A series of disaggregation experiments on the use of travel-forecasting procedures aimed at examining the impacts of disaggregation on estimates of mobile source emissions is described. The research used a case study of Baton Rouge, Louisiana, which is an ozone nonattainment area. The procedures in recent use in Baton Rouge were largely adopted as the basis for the research. The steps of disaggregation included estimating emissions at the link level in the loaded network, applying diurnal factors in two alternative methods, applying seasonal and day-of-week factors to standard fall/spring travel-forecasting results, and performing speed postprocessing after assignment. The results show that there are a number of areas in which significant impacts on emissions estimates arise, particularly from the seasonal and day-of-week adjustments and from speed postprocessing. Impacts of diurnal factoring appear less than expected, but are likely to be a function of levels of congestion. The paper concludes by noting some important issues relating to standard modeling practice and also suggests a number of areas for further research.