.09.006Lin Z Q,De Q,Bianba Z X,et al. 2015. Dynamical analogue precipitation prediction over Tibetan Plateau effects by tropical storm of the Bay of Bengal. Bull Sci Technol,31(9):25-28,35 (in Chinese) doi: 10.13774/j.cnki.kjtb.2015.09.006 [14] 刘春霞. 2002. 广东热带气旋短期气候预测——相空间相似预报方法的应用. 热带气象学报,18(1):83-90 doi: 10.3969/j.issn.1004-4965.2002.01.010Liu C X. 2002. The short-term climate forecasting of tropical cyclone in Guangdong — In the phase space similarity method. J Trop Meteor,18(1):83-90 (in Chinese) doi: 10.3969/j.issn.1004-4965.2002.01.010 [15] 邱崇践,丑纪范. 1989. 天气预报的相似-动力方法. 大气科学,13(1):22-28Qiu C J,Chou J F. 1989. Similarity-dynamical method for weather forecast. Sci Atmos Sinica,13(1):22-28 (in Chinese) [16] 任福民,杨慧. 2019. 1949年以来我国台风暴雨及其预报研究回顾与展望. 暴雨灾害,38(5):526-540Ren F M,Yang H. 2019. An overview of advances in typhoon rainfall and its forecasting researches in China during the past 70 years and future prospects. Torrential Rain Disaster,38(5):526-540 (in Chinese) [17] 任宏利,封国林,张培群. 2007. 论动力相似预报的物理基础问题. 地球科学进展,22(10):1027-1035Ren H L,Feng G L,Zhang P Q. 2007. Physical basis of dynamical analogue prediction. Adv Earth Sci,22(10):1027-1035 (in Chinese) [18] 阎惠芳,李社宗,黄跃青等. 2003. 常用相似性判据的检验和综合相似系数的使用. 气象科技,31(4):211-215 doi: 10.3969/j.issn.1671-6345.2003.04.004Yan H F,Li S Z,Huang Y Q,et al. 2003. Tests for conventional similarity criterions and application of composite similar coefficient. Meteor Sci Technol,31(4):211-215 (in Chinese) doi: 10.3969/j.issn.1671-6345.2003.04.004 [19] 杨仁勇,陈有龙,符式红. 2012. 用T213产品动力过程相似释用法制作暴雨预报. 气象科技,40(3):401-405 doi: 10.3969/j.issn.1671-6345.2012.03.015Yang R Y,Chen Y L,Fu S H. 2012. Prediction of heavy rainfall based on similarity between dynamic processes from T213 products. Meteor Sci Technol,40(3):401-405 (in Chinese) doi: 10.3969/j.issn.1671-6345.2012.03.015 [20] 于海鹏,黄建平,李维京等. 2014. 数值预报误差订正技术中相似-动力方法的发展. 气象学报,72(5):1012-1022 doi: 10.11676/qxxb2014.082Yu H P,Huang J P,Li W J,et al. 2014. Development of the analogue-dynamical method for error correction of numerical forecasts. Acta Meteor Sinica,72(5):1012-1022 (in Chinese) doi: 10.11676/qxxb2014.082 [21] 张大林. 2005. 大气科学的世纪进展与未来展望. 气象学报,63(5):812-824 doi: 10.3321/j.issn:0577-6619.2005.05.025Zhang D L. 2005. An overview of centenary advances and prospects in atmospheric sciences. Acta Meteor Sinica,63(5):812-824 (in Chinese) doi: 10.3321/j.issn:0577-6619.2005.05.025 [22] 郑祖光. 1981. 苏联统计相似预报方法的新进展——群相似原理及其应用. 气象科技,(S4):8-11 doi: 10.19517/j.1671-6345.1981.s4.002Zheng Z G. 1981. The new development of statistical analogue prediction method in the Soviet Union — Group analogue principle and its application. Meteor Sci Technol,(S4):8-11 (in Chinese) doi: 10.19517/j.1671-6345.1981.s4.002 [23] 钟元. 2003. 多元判据综合评估中期天气客观相似预报模式. 气象,29(4):3-9 doi: 10.7519/j.issn.1000-0526.2003.04.001Zhong Y. 2003. An objective analogue model for medium-range weather forecast considered synthetic evaluation by multi-criterion. Meteor Mon,29(4):3-9 (in Chinese) doi: 10.7519/j.issn.1000-0526.2003.04.001 [24] 钟元,胡波. 2003. 