Variabilidade espacial da fertilidade do solo e mapas de recomendação

Abstract

Among the technologies that can contribute to improve the efficiency of agricultural production systems is precision agriculture, which provides for the specialized management of production factors, based on information collected georeferenced. The fertility maps bring the possibility of identifying the spatial variation of the nutrients available in the soil for the plant, so that fertilizers can be added at a variable rate. In this context, this work aimed to evaluate the spatial variability of soil fertility and to make recommendation maps for the use of fertilizers and correction agents at variable rates to meet their demands. The characterization of the chemical attributes of the soil was in an area of 132 ha through georeferenced points and two sample grids were generated by the geostatistical software on the free platform QGIS (version 2.18.24), with the first grid with 64 points allocated on a regular basis. point every 2.06 hectares (143.61x143.61m), and the second grid with 13 sample points, randomly allocated. The final sample density was 01 sample for each 1.78 hectares, totaling 77 sample points. The data were submitted to descriptive statistics and the maps were interpolated / processed using the IDW interpolator sized for a spatial resolution of 10 meters. According to the generated maps, there is a low need for liming, since the maps correlated to the chemical attributes of V% are not very far from the desired. Analyzing the H + Al maps, it is observed that the values are not very high and the Ca + Mg maps show that the values are good. For the phosphorus recommendation, it is observed that the central region of the field will receive the lowest doses, and around it, will be the highest doses. This can be observed by analyzing the available P layout map. For the potassium recommendation map, it is noticed that there is a wide variability in the application, since the values of the potassium disposition map mostly present low values.