基于Copula函数的华南台风极端灾害的重现期研究
|
摘要
利用1981-2014年华南地区的台风灾害数据,通过计算致灾因子的重现期来预测台风极端灾害频率,评估台风极端灾害的受灾程度.选取最大风速极值(X)和中心气压极值(Y)为研究变量,经过K-S检验选取最优的对数正态分布、正态分布分别作为X、Y的边缘分布,再通过Frank Copula函数构造联合分布,计算单变量重现期、联合重现期及同现重现期,并求出变量X与Y的设计值.计算结果发现,重现期不超过两年时,联合重现期的设计值要优于单变量重现期和同现重现期的设计值"重现期超过两年时,同现重现期的设计值要优于单变量重现期和联合重现期的设计值.最后以2015年至2017年台风中的X与Y作为输入,其重现期作为输出进行检验,结果与设计的组合重现期对比最优.研究结果表明,重现期小于等于两年时,选取联合重现期的设计值作为防灾标准"重现期大于两年时,选取同现重现期的设计值作为防灾标准,且致灾因子的重现期越长,台风极端灾害就越严重.
Using the extreme typhoon disaster data of South China from 1981 to 2014,the frequency of extreme typhoon disasters was estimated by calculating the return period of the hazards, and the extent of typhoon disasters was assessed. Select the maximum wind speed(X) and center pressure(Y) as research variables. After K-S test, select the optimal lognormal distribution and normal distribution as the edge distribution of X and Y, and then pass the Frank Copula function. Construct a joint distribution, calculate the univariate return period, joint return period, and co-occurrence return period, and calculate the design values of variables X and Y. The calculation results show that when the return period is not more than two years, the design value of the joint return period is better than the design value of the return period and co-occurrence return period of the univariate; when the return period exceeds two years, the co-occurrence reappears. The design value of the period is better than the design value of the univariate return period and the joint return period. Finally, taking the X and Y in the typhoon from 2015 to 2017 as input, the return period is used as an output test, and the result is optimal compared with the combined return period of the design. The results of the study indicate that when the return period is less than or equal to two years,the design value of the joint return period is selected as the disaster prevention standard;when the return period is greater than two years, the design value of the return period of co-occurrence is selected as the disaster prevention standard. The longer the return period of disaster factors, the more severe the typhoon disaster.
Using the extreme typhoon disaster data of South China from 1981 to 2014,the frequency of extreme typhoon disasters was estimated by calculating the return period of the hazards, and the extent of typhoon disasters was assessed. Select the maximum wind speed(X) and center pressure(Y) as research variables. After K-S test, select the optimal lognormal distribution and normal distribution as the edge distribution of X and Y, and then pass the Frank Copula function. Construct a joint distribution, calculate the univariate return period, joint return period, and co-occurrence return period, and calculate the design values of variables X and Y. The calculation results show that when the return period is not more than two years, the design value of the joint return period is better than the design value of the return period and co-occurrence return period of the univariate; when the return period exceeds two years, the co-occurrence reappears. The design value of the period is better than the design value of the univariate return period and the joint return period. Finally, taking the X and Y in the typhoon from 2015 to 2017 as input, the return period is used as an output test, and the result is optimal compared with the combined return period of the design. The results of the study indicate that when the return period is less than or equal to two years,the design value of the joint return period is selected as the disaster prevention standard;when the return period is greater than two years, the design value of the return period of co-occurrence is selected as the disaster prevention standard. The longer the return period of disaster factors, the more severe the typhoon disaster.
引文
[1]雷小途,陈佩燕,杨玉华,等.中国台风灾情特征及其灾害客观评估方法[J].气象学报,2009, 67(05):875-883.
[2]梁必骐,梁经萍,温之平.中国台风灾害及其影响的研究[J].自然灾害学报,1995(1):84-91.
[3]陈联寿等主编.台风预报及其灾害[M].浙江大学学报,北京:气象出版社,2009.
[4]牛海燕,刘敏,陆敏,等.中国沿海地区台风灾害损失评估研究[J].灾害学,2011, 26(3):61-64.
[5]丁燕,史培军.台风灾害的模糊风险评估模型[J].自然灾害学报,2002, 11(1):34-43.
