Robust water temperature retrieval using multi-spectral and multi-angular IR measurements

Temperature retrievals over water surfaces are possible because of the almost constant emissivity of water. As an example the authors discuss the channel selection for a robust sea surface temperature retrieval under various atmospheric conditions (water vapor content, cirrus clouds and aerosols). The presented model uses simulated data from MODTRAN runs and detector noise. It is based on a linear regression fit of the measured radiances to the sea surface temperatures. Results indicate that the sea surface temperature can be retrieved to within less than one degree absolute accuracy using the nadir and one off-nadir measurement. The authors compare the procedure to inversion of an analytical model and discuss sensitivities of the model