<title>Sensitivities of an atmospheric profiling retrieval to the microwave land emissivity</title>

Abstract

This study focuses on microwave land surface emissivity estimation over Northern Africa and the Middle East and the related impact on temperature and moisture retrievals. Land surface temperature retrievals are performed using a plane-parallel radiative transfer model, analyses from the Navy Operational Global Atmospheric Prediction System (NOGAPS) and data from the High Resolution Infrared Radiation Sounder Version 3 (HIRS/3). Infrared surface emissivity is indexed to each location using soil and vegetation databases provided by the Global Land Data Assimilation System (GLDAS), and spectral reflectance libraries of soil and vegetation. Initial microwave land emissivity estimates are made using a plane-parallel radiative transfer model, the infrared retrieved land surface temperatures, analyses from NOGAPS, and data from the Advanced Microwave Sounding Unit (AMSU). Perturbations of the atmospheric profiles and land surface temperatures provide estimates of the microwave emissivity error covariances necessary for retrievals and radiance assimilation. The error estimation is used in both the Naval Research Laboratory (NRL) 1DVAR retrieval, and for future use in 3DVAR radiance assimilation with the NRL Atmospheric Variational Data Assimilation System (NAVDAS). The window channels on AMSU/A have shown sensitivity to both temperature and moisture in the lowest five kilometers of the atmospheric profile, with these sensitivities strongly correlated to the estimate of the microwave land emissivity. Though the sensitivities are strongly correlated in the vertical dimension, an ability to extract meaningful profiling information from the microwave data is displayed. Further, the atmospheric sensitivity is linked to the precision to which the microwave radiances are estimated.