Atmos. Meas. Tech., 10, 1403–1424, 2017 www.atmos-meas-tech.net/10/1403/2017/ doi:10.5194/amt-10-1403-2017 © Author(s) 2017. CC Attribution 3.0 License. Long-term assessment of the CALIPSO Imaging Infrared Radiometer (IIR) calibration and stability through simulated and observed comparisons with MODIS/Aqua and SEVIRI/Meteosat Anne Garnier1,2, Noëlle A. Scott3, Jacques Pelon4, Raymond Armante3, Laurent Crépeau3, Bruno Six5, and Nicolas Pascal6 1Science Systems and Applications, Inc., Hampton, VA 23666, USA 2NASA Langley Research Center, Hampton, VA 23681, USA 3Laboratoire de Météorologie Dynamique, Ecole Polytechnique–CNRS, 91128 Palaiseau, France 4Laboratoire Atmosphères, Milieux, Observations Spatiales, UPMC–UVSQ–CNRS, 75252 Paris, France 5Université Lille 1, AERIS/ICARE Data and Services Center, 59650 Lille, France 6Hygeos, AERIS/ICARE Data and Services Center, 59650 Lille, France Correspondence to: Anne Garnier ([email protected]) Received: 13 October 2016 – Discussion started: 14 November 2016 Revised: 6 March 2017 – Accepted: 21 March 2017 – Published: 13 April 2017 Abstract. The quality of the calibrated radiances of the and for all types of scenes in a wide range of brightness tem- medium-resolution Imaging Infrared Radiometer (IIR) on- peratures. The relative approach shows an excellent stability board the CALIPSO (Cloud-Aerosol Lidar and Infrared of IIR2–MODIS31 and IIR3–MODIS32 brightness temper- Pathfinder Satellite Observation) satellite is quantitatively ature differences (BTDs) since launch. A slight trend within evaluated from the beginning of the mission in June the IIR1–MODIS29 BTD, that equals −0.02 K yr−1 on av- 2006. Two complementary “relative” and “stand-alone” ap- erage over 9.5 years, is detected when using the relative ap- proaches are used, which are related to comparisons of mea- proach at all latitudes and all scene temperatures. For very sured brightness temperatures and to model-to-observations cold scene temperatures (190–200 K) in the tropics, each IIR comparisons, respectively. In both cases, IIR channels channel is warmer than its MODIS companion channel by 1 (8.65 µm), 2 (10.6 µm), and 3 (12.05 µm) are paired 1.6 K on average. For the stand-alone approach, clear sky with the Moderate Resolution Imaging Spectroradiometer measurements only are considered, which are directly com- (MODIS)/Aqua Collection 5 “companion” channels 29, 31, pared with simulations using 4A/OP and collocated ERA- and 32, respectively, as well as with the Spinning Enhanced Interim (ERA-I) reanalyses. The clear sky mask is derived Visible and Infrared Imager (SEVIRI)/Meteosat companion from collocated observations from IIR and the CALIPSO li- channels IR8.7, IR10.8, and IR12, respectively. These pairs dar. Simulations for clear sky pixels in the tropics reproduce were selected before launch to meet radiometric, geometric, the differences between IIR1 and MODIS29 within 0.02 K and space-time constraints. The prelaunch studies were based and between IIR2 and MODIS31 within 0.04 K, whereas on simulations and sensitivity studies using the 4A/OP ra- IIR3–MODIS32 is larger than simulated by 0.26 K. The diative transfer model and the more than 2300 atmospheres stand-alone approach indicates that the trend identified from of the climatological Thermodynamic Initial Guess Retrieval the relative approach originates from MODIS29, whereas no (TIGR) input dataset further sorted into five air mass types. trend (less than ±0.004 K yr−1) is identified for any of the Using data from over 9.5 years of on-orbit operation, and IIR channels. Finally, using the relative approach, a year- following the relative approach technique, collocated mea- by-year seasonal bias between nighttime and daytime IIR– surements of IIR and of its companion channels have been MODIS BTD was found at mid-latitude in the Northern compared at all latitudes over ocean, during day and night, Hemisphere. It is due to a nighttime IIR bias as determined Published by Copernicus Publications on behalf of the European Geosciences Union. 1404 A. Garnier et al.: Long-term assessment of the CALIPSO Imaging Infrared Radiometer by the stand-alone approach, which originates from a cali- MetOpB satellites in cooperation with CNES (e.g., Jouglet et bration drift during day-to-night transitions. The largest bias al., 2014) within the frame of its GSICS activities. is in June and July when IIR2 and IIR3 are warmer by 0.4 K The first approach, called the “relative” approach, is based on average, and IIR1 is warmer by 0.2 K. on a channel to channel intercomparison of radiances, fur- ther converted into equivalent brightness temperatures, with collocated measurements in companion channels of com- panion instruments under controlled conditions. The rela- tive approach, sometimes referred to as the “inter-channel” 1 Introduction or “inter-calibration” approach, was initially developed in a geostationary–low Earth orbit (GEO–LEO) combination for The Cloud-Aerosol Lidar and Infrared Pathfinder