Development of the PCRTM-Based All-Sky, Single Field Retrieval Algorithm for Multiplatform, Long-Term Applications

This presentation demonstrates a single field of view (FOV) retrieval algorithm that can use various measurements from various hyper-spectral sounders (AIRS, IASI and CrIS) for both weather and climate study. The physical retrieval algorithm uses the Principal Component based Radiative Transfer Model (PCRTM) for the forward simulation. The incorporation of multiple scattering simulation in PCRTM allows the direct radiative relationship between single FOV radiance observations and corresponding thermal dynamic variables including cloud properties to be established. Therefore, geophysical data products that match the spatial resolution of sounder observations can be obtained via the single FOV retrieval. As a comparison, the operational retrieval algorithms for current sounders use cloud-cleared radiances generated from multiple FOV observations and therefore cannot yield data products of the native spatial resolution. By directly fitting the spectral radiance obtained via an accurate radiative transfer calculation to real satellite observations, a closure between the radiance observation and the retrieved geophysical parameters can be achieved under all-sky conditions. The retrieved geophysical parameter can be used to build radiative kernels that are valuable for the study of long-term climate trends and feedbacks. The radiometric consistent retrieval scheme also provides a mean of consistency check between different sounder measurements, i.e., systematic radiance difference between overlapping observations can be evaluated using the corresponding retrieved geophysical variables and radiative kernels. The presentation will focus on the error analysis of the retrieval algorithm, the validation for the simultaneous retrieval application on real data, and the construction of radiative kernels using multi-year observations.