Global nitrous oxide (N2O) inventory estimates for pasture systems are refined based on measurements of N2O loss from simulated urine patches. A variety of methods are used for patch simulation but they frequently use a uniform wetted area (UWA), often smaller than a bovine urine patch. However, natural patches follow non-uniform infiltration patterns expanding naturally from a point of deposit with a non-wetted zone of influence. Using 2 litres of urine the UWA method was compared, using a 0·156 m2 collar, with a naturally expanding effective area (NEEA) method, using a 0·462 m2 collar under high (HL) and low (LL) N2O loss conditions. The method chosen affects urine nitrogen (N) loading to the soil. Under HL the UWA method induced a N2O-N loss of 280·6 mg/patch, significantly less than the 434·8 mg/patch loss for the NEEA method, for the same simulated urination. Under LL there was no method effect. Efforts should be made to employ patch simulation methods, which mimic natural deposits and can be achieved, at least in part, by: (a) Using a urine volume and N content similar to that of the animal of interest. (b) Allowing natural infiltration of the chosen urine volume to permit tapering towards the edges. (c) Measuring from the zone of influence in addition to the wetted area, i.e. the patch effective area.