Satel- the calibration of Meteosat-1, based on space and time collo- lite Observation (CALIPSO) satellite (Winker et al., 2010), cations with TOVS on the NOAA TIROS-N series (Bériot et launched in April 2006, includes a payload of three instru- al., 1982). The second approach, called the “stand-alone” ap- ments, the Cloud-Aerosol Lidar with Orthogonal Polariza- proach, is based on comparisons between measured and sim- tion (CALIOP), a visible wide-field camera, and the Imag- ulated brightness temperatures for each companion channel ing Infrared Radiometer (IIR). The IIR was built in France of each companion instrument. The two approaches are com- by the Centre National d’Études Spatiales (CNES), the So- plementary: the inter-calibration approach studies the behav- ciété d’Études et de Réalisations Nucléaires (SODERN), and ior of one channel relative to its companion regardless of the Institut Pierre Simon Laplace (IPSL) (Corlay et al., 2000). underlying clear or cloudy scene and therefore allows for the It includes three spectral bands in the thermal infrared at- study of a wide range of brightness temperatures. The stand- mospheric window, at 8.65 µm (IIR1), 10.6 µm (IIR2), and alone approach screens each channel of each instrument, in- 12.05 µm (IIR3) with bandwidths of 0.85, 0.6, and 1 µm, re- dividually, for clear sky scenes and helps to identify which spectively. These three channels were chosen to optimize re- channel deviates from the other(s). In this study, the clear trievals of ice cloud properties in synergy with collocated sky mask is derived from collocated observations from the observations from the CALIOP lidar (Garnier et al., 2012, IIR and CALIOP. 2013). The IIR companion instruments chosen for this study It is now well recognized by the international community are the Moderate Resolution Imaging Spectroradiometer that to be fully useful for climate and meteorological applica- (MODIS) on-board Aqua and the Spinning Enhanced Visi- tions, satellite observations require quality control during the ble and Infrared Imager (SEVIRI) on-board the geostationary instrument’s lifetime. Indeed, any systematic error or spuri- second generation satellites Meteosat 8, 9, and 10. Both the ous trend not identified in the calibrated radiances may in- MODIS and SEVIRI instruments, in operation since 2002 duce artifacts in the retrieved variables. In the mid-1990s, and 2004, respectively, include medium-resolution spectral the NOAA/NASA Pathfinder Program and, later in 2005, the bands similar to IIR channels. Global Space-based Inter-Calibration System (GSICS) initi- An assessment of IIR radiances after 9.5 years of nearly ated international collaborative efforts to improve and har- continuous operation is presented, thereby updating the first monize the quality of observations from operational weather results published in Scott (2009). This paper presents a brief and environmental satellites in order to create climate data description of the IIR instrument in Sect. 2, with prelaunch records (e.g., Goldberg et al., 2011). A recent update on the studies for the selection of IIR companion instrument and GSICS vision is given in a 2015 World Meteorological Or- channels identified in Sect. 3, and the implementation of the ganization (WMO) report (GSICS, 2015). relative and stand-alone approaches described in Sect. 4. Re- In this paper, IIR observations since launch are monitored sults and findings from the relative and the stand-alone ap- and characterized using two complementary approaches, proaches are presented in Sect. 5 and Sect. 6, respectively, which are the heritage of the processing of numerous years and our assessment is summarized in the last section. of satellite radiances for the restitution of climate variables (Chédin et al., 1985, and papers following on similar top- ics referenced and regularly updated at http://ara.abct.lmd. 2 The IIR/CALIPSO instrument polytechnique.fr/index.php?page=publications). The moni- toring of observational and computational biases or trends The entire CALIPSO payload flies in a near-nadir, down- over long periods of time started with the NOAA/NASA looking configuration, with a 0.3◦ angle off-nadir in the for- TIROS Operational Vertical Sounder (TOVS) Pathfinder Pro- ward direction. The off-nadir angle was increased to 3◦ at the gram (Scott et al., 1999). Since then, it has been implemented end of November 2007 to reduce specular reflections from for more recent hyperspectral sounders, namely the Atmo- horizontally oriented ice crystals in CALIOP observations spheric Infrared Sounder (AIRS) on-board the Aqua satel- (Hu et al., 2009). The IIR instantaneous field of view is ±2.6◦ lite since 2003, and since 2007 for the Interféromètre Atmo- or 64 × 64 km on the ground, with 1 km size pixels. Thus, the sphérique de Sondage Infrarouge (IASI) on the MetOpA and viewing angles from nadir range from 0 to 3◦ until Novem- Atmos. Meas. Tech., 10, 1403–1424, 2017 www.atmos-meas-tech.net/10/1403/2017/ A.