Trade study of substituting VIIRS M10 with aggregated I3 to enable addition of a water vapor channel

The U.S. National Weather Service currently assimilates into its numerical weather prediction models satellite observations from the aging MODIS instruments that track polar winds from motion of both clouds and atmospheric moisture. Next generation weather observations are provided by VIIRS instruments, but VIIRS lacks a water vapor channel at 6.7 μm, allowing for only cloud-tracking of winds. An addition of the 6.7 μm channel to future VIIRS instruments has been proposed. The additional channel could replace a 750-m channel at 1.6 μm (M10) that shares spectral response characteristics with a 375-m channel (I3). M10 data would then be synthesized by the 2-by-2 aggregation of I3 pixels. Radiometric response of such a synthesized channel is very similar to the actual one, although some differences exist. In this study, SNR (signal-to-noise ratio) for the M10 data simulated by the aggregation of the I3 pixels was compared with SNR for the actual M10 data. SNR for the simulated M10 was found to be always lower than SNR for the actual M10. This result contrasts with results of an analogous SNR comparison for bands I2 and M7 that share the same spectral response at 865 nm. Aggregated I2 data have SNR comparable to actual M7 data measured with the low gain, although lower than high-gain M7. The main reason for the different SNR behavior may be the use of microlenses with the I3 and M10 detectors, but not with the I2 and M7 ones.