基于自组织映射神经网络的吉林省春夏期降水统计模拟研究
|
摘要
利用1997—2015年吉林省春夏期(4—7月)逐日气象站地面观测资料,以气温、气压、相对湿度、水汽压、风速为协变量,建立各站点逐日降水量的基于自组织映射神经网络(Self-Organizing Maps,SOM)的统计预测模型;分析吉林省春夏期的主要天气模态,研究逐日降水和天气模态之间的关系,并基于此关系提出逐日降水量的蒙特卡罗模拟方法。结果表明:SOM对天气模态的分型质量较好,邻近天气模态的累积概率分布较相似,距离较远的天气模态累计概率分布差异较大。各天气模态下无降水的概率与日降水量区间宽度的相关系数为-0. 94,显著性水平小于0. 01。基于降水量累积概率分布,20种天气模态被划分成4类,并与降水易发程度和逐日降水量完全对应。在此基础上,对吉林省24个站点逐日降水量进行蒙特卡罗模拟,并进行预测性能分析。平均绝对误差(Mean Absolute Error,MAE)和均方根误差(Root Mean Square Error,RM SE)的中位数分别为3. 12 mm和6. 13 mm,SBrier和Ssig分别为0. 06和0. 51,站点的逐日降水量预测性能整体较好。MAE和RMSE分布呈现东南大西北小,去除降水自然变异差异的影响,所有站点的误差都较小; SBrier和Ssig没有明显的空间分布特征。
Based on the daily ground observations in meteorological stations of Jilin Province during April-July 1997—2015,taking temperature,air pressure,relative humidity,water vapor pressure and wind speed as covariates,this paper established a statistical prediction model of daily precipitation based on self-organizing maps( SOM). This paper studied major synoptic patterns in Jilin Province and the relationship between daily precipitation and the patterns,and based on this relationship,proposed a Monte Carlo simulation method for daily precipitation. Results demonstrate that SOMhas high classification quality of synoptic patterns,and the accumulative probability distributions of adjacent synoptic patterns are similar,while those of synoptic patterns far away are quite different.The correlation coefficient between the probability of no precipitation and the corresponding width of daily precipitation interval in the synoptic patterns is-0. 94,and the significance level is less than 0. 01.According to the accumulative probability distribution of precipitation,20 types of synoptic patterns are divided into four categories,which match the occurrence rate of precipitation and the daily precipitation. On this basis,this paper carried out Monte Carlo simulation of daily precipitation in 24 stations of Jilin Province,and analyzed the forecast performance.The median values of MAE( mean absolute error),RMSE( root mean square error),SBrierand Ssigare 3. 12 mm,6. 13 mm,0. 06 and 0. 51,respectively,which indicates that the method has a good forecast performance in general.The distribution of MAE and RMSE is large in the southeast and small in the northwest,and all stations have smaller errors after removing the effect of the natural fluctuation of precipitation.SBrierand Ssighave no obvious spatial distribution characteristics.
Based on the daily ground observations in meteorological stations of Jilin Province during April-July 1997—2015,taking temperature,air pressure,relative humidity,water vapor pressure and wind speed as covariates,this paper established a statistical prediction model of daily precipitation based on self-organizing maps( SOM). This paper studied major synoptic patterns in Jilin Province and the relationship between daily precipitation and the patterns,and based on this relationship,proposed a Monte Carlo simulation method for daily precipitation. Results demonstrate that SOMhas high classification quality of synoptic patterns,and the accumulative probability distributions of adjacent synoptic patterns are similar,while those of synoptic patterns far away are quite different.The correlation coefficient between the probability of no precipitation and the corresponding width of daily precipitation interval in the synoptic patterns is-0. 94,and the significance level is less than 0. 01.According to the accumulative probability distribution of precipitation,20 types of synoptic patterns are divided into four categories,which match the occurrence rate of precipitation and the daily precipitation. On this basis,this paper carried out Monte Carlo simulation of daily precipitation in 24 stations of Jilin Province,and analyzed the forecast performance.The median values of MAE( mean absolute error),RMSE( root mean square error),SBrierand Ssigare 3. 12 mm,6. 13 mm,0. 06 and 0. 51,respectively,which indicates that the method has a good forecast performance in general.The distribution of MAE and RMSE is large in the southeast and small in the northwest,and all stations have smaller errors after removing the effect of the natural fluctuation of precipitation.SBrierand Ssighave no obvious spatial distribution characteristics.
