In order to acquire baseline data for prospective uranium mines, environmental radioactivity measurements are usually performed over large areas. This leads to a huge number of sampling points with a time-consuming and expensive data-acquisition process. For this study, probabilistic and non-probabilistic sampling strategies were used to determine the fewest possible sampling points needed to establish baseline data around the prospective uranium mine of Manyoni (Tanzania). The probabilistic approach used stratified and systematic grid sampling to obtain 32 sampling points at a 27.96 km 2 site prone to pollution from the prospective mining activities. These points were proportionally distributed based on pre-determined strata, geological areas, and administrative wards. Non-probabilistic sampling used judgmental sampling to allocate 8 sampling points in a region expected to receive higher levels of mining dust pollution than the WHO guidelines. We present a more representative and less expensive sampling approach with fewer sampling points around the prospective mine.