Publications

Doelling, David R.; Khlopenkov, Konstantin V.; Okuyama, Arata; Haney, Conor O.; Gopalan, Arun; Scarino, Benjamin R.; Nordeen, Michele; Bhatt, Rajendra; Avey, Lance (2015). MTSAT-1R Visible Imager Point Spread Correction Function, Part I: The Need for, Validation of, and Calibration With. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 53(3), 1513-1526.

Abstract
The multifunctional transport satellite (MTSAT)-1R imager was launched in 2005 and is operated by the Japan Meteorological Agency (JMA). A nonlinear behavior in the MTSAT-1R visible sensor response is observed when the instrument is intercalibrated with coincident moderate resolution imaging spectroradiometer (MODIS) ray-matched radiances. Analysis reveals that the nonlinear behavior is not a result of imager navigation, sensor spectral response difference, nor scan pattern. Examination of coincident MTSAT-1R and MTSAT-2 images reveals that MTSAT-1R dark ocean radiances are affected by neighboring bright clouds, whereas large regions of dark ocean radiances are not impacted. Although the IR and visible optical paths are shared, the MTSAT-1R brightness temperatures are not affected. A dust contaminant coating the mirror, which only affects certain wavelengths, may be one explanation. To address the nonlinearity, a pixel point spread function (PSF) correction algorithm is implemented, wherein most of the radiance contribution is from the pixel field of view itself, as well as including a small contribution from all pixels within a radii of several hundred kilometers. The application of the PSF-corrected similar to 80% of the affected pixel radiances. After application, a near linear response is observed between the coincident MTSAT-1R and Aqua-MODIS ray-matched radiances, and the intercept is now near the predicted space count of zero. The monthly calibration gain noise is reduced by one-third when compared with the non-PSF-corrected gains. The monthly gains are the most erratic during the first two years of operation, and the MTSAT-1R visible sensor is degrading at similar to 1.9% decade.

DOI:
10.1109/TGRS.2014.2344678

ISSN:
0196-2892