引文
Akande A,Costa A C,Mateu J,et al.,2017.Geospatial analysis of extreme weather events in Nigeria(1985-2015)using self-organizing maps[J].Advances in Meteorology(2):1-11.
Alexander L V,Uotial P,Nicholls N,et al.,2010.A newdaily pressure dataset for Australia and its application to the assessment of changes in synoptic patterns during the last century[J].J Climate,23(5):1111-1126.
Alvarez-Guerra M,Gonzlez-Pi1uela C,Andrés A,et al.,2008.Assessment of self-organizing map artificial neural networks for the classification of sediment quality[J].Environment International,34(6):785-789.
陈潇潇,石银,张楠楠,等,2016.包含外强迫因子的平潭旬降水量预测[J].气象与环境学报,39(2):99-102.Chen X X,Shi Y,Zhang N N,et al.,2016.Prediction of ten-days precipitation in Pingtan containing the effect of the external forcing factors[J].Meteorological and Environmental Sciences,39(2):99-102.(in Chinese).
初征,郭建平,赵俊芳,2017.东北地区未来气候变化对农业气候资源的影响[J].地理学报,72(5):1248-1249.Chu Z,Guo J P,Zhao J F,2017.Impacts of projected climate change on agricultural climate resources in Northeast China[J].Journal of Geographical Sciences,72(5):1248-1249.(in Chinese).
Fu G B,Liu Z F,Charles S P,et al.,2013.A score-based method for assessing the performance of GCMs:a case study of southeastern Australia[J].JGeophys Res Atmos,118(10):4154-4167.
顾伟宗,张莉,伯忠凯,等,2018.不同时间尺度上黄淮地区夏季降水异常成因及预测研究[J].大气科学学报,41(3):388-394.Gu W Z,Zhang L,Bo Z K,et al.,2018.Study on the causes and prediction of summer precipitation in the Huang-huai area on different time scales[J].Trans Atmos Sci,41(3):388-394.(in Chinese).
Hewitson B C,Crane R G,2006.Consensus between GCMclimate change projections with empirical downscaling:precipitation downscaling over South Africa[J].Int J Climatol,26(10):1315-1337.
Hsu K C,Li S T,2010.Clustering spatial-temporal precipitation data using wavelet transform and self-organizing map neural network[J].Advances in Water Resources,33(2):190-200.
黄瑜,秦正坤,2017.多模式热带西太平洋夏季降水可预测性比较分析[J].大气科学学报,40(4):508-518.Huang Y,Qin Z K,2017.Comparative evaluation of summer rainfall predictability based on 10-models over the tropical western Pacific[J].Trans Atmos Sci,40(4):508-518.(in Chinese).
Kohonen T,1982.Self-organized formation of topologically correct feature maps[J].Biological Cybernetics,43(1):59-69.
黎玥君,郭品文,2017.基于BP神经网络的浙北夏季降尺度降水预报方法的应用[J].大气科学学报,40(3):425-432.Li Y J,Guo P W,2017.Application of downscaling forecast for the North of Zhejiang precipitation in summer based on the BP neural network model[J].Trans Atmos Sci,40(3):425-432.(in Chinese).
刘婷婷,陈海山,蒋薇,等,2016.基于土壤湿度和年际增量方法的我国夏季降水预测试验[J].大气科学,40(3):591-603.Liu T T,Chen H S,Jiang W,et al.,2016.Summer precipitation prediction in China using soil moisture and the year-to-year increment approach[J].Chin J Atmos Sci,40(3):591-603.(in Chinese).
马建勇,许吟隆,潘婕,2012.东北地区农业气象灾害的趋势变化及其对粮食产量的影响[J].中国农业气象,33(2):283-288.Ma J Y,Xu Y L,Pan J,2012.Analysis of agro-meteorological disasters trendency variation and the impacts on grain yield over northeast China[J].Chinese Journal of Agrometeorology,33(2):283-288.(in Chinese).
Ma S,Zhou T,Stone D A,et al.,2017.Detectable anthropogenic shift toward heavy precipitation over eastern China[J].J Climate,30(4):1381-1382.
孟翊星,2017.近50年新疆地区降水变化及天气分型[D].南京:南京大学:5-46.Meng Y X,2017.Change of precipitation and synoptic weather typing over Xinjiang in the last 50 years[D].Nanjing:Nanjing University:5-46.(in Chinese).
