ABSTRACT Obtaining knowledge about the distribution of spatial variability of soil properties is crucial to the proper site-specific management. One way to improve the quality of soil mapping is by using auxiliary information (covariate). The objective of this study was to test whether remote and proximal sensing data can assist in soil properties mapping through geostatistical prediction. We worked in an experimental area cultivated with sugarcane located in Sao Paulo State, Brazil, and selected five soil properties: organic matter, CEC, base saturation, K and P availability. Two covariates often used to express soil variation were chosen, one obtained by remote sensing (SWIR2 band) and the other by proximal sensing (apparent soil electrical conductivity – ECa). These covariates were individually and together used in geostatistical interpolation method (kriging with external drift). We found that ECa is a more promising covariate than SWIR2 band from orbital imaging. Such proximal sensing can identify the soil short-range spatial variability. However, when the soil property variability is well explained by the sampling procedure, multivariate geostatistical methods may not improve the mapping accuracy.

ABSTRACT Data interpolation is widely required in precision agriculture. However, its effectiveness is a function of the characteristics of the dataset, being necessary for the evaluation of several parameters. This study aimed to identify how the spatial interpolators, Kriging, and Inverse Distance Weighting, are influenced by the degree of spatial dependence of the variables analyzed and the number of neighbors considered in the interpolation process (sampling neighborhood). Soil samples were collected from three sugarcane fields. By the optimization process, we verified that the sampling neighborhood influences the accuracy of interpolations, but there is not a standard recommendation to follow. Thus, the best sampling neighborhood must ever be optimized for each case when preparing fertilizer prescription maps. Evaluating the performance of interpolations is always important to infer the reliability of the prescription maps, since no index that measures the degree of spatial dependence is effective. Because high prediction errors can occur when spatial dependence is poorly modeled, one cannot expect crop response with the continuous application of fertilizers in variable rates. Thus, work with homogeneous soil zones can be an interesting palliative approach. This study guides precision agriculture practitioners on some points that should be carefully considered in the data interpolation process for generating fertilizer prescription maps.

Técnicas de agricultura de precisão são alternativas para otimizar a utilização de insumos, dentre eles o nitrogênio. O presente trabalho teve como objetivo verificar a possibilidade de uso de um sensor ótico ativo comercial na cultura da cana-de-açúcar, verificando sua capacidade em identificar a resposta da cultura a diferentes doses de nitrogênio (N). Para tanto um experimento em blocos ao acaso foi instalado em área experimental de cana-de-açúcar de terceiro corte, colhida mecanicamente, com diferentes doses de N (0, 50, 100, 150 e 200 kg ha-1) e quatro repetições. Aos 30, 60 e 90 dias após o corte (DAC) realizou-se a mensuração do NDVI (índice de vegetação por diferença normalizada) a partir do sensor ótico. Não se obteve resposta do NDVI mensurado aos 30 e 60 DAC, porém aos 90 DAC se verificou que o sensor foi capaz de diferenciar as doses de N aplicadas, correlacionando-se positivamente com a produtividade final de colmos obtida. Este trabalho dá subsídios a trabalhos de mesma natureza que visem a gestão da adubação nitrogenada em cana-de-açúcar a partir de sensores óticos.

Techniques of precision agriculture are alternatives to optimize the use of inputs, among them nitrogen. The present work had as its objective to verify the possibility of use of a commercial optical sensor in sugarcane, verifying its capability in identification of the crop response to different nitrogen (N) rates. An experiment was installed in randomized blocks in an experimental area with third sugarcane ratoon, harvested mechanically, with different N rates (0, 50, 100, 150 and 200 kg ha-1) and four repetitions. At 30, 60 and 90 days after the harvest (DAH) the measuring of NDVI (normalized difference vegetation index) was accomplished starting from the optical sensor. No response was verified of NDVI on measurements at the 30 and 60 DAH, however at 90 DAH the sensor was capable to differentiate the doses of applied N, positively correlated with the sugarcane final production. This work gives subsidies to studies of same nature that seek the management of the nitrogen fertilization in sugarcane using optical sensor.