[6]殷洁,吴绍洪,戴尔阜.广东省台风灾害经济损失风险评估(英文)[J]. Journal of Resources and Ecology,2012, 3(02):144-150.
[7]陈佩燕,杨玉华,雷小途,等.我国台风灾害成因分析及灾情预估[J].自然灾害学报,2009,18(1):64-73.
[8] Yu M, Yu W, Cheng J. The influencing factors of typhoon disaster to populated islands and its defensive countermeasures research[J]. Journal of Coastal Research, 2015, 73:348-352.
[9]张永恒,范广洲,马青云,等.浙江省台风灾害影响评估模型[J].应用气象学报,2009, 20(6):772-776.
[10]刘少军,张京红,何政伟,等.改进的物元可拓模型在台风灾害预评估中的应用[J].自然灾害学报,2012,21(2):135-141.
[11]娄伟平,陈海燕,邱新法,等.基于粒子群优化神经网络的台风灾情定量评估[J].自然灾害学报,2010,19(4):135-140.
[12]赵飞,廖永丰,张妮娜,等.登陆中国台风灾害损失预评估模型研究[J].灾害学,2011,26(2):81-85.
[13]国家统计局.中国统计年鉴1981-2017[Z].北京:中国统计出版社,2017.
[14]经广,杨武志.台风“海鸥”与“威马逊”风暴增水的差异分析[J].广东水利水电,2015(3):20-24.
[15]张京红,刘少军,蔡大鑫.基于GIS的海南岛橡胶林风害评估技术与应用[J].自然灾害学报,2013(4):177-183.
[16]郭爱军,黄强,畅建霞,等.基于Copula函数的泾河流域水沙关系演变特征分析[J].自然资源学报,2015,30(4):673-683.
[17] Singh V P, Zhang L. Bivariate flood frequency analysis using the copula method[J]. Journal of Hydrologic Engineering, 2006, 11(2):150-164.
[18]万永静,刁秀媚,刘俊,等.基于Copula函数的暴雨潮位组合分析[J].河海大学学报(自然科学版),2017,45(3):211-217.
[19] Song S, Singh V P. Meta-elliptical copulas for drought frequency analysis of preiodic data[J].Stochastic Environmental Research and Risk Assessment, 2010, 24(3):425-444.
[20]宋松柏,蔡焕杰,金菊良,等.Copula函数及其在水文中的应用[M].北京:科学出版社,2012.
[21]倪增华,刘合香,罗彦丽等.Copula函数在广西洪涝灾害的降水概率预测中的应用[J].气象研究与应用,2014, 35(2):32-39.
[22]史黎翔,宋松柏.基于Copula函数的两变量洪水重现期与设计值计算研究[J].水力发电学报,2015,34(10):27-34.
[23]潘璀林,陈子.基于GH Copula的韩江水文干旱联合概率分布研究[J].中山大学学报(自然科学版),2015, 54(1):110-115.
[24] Zhang L, Singh V P. Bivariate rainfall frequency distributions using Archimedean copulas[J]. Joumal of Hydrology, 2007, 332(2):93-109.
[25] Grimaldi S, Serinaldi F. Asymmetric copula in multivariate flood frequency analysis[J]. Advances in Water Resources, 2006, 29(8):1155-1167.
[26]中国气象局.热带气旋年鉴1981-2014年[Z].北京:气象出版社,2014.
[27]中国台风网.http://typhoon.weather.com.cn/index.shtml/2017/07/14.
[28]温克刚.中国气象灾害大典1981-1999年[Z].北京:气象出版社,2008.
[29] Nelsen R B. An Introduction to Copula[M].New York:Springer, 2006.
[30] Sklar A. Fonctions de repartitionan dimensionsetleur smarges[J]. Pub Inst Statist Univ Paris, 1959,8:229-231.
[31]郭生练,闫宝伟,肖义,等.Copula函数在多变量水文分析计算中的应用及研究进展[J].水文,2008,28(3):1-7.
[2]梁必骐,梁经萍,温之平.中国台风灾害及其影响的研究[J].自然灾害学报,1995(1):84-91.
[3]陈联寿等主编.台风预报及其灾害[M].浙江大学学报,北京:气象出版社,2009.
