A simulation study of 1D variational cloud retrieval with infrared satellite data from multiple fields of view

A method for retrieving cloud-top height using one-dimensional variational analysis with multiple pixels is examined. The method is tested for a range of cloud scenarios using simulated multi-spectral infrared imagery. The skill of the multiple pixel method is compared to the skill of the single-field-of-view method. It is found that, when applied to two pixels, the multiple-field-of-view method considerably increases the skill of retrieval for the less cloudy pixel, and marginally improves the skill of retrieval for the cloudier pixel in the majority of cases. It is also found that, when applied to six pixels, the multiple-field-of-view method has a positive impact in many cases.