统计降尺度方法在天津小时降水和气温精细化预报中的应用

doi:10.11755/j.issn.1006-7639(2022)-01-0135

摘要/Abstract

摘要：

基于ECWMF模式预报数据对2018年3—11月降水和2 m温度进行统计降尺度,利用先频率匹配法、再阈值法对插值后的降水订正,利用Kalman滤波型的递减平均统计降尺度法对插值后的温度订正,最终获得逐小时降水量和温度的预报。结果表明:(1)对于晴雨预报准确率,绝大多数预报时效频率匹配法和阈值法均对其有明显提高,前者最大改进幅度可达20%以上。对于相对误差,阈值法对空报现象有较显著改进。对于1 h降雨量大于等于20 mm的短时强降水,频率匹配法订正后的TS评分有明显提高。对2018年“安比”台风事件,除具有以上改进效果外,频率匹配法提高了降水主体形态和量级的预报水平,阈值法对空报站订正正确。(2)对于温度的ECWMF模式预报检验,几乎在任何预报时效内都是3月的绝对误差最大。通过Kalman滤波型的递减平均统计降尺度法后,各月的绝对误差都有不同程度减小。总体上,订正后的绝对误差曲线仍具有订正前的周期性波动,波峰、波谷位置也与订正前基本一致,且绝对误差越大,订正幅度越大。个例分析也表明订正后保留了温度预报空间分布的准确性,且绝对误差有明显下降。

关键词: 天津, 统计降尺度, 频率匹配法, 阈值法, Kalman滤波型的递减平均统计降尺度法

Abstract:

Statistical downscaling forecasting of precipitation and 2 m air temperature was obtained based on ECMWF model forecast data from March to November 2018. The interpolated precipitation was corrected using the frequency matching method firstly and then the threshold method, the interpolated temperature was corrected by using the Kalman filter-type decreasing average statistical downscaling technique, finally the hourly precipitation and temperature prediction were obtained. The results are as follows: (1) For the accuracy of rain probability forecast, it was obviously improved by using the frequency matching method and the threshold method for most forecasting time, and the maximum improvement range was more than 20% for the former. For the relative error, the threshold method had reduced the occurrence of false alarms considerably. For the short-term heavy rainfall with 1 h rainfall greater than or equal to 20 mm, the TS score was also improved significantly after using the frequency matching method. For the Typhoon “Amby” event in 2018, in addition to the above improvement effects, the frequency matching method improved the prediction capacity of the model about the amount and patterns of rainfalls, and the threshold method corrected false-alarm station completely. (2) For the test of temperature forecast of ECMWF model, the absolute error was the largest in March for almost forecast time. After using the Kalman filter-type decreasing average statistical downscaling technique, the absolute error of temperature in different months decreased to varying degrees. In general, the absolute error curve after correction still had the periodic fluctuation with the extension of forecast period, and the position of wave peak and trough was basically the same as those before correction, and the greater the absolute error, the greater the correction range was. For the temperature case, the accuracy of the spatial distribution of temperature prediction was retained, and the absolute error decreased significantly after correction.

Key words: Tianjing, statistical downscaling, frequency matching method, threshold method, Kalman filter-type decreasing average statistical downscaling technique

中图分类号:

P457.3

田笑, 余文韬, 从靖, 周红梅. 统计降尺度方法在天津小时降水和气温精细化预报中的应用[J]. 干旱气象, 2022, 40(1): 135-145.

TIAN Xiao, YU Wentao, CONG Jing, ZHOU Hongmei. Fine prediction of hourly precipitation and air temperature of Tianjin based on statistical downscaling in ECMWF model[J]. Journal of Arid Meteorology, 2022, 40(1): 135-145.

