A Simulation Test of Three-Dimensional Temperature Retrievals

Retrieving information from remotely sensed data and analyzing the resulting geophysical parameters on a regular grid may be combined using a variational analysis method. This approach is applicable to the problem of retrieving temperature and cloud parameters within a three-dimensional volume from observations of infrared radiances at several frequencies and locations. The feasibility of such a three-dimensional retrieval method is demonstrated using simulated HIRS2 data. The method is successful in fitting the radiance data with a three-dimensional temperature representation. When clouds are included in the problem the results are sensitive to the initial estimates of the cloud parameters.