The goal of the Clouds and the Earth's Radiant Energy System (CERES) project is to provide a long-term record of radiation budget at the top-of-atmosphere (TOA), within the atmosphere, and at the surface with consistent cloud and aerosol properties at climate accuracy. To date, five CERES instruments (PFM, FM1-FM4) have flown on three different spacecraft: TRMM, EOS-Terra and EOS-Aqua. NASA and NOAA have agreed to fly the final existing CERES Flight Model (FM-5) on the NPP spacecraft for launch in 2011 and to procure an additional CERES Sensor with modest performance upgrades for flight on the first NASA/NOAA JPSS spacecraft in 2016, followed by a new CERES follow-on sensor for flight in 2020 on the second JPSS spacecraft.

Each CERES instrument is a scanning broadband radiometer that measures filtered radiances in the reflected solar region (wavelengths between 0.3-5 μm), total (TOT) (wavelengths between 0.3-200 μm) and emitted thermal region (wavelengths between 8-12 μm) regions. A Rigorous pre-launch radiometric ground calibration is performed on each CERES sensor to ensure accuracy requirements of 1% and 0.5% (1-sigma) for SW and LW radiance observations respectively are met. Any ground to flight or in-flight changes in radiometric response are monitored using a protocol employing both onboard and vicarious calibration sources and experiments. Studies of FM-1 through FM-4 flight observations have shown that the SW response of space based broadband radiometers can change dramatically due to optical contamination in the operational environment. The changes are greatest for wavelengths below 700 nm, and are particularly difficult to monitor using in-flight tungsten calibration lamps that are devoid of output in this spectral region.

While science goals remain unchanged for the long-term ERB Climate Data Record, it is now understood that the task of achieving these goals is more difficult for two reasons. The first is an increased understanding that rigorous separation of natural variability from anthropogenic change on decadal time scales requires observations with higher accuracy and stability than originally envisioned. Secondly, future implementation scenarios involve less redundancy in flight hardware (1 vs. 2 orbits and operational sensors) resulting in higher risk of loss of continuity and reduced number of independent observations to characterize performance of individual sensors. Although the EOS CERES Climate Data Records realize a factor of 2 to 4 improvement in accuracy and stability over previous ERBE Records, future sensors will require an additional factor of 2 improvement to rigorously answer the science questions. Implementing modest improvements, defined through the CERES Science Team's 30-year operational history of the EOS CERES sensors, to onboard calibration hardware and pre-flight calibration and test programs will ensure meeting these requirements while reducing costs in re-processing scientific datasets.

The current effort summarizes the radiometric performance of, and lessons learned from, of the four CERES sensors which are currently flying on the EOS Terra and Aqua Spacecraft.