Mioduszewski J R,Rennermalm A K,Hammann A,2016.Atmospheric drivers of greenland surface melt revealed by self organizing maps[J].J Geophys Res Atmos,121(10):2-31.
Ning L,Mann ME,Crane R,et al.,2012.Probabilistic projections of climate change for the mid-Atlantic region of the United States:validation of precipitation downscaling during the historical Era[J].J Climate,25(2):509-526.
Perkins S E,Pitman A J,Holbrook N J,et al.,2007.Evaluation of the AR4 climate models’simulated daily maximum temperature,minimum temperature,and precipitation over Australia using probability density functions[J].J Climate,20(17):4356-4376.
Reusch D B,Alley R B,Hewitson B C,2007.North Atlantic climate variability from a self-organizing map perspective[J].J Geophys Res Atmos,112(D2):112-119.
沈伟,袁慧玲,陈曦,等,2017.江苏暖季短时强降水的时空不均匀特征分析[J].大气科学学报,40(4):453-462.Shen W,Yuan H L,Chen X,et al.,2017.Temporal and spatial heterogeneity of warm-season short-time heavy rainfall in Jiangsu Province[J].Trans Atmos Sci,40(4):453-462.(in Chinese).
Sun Q,Miao C,Duan Q,et al.,2017.Changes in the spatial heterogeneity and annual distribution of observed precipitation across China[J].J Climate,30(23):9399-9400.
Uysal G,Ensoy A,Orman A A,2016.Improving daily streamflowforecasts in mountainous Upper Euphrates basin by multi-layer perceptron model with satellite snowproducts[J].Journal of Hydrology,543:630-650.
王姝苏,智协飞,俞剑蔚,等,2018.降水的统计降尺度预报及其空间相关性和时间连续性重建[J].大气科学学报,41(1):36-45.Wang S S,Zhi X F,Yu J W,et al.,2018.Statistical downscaling forecast and reconstruction of spatial and temporal correlation of the precipitation[J].Trans Atmos Sci,41(1):36-45.(in Chinese).
韦庆,卢文喜,田竹君,2004.运用蒙特卡罗方法预报年降水量研究[J].干旱区资源与环境,18(4):144-146.Wei Q,Lu W X,Tian Z J,2004.Application of Monte-Carlo to annual precipitation forecast[J].Journal of Arid Land Resources and Environment,18(4):144-146.(in Chinese).
Wu X H,Niu S J,Jin D Z,et al.,2015.Influence of natural rainfall variability on the evaluation of artificial precipitation enhancement[J].Science China:Earth Sciences,58(6):906-914.
Yin C H,Li Y P,Ye W,et al.,2011.Statistical downscaling of regional daily precipitation over southeast Australia based on self-organizing maps[J].Theoretical and Applied Climatology,105:12-24.
于雪莹,2017.广州地区天气分型和降水预估[D].南京:南京大学:3-56.Yu X Y,2017.Synoptic weather typing and precipitation projection of Guangzhou area[D].Nanjing:Nanjing University:3-56.(in Chinese).
张舰齐,王丽琼,左瑞亭,等,2017.基于幅频分离的气候时间序列预测试验[J].大气科学,41(3):501-513.Zhang J Q,Wang L Q,Zuo R T,et al.,2017.A predicting test on climatic time series based on amplitude-frequency separation[J].Chin J Atmos Sci,41(3):501-514.(in Chinese).
赵中军,刘善亮,游大鸣,等,2015.偏最小二乘回归模型在辽宁汛期降水预测中的应用[J].干旱气象,33(6):1038-1044.Zhao Z J,Liu S L,You D M,et al.,2015.Application of partial least square regression on precipitation prediction during the flood season in Liaoning Province[J].Journal of Arid Meteorology,33(6):1038-1044.(in Chinese).
周璞,江志红,2016.自组织映射神经网络(SOM)降尺度方法对江淮流域逐日降水量的模拟评估[J].气候与环境研究,21(5):513-523.Zhou P,Jiang Z H,2016.Simulation and evaluation of statistical downscaling of regional daily precipitation over Yangtze-Huaihe River basin based on self-organizing maps[J].Climatic and Environmental Research,21(5):513-523.(in Chinese).
朱灵子,2015.基于统计模型的汛期降水预测研究---以义乌市为例[D].杭州:浙江师范大学:27-31.Zhu L Z,2015.Study of flood season precipitation prediction based on statistical model-taking Yiwu city as an examlpe[D].Hangzhou:Zhejiang Normal University:27-31.
