Performance studies of CERES sensors on Earth Science Enterprise (ESE) Terra mission using onboard calibrations and other validation methods

Abstract

Clouds and the Earth's Radiant Energy system (CERES) sensors provide accurate measurements for the long-term monitoring of the Earth's radiation budget components such as reflected shortwave and emitted longwave radiances. CERES instruments has three scanning thermistor bolometers that measure broadband radiances in the shortwave (0.3 to 5.0 micrometer), total (0.3 to >100 micrometer) and in 8 - 12 micrometer water vapor window regions. Two of the CERES instruments (Flight Models 1 and 2(FM1&2)) are part of ESE Terra mission and has been successfully making Earth radiance measurements for the past two years. The CERES sensors are calibrated in flight using the on-board blackbody sources and a tungsten lamp known as internal calibration module (ICM) as well as a solar diffuser plate known as the Mirror Attenuator Mosaic (MAM). The ICM calibrations is used to determine the sensor measurement precisions during the ground, ground to orbit, and the on-orbit phases of the sensor calibrations. The MAM calibrations define on-orbit shifts or drifts in shortwave and total sensor responses. The traditional validation studies conducted to understand the stability of the sensors' performance include the analysis of Tropical Mean (TM) value using nadir tropical ocean measurements and three channel intercomparison between sensors of the same instrument. With two CERES instruments on the same platform, an additional study utilizing direct comparison of similar sensor measurements viewing the same geolocation is also conducted. The ICM results have shown that total sensors on both instruments have shown a variation of 0.25 and 0.5 percent respectively, whereas the shortwave sensors show a minimal change of 0.2 percent each. With high variability during the initial year, the MAM results have stabilised within the 0.5 percent precision range in the second year. The validation studies have shed additional light into the behaviour of total sensors in the shortwave and longwave spectral regions. The TM longwave day night difference (DN) derived from longwave sensors of each CERES instrument indicate that the FM1 sensors have varied about 0.25 watts m^-2sr^-1, whereas the total sensor in FM2 has a gradual rise of 1.4 watts m^-2sr^-1 in a 30 month period. This paper discusses briefly the contribution of each calibration and validation study in understanding the CERES sensors' behavior and the results from both Terra instruments' sensors. It also discusses how the various analyses are put together in understanding the rise seen in the total sensor of FM2 instrument.