Geração de mapas de produtividade para experimentos com milho no âmbito de fazenda.
2003
SHIRATSUCHI, L. S.
ABSTRACT: Many farmers do not use precision agriculture tools due to high equipment costs and lack of knowledge about this new approach. A number of farmers have, however, conducted on-farm experiments without the help of these new and still expensive tools. One of most important equipments for precision agriculture practices is the crop yield monitor. It can also be used for on-farm experiments to collect yield data, reducing costs and paybacks. The aim of this paper is to evaluate the effects of combine header distance on corn harvest accuracy of yield maps. Conventional yield map was statistically compared with five maps generated by removing 2, 4, 6, 8 e 10 combine header width generated by a GIS-based software package. Two interpolations were made, i. e., ordinary block kriging and optimal inverse distance weighing, with and without MCF. This a new procedure that uses 80% of combine header width as a circular mean diameter to generate a sampling grid. For the MCF, before kriging, good results were obtained until six times the combine header width (five treatments, one control) based on yield comparison map. For the MCF method with inverse distance, a reliabe yield map was obtained until four times the combine header width.
显示更多 [+] 显示较少 [-]