综合评估环境场影响的热带气旋路径客观相似预报模式. 热带气象学报,19(2):147-156Zhong Y,Hu B. 2003. The objective analogue prediction model of tropical cyclone track considering synthetical evaluation environment. J Trop Meteor,19(2):147-156 (in Chinese) [25] 钟元,余晖,滕卫平等. 2009. 热带气旋定量降水预报的动力相似方案. 应用气象学报,20(1):17-27 doi: 10.3969/j.issn.1001-7313.2009.01.003Zhong Y,Yu H,Teng W P,et al. 2009. A dynamic similitude scheme for tropical cyclone quantitative precipitation forecast. J Appl Meteor Sci,20(1):17-27 (in Chinese) doi: 10.3969/j.issn.1001-7313.2009.01.003 [26] Bauer P,Thorpe A,Brunet G. 2015. The quiet revolution of numerical weather prediction. Nature,525(7567):47-55 doi: 10.1038/nature14956 [27] Cangialosi J P,Franklin J L. 2015. 2014 National Hurricane Center Forecast Verification Report. NOAA/NWS/NCEP/National Hurricane Center,Miami:Florida [28] Charney J,Halem M,Jastrow R. 1969. Use of incomplete historical data to infer the present state of the atmosphere. J Atmos Sci,26(5):1160-1163 doi: 10.1175/1520-0469(1969)026<1160:UOIHDT>2.0.CO;2 [29] Chen L S,Li Y,Cheng Z Q. 2010. An overview of research and forecasting on rainfall associated with landfalling tropical cyclones. Adv Atmos Sci,27(5):967-976 doi: 10.1007/s00376-010-8171-y [30] Ding C C,Ren F M,Liu Y N,et al. 2020. Improvement in the forecasting of heavy rainfall over south China in the DSAEF_LTP model by introducing the intensity of the tropical cyclone. Wea Forecasting,35(5):1967-1980 doi: 10.1175/WAF-D-19-0247.1 [31] Ebert E E,Turk M,Kusselson S J,et al. 2011. Ensemble tropical rainfall potential (eTRaP) forecasts. Wea Forecasting,26(2):213-224 doi: 10.1175/2010WAF2222443.1 [32] Fraedrich K,Rückert B. 1998. Metric adaption for analog forecasting. Phys A,253(1-4):379-393 doi: 10.1016/S0378-4371(97)00668-7 [33] Fraedrich K,Raible C C,Sielmann F. 2003. Analog ensemble forecasts of 660 tropical cyclone tracks in the Australian region. Wea. Forecasting,18:3-11 doi: 10.1175/1520-0434(2003)018<0003:AEFOTC>2.0.CO;2 [34] Jia L,Jia Z,Ren F M,et al. 2020. Introducing TC intensity into the DSAEF_LTP model and simulating precipitation of super-typhoon Lekima (2019). Quart J Roy Meteor Soc,146(733):3965-3979 doi: 10.1002/qj.3882 [35] Jia L,Ren F M,Ding C C,et al. 2022. Improvement of the ensemble methods in the dynamical-statistical-analog ensemble forecast model for land-falling typhoon precipitation. J Meteor Soc Japan,100(3):575-592 doi: 10.2151/jmsj.2022-029 [36] Kidder S Q,Knaff J A,Kusselson S J,et al. 2005. The tropical rainfall potential (TRaP) technique. Part Ⅰ:Description and examples. Wea Forecasting,20(4):456-464 doi: 10.1175/WAF860.1 [37] Langmack H,Fraedrich K,Sielmann F. 2012. Tropical cyclone track analog ensemble forecasting in the extended Australian basin:NWP combinations. Quart J Roy Meteor Soc,138(668):1828-1838 doi: 10.1002/qj.1915 [38] Lonfat M,Rogers R,Marchok T,et al. 2007. A parametric model for predicting hurricane rainfall. Mon Wea Rev,135(9):3086-3097 doi: 10.1175/MWR3433.1 [39] Luitel B,Villarini