HR: 0800h
AN: B51B-0203    [Abstracts]
TI: Fusion of Moderate and Imagery Resolution VIIRS LWIR Brightness Temperature to Produce Imagery Resolution Surface Temperature
AU: * Ip, J
EM: justin.ip@ngc.com
AF: Northrop Grumman Space Technology, One Space Park Drive, Redondo Beach, CA 90278 United States
AU: Hauss, B
EM: bruce.hauss@ngc.com
AF: Northrop Grumman Space Technology, One Space Park Drive, Redondo Beach, CA 90278 United States
AB: The Visible/Infrared Imager/Radiometer Suite (VIIRS) Sea Ice algorithm chain requires surface temperature (on sea ice and ocean surfaces) at 375 m imagery resolution to produce Environmental Data Records (EDR) at that resolution. The VIIRS Ice Surface Temperature (IST) and Sea Surface Temperature EDR are produced from 750 m moderate resolution data, and thus are not adequate for this purpose. To address the imagery resolution requirement, the VIIRS Surface Temperature Intermediate Product (ST IP) has been created to provide the needed surface temperature at imagery resolution. The VIIRS ST IP algorithm is based on the same 2-band split-window water vapor correction method as in the VIIRS IST moderate resolution algorithm. It utilizes brightness temperatures from both the VIIRS moderate resolution Long-wave Infrared (LWIR) channels at 10.76 æm (M15) and 12.01 æm (M16) and the VIIRS imagery resolution LWIR band at 11.45 æm (I5). The algorithm employs fusion of the moderate resolution M15 and M16 brightness temperatures with the imagery resolution I5 brightness temperature in the split window algorithm to produce surface temperatures for the imagery pixels. The algorithm is based on two key assumptions that (1) the brightness temperature in the LWIR imagery band (I5) can be written spectrally as a linear combination of the brightness temperatures in the moderate resolution LWIR bands (M15 and M16), and (2) the brightness temperature difference is primarily determined by atmospheric conditions that are stable on a spatial scale of a few km, which allows the approximation of the brightness temperature difference at imagery resolution as the difference in the nearest moderate resolution pixel. Performance based on a wide range of Moderate Resolution Imaging Spectroradiometer (MODIS) global sea ice proxy data sets has been calculated to be < 0.2 K in accuracy, precision, and uncertainty for sea ice surface temperature. It is also demonstrated that the temperature variations among the four so-called imagery pixels (at 1 km resolution) in the MODIS data are correctly retrieved and the VIIRS ST IP retrieval captures the MODIS temperature variations with the derived regression coefficients.
DE: 1855 Remote sensing (1640)
SC: Biogeosciences [B]
MN: Fall Meeting 2005