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.

Yield, nutrition, and leaf gas exchange of lettuce plants in a hydroponic system in response to Bacillus subtilis inoculation

Inoculation with Bacillus subtilis is a promising approach to increase plant yield and nutrient acquisition. In this context, this study aimed to estimate the B. subtilis concentration that increases yield, gas exchange, and nutrition of lettuce plants in a hydroponic system. The research was carried out in a greenhouse in Ilha Solteira, Brazil. A randomized block design with five replications was adopted. The treatments consisted of B. subtilis concentrations in nutrient solution [0 mL "non-inoculated", 7.8 × 103, 15.6 × 103, 31.2 × 103, and 62.4 × 103 colony forming units (CFU) mL-1 of nutrient solution]. There was an increase of 20% and 19% in number of leaves and 22% and 25% in shoot fresh mass with B. subtilis concentrations of 15.6 × 103 and 31.2 × 103 CFU mL-1 as compared to the non-inoculated plants, respectively. Also, B. subtilis concentration at 31.2 × 103 CFU mL-1 increased net photosynthesis rate by 95%, intercellular CO2 concentration by 30%, and water use efficiency by 67% as compared to the non-inoculated treatments. The concentration of 7.8 × 103 CFU mL-1 improved shoot accumulation of Ca, Mg, and S by 109%, 74%, and 69%, when compared with non-inoculated plants, respectively. Inoculation with B. subtilis at 15.6 × 103 CFU mL-1 provided the highest fresh leaves yield while inoculation at 15.6 × 103 and 31.2 × 103 CFU mL-1 increased shoot fresh mass and number of leaves. Concentrations of 7.8 × 103 and 15.6 × 103 increased shoot K accumulation. The concentrations of 7.8 × 103, 15.6 × 103, and 31.2 × 103 CFU mL-1 increased shoot N accumulation in hydroponic lettuce plants.

Inoculation with Azospirillum brasilense Strains AbV5 and AbV6 Increases Nutrition, Chlorophyll, and Leaf Yield of Hydroponic Lettuce

Inoculation with Azospirillum brasilense has promisingly increased plant yield and nutrient acquisition. The study aimed to estimate the dose of A. brasilense that increases yield, gas exchange, nutrition, and foliar nitrate reduction. The research was carried out in a greenhouse at Ilha Solteira, in a hydroponic system in randomized blocks with four replicates. The treatments consisted of doses of inoculation with A. brasilense strains AbV5 and AbV6 via nutrient solution (0, 8, 16, 32, and 64 mL 100 L-1). Inoculation with A. brasilense at calculated doses between 20 and 44 mL provided the highest fresh and dry mass of shoots and roots, number of leaves, and leaf yield. In addition, the calculated doses of inoculation with A. brasilense increased the accumulation of N, P, K, Ca, Mg, S, B, Fe, Mn, and Zn in shoots and roots, except the accumulation of Ca in roots. It also increased cell membrane integrity index (15%), relative water content (13%), net photosynthesis rate (85%), intracellular CO2 concentration (15%), total chlorophyll (46%), stomatal conductance (56%), transpiration (15%), and water use efficiency (59%). Hence, inoculation with A. brasilense at doses between 20 and 44 mL 100 L-1 is considered the best approach for increasing the growth, yield, accumulation of nutrients, and gas exchange of hydroponically grown iceberg lettuce.

A site suitability analysis for castor (Ricinus communis L.) production during Brazil's second harvest incorporating disease prediction

