Since mineral aerosols absorb significant amounts of infrared radiation, they may contribute to errors in the retrievals of atmospheric and surface parameters from the TIROS Operational Vertical Sounder (TOVS) High-Resolution Infrared Radiation Sounder (HIRS) if the atmosphere is assumed clear. TOVS is an operational sounder on NOAA polar satellites. To see if observed brightness temperatures are reduced by mineral aerosol, we analyzed results from the Data Assimilation Office (DAO) Finite Volume Data Assimilation System (fvDAS). Every 6 hours the assimilated temperature and moisture profiles are used as a first guess in the DAO interactive cloud-clearing TOVS retrieval system. The observed minus the forecast (O--F) brightness temperature, which is a measure of the accuracy of the first guess and radiative transfer parameters, becomes more negative with increasing dust concentrations. Dust concentrations are from the Goddard Ozone Chemistry Aerosol Radiation Transport (GOCART) model. Since there was no account of dust during this fvDAS run, the dependence of O--F on the estimated atmospheric dust concentrations from GOCART indicates that the dust is affecting the TOVS brightness temperatures. HIRS channels that are sensitive to the surface temperature, lower tropospheric temperature, and moisture are subject to a 0.5 K or more reduction in the brightness temperature during heavy dust loading conditions. The radiative transfer module used in the TOVS retrieval system was modified to account for dust assuming a composition of illite, and the fvDAS run was repeated. Accounting for dust absorption in the retrieval system yields warmer surface temperatures (0.4 K) and warmer lower tropospheric temperatures in regions of moderate dust loading over the tropical Atlantic.