图/表 10

图1 2018年7月1日至8月31日1~96 h时效1 h累计降水的ECMWF模式预报(蓝线),频率匹配法订正后(红线)及二次订正后(绿线)的晴雨预报准确率曲线

Fig.1 The accuracy of rain probability forecast by ECMWF model (blue line) and its revised curves by using frequency matching method (red line) and secondary correction (green line) about 1-hour accumulated precipitation for 1-96 h forecast time from July 1 to August 31, 2018

图2 2018年7月1日至8月31日1~96 h时效1 h累计降水的ECMWF模式预报(蓝线)及频率匹配法订正后(红线)和二次订正后(绿线)的平均相对误差曲线

Fig.2 The average relative error of ECMWF model forecast (blue line) and its revised curves by using frequency matching method (red line) and secondary correction (green line) about 1-hour accumulated precipitation for 1-96 h forecast time from July 1 to August 31, 2018

图3 2018年7月1日至8月31日1~96 h时效1 h强降水(大于等于20 mm)的ECMWF模式预报(蓝线)和频率匹配法订正后(红线)的TS评分曲线

Fig.3 The TS score of ECMWF model forecast (blue line) and its revised curve by using frequency matching method (red line) of 1-hour heavy rainfall (equal to or more than 20 mm) for 1-96 h forecast time from July 1 to August 31, 2018

图4 2018年7月23—24日1~96 h时效1 h累计降水的ECMWF模式预报(蓝线)及频率匹配法订正后(红线)和二次订正后(绿线)的晴雨预报准确率曲线

Fig.4 The accuracy of rain probability forecast by ECMWF model (blue line) and its revised curves by using frequency matching method (red line) and secondary correction (green line) about 1-hour accumulated precipitation for 1-96 h forecast time from July 23 to 24, 2018

图5 2018年7月23—24日1~96 h时效1 h累计降水的ECMWF模式预报(蓝线)及频率匹配法订正后(红线)和二次订正后(绿线)的平均相对误差曲线

Fig.5 The average relative error of ECMWF model forecast (blue line) and its revised curves by frequency matching method (red line) and secondary correction (green line) about 1-hour accumulated precipitation for 1-96 h forecast time from July 23 to 24, 2018

图6 2018年7月23—24日1~96 h时效1 h强降水(大于等于20 mm)的ECMWF模式预报(蓝线)和频率匹配法订正后(红线)的TS评分曲线

Fig.6 The TS score of ECMWF model forecast (blue line) and its revised curve by using frequency matching method (red line) about 1-hour heavy rainfall (equal to or more than 20 mm) for 1-96 h forecast time from July 23 to 24, 2018

图7 2018年7月24日06:00降水实况(a),ECMWF模式7月22日预报的34 h时效降水预报(b)及频率匹配法订正(c)和二次订正(d)后的降水空间分布(单位:mm)

Fig.7 The spatial distribution of observed precipitation at 06:00 BST on July 24, 2018 (a), the 34 h forecast time precipitation of ECMWF model forecasted on July 22, 2018 (b), precipitation after frequency matching method correction (c) and after the second correction (d) (Unit: mm)

图8 2018年春季(a)、夏季(b)与秋季(c)的1~96 h时效1 hECMWF模式温度预报统计降尺度前(虚线)、后(实线)的平均绝对误差

Fig.8 The average absolute error of one hour temperature forecast of ECMWF model for 1-96 h forecast time in spring (a), summer (b) and autumn (c) in 2018 before (dotted lines) and after (solid lines) down-scaling

图9 2018年3月31日(a)与4月1日预报(b)的1~96 h时效1 h ECMWF模式温度预报统计降尺度前(虚线)、后(实线)的平均绝对误差

Fig.9 The average absolute error of 1-hour temperature of ECMWF model for 1-96 h forecast time forecasted on March 31, 2018 (a) and on April 1, 2018 (b)before (dotted line) and after (solid line) down-scaling

图10 2018年4月3日08:00天津气温分布实况(a), ECMWF模式3月31日预报的4月3日08:00统计降尺度前(b、d)、后(c、e)的气温分布(b、c)和绝对误差分布(d、e)(单位:℃)

Fig.10 The spatial distribution of observed temperature in Tianjin at 08:00 BST on April 3, 2018 (a), and forecasted temperature (b, c) by ECMWF model on March 31, 2018 and their absolute errors (d, e) before (b, d) and after (c,e) down-scaling (Unit: ℃)

