卫星的射出长波辐射OLR(Outgoing Long-wave Radiation)数据具有不同程度的误差。为满足业务和科研工作的需要,对国产卫星FY-3 A/VIRR的OLR产品与其他同类卫星产品进行一致性和差异性分析是非常必要的。采用风云三号A星(FY-3A)扫描辐射计(VIRR)的OLR日平均产品作为被检验数据,美国大气海洋局NOAA18卫星上搭载的甚高分辨率扫描辐射计(AVHRR)OLR日平均产品作为检验源数据,使用相关系数、平均偏差、均方根误差、相对误差等检验方法,对两种产品进行一致性和差异性分析。结果表明,两种OLR资料大部分相关系数较大,平均偏差、均方根误差和相对误差较小,个别资料相关系数较小,平均偏差、均方根误差和相对误差较大。如2010年4月23日、7月13日和10月13日,相关系数、平均偏差、均方根误差、相对误差分别为(0.63,7 Wm^-2,31 Wm^-2,0.12)、(0.5,-5 Wm^-2,45 Wm^-2,0.125)和(0.3,-200 Wm^-2,225 Wm^-2,0.85)。这些偏差主要发生在高山和海洋地区,并且暖季相对冷季偏差较大,可能是由于高山、海洋地区和暖季较强的对流活动造成两种资料的对比结果存在一定的差异;此外,由于FY-3A为上午星,NOAA18为下午星,过境时间存在一定的差异,也给二者的对比结果带来一定的差异。

The Outgoing Long-wave Radiation (OLR) derived from the FY-3A/VIRR and NOAA19/AVHRR satellites indicated the presence of some errors. To satisfy the needs of the operational and scientific studies, it was necessary to analyze the similarity and differences in the FY-3A/VIRR-OLR and the OLR derived from other satellites.Using the daily average OLR of FY-3A/VIRR as the test data, the daily average OLR of NOAA18/AVHRR as the test source data, and using the methods to determine the correlation coefficient, deviation, root mean square error (RMSE), and relative error, the similarity and differences of the daily average OLR of the satellites FY-3A/VIRR and NOAA18/AVHRR were analyzed.The results showed that the correlation coefficient was large and the deviation, RMSE and relative error were small for the majority of the two sets of data. However, the correlation coefficient was small, and the deviation, RMSE and relative error were large for a small proportion of the two sets of data. For example, on April 23, July 13, and October 13, 2010, the correlation coefficient was 0.63, 0.5, and 0.3, respectively; the deviation was 7 Wm^-2, -5 Wm^-2, and -200 Wm^-2, respectively; the RMSE was 31, 45, and 225 Wm^-2, respectively; and the relative error was 0.12, 0.125, and 0.85, respectively. The difference occurred mainly in the mountains and the tropical oceans, and the difference was larger in the warm season than in the cold season. The difference in the two sets of data may have also been caused by the stronger convective flow in the mountains and tropical oceans in the warm season. Because the FY-3A Satellite acquired the data in the morning and the NOAA18 Satellite acquired the data in the afternoon, there were different orbiting periods, which may have led to differences for the two sets of data.