Another very important remark to make is that the sampling strategies proposed here, are only valid to estimate an EF for a certain barn and cannot be generalised to all buildings using the same housing system. In contrast, Ogink et al. (2011) and Mosquera and Ogink (2011) proposed a reduced sampling strategy based on a limited number of measurements at multiple farms, in order to take into account the between-farm variance. Furthermore, these measurements have to be spread out over time.