水文集合数据同化系统参数对同化效应的影响研究
|
摘要
集合数据同化近年来被引入水文研究中,并得到了较为广泛的应用。而集合数和误差是该类算法同化时的重要变量,影响同化的效果和决定计算的消耗。以一个水文模型蒸散发数据同化系统(Evapotranspiration Data Assimilation System,EDAS)为例,借助均方根误差(Root Mean Square Error,RMSE)、皮尔逊相关系数(Pearson Correlation Coefficient,PCC)、平均绝对百分比误差(Mean Absolute Percentage Error,MAPE)等作为评价指标,利用移动步距法研究了集合数、误差的大小变化过程对数据同化结果的影响。研究发现,当集合数大于100时、模型误差方差和观测误差方差分别稳定在0. 05和0. 006时同化结果较为合理。研究成果可为集合同化的参数分析与率定提供借鉴。
Ensemble filtering algorithm is more and more widely used in remote sensing-hydrological model data assimilation in recent years. The size of the ensemble and the value of the errors for the algorithm are the important parameters affecting assimilation,which decides the assimilation effect and computational cost. An evapotranspiration data assimilation system(EDAS) was employed as a case to test the extent to which assimilation effects are affected by assimilation parameters. The assimilation system took the evapotranspiration observation by remote sensing, Distributed Time Variant Gain Model(DTVGM) as the model operator, and the Ensemble Kalman Filter(En KF) as the assimilation algorithm. By means of the Root Mean Square Error(RMSE), the Pearson Correlation Coefficient(PCC) and the Mean Absolute Percentage Error(MAPE) as the evaluation indexes, the influence of the size of the ensemble and the error values on the data assimilation results is studied by using the moving step method. It is found that when the ensemble size is greater than 100, the model error variance and observation error variance are stable at 0. 05 and 0. 006 respectively, the assimilation results are reasonable. The research can provide reference for parameter analysis and calibration of the ensemble data assimilation.
Ensemble filtering algorithm is more and more widely used in remote sensing-hydrological model data assimilation in recent years. The size of the ensemble and the value of the errors for the algorithm are the important parameters affecting assimilation,which decides the assimilation effect and computational cost. An evapotranspiration data assimilation system(EDAS) was employed as a case to test the extent to which assimilation effects are affected by assimilation parameters. The assimilation system took the evapotranspiration observation by remote sensing, Distributed Time Variant Gain Model(DTVGM) as the model operator, and the Ensemble Kalman Filter(En KF) as the assimilation algorithm. By means of the Root Mean Square Error(RMSE), the Pearson Correlation Coefficient(PCC) and the Mean Absolute Percentage Error(MAPE) as the evaluation indexes, the influence of the size of the ensemble and the error values on the data assimilation results is studied by using the moving step method. It is found that when the ensemble size is greater than 100, the model error variance and observation error variance are stable at 0. 05 and 0. 006 respectively, the assimilation results are reasonable. The research can provide reference for parameter analysis and calibration of the ensemble data assimilation.
引文
[1]孟现勇,师春香,刘时银,等. CMADS数据集及其在流域水文模型中的驱动作用——以黑河流域为例[J].人民珠江,2016,37(7):1-19.
[2]陈鹤,杨大文,刘钰,等.集合卡尔曼滤波数据同化方法改进土壤水分模拟效果[J].农业工程学报,2016,32(2):99-104.
[3]梁顺林,李新,谢先红.陆面过程观测、模拟与数据同化[M].北京:高等教育出版社,2013.
[4]LIU Y,WEERTS A H,CLARK M,et al. Toward advancing data assimilation in operational hydrologic forecasting and water resources management:current status,challenges,and emerging opportunities[J]. Hydrology&Earth System Sciences,2012,16(3):3863-3887.
[5]SUN L,SEIDOU O,NISTOR I,et al. Review of the Kalman-type hydrological data assimilation[J]. Hydrological Sciences Journal,2016,61(13):2348-2366.
[6]张苹苹,童菊秀,廖卫红.基于EnKF的新安江模型参数和变量同步估计方法[J].南水北调与水利科技,2018,16(2):51-58.
[7] CAMMALLERI C,CIRAOLO G. State and parameter update in a coupled energy/hydrologic balance model using ensemble Kalman filtering[J]. Journal of Hydrology,2012,416–417(2):171-181.
[8]EVENSEN G. Sequential data assimilation with a nonlinear quasigeostrophic model using Montel Carlo methods to forecast error statistics[J]. Journal of Geophysical Research, 1994, 99:10143-10162.
[9]黄春林,李新.土壤水分同化系统的敏感性试验研究[J].水科学进展,2006,17(4):457-465.
[10]PIPUNIC C,WALKER P,WESTERN A. Assimilation of remotely sensed data for improved latent and sensible heat flux prediction:A comparative synthetic study[J]. Remote Sensing of Environment,2008,112(4):1295-1305.
[11]杨娜.基于云参数干旱遥感监测模型与集合卡尔曼滤波的土壤湿度同化研究[D].武汉:武汉大学,2010.
[12]WU G,ZHENG X,ZHANG S,et al. Maximum likelihood estimation of inflation factors on error covariance matrices for ensemble Kalman filter assimilation[J]. 2012,138:263–273.
