One of our main challenges in meteorology and environment research is that in many important remote areas, sensor coverage is sparse, leaving us with numerous blind spots. Placement and maintenance of sensors in these areas are expensive. It is therefore desirable to find out how, within a given budget, we can design a sensor network are important activities was developing reasonable techniques for sensor that would provide us with the largest amount of useful information while minimizing the size of the "blind spot" areas which is not covered by the sensors.
This problem is very difficult even to formulate in precise terms because of the huge uncertainty. There are two important aspects of this problem: (1) how to best distribute the sensors over the large area, and (2) what is the best location of each sensor in the corresponding zone. There is some research on the first aspect of the problem.
In this paper, we illustrate the second aspect of the problem, on the example of optimal selection of locations for the Eddy towers, an important micrometeorological instrument.