Ammonia emissions were estimated at five microclimatic locations in a free stall dairy building using four different methods. Measurements were performed simultaneously with the different methods to enable comparison. In the first method, the rate of ammonia emission from manure was theoretically modeled utilizing Fick's law and boundary layer theory. In the second method, recirculation flux-chamber technique was used to model ammonia emission from manure. In the third and fourth methods, respectively, carbon dioxide and methane balances were employed to calculate ammonia emission. The mean ammonia emissions measured from the five locations using the different methods ranged from 0.10 to 0.15 g m(-2) h(-1). The percentage of variation of ammonia emission from the different location ranged between 8 and 52% for the different methods. Recorded ammonia emission rates in the dairy building were from 0.04 to 0.25 g m(-2) h(-1). The percentage of variation in ammonia, carbon dioxide, methane, and manure properties in the building was 50%. Two-way statistical analysis of variance showed that there were no significant differences (p > 0.63) between the four different methods or between the measurements obtained at the five locations (p > 0.90).