The ability of inverse modeling to deduce the sources of CCl3F using a chemical transport model based on assimilated winds is examined. The sources of CCl3F are relatively well known and thus offer an opportunity to test methodologies that can be used to estimate the source strengths of trace gases whose sources are less well constrained. The Model of Atmospheric Transport and Chemistry (MATCH) is used in combination with assimilated winds from the European Center for Medium-Range Weather Forecasts (ECMWF) operational analysis and National Center for Environmental Prediction (NCEP)/National Center for Atmospheric Research reanalysis. Using our best a priori information about the emissions, comparisons are made between model simulations and observations at nine observing stations from the Atmospheric Lifetime Experiment/Global Atmospheric Gases Experiment and Climate Monitoring and Diagnostics Laboratory networks. The model simulates many features of pollution events and seasonal variability with both wind data sets. However, the interhemispheric gradient is too strong in the simulations with the ECMWF winds, although it is accurate with the NCEP winds. A recursive weighted least squares inverse method is used to determine the magnitude of emissions from five regions. The total magnitude as well as the hemispheric distribution of the sources of CCl3F are correctly estimated using the combination of the observations, model transport, and assumed a priori emission distribution. However, longitudinal source information is more difficult to estimate from observations. A sensitivity study suggests that locating the observing stations closer to the source regions would improve the ability of the inverse method to deduce longitudinal information about the sources. ¿ 1997 American Geophysical Union