参考文献 23

[1]	刘永和, 郭维栋, 冯锦明, 等. 气象资料的统计降尺度方法综述[J]. 地球科学进展, 2011, 26(8):837-847.
[2]	WILBY R L, WIGLEY T M L. Downscaling general circulation model output: A review of methods and limitations[J]. Progress in Physical Geography, 1997, 21:530-548.
[3]	范丽军, 符淙斌, 陈德亮. 统计降尺度法对未来区域气候变化情景预估的研究进展[J]. 地球科学进展, 2005, 20(3):320-329.
[4]	REBORA N, PROCENZALE L. Rain FARM: Rainfall downscaling by a filtered autoregressive model[J]. Journal of Hydrometeorolgy, 2006, 7(4):724-738.
[5]	BRANKOVIĆ Č, MATJAČIĆ B, IVATEK-ŠAHDAN S, et al. Downscaling of ECMWF ensemble forecasts for cases of severe weather: Ensemble statistics and cluster analysis[J]. Monthly Weather Review, 2008, 136(9):3323-3342. DOI URL
[6]	CAÑÓN J, DOMÍNGUEZ F, VALDÉ S. Downscaling climate variability as sociated with quasi-periodic climate signals: a new statistical approach using MSSA[J]. Journal of Hydrometeorolgy, 2011, 398(1/2):65-75.
[7]	KANTNALAL H, KANAMITSU M. Fifty-seven-year California reanalysis downscaling at 10 km (CaRD10). Part II: Comparison with North American regional reanalysis[J]. Journal of Climate, 2007, 20(20):5572-5592. DOI URL
[8]	YOSHIMURA K, KANAMITSU M. Dynamical global downscaling of global reanalysis[J]. Monthly Weather Review, 2008, 136(8):242-243. DOI URL
[9]	舒建川, 蒋兴文, 黄小梅, 等. 中国西南夏季降水预测的统计降尺度建模分析[J]. 高原气象, 2019, 38(2):349-358. DOI
[10]	徐一剑. 我国沿海城市应对气候变化的发展战略[J]. 气候变化研究进展, 2020, 16(1):88-98.
[11]	谢正辉, 刘斌, 延晓冬, 等. 应对气候变化的城市规划实施效应评估研究[J]. 地理科学进展, 2020, 39(1):120-131. DOI
[12]	唐文苑, 郑永光. 基于快速更新同化数值预报的小时降水量时间滞后集合订正技术[J]. 气象, 2019, 45(3):305-317.
[13]	李江萍, 王式功. 统计降尺度法在数值预报产品释用中的应用[J]. 气象, 2008, 34(6):41-45.
[14]	SAILOR D J, LI X. A semiempirical downscaling approach for predicting regional temperature impacts associated with climatic change[J]. Journal of Climate, 1999, 12(1):103-114. DOI URL
[15]	高红霞, 汤剑平. 华东地区月平均气温统计降尺度方法比较[J]. 气象科学, 2015, 35(6):760-768.
[16]	范丽军, 符淙斌, 陈德亮. 统计降尺度法对华北地区未来区域气温变化情景的预估[J]. 大气科学, 2007, 31(5):887-897.
[17]	陈国华, 郭品文. 中国地面气温统计降尺度预报方法研究[J]. 大气科学学报, 2016, 39(4):569-575.
[18]	肖瑶, 郭品文. 中国地面气温自适应递减平均降尺度方法的改进研究[J]. 气象科学, 2016, 26(5):606-613.
[19]	武杨, 李晴岚. 深圳地区日极值气温的降尺度研究[J]. 集成技术, 2014, 3(2):53-67.
[20]	王亚男, 智协飞. 多模式降水集合预报的统计降尺度研究[J]. 暴雨灾害, 2012, 31(1):1-7.
[21]	CUI B, TOTH Z, ZHU Y, et al. Bias Correction for Global Ensemble Forecast[J]. Weather & Forecasting, 2012, 27(2):396-410.
[22]	李莉, 李应林, 田华, 等. T213全球集合预报系统性误差订正研究[J]. 气象, 2011, 37(1):31-38.
[23]	邱学兴, 王东勇, 陈宝峰. T639模式预报系统误差统计和订正方法研究[J]. 气象, 2012, 38(5):526-532.