Alexander L V,Uotial P,Nicholls N,et al.,2010.A newdaily pressure dataset for Australia and its application to the assessment of changes in synoptic patterns during the last century[J].J Climate,23(5):1111-1126.
Alvarez-Guerra M,Gonzlez-Pi1uela C,Andrés A,et al.,2008.Assessment of self-organizing map artificial neural networks for the classification of sediment quality[J].Environment International,34(6):785-789.
陈潇潇,石银,张楠楠,等,2016.包含外强迫因子的平潭旬降水量预测[J].气象与环境学报,39(2):99-102.Chen X X,Shi Y,Zhang N N,et al.,2016.Prediction of ten-days precipitation in Pingtan containing the effect of the external forcing factors[J].Meteorological and Environmental Sciences,39(2):99-102.(in Chinese).
初征,郭建平,赵俊芳,2017.东北地区未来气候变化对农业气候资源的影响[J].地理学报,72(5):1248-1249.Chu Z,Guo J P,Zhao J F,2017.Impacts of projected climate change on agricultural climate resources in Northeast China[J].Journal of Geographical Sciences,72(5):1248-1249.(in Chinese).
Fu G B,Liu Z F,Charles S P,et al.,2013.A score-based method for assessing the performance of GCMs:a case study of southeastern Australia[J].JGeophys Res Atmos,118(10):4154-4167.
顾伟宗,张莉,伯忠凯,等,2018.不同时间尺度上黄淮地区夏季降水异常成因及预测研究[J].大气科学学报,41(3):388-394.Gu W Z,Zhang L,Bo Z K,et al.,2018.Study on the causes and prediction of summer precipitation in the Huang-huai area on different time scales[J].Trans Atmos Sci,41(3):388-394.(in Chinese).
Hewitson B C,Crane R G,2006.Consensus between GCMclimate change projections with empirical downscaling:precipitation downscaling over South Africa[J].Int J Climatol,26(10):1315-1337.
Hsu K C,Li S T,2010.Clustering spatial-temporal precipitation data using wavelet transform and self-organizing map neural network[J].Advances in Water Resources,33(2):190-200.
黄瑜,秦正坤,2017.多模式热带西太平洋夏季降水可预测性比较分析[J].大气科学学报,40(4):508-518.Huang Y,Qin Z K,2017.Comparative evaluation of summer rainfall predictability based on 10-models over the tropical western Pacific[J].Trans Atmos Sci,40(4):508-518.(in Chinese).
Kohonen T,1982.Self-organized formation of topologically correct feature maps[J].Biological Cybernetics,43(1):59-69.
黎玥君,郭品文,2017.基于BP神经网络的浙北夏季降尺度降水预报方法的应用[J].大气科学学报,40(3):425-432.Li Y J,Guo P W,2017.Application of downscaling forecast for the North of Zhejiang precipitation in summer based on the BP neural network model[J].Trans Atmos Sci,40(3):425-432.(in Chinese).
刘婷婷,陈海山,蒋薇,等,2016.基于土壤湿度和年际增量方法的我国夏季降水预测试验[J].大气科学,40(3):591-603.Liu T T,Chen H S,Jiang W,et al.,2016.Summer precipitation prediction in China using soil moisture and the year-to-year increment approach[J].Chin J Atmos Sci,40(3):591-603.(in Chinese).
马建勇,许吟隆,潘婕,2012.东北地区农业气象灾害的趋势变化及其对粮食产量的影响[J].中国农业气象,33(2):283-288.Ma J Y,Xu Y L,Pan J,2012.Analysis of agro-meteorological disasters trendency variation and the impacts on grain yield over northeast China[J].Chinese Journal of Agrometeorology,33(2):283-288.(in Chinese).
Ma S,Zhou T,Stone D A,et al.,2017.Detectable anthropogenic shift toward heavy precipitation over eastern China[J].J Climate,30(4):1381-1382.
孟翊星,2017.近50年新疆地区降水变化及天气分型[D].南京:南京大学:5-46.Meng Y X,2017.Change of precipitation and synoptic weather typing over Xinjiang in the last 50 years[D].Nanjing:Nanjing University:5-46.(in Chinese).
Mioduszewski J R,Rennermalm A K,Hammann A,2016.Atmospheric drivers of greenland surface melt revealed by self organizing maps[J].J Geophys Res Atmos,121(10):2-31.
