Estimating yield is a major challenge for the majority of agricultural crops. With the advancement of field technologies however, especially those related to the use of Unmanned Aerial Vehicles (UAV) or Drones, the quality of available information has increased, making it possible to overcome technological bottlenecks. However, drone technologies have advanced much faster than studies dealing with the treatment and analysis of information, which can represent an obstacle to the complete adoption of such technologies in sugarcane fields. The objective of the present study was to evaluate the potential for UAV images to assess the degree of canopy closure from different planting approaches and row-spacing treatments applied to sugarcane crop, in order to assess the potential of these tools to predict crop yield. The vegetative growth of the crop was evaluated and the images were obtained at the point of maximum tillering and the inflection point of the biomass accumulation curve. The evaluations included the index; LAI (Leaf Area Index) and GRVI (Green-Red Vegetation Index) obtained by field sensor and UAV, respectively. Because the images from UAV cover the total area, the results revealed that GRVI appears to be much better able to reflect the whole condition of the crop yield (R² = 0.69) in the field when compared to LAI (R² = 0.34); demonstrated convincingly by the high spatial resolution capacity of the technology. When integrated, these two indices were able to improve yield estimates by 10% (R² = 0.79). Images obtained using UAV can represent a low-cost tool for obtaining high-precision remote data that can be used to estimate the agricultural yield of sugarcane fields; and in this way are an effective tool to aid decision making by growers.

Sugarcane (saccharum spp.) in Brazil is managed on the basis of “production environments”. These “production environments” are used for many purposes, such as variety allocation, application of fertilizers and definition of the planting and harvesting periods. A quality classification is essential to ensure high economic returns. However, the classification is carried out by few and, most of the time, non-representative soil samples, showing unreal local conditions of soil spatial variability and resulting in classifications that are imprecise. One of the important tools in the precision agriculture technological package is the apparent electrical conductivity (ECa) sensors that can quickly map soil spatial variability with high-resolution and at low-cost. The aim of the present work was to show that soil ECa maps are able to assist classification of the “production environments” in sugarcane fields and rapidly and accurately reflect the yield potential. Two sugarcane fields (35 and 100 ha) were mapped with an electromagnetic induction sensor to measure soil ECa and were sampled by a dense sampling grid. The results showed that the ECa technique was able to reflect mainly the spatial variability of the clay content, evidencing regions with different yield potentials, guiding soil sampling to soil classification that is both more secure and more accurate. Furthermore, ECa allowed for more precise classification, where new “production environments”, different from those previously defined by the traditional sampling methods, were revealed. Thus, sugarcane growers will be able to allocate suitable varieties and fertilize their agricultural fields in a coherent way with higher quality, guaranteeing greater sustainability and economic return on their production.

The high prevalence of Haemonchus contortus and its anthelmintic resistance have affected sheep production worldwide. Machine learning approaches are able to investigate the complex relationships among the factors involved in resistance. Classification trees were built to predict multidrug resistance from 36 management practices in 27 sheep flocks. Resistance to five anthelmintics was assessed using a fecal egg count reduction test (FECRT), and 20 flocks with FECRT < 80% for four or five anthelmintics were considered resistant. The data were randomly split into training (75%) and test (25%) sets, resampled 1,000 times, and the classification trees were generated for the training data. Of the 1,000 trees, 24 (2.4%) showed 100% accuracy, sensitivity, and specificity in predicting a flock as resistant or susceptible for the test data. Forage species was a split common to all 24 trees, and the most frequent trees (12/24) were split by forage species, grazing pasture area, and fecal examination. The farming system, Suffolk sheep breed, and anthelmintic choice criteria were practices highlighted in the other trees. These management practices can be used to predict the anthelmintic resistance status and guide measures for gastrointestinal nematode control in sheep flocks.