Sea Surface Temperature (SST) plays a critical role in regulating the flux exchange between the atmosphere and oceans. It may vary rapidly in space and time. Although SSTs may be retrieved accurately from polar orbiting satellites, such estimates are limited in space and time. Less accurate estimates can be made using the Geostationary satellites, but give much more frequent coverage than polar orbiting satellites, and therefore more chance of cloud free data. Apart from this, major application of SSTs from Geostationary satellites will improve our understanding of thermocline of the ocean. The thermocline layer is the transition layer between the mixed layer at the surface and the deep water layer. Frequent observations from Geostationary satellites will improve understanding of thermal structure of the ocean layers.
In this study a paramerisation approach has been carried out to retrieve SST from GOES8 & 9 satellite data. This was achieved by using atmospheric profiles with MODTRAN 3.7 Radiative transfer code. Independent profiles were used for validation, i.e. for checking the accuracy with the parametrized relations. The accuracy of SST estimate involves atmospheric and surface effects. The atmospheric effects is primarily by water vapor. The surface effect is caused by the reflected downwelling radiation and emissivity. The emissivity varies with the view angle of the sensor. The MODTRAN-3.7 code is employed to simulate the parameters for the satellite brightness temperature. The appropriate emissivity values for the view angles are incorporated in the simulation. The parametric relationship is derived in this study to retrieve SST from GOES8 &9 11 and 12 micron channels. The error estimates have been calculated with the independent set of atmospheric profiles. The errors in the SST estimates are atmospheric error and the instrumental noise. The total error is within 0.7K. An extensive comparison/validation with in-situ and polar orbiting satellites are being investigated.