Abstract

The CERES project utilizes geostationary-derived broadband measurements to infer the regional diurnal flux in between CERES instantaneous measurements to estimate the daily-averaged flux. The geostationary (GEO) imager radiances must be calibrated to the same reference to ensure spatial and temporal consistency of cloud properties and fluxes across the contiguous GEO domains. In order to place all of the GEO visible imagers on the same radiometric scale, the CERES project inter-calibrates the GEO imagers with Aqua-MODIS using multiple independent approaches. The primary inter-calibration approach relies on coincident, ray-matched GEO and MODIS radiance pairs over all-sky tropical ocean scenes (ATO). Another ray-matching approach was recently developed to take advantage of the visible spectral uniformity and near-Lambertian reflectance of deep convective clouds (DCC). The success of the DCC raymatching (DCC-RM) approach has been demonstrated by comparing the calibration with the ATO ray-matching (ATORM) approach for the 0.65-μm GEO and MODIS bands. Now that many of the recently launched GEO imagers have multiple reflective solar band channels, the DCCRM algorithm is being modified to inter-calibrate those channels as well, especially for the SWIR bands. The spectral uniformity of the DCC over the SWIR bands is not uniform, given that the ice particle absorption is a function of wavelength. New Spectral Band Adjustment Factor (SBAF) strategies will need to be developed. DCC-RM is also wellsuited to inter-calibrate historical near-broadband visible GEO imagers. DCC are spectrally flat across the visible spectrum, which reduces the SBAF uncertainty between two ray-matched sensors. Applying the DCC-RM technique on historical GEO imagers is challenging due to the coarser pixel and temporal resolution of the ISCCP B1U formatted dataset. The ATO-RM and DCC-RM calibration methods were applied to multiple visible bands on Himawari-8 using MODIS as the calibration reference. The Aqua-MODIS and Himawari-8 calibration difference was less than 0.4% for wavelengths less than 1 µm and for the Terra-MODIS 0.65-μm channel. Other channel combinations would need further examination to obtain consistent ATO and DCC gain results. The ATO-RM and DCC-RM calibration methods were also applied to GOES-8 in the ISCCP B1U format with NOAA-14 AVHRR as the calibration reference. The GOES-8 ATO and DCC calibration gain difference was within 0.15%. The agreement between ATO and DCC gains provides confidence in both methods.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call