ON STATISTICAL SAMPLING DISTRIBUTION IN EXPERIMENTAL BONIS RIVER BASIN (SILA GRECA - CALABRIA - ITALY)

The objective of this paper is to spread a optimized statistical sampling distribution in a representative river basin to evaluate the soil parameters in researching activities. The proposed approach was applied to starting two researching activities about interrill and rill soil erosion. In both researches is needed an optimized sampling distribution, which minimizes the spatial uncertainty in soil proprieties mapping to have a soil characteristic map which is the most accurate feasible. The sampling procedure uses an algorithm named Spatial Simulated Annealing (SSA) that was designed for optimization of spatial sampling schemes. This is done by minimizing the expectation of the distance between an arbitrarily chosen location, and its nearest observation point. This criterion is named Minimization of the Mean of the Shortest Distances. The elaboration was done using a Van Groenigen and Stein application named SANOS, using the sampling field GRID file and the list of the geographic coordinate of the first 55 samples points used as earlier observations. The result consist of 110 samples points to locate on the landscape. This sampling distribution will have advantages in accuracy of final soil properties mapping.