Ground-truths or Ground-lies? : environmental sampling for remote sensing application exemplified by vegetation cover data

During time of fast development of computer and sensor technology, ground data sampling
strategies have achieved diminished attention in many remote sensing studies. This paper
discusses the importance of designing an appropriate sampling scheme of ground data collection
for remote sensing applications. The difficulties of achieving a balance between the size and the
error of the samples are identified. Different techniques of vegetation cover estimations are
evaluated to illustrate parts of the proposed sampling design. The study indicates that traditional methods of ground data collection for remote sensing applications do not have to result in "ground lies". Determination of a reliable and appropriate sampling scheme for the ground data collection should be given a more attention when assuring accurate results in remote sensing
studies.