This paper introduces a radiative transfer model for the 9.6-μm ozone band that specifically matches the TIROS operational vertical sounder (TOVS) channel 9. The model is based on a spectral Malkmus band model for transmission. Band parameters were calculated by comparing to MODTRAN3 derived radiances. The effect of pressure on absorption is dealt with using a four-parameter approximation, and the improvements of this approximation over the more common Van de Hulst-Curtis-Godson scaling approximation are assessed. While this new model is exploited in the development of the retrieval described in this paper, the result has wider applicability to ozone climate problems requiring calculation of the radiative forcing associated with changing ozone. A two-layer version of the radiative transfer model is implemented in a retrieval scheme to obtain total ozone amounts from radiances measured by the TOVS instrument. Because the TOVS ozone channel is mainly sensitive to lower stratospheric ozone, ozone columns of the upper layer (above 30 hPa and with mean pressure of 10 hPa) are prescribed as a function of latitude. Ozone columns of the lower layer (mean pressure of 105 hPa) are then retrieved. The retrieval is based on a nonlinear optimal estimation algorithm and provides definition of error characteristics for every retrieval, which makes it possible to obtain a spatial distribution of the errors in the retrieval together with the spatial distribution of the retrieved total ozone column itself. This global distribution of the retrieval error and also of the contribution of a priori knowledge to the retrieval is presented to provide a validity of the ozone retrievals. Total ozone mapping spectrometer (TOMS) statistics are used as a priori information in the retrieval, and the 40-layer model is used to estimate the forward model error of the two-layer model. Comparisons of ozone retrievals for 1989 with TOMS total ozone measurements show good agreement both in time and in space with a rms difference between 1% and 3% for zonal means and 10% for global gridded measurements.¿ 1997 American Geophysical Union