IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 51, NO. 1, JANUARY 2013 671 Diurnal and Scan Angle Variations in the Calibration of GOES Imager Infrared Channels Fangfang Yu, Member, IEEE, Xiangqian Wu, Member, IEEE, M. K. Rama Varma Raja, Yaping Li, Member, IEEE, Likun Wang, and Mitchell Goldberg Abstract—The current Geostationary Operational Environ- mental Satellite (GOES) Imager infrared (IR) channels experience a midnight effect that can result in erroneous instrument re- sponsivity around satellite midnight. An empirical method named the Midnight Blackbody Calibration Correction (MBCC) was developed and implemented in the GOES Imager IR operational calibration, aiming to correct the midnight calibration errors. The main objective of this study is to evaluate the MBCC performance for the GOES-11/-12 Imager IR channels by examining the diurnal variation of the mean brightness temperature (Tb) bias with respect to reference instruments. Two well-calibrated hyperspec- tral radiometers on low Earth orbits (LEOs), the Atmospheric Infrared Sounder on the Aqua satellite and the Infrared Atmo- spheric Sounding Interferometer (IASI) on the Metop-A satellite, are used as the reference instruments in this study. However, as the timing of the collocated geostationary–LEO intercalibration data is related to the GOES scan angle, it is then necessary to assess the GOES scan angle calibration variations, which becomes the sec- ond objective of this study. Our results show that the applications and performance of the MBCC method varies greatly between the different channels and different times. While it is usually applied with high frequency for about 8 h around satellite midnight for the short-wave channels (Ch2), it may only be intensively used right after satellite midnight or even barely used for the other IR channels. The MBCC method, if applied with high frequency, can reduce the mean day/night calibration difference to less than 0.15 K in almost all the GOES IR channels studied in this paper Manuscript received July 10, 2011; revised December 7, 2011 and March 3, 2012; accepted April 7, 2012. Date of publication June 11, 2012; date of current version December 19, 2012. This work was supported by the Center for Satellite Applications and Research, National Environmental Satellite, Data, and Information Service, National Oceanic and Atmospheric Administration (NOAA), through the Calibration and Validation project. The contents of this paper are solely the opinions of the authors and do not constitute a statement of policy, decision, or position on behalf of NOAA or the U.S. government. F. Yu is with Earth Resources Technology, Inc., Laurel, MD 20707 USA (e-mail: Fangfang.Yu@noaa.gov). X. Wu and M. Goldberg are with the Center for Satellite Applications and Research, National Environmental Satellite, Data, and Information Service, National Oceanic and Atmospheric Administration, Camp Springs, MD 20746 USA (e-mail: xiangqian.wu@noaa.gov; mitch.goldberg@noaa.gov). M. K. Rama Varma Raja was with I.M. Systems Group, Inc., Rockville, MD 20852 USA. He is now with Science Systems and Applications, Inc., Lanham, MD 20706 (email: rama.mundakkara@ssaihq.com). Y. Li is with I.M. Systems Group, Inc., Rockville, MD 20852 USA (e-mail: Yaping.li@noaa.gov). L. Wang was with Dell, Round Rock, TX 78682-7000 USA. He is now with University of Maryland, College Park, MD 20742 USA (email: Likun. Wang@noaa.gov). Digital Object Identifier 10.1109/TGRS.2012.2197627 except for Ch4 (10.7 μm). The uncertainty of the nighttime GOES and IASI Tb difference for different scan angles is less than 0.1 K in each IR channel, indicating that there is no apparent systematic variation with the scan angle, and therefore, the estimated diurnal cycles of GOES Imager calibration is not prone to the systematic effects due to scan angle. Index Terms—Diurnal calibration variation, GEO-LEO inter– calibration, Geostationary Operational Environmental Satellite (GOES) Imager, GSICS, infrared channels, radiometric calibra- tion, scan angle variation. I. I NTRODUCTION T HE Geostationary Operational Environmental Satellite (GOES) is a series of three-axis body-stabilized geosta- tionary (GEO) satellites operated by the National Oceanic and Atmospheric Administration (NOAA) of the U.S. Usually, two GOES satellites, located at 75 ◦ W (GOES-East) and 135 ◦ W (GOES-West) about 35800 km above the equator, are operated simultaneously to provide continuous streams of satellite data for weather monitoring and forecasting operations of the U.S. and its neighboring environments. The satellite radiances and the resulting quantitative meteorological products, such as sur- face temperature, winds, and others, are also important inputs to the numerical weather prediction systems and global climate change studies. Compared to the instruments onboard spin-stabilized satel- lites which spend more than 90% of their time scanning space (SP), the GOES instruments can continuously observe Earth targets except for routine brief periods to scan SP and the onboard blackbody (BB) for the radiometric calibration of infrared (IR) data and other calibration purposes [1]. However, several calibration issues, such as the midnight calibration anomaly [2] and the angle-dependent variation of the scan- mirror emissivity [1], [3], were identified for the instruments onboard GOES and other three-axis stabilized satellite in- struments. Around satellite midnight, the IR data experience erroneous calibration slopes (responsivity) derived from the standard IR calibration algorithm. This midnight slope error was demonstrated to cause up to an approximate 1 K brightness temperature (Tb) depression in the Imager short-wave channel (3.8 μm) from the scene over the Gulf of Mexico, where diurnal variation of ocean temperature should not exceed a few tenths of a degree [2]. It was also found that the erroneous midnight calibration slopes of the long-wave channels can result in a systematic bias of over 0.6 K of sea surface temperature when calculated using the split window technique [4]. To compensate for the satellite midnight calibration anomaly, an 0196-2892/$31.00 © 2012 IEEE