Francisco de Assis de Carvalho Pinto

The search for more sustainable agricultural production systems is a goal of both agroforestry systems (AFS) and precision agriculture (PA). In this context, integrating PA technologies into coffee cultivation under AFS presents the potential to increase productive efficiency and sustainability. Thus, the objective of this study was to evaluate whether the variability of environmental conditions in coffee agroforestry systems can be estimated using technologies associated with PA. The studies were conducted on three coffee farms in Coimbra and Araponga, Minas Gerais, Brazil. For management zone delineation, combinations of apparent soil electrical conductivity (ECa), Normalized Difference Vegetation Index (NDVI), and Digital Terrain Model (DTM) were used. Microclimatic variability was evaluated using weather stations, considering different agroforestry management practices and altitude conditions in mountain coffee cultivation. Canopy cover was estimated by digital image processing techniques applied to aerial images obtained by an Unmanned Aerial Vehicle (UAV). The combination ECa + NDVI was efficient in delineating zones with distinct physical and chemical attributes. Agroforestry management practices and altitude influenced temperature and relative humidity. Tree canopy estimation proved viable to support management. It is concluded that PA is strategic for coffee cultivation in AFS, favoring site-specific and more sustainable management.

Wheat provides one fifth of all human protein and calories and is the most widely grown crop globally. Climate change induced heat stress increasingly threatens global wheat production, yet the role of root associated microbial communities in mitigating heat related yield losses remains poorly understood under field conditions. This study investigated how plant heat stress affects rhizosphere bacteriome composition across four wheat genotypes with contrasting sensitivity to heat stress in terms of yield and root biomass formation. Using 16S rRNA amplicon sequencing, we characterized microbial community structure during flowering under heat stress and optimal conditions over two consecutive years in the Yaqui Valley in northwestern Mexico. Surprisingly, rhizosphere bacteriome composition and diversity were more strongly influenced by environmental factors, particularly year and heat stress treatment, than by wheat genotype. An increase in Bacillus and stable levels of stress-tolerant taxa such as Rubrobacter and Arthrobacter, indicating a shift toward more resilient bacteria and a shared microbial response to elevated temperature. Despite differences to plant heat stress susceptibility, all genotypes converged toward a shared core microbiome dominated by Proteobacteria, Actinobacteria, and Firmicutes under heat stress. These findings suggest that environmental filtering under high temperature outweighs host genotype in shaping rhizosphere bacterial assembly. For subsequent studies it is suggested to include sampling of endophytic bacteria in order to study the contribution of bacteria to plant heat stress resilience. Our study provides a field based baseline for harnessing heat adapted microbiomes to support wheat resilience in a warming world.

Developing breakthrough products that deliver a minimum 25% productivity improvement in farmers’ fields requires that breeding pipelines more closely align with farmer and end-user needs and the target population of environments (TPEs). In the case of maize (Zea mays), climate change is accelerating climatic shifts within the TPE, potentially limiting future realized genetic gain. On-farm breeding approaches can play a valuable role in supporting breeding programs transitioning toward breakthrough products. Increased sampling of the TPE by moving selection into farmers’ fields can overcome the challenges of climatic changes within the TPE and extensive farmer diversity, whilst providing information to refine managed environment trials designed to sample the TPE. This review highlights advances in inclusive host farmer sampling, sparse testing strategies and phenotyping to facilitate scaling of on-farm testing within breeding pipelines.

Coffee production in Brazil requires efficient monitoring tools, especially in regions with heterogeneous terrain and planting patterns. Accurate delineation of coffee canopies from UAV imagery is essential for extracting reliable agronomic information and supporting precision agriculture practices. Multispectral UAV images were acquired over a commercial coffee farm in the Matas de Minas region, Brazil. An orthomosaic was generated and manually annotated to create binary ground-truth masks of coffee canopies. A U-Net convolutional neural network was trained using multispectral bands and the Normalized Difference Vegetation Index (NDVI) as input. Model performance was evaluated using Dice coefficient, Intersection over Union (IoU), precision, recall, and accuracy. The proposed model achieved a Dice coefficient of 0.89, IoU of 0.83, and overall accuracy above 0.92. The predicted coffee canopy area (1.50 ha) was consistent with the reference area obtained from manual annotation (1.45 ha), indicating reliable canopy delineation under heterogeneous field conditions. The results demonstrate that semantic segmentation using multispectral UAV imagery and a U-Net model is an effective and applicable approach for coffee canopy mapping, with potential use in precision agriculture and farm-scale decision-support systems.

