Models for predicting coffee yield from chemical characteristics of soil and leaves using machine learning.

BACKGROUND: Coffee farming constitutes a substantial economic resource, representing a source of income for several countries due to the high consumption of coffee worldwide. Precise management of coffee crops involves collecting crop attributes (characteristics of the soil and the plant), mapping, and applying inputs according to the plants' needs. This differentiated management is precision coffee growing and it stands out for its increased yield and sustainability. RESULTS: This research aimed to predict yield in coffee plantations by applying machine learning methodologies to soil and plant attributes. The data were obtained in a field of 54.6 ha during two consecutive seasons, applying varied fertilization rates in accordance with the recommendations of soil attribute maps. Leaf analysis maps also were monitored with the aim of establishing a correlation between input parameters and yield prediction. The machine-learning models obtained from these data predicted coffee yield efficiently. The best model demonstrated predictive fit results with a Pearson correlation of 0.86. Soil chemical attributes did not interfere with the prediction models, indicating that this analysis can be dispensed with when applying these models. CONCLUSION: These findings have important implications for optimizing coffee management and cultivation, providing valuable insights for producers and researchers interested in maximizing yield using precision agriculture. © 2024 Society of Chemical Industry.

Understanding the contribution of soybean crop residues inoculated with Bradyrhizobium spp. and not harvested on nitrogen supply in off-season corn cultivars.

Excessive rainfall in the soybean preharvest period can make mechanized crop harvesting technically and economically unfeasible, causing 100% losses in soybean grain yield. An alternative to reduce the economic losses of farmers would be using unharvested soybean crop residues as a source of nitrogen (N) for the subsequent corn crop. However, a question that still needs to be understood is whether the amount of N released from unharvested soybean residues (straw and grains) is sufficient to meet all the nutritional demand for this nutrient in the off-season corn. Therefore, this study investigated the impact of unharvested soybean crop residue persistence on the yield response of off-season corn crop (Zea mays L.) to the application of N fertilizer rates when grown in tropical Cerrado soils of medium and high fertility. Four simple corn hybrids (SYN7G17 TL, 30F53VYHR, B2433PWU, and AG 8700 PRO3) were grown in soils of medium fertility and medium acidity level (UFMS 1) and high fertility and low acidity level (UFMS 2) and fertilized with five of N fertilizer rates (0, 40, 80, 120, and 160 kg ha-1 of N) applied at 30 days after emergence (DAE). Canonical correlation analysis (CCA) was used to investigate the interrelationships between the groups of independent (agricultural production areas, corn cultivars, and N application rates) and dependent (corn agronomic traits) variables. Crop residues remaining on the soil surface from soybeans not harvested and inoculated with Bradyrhizobium spp. can supply most of the nitrogen requirement of off-season corn grown in succession, especially in tropical soils of medium fertility. However, in high-fertility tropical soils, the maximum grain yield potential of off-season corn cultivars can be obtained with the application of mineral N fertilizer in supplement the amount of nitrogen released from unharvested soybean residues. Therefore, the N requirement depends on the corn cultivar and the agricultural production area. However, our results show that when off-season corn is grown on unharvested soybean residues, nitrogen fertilization in topdressing can be dispensed. The agricultural area with high fertility soil (UFMS 2) enhances the grain yield of the off-season corn crop. The corn cultivar AG 8700 PRO3 has a higher thousand-grain mass and high grain yield potential under Brazilian Cerrado conditions.

Adaptability and Stability of Safflower Genotypes for Oil Production.

The study aimed to analyze the agronomic performance of 11 safflower genotypes using adaptability and stability methods, while identifying safflower genotypes with stable behavior and a high grain yield in different environments of the Brazilian Cerrado. Ten lines and a cultivar of safflower were evaluated in four environments in the Brazilian conditions. Our results revealed the genotypes P30, P35, P9, P11, and P31 to be superior for grain yield and P43, P7, P11, and P31 to be superior for oil content. The lowest Wricke index, an indication of genotype stability, was observed for P9 (0.41%), which is considered the most stable genotype, followed by P35 (1.29%) and P31 (1.98%). For the predictability of the behavior of genotypes in the environments, P7 (80.85%), P35 (86.10%), P31 (85.90%), and P9 (97.42%) were considered predictable genotypes. The genotypes P11 (1045.6 kg ha-1 and 19.7%) and P21 (952.7 kg ha-1 and 20.6%) are recommended for cultivation in this region, considering both their grain yield and oil content. Safflower is viable to use out of season in the Brazilian Cerrado. The crop can generate profits for farmers and be used for oil production in periods of uncertain corn production.