Entropy is not only a useful measure in descriptive studies, but may also be employed as auxiliary information for properly sampling from georeferenced environmental data. Techniques are borrowed from the branch of spatial sampling, in particular, from sequential procedures based on a system of weights, assigned according to the distance between units. Such weights sequentially modify the inclusion probabilities, i.e. the probabilities to include units in the sample. In our proposal, the procedures are adapted to include the spatial entropy of the study variable in the weighting system. The traditional spatial sampling methods perform well over data with a compact spatial structure, while random or repulsive behaviours benefit from the direct inclusion of spatial entropy in the weighting system of the selection probabilities. The procedure is successfully applied to a point dataset for the estimation of the total number of spruce trees in a study area.