Abstract
The greenhouse gases observing satellite (GOSAT) was launched in 2009 to measure the global concentrations of atmospheric CO2 and CH4. GOSAT is equipped with two earth-observing instruments : the thermal and near-infrared sensor for carbon observation Fourier transform spectrometer (TANSO-FTS), and the cloud and aerosol imager (TANSO-CAI). The presence —even partial— of clouds in the instantaneous field-of-view (IFOV) of the FTS can lead to incorrect estimates of CO2 or CH4 concentrations. Thus, it is necessary that cloud-contaminated FTS data be identified and rejected. A cloud-screening algorithm (CLAUDIA : cloud and aerosol unbiased decision intellectual algorithm) was developed for the CAI to identify cloud-free FTS data (CLAUDIA-CAI). The result is publicly available as the “CAI L2 cloud flag” product. We evaluated the accuracy of the GOSAT CAI L2 cloud flag product by comparing readings from the product with visual inspections of the same CAI images in the Amazon. We found the accuracy of the algorithm to be approximately 80 %, and the accuracy of obtaining cloud-free FTS data in the Amazon to be approximately 30 %. Various tests were then performed to increase cloud-free FTS data for the GOSAT-2. One experiment narrowed the IFOV of the FTS to increase the frequency. We also evaluated how the accuracy of obtaining cloud-free FTS data changed with a narrower IFOV. The result showed that a narrower IFOV of FTS was effective in increasing not only the frequency but also the accuracy of obtaining cloud-free data. However, it could also potentially increase underestimation of CO2 concentrations by overlooking small clouds. We could not confirm the difference in the dry air column-averaged mixing ratio of CO2 in the FTS data, which were judged cloud-contamination by visual inspection of CAI L1B+ images in IFOV of FTS, depending on the clouds.
