FY-3C/VIRR西北太平洋区域海表温度精度评估  

Evaluation of Sea Surface Temperature from FY-3C/VIRR Observations in the Northwest Pacific

作　　者：杨航 王素娟[2] 刘铭坤 管磊[1,3] YANG Hang;WANG Su-Juan;LIU Ming-Kun;GUAN Lei(Department of Marine Technology, College of Information Science & Engineering, Ocean University of China, Qingdao 266100, China;National Meteorological Satellite Center, China Meteorological Administration, Beijing 100081, China;Laboratory for Regional Oceanography and Numerical Modeling, Pilot National Laboratory for Marine Science and Technology(Qingdao), Qingdao 266237, China)

机构地区：[1]中国海洋大学信息科学与工程学院海洋技术系,山东青岛266100 [2]中国气象局国家卫星气象中心,北京100081 [3]青岛海洋科学与技术试点国家实验室区域海洋动力学与数值模拟功能实验室,山东青岛266237

出　　处：《中国海洋大学学报（自然科学版）》2020年第12期151-159,共9页Periodical of Ocean University of China

基　　金：自然资源部“全球变化与海气相互作用”专项(GASI-02-PACIND-YGST2-03);全球变化研究国家重大科学研究计划项目(2015CB953901);国家自然科学基金项目(41376105);中国海洋大学研究生自主科研项目(201861008)资助。

摘　　要：本文对我国第二代极轨气象卫星风云三号C星(Fengyun-3C,FY-3C)上的可见光红外扫描辐射计(Visible and Infra-Red Radiometer,VIRR)测得的西北太平洋地区的海表温度数据精度进行了评估,该评估结果可为风云卫星遥感资料的使用提供参考。将FY-3C/VIRR Level-2(二级数据)海表温度数据与实测浮标海表温度数据进行比较,两者匹配的时间窗口为1 h,空间窗口为0.01°,分别对FY-3C/VIRR全样本和质量等级为优的海表温度数据进行评估。FY-3C/VIRR全样本海表温度数据白天平均偏差为-0.12℃(FY-3C/VIRR海表温度减去实测海表温度),标准偏差为0.79℃;夜晚平均偏差为0.03℃,标准偏差为0.76℃。FY-3C/VIRR最优样本海表温度数据白天平均偏差为-0.01℃,标准偏差为0.64℃;夜晚平均偏差为0.16℃,标准偏差为0.58℃。针对部分云漏检及温度梯度大区域卫星与实测海表温度空间不匹配的问题,对匹配点周围3×3网格的有效海表温度数据进行统计分析,统计结果表明FY-3C/VIRR海表温度数据在质量为优、且匹配点周围3×3网格数据的局部标准偏差在阈值为0.3℃时,白天匹配数据的平均偏差为0.03℃,标准偏差为0.51℃;夜晚匹配数据的平均偏差为0.23℃,标准偏差为0.55℃。结果表明FY-3C/VIRR测得的海表温度数据精度较高,可用于全球变化与海气相互作用等研究。In this paper,the sea surface temperature(SST)the visibleand infrared scanning radiometer(VIRR)onboard the Fengyun-3C(FY-3C)satellite is evaluated in the Northwest Pacific,and the results can be usedto analyze Fengyun remote sensing data.The FY-3C/VIRR SST data are compared with the buoy SST measurements.The collocations of FY-3C/VIRR and buoy SST data are generated with the spatial window of 0.01°and the temporal window of 1h.All valid FY-3C/VIRR and the optimal SST data are evaluated separately.The bias of all valid FY-3C/VIRR data is-0.12℃against buoy measurements(FY-3C/VIRR minus buoy)and the standard deviation is 0.79℃during the daytime.The bias of all valid FY-3C/VIRR data is 0.03℃and thestandard deviation is 0.76℃at night.The bias of the optimal FY-3C/VIRR data is-0.01℃and the standard deviation is 0.64℃during the daytime.The bias of the optimal FY-3C/VIRR data is 0.16℃and the standard deviation is 0.58℃at night.In order to reduce the influence of cloud contamination and the spatial discrepancy,the error analysisof SST difference between FY-3C/VIRR and buoy data is investigated when the 3×3 pixels around FY-3C/VIRR matching points are all valid and the local standard deviation in the 3×3 box is under certain threshold.The statistics show that the product accuracy isthe bestwhen the local standard deviation of the 3×3 data around the optimal FY-3C/VIRRmatching point is 0.3℃.The bias is 0.03℃,and the standard deviation is 0.51℃during the daytime.The bias is 0.23℃,and the standard deviation is 0.55℃at night.The results indicate that FY-3C SST has a high accuracy around 0.5℃.The FY-3C SST data can be uesd to study global change and air-sea interaction.

关 键 词：FY-3C/VIRR 浮标 海表温度 精度 质量控制 误差分析 

分 类 号：TP701[自动化与计算机技术—检测技术与自动化装置]