Rising nighttime temperatures pose a growing threat to wheat productivity, yet their physiological impacts remain poorly characterized under field conditions. This study evaluated whether vegetation indices (VIs) derived from nocturnal hyperspectral reflectance can be linked to wheat physiological and agronomical responses to elevated nocturnal temperatures. A panel of 12 wheat genotypes was grown over three seasons under control and +2 °C nighttime warming conditions. Leaf hyperspectral reflectance was measured during day and night across two phenological stages, and leaf sides, and 27 VIs and modelled physiological traits were analysed. Nighttime warming altered leaf spectral properties, particularly in wavelength regions associated with pigment and water content. Several greenness, pigment, and water-related VIs showed significant treatment responses (p 0.90) and lower variability, indicating more stable spectral signals. NDVI was positively associated with yield under control and nighttime warming (NDVI705: r = 0.62, 0.61, respectively) and NWI related to water status under heat stress (NWI vs Ψ, r = −0.60). These findings demonstrate that nocturnal hyperspectral phenotyping provides complementary insights into wheat responses to warming and can support the identification of heat-resilient genotypes.

Precision agriculture, including sensors and artificial intelligence, is transforming agricultural monitoring. This study developed predictive models for fresh and dry forage mass, canopy height, forage density, dry matter (%DM), and crude protein (%CP) concentrations in signalgrass [Urochloa decumbens (Stapf) R.D. Webster] pastures using machine learning and UAV-based multispectral imagery. The experiment was conducted at the Federal University of Viçosa (2019–2020), applying nitrogen doses after each harvest to promote variability. Multiple Linear Regression (MLR), Support Vector Regressor (SVR), and Random Forest Regressor (RFR) models were trained with multispectral and meteorological data. The best results were obtained for fresh forage mass with RFR (R² = 0.82, RMSE = 2894.10 kg ha⁻¹), dry forage mass with SVR (R² = 0.68, RMSE = 719.87 kg ha⁻¹), and dry matter concentration with MLR (R² = 0.64, RMSE = 3.83%). Forage density showed moderate performance (R² = 0.56), while canopy height demonstrated limited accuracy (R² = 0.44). Crude protein was not adequately predicted by any model, highlighting multispectral sensor limitations and suggesting hyperspectral sensors usage. Results demonstrate the applicability of remote sensing combined with machine learning in forage management, but indicate the need to expand temporal and spatial data variability and integrate different sensor types to increase model robustness.

The characterization of soil attribute variability often requires dense sampling grids, which can be economically unfeasible. A possible solution is to perform targeted sampling based on previously collected data. The objective of this research was to develop a method for mapping soil attributes based on Management Zones (MZs) delineated from Sentinel-1 radar data. Sentinel-1 images were used to create time profiles of six indices based on VV (vertical–vertical) and VH (vertical–horizontal) backscatter in two agricultural fields. MZs were delineated by analyzing indices and VV/VH backscatter bands individually through two approaches: (1) fuzzy k-means clustering directly applied to the indices’ time series and (2) dimensionality reduction using deep-learning autoencoders followed by fuzzy k-means clustering. The best combination of index and MZ delineation approaches was compared with four soil attribute mapping methods: conventional (single composite sample), high-density uniform grid (one sample per hectare), rectangular cells (one composite sample per cell of 5 to 10 hectares), and random cells (one composite sample per cell of varying sizes). Leave-one-out cross-validation evaluated the performance of each sampling method. Results showed that combining the VV/VH index and autoencoders for MZ delineation provided more accurate soil attribute estimates, outperforming the conventional, random cells, and often the rectangular cell method. In conclusion, the proposed methodology presents scalability potential, as it does not require prior calibration and was validated on soil types commonly found across Brazil’s agricultural regions, making it suitable for integration into digital platforms for broader application in precision agriculture.

Characterization of soil attribute variability often requires dense sampling grids, which can be economically unfeasible. A possible solution is to perform targeted sampling based on previously collected data. The objective of this research was to develop a method for mapping soil attributes based on Management Zones (MZs) delineated from Sentinel-1 radar data. Sentinel-1 images were used to create time profiles of six indices based on VV (vertical-vertical) and VH (vertical-horizontal) backscatter in two agricultural fields. MZs were delineated by analyzing indices and VV/VH backscatter bands individually through two approaches: (1) fuzzy k-means clustering directly applied to the indices' time series, and (2) dimensionality reduction using deep-learning autoencoders followed by fuzzy k-means clustering. The best combination of index and MZs delineation