廖志宏,徐宾,张雷,师春香,周自江. 2020. 基于FY-3C MWRI的海表温度产品偏差订正方法研究[J]. 气象学报, (0):-, doi:10.11676/qxxb2020.051
基于FY-3C MWRI的海表温度产品偏差订正方法研究
投稿时间:2019-11-29  修订日期:2020-04-22
DOI:10.11676/qxxb2020.051
中文关键词:  偏差订正  海表温度  FY-3C MWRI  分段回归
英文关键词:Bias  Correction, Sea  Surface Temperature (SST), Fengyun-3C (FY-3C)MicroWave  Radiation Imager(MWRI), Picece-Wise  Regression(PWR)
基金项目:(41806213),国家重点研发计划项目(2018YFC1506600)。
作者单位E-mail
廖志宏 国家气象信息中心 liaozh@cma.gov.cn 
徐宾 国家气象信息中心 xubin@cma.gov.cn 
张雷  zhanglei@cma.gov.cn 
师春香  shicx@cma.gov.cn 
周自江  zhouzj@cma.gov.cn 
摘要点击次数: 143
全文下载次数: 86
中文摘要:
      本文提出一种针对FY-3C MWRI的海表温度(Sea Surface Temperature,SST)产品的分段回归(Piece-Wise Regression,PWR)偏差订正方法,该方法通过引进气候态SST数据,建立与关联实测SST相匹配的回归模型,并通过对模型中关联变量的误差分析,选择最优样本进行分段回归,以实现对SST数据的重新估计。通过对MWRI SST数据的偏差订正试验表明,采用PWR方法获得的订正结果无论在误差指标的空间分布还是时间序列上,都要明显优于采用传统概率密度函数(Probability Density Function,PDF)偏差订正方法的结果。其中,采用PWR方法订正后的SST产品误差标准差(SD)和均方根误差(RMSE)值从订正前的0.9℃~1.0℃,减小到0.6℃以下,而采用PDF方法获得相应的订正误差仅为0.8℃左右,订正效果得到了明显改善。
英文摘要:
      A Picece-Wise Regression(PWR)method was proposed for Fengyun-3C (FY-3C)MicroWave Radiation Imager(MWRI) products. This method developed a regression model that matches the associated in-situ Sea Surface Temperature (SST) with the daily climatology SST, and selcted the optimal matchups through the error analysis of the regressors in the model, then the SSTs were recalculated by using these optimal matchups for the Piece-Wise regression. Compared with the traditional probability density function (PDF) matching technique for bias correction, the PWR method can better remove the biases in the space-time domain, and the SDs and RMSEs were decreased from 0.9℃~1.0℃ to 0.6℃, which were much better than the results from the PDF method with the values of about 0.8℃.
查看全文   查看/发表评论  下载PDF阅读器
分享按钮