Abstract
Maps used in the precision farming include soil parameters related to moisture, pH, electric conductivity, soil fertility, inorganic nitrogen, NH4 and yields. We used the kriging method on a geostatistical basis to convert soil sampling data from paddy field and farms into two-dimensional maps containing position information. In this study, we developed a method to interpolate the selected soil sampling data into that needed to make a map by solving the problem of two-dimensional potential through an application of the finite element method. Soil fertility mapped by FEM, and kriging (exponential, gaussian and spherical models) were investigated. There were slight differences among the soil fertility maps in terms of frequency distribution and statistics. The average maps derived from FEM for each 10 meter-sized mesh were found to be similar to those derived by kriging.