[13]占车生,尹剑,王会肖,等.基于双层模型的沙河流域蒸散发定量遥感估算[J].自然资源学报,2013,28(1):161-170.
[14]夏军,叶爱中,王蕊,等.跨流域调水的大尺度分布式水文模型研究与应用[J].南水北调与水利科技,2011,9(1):1-7,95.
[15]YIN J,ZHAN C,YE W. An experimental study on evapotranspiration data assimilation based on the hydrological model[J]. Water Resources Management,2016,30(14):5263-5279.
[16]DUAN Q,SOROOSHIAN S,GUPTA V. Effective and efficient global optimization for conceptual rainfall-runoff models[J]. Water Resources Research,1992,28(4):1015-1031.
[17]王纲胜.分布式时变增益水文模型理论与方法研究[D].北京:中国科学院研究生院,2005.
[18]王卫光,李进兴,魏建德,等.基于蒸散发数据同化的径流过程模拟[J].水科学进展,2018,29(2):159-168.
[19]陈鹤.基于遥感蒸散发的陆面过程同化方法研究[D].北京:清华大学,2013.
[20]ZHAN C S,SONG X M,XIA J,et al. An efficient integrated approach for global sensitivity analysis of hydrological model parameters[J]. Environmental Modelling&Software,2013,41(41):39-52.
[21]ZOU L,ZHAN C,XIA J,et al. Implementation of evapotranspiration data assimilation with catchment scale distributed hydrological model via an ensemble Kalman filter[J]. Journal of Hydrology,2017,549:685-702.
[2]陈鹤,杨大文,刘钰,等.集合卡尔曼滤波数据同化方法改进土壤水分模拟效果[J].农业工程学报,2016,32(2):99-104.
[3]梁顺林,李新,谢先红.陆面过程观测、模拟与数据同化[M].北京:高等教育出版社,2013.
[4]LIU Y,WEERTS A H,CLARK M,et al. Toward advancing data assimilation in operational hydrologic forecasting and water resources management:current status,challenges,and emerging opportunities[J]. Hydrology&Earth System Sciences,2012,16(3):3863-3887.
[5]SUN L,SEIDOU O,NISTOR I,et al. Review of the Kalman-type hydrological data assimilation[J]. Hydrological Sciences Journal,2016,61(13):2348-2366.
[6]张苹苹,童菊秀,廖卫红.基于EnKF的新安江模型参数和变量同步估计方法[J].南水北调与水利科技,2018,16(2):51-58.
[7] CAMMALLERI C,CIRAOLO G. State and parameter update in a coupled energy/hydrologic balance model using ensemble Kalman filtering[J]. Journal of Hydrology,2012,416–417(2):171-181.
[8]EVENSEN G. Sequential data assimilation with a nonlinear quasigeostrophic model using Montel Carlo methods to forecast error statistics[J]. Journal of Geophysical Research, 1994, 99:10143-10162.
[9]黄春林,李新.土壤水分同化系统的敏感性试验研究[J].水科学进展,2006,17(4):457-465.
[10]PIPUNIC C,WALKER P,WESTERN A. Assimilation of remotely sensed data for improved latent and sensible heat flux prediction:A comparative synthetic study[J]. Remote Sensing of Environment,2008,112(4):1295-1305.
[11]杨娜.基于云参数干旱遥感监测模型与集合卡尔曼滤波的土壤湿度同化研究[D].武汉:武汉大学,2010.
[12]WU G,ZHENG X,ZHANG S,et al. Maximum likelihood estimation of inflation factors on error covariance matrices for ensemble Kalman filter assimilation[J]. 2012,138:263–273.
[13]占车生,尹剑,王会肖,等.基于双层模型的沙河流域蒸散发定量遥感估算[J].自然资源学报,2013,28(1):161-170.
[14]夏军,叶爱中,王蕊,等.跨流域调水的大尺度分布式水文模型研究与应用[J].南水北调与水利科技,2011,9(1):1-7,95.
[15]YIN J,ZHAN C,YE W. An experimental study on evapotranspiration data assimilation based on the hydrological model[J]. Water Resources Management,2016,30(14):5263-5279.
[16]DUAN Q,SOROOSHIAN S,GUPTA V. Effective and efficient global optimization for conceptual rainfall-runoff models[J]. Water Resources Research,1992,28(4):1015-1031.
[17]王纲胜.分布式时变增益水文模型理论与方法研究[D].北京:中国科学院研究生院,2005.
[18]王卫光,李进兴,魏建德,等.基于蒸散发数据同化的径流过程模拟[J].水科学进展,2018,29(2):159-168.
[19]陈鹤.基于遥感蒸散发的陆面过程同化方法研究[D].北京:清华大学,2013.
[20]ZHAN C S,SONG X M,XIA J,et al. An efficient integrated approach for global sensitivity analysis of hydrological model parameters[J]. Environmental Modelling&Software,2013,41(41):39-52.
[21]ZOU L,ZHAN C,XIA J,et al. Implementation of evapotranspiration data assimilation with catchment scale distributed hydrological model via an ensemble Kalman filter[J]. Journal of Hydrology,2017,549:685-702.