G,Vecchi G A. 2018. Verification of the skill of numerical weather prediction models in forecasting rainfall from U. S. landfalling tropical cyclones. J Hydrol,556:1026-1037 doi: 10.1016/j.jhydrol.2016.09.019 [40] Ma Y Q,Jia Z,Ren F M,et al. 2022a. Introducing TC translation speed into the dynamical-statistical-analog ensemble forecast for landfalling typhoon daily precipitation model and simulating the daily precipitation of supertyphoon Lekima (2019). Wea Forecasting,37(11):2005-2020 doi: 10.1175/WAF-D-21-0135.1 [41] Ma Y Q,Ren F M,Jia L,et al. 2022b. Experiments with the improved dynamical-statistical-analog ensemble forecast model for landfalling typhoon precipitation over South China. J Trop Meteor,28(2):139-153 doi: 10.46267/j.1006-8775.2022.011 [42] Marks Jr F D, Kappler G, DeMaria M. 2002. Development of a tropical cyclone rainfall climatology and persistence （R-CLIPER） model∥Pre- prints, 25th Conference on Hurricanes and Tropical Meteorology. San Diego: American Meteorological Society, 327-328 [43] Qin S,Jia L,Ding C C,et al. 2022. Experiments of DSAEF_LTP Model with two improved parameters for accumulated precipitation of landfalling tropical cyclones over southeast China. J Trop Meteor,28(3):286-296 doi: 10.46267/j.1006-8775.2022.022 [44] Ren F M,Wang Y M,Wang X L,et al. 2007. Estimating tropical cyclone precipitation from station observations. Adv Atmos Sci,24(4):700-711 doi: 10.1007/s00376-007-0700-y [45] Ren F M,Qiu W Y,Ding C C,et al. 2018. An objective track similarity index and its preliminary application to predicting precipitation of landfalling tropical cyclones. Wea Forecasting,33(6):1725-1742 doi: 10.1175/WAF-D-18-0007.1 [46] Ren F M,Ding C C,Zhang D L,et al. 2020. A dynamical-statistical-analog ensemble forecast model:Theory and an application to heavy rainfall forecasts of landfalling tropical cyclones. Mon Wea Rev,148(4):1503-1517 doi: 10.1175/MWR-D-19-0174.1 [47] Rogers R. 2018. Summary of fourth international workshop on tropical cyclone landfall processes (IWTCLP-4). Honolulu:WMO/IWTC-9,10:3-7 [48] Tuleya R E,DeMaria M,Kuligowski R J. 2007. Evaluation of GFDL and simple statistical model rainfall forecasts for U. S. landfalling tropical storms. Wea Forecasting,22(1):56-70 doi: 10.1175/WAF972.1 [49] Wang M Y, Ren F M, Zhang D L, et al. 2022. Tropical cyclone heavy rainfall forecasts over China with an upgraded DSAEF_LTP model. Weather and Climate Extremes, Under Revision [50] Wang Y,Shen X S,Chen D H. 2012. Verification of tropical cyclone rainfall predictions from CMA and JMA global models. J Trop Meteor,18(4):537-542 [51] WMO. 2014. Pre-workshop topic reports: Eighth WMO international workshop on tropical cyclones（IWTC-Ⅲ）. Jeju, Republic of Korea, 10: 2-10 [52] Woo W C, Hogsett W, Mohapatra M, et al. 2014. Challenges and advances related to TC rainfall forecast. The Third International Workshop on Tropical Cyclone Landfall Processes （IWTCLP-Ⅲ） -
下载:
点击查看大图
图(9) / 表(1)
计量
- 文章访问数: 376
- HTML全文浏览量: 75
- PDF下载量: 113
- 被引次数: 0
