Environmental Thematic Map Prediction and Easy Probabilistic Classification within GRASS

Environmental observations are usually sampled in irregularly spaced points even though the information we need must be, in most cases, continuous and representative of continuous phenomena. To create the thematic map, the stochastic approach is often the best interpolating method to apply. One of the advantages of the application of this theory consists in making available not only a prediction map (thematic map), but also an error prediction map (accuracy map); starting from these two information layers a procedure can be implemented to verify if in some areas the predicted data are within an assigned interval or exceed a threshold value at certain degree of probability. The methodology adopted, even though simplified from the theoretical point of view, gives in detecting risk areas, information that is more accurate than that deriving from usual deterministic approaches. The procedures were developed by creating GRASS commands calling upon and integrating the statistic Gstat library, and commands which allow the computing of probabilistic risk areas.