Ning L,Mann ME,Crane R,et al.,2012.Probabilistic projections of climate change for the mid-Atlantic region of the United States:validation of precipitation downscaling during the historical Era[J].J Climate,25(2):509-526.
Perkins S E,Pitman A J,Holbrook N J,et al.,2007.Evaluation of the AR4 climate models’simulated daily maximum temperature,minimum temperature,and precipitation over Australia using probability density functions[J].J Climate,20(17):4356-4376.
Reusch D B,Alley R B,Hewitson B C,2007.North Atlantic climate variability from a self-organizing map perspective[J].J Geophys Res Atmos,112(D2):112-119.
沈伟,袁慧玲,陈曦,等,2017.江苏暖季短时强降水的时空不均匀特征分析[J].大气科学学报,40(4):453-462.Shen W,Yuan H L,Chen X,et al.,2017.Temporal and spatial heterogeneity of warm-season short-time heavy rainfall in Jiangsu Province[J].Trans Atmos Sci,40(4):453-462.(in Chinese).
Sun Q,Miao C,Duan Q,et al.,2017.Changes in the spatial heterogeneity and annual distribution of observed precipitation across China[J].J Climate,30(23):9399-9400.
Uysal G,Ensoy A,Orman A A,2016.Improving daily streamflowforecasts in mountainous Upper Euphrates basin by multi-layer perceptron model with satellite snowproducts[J].Journal of Hydrology,543:630-650.
王姝苏,智协飞,俞剑蔚,等,2018.降水的统计降尺度预报及其空间相关性和时间连续性重建[J].大气科学学报,41(1):36-45.Wang S S,Zhi X F,Yu J W,et al.,2018.Statistical downscaling forecast and reconstruction of spatial and temporal correlation of the precipitation[J].Trans Atmos Sci,41(1):36-45.(in Chinese).
韦庆,卢文喜,田竹君,2004.运用蒙特卡罗方法预报年降水量研究[J].干旱区资源与环境,18(4):144-146.Wei Q,Lu W X,Tian Z J,2004.Application of Monte-Carlo to annual precipitation forecast[J].Journal of Arid Land Resources and Environment,18(4):144-146.(in Chinese).
Wu X H,Niu S J,Jin D Z,et al.,2015.Influence of natural rainfall variability on the evaluation of artificial precipitation enhancement[J].Science China:Earth Sciences,58(6):906-914.
Yin C H,Li Y P,Ye W,et al.,2011.Statistical downscaling of regional daily precipitation over southeast Australia based on self-organizing maps[J].Theoretical and Applied Climatology,105:12-24.
于雪莹,2017.广州地区天气分型和降水预估[D].南京:南京大学:3-56.Yu X Y,2017.Synoptic weather typing and precipitation projection of Guangzhou area[D].Nanjing:Nanjing University:3-56.(in Chinese).
张舰齐,王丽琼,左瑞亭,等,2017.基于幅频分离的气候时间序列预测试验[J].大气科学,41(3):501-513.Zhang J Q,Wang L Q,Zuo R T,et al.,2017.A predicting test on climatic time series based on amplitude-frequency separation[J].Chin J Atmos Sci,41(3):501-514.(in Chinese).
赵中军,刘善亮,游大鸣,等,2015.偏最小二乘回归模型在辽宁汛期降水预测中的应用[J].干旱气象,33(6):1038-1044.Zhao Z J,Liu S L,You D M,et al.,2015.Application of partial least square regression on precipitation prediction during the flood season in Liaoning Province[J].Journal of Arid Meteorology,33(6):1038-1044.(in Chinese).
周璞,江志红,2016.自组织映射神经网络(SOM)降尺度方法对江淮流域逐日降水量的模拟评估[J].气候与环境研究,21(5):513-523.Zhou P,Jiang Z H,2016.Simulation and evaluation of statistical downscaling of regional daily precipitation over Yangtze-Huaihe River basin based on self-organizing maps[J].Climatic and Environmental Research,21(5):513-523.(in Chinese).
朱灵子,2015.基于统计模型的汛期降水预测研究---以义乌市为例[D].杭州:浙江师范大学:27-31.Zhu L Z,2015.Study of flood season precipitation prediction based on statistical model-taking Yiwu city as an examlpe[D].Hangzhou:Zhejiang Normal University:27-31.