[4]牛海燕,刘敏,陆敏,等.中国沿海地区台风灾害损失评估研究[J].灾害学,2011, 26(3):61-64.
[5]丁燕,史培军.台风灾害的模糊风险评估模型[J].自然灾害学报,2002, 11(1):34-43.
[6]殷洁,吴绍洪,戴尔阜.广东省台风灾害经济损失风险评估(英文)[J]. Journal of Resources and Ecology,2012, 3(02):144-150.
[7]陈佩燕,杨玉华,雷小途,等.我国台风灾害成因分析及灾情预估[J].自然灾害学报,2009,18(1):64-73.
[8] Yu M, Yu W, Cheng J. The influencing factors of typhoon disaster to populated islands and its defensive countermeasures research[J]. Journal of Coastal Research, 2015, 73:348-352.
[9]张永恒,范广洲,马青云,等.浙江省台风灾害影响评估模型[J].应用气象学报,2009, 20(6):772-776.
[10]刘少军,张京红,何政伟,等.改进的物元可拓模型在台风灾害预评估中的应用[J].自然灾害学报,2012,21(2):135-141.
[11]娄伟平,陈海燕,邱新法,等.基于粒子群优化神经网络的台风灾情定量评估[J].自然灾害学报,2010,19(4):135-140.
[12]赵飞,廖永丰,张妮娜,等.登陆中国台风灾害损失预评估模型研究[J].灾害学,2011,26(2):81-85.
[13]国家统计局.中国统计年鉴1981-2017[Z].北京:中国统计出版社,2017.
[14]经广,杨武志.台风“海鸥”与“威马逊”风暴增水的差异分析[J].广东水利水电,2015(3):20-24.
[15]张京红,刘少军,蔡大鑫.基于GIS的海南岛橡胶林风害评估技术与应用[J].自然灾害学报,2013(4):177-183.
[16]郭爱军,黄强,畅建霞,等.基于Copula函数的泾河流域水沙关系演变特征分析[J].自然资源学报,2015,30(4):673-683.
[17] Singh V P, Zhang L. Bivariate flood frequency analysis using the copula method[J]. Journal of Hydrologic Engineering, 2006, 11(2):150-164.
[18]万永静,刁秀媚,刘俊,等.基于Copula函数的暴雨潮位组合分析[J].河海大学学报(自然科学版),2017,45(3):211-217.
[19] Song S, Singh V P. Meta-elliptical copulas for drought frequency analysis of preiodic data[J].Stochastic Environmental Research and Risk Assessment, 2010, 24(3):425-444.
[20]宋松柏,蔡焕杰,金菊良,等.Copula函数及其在水文中的应用[M].北京:科学出版社,2012.
[21]倪增华,刘合香,罗彦丽等.Copula函数在广西洪涝灾害的降水概率预测中的应用[J].气象研究与应用,2014, 35(2):32-39.
[22]史黎翔,宋松柏.基于Copula函数的两变量洪水重现期与设计值计算研究[J].水力发电学报,2015,34(10):27-34.
[23]潘璀林,陈子.基于GH Copula的韩江水文干旱联合概率分布研究[J].中山大学学报(自然科学版),2015, 54(1):110-115.
[24] Zhang L, Singh V P. Bivariate rainfall frequency distributions using Archimedean copulas[J]. Joumal of Hydrology, 2007, 332(2):93-109.
[25] Grimaldi S, Serinaldi F. Asymmetric copula in multivariate flood frequency analysis[J]. Advances in Water Resources, 2006, 29(8):1155-1167.
[26]中国气象局.热带气旋年鉴1981-2014年[Z].北京:气象出版社,2014.
[27]中国台风网.http://typhoon.weather.com.cn/index.shtml/2017/07/14.
[28]温克刚.中国气象灾害大典1981-1999年[Z].北京:气象出版社,2008.
[29] Nelsen R B. An Introduction to Copula[M].New York:Springer, 2006.
[30] Sklar A. Fonctions de repartitionan dimensionsetleur smarges[J]. Pub Inst Statist Univ Paris, 1959,8:229-231.
[31]郭生练,闫宝伟,肖义,等.Copula函数在多变量水文分析计算中的应用及研究进展[J].水文,2008,28(3):1-7.