Castor (Ricinus communis L.) is an important industrial crop with a wide range of industrial and pharmaceutical applications. Brazil is among the largest castor-producing countries. Between 2004 and 2010, castor cultivation was stimulated with an emphasis towards biodiesel production. However, this was not enough to leverage the production of castor in Brazil, mainly due to the lack of structured trade and the competition with other cheaper raw materials for the production of biodiesel. Despite this failure, the species presents itself as an excellent alternative for crop rotation in the second crop among soybean, corn, beans, and cotton cultivation areas as the oil is highly valuable for other products. Moreover, it has drawn the attention of producers and researchers in Brazil for this potential rotation as it is considered a plant tolerant of water-deficiency and is highly susceptible to gray mold, a disease favored by high humidity in the final stages of the crop. For instance, its cultivation in the second crop in Cerrado regions, where rains occur in the early stages of the crop and cease when the plants reach the final stage of production, has been successful and shows great promise. The current study aimed to evaluate the suitability of environments throughout Brazil to grow castor, incorporating variables associated with the incidence of gray mold and confirm these findings based on existing castor trial data obtained from the literature. The site suitability analysis determined that 74.99 million hectares - 8.8% of Brazilian territory - are highly suitable for castor production during second harvest, mostly located in the Northeastern and Midwestern regions. These results are surprising since Brazil currently has around 7.8% (∼66.81 million hectares) of its territory occupied with agriculture (grains, fruits, vegetables, and perennial crops). The findings of this study provide a method to perform site suitability for crops using data associated with agronomic and disease characteristics, as is the case with gray mold that often results in significant losses in castor production. Also, this analysis provides evidence for the great potential of Brazil to increase castor production and meet the world demand for its oil through utilization of second-crop cultivation.

Can co-inoculation of Bradyrhizobium and Azospirillum alleviate adverse effects of drought stress on soybean (Glycine max L. Merrill.)?

Harnessing the beneficial potential of plant growth-promoting rhizobacteria may be an alternative strategy to improve plant tolerance to drought stress. The effect of inoculation with Bradyrhizobium japonicum and Azospirillum brasilense either alone or in combination on the plant growth and drought tolerance of soybean [Glycine max (L.) Merrill.] was investigated in this study in greenhouse conditions. Treatments were arranged in a randomized block design in a 3 × 4 factorial: three irrigation regimes [100% of pot capacity-PC (well-watered control), 50% of PC (moderate stress) and 25% of PC (severe stress)] and four inoculation treatments [control (non-inoculated), inoculation with B. japonicum, inoculation with A. brasilense, and co-inoculation with B. japonicum and A. brasilense]. Leaf relative water content, cell membrane stability, root nodulation, plant growth, and morphophysiological indexes were recorded. The inoculation of soybean plants with B. japonicum and A.brasilense either alone or in combination improved leaf membrane stability under drought stress conditions when compared to non-inoculated plants; however, this lower damage to cell membranes was not sufficient to maintain the leaf water content of the plant under drought stress. Plants co-inoculated with B. japonicum and A.brasilense improved the root nodulation under severe drought conditions. Inoculation of B. japonicum and A. brasilense either alone or in combination reduced the pod abortion rate under moderate drought stress, but had no effect under severe drought stress. In summary, the co-inoculation of A. brasilense and B. japonicum alleviate adverse effects limited by drought stress to the growth of soybeans.Author: Please check and confirm that the authors [Elijanara Raissa Silva, Carlos Eduardo Silva Oliveira, Alan Mario Zuffo, Eduardo Pradi Vendruscolo] and their initials have been correctly identified and amend if necessary.The authors were correctly identified.

Feeding preference of Spodoptera frugiperda on different sorghum genotypes

ABSTRACT The resistance of plants to pests is a feature of great importance for agriculture, as it reduces costs with insecticides and promotes increased yield resulting in higher profits. This work aimed to evaluate the feeding preference of the Spodoptera frugiperda caterpillar by different sorghum genotypes. The experiment was carried out at the Entomology Laboratory of Universidade Estadual de Mato Grosso do Sul, at University Unit of Cassilândia, from March to June 2016. The experimental design was completely randomized with ten replicates. The treatments were composed by seven sorghum genotypes: Agromen 50A40, Agromen 50A50, DOW 1G100, DOW 1G220, DOW 1G233, XB 6022 and LG 6310. Evaluations were performed with 1st instar caterpillars. The number of caterpillars that settled in the genotypes at 5, 10, 15, 20, 25, 30, 60 minutes and 24 hours after the infestation was recorded. The feeding preference index and the fresh leaf mass consumed were estimated. The Agromen 50A40 genotype showed lower attractiveness rating for the S. frugiperda among all evaluated sorghum genotypes.