Twenty nine field experiments were carried out to study the ventilation rate in a naturally ventilated dairy barn during 4 summer seasons and 3 winter seasons. The air exchange rates (AER) were determined by the tracer gas technique (TGT), which was the decay of the radioactive tracer gas Krypton-85, and the CO2-balance was set as reference method (RM). During each field experiment, continuous measurements of the gaseous concentrations (NH3, CO2, CH4, and N2O) inside and outside the building and 85Kr tracer gas experiments inside the building were carried out. The TGT was further developed by working on the gas release location and the calculation method. However, the TGT still overestimating the AER by 1.63±0.14 fold the RM in summer (R2=0.94) and 1.19±0.15 fold the RM in winter (R2=0.91). The tracer gas technique is a promising method; however, it overestimates the AER. On the other hand, the CO2-balance has several error sources that depend on animal production of CO2 which varies in function of animal weight, productivity and pregnancy. Therefore, the tracer gas technique should be further developed, where it delivers comparable results and is independent on the physiological changes. The average gaseous emissions through summer seasons, subject to the reference method, were 124, 538, 45610, and 28 g d-1 AU-1 for NH3, CH4, CO2, and N2O respectively. On the other hand, they were average of 64, 348, 42760, and 39 g d-1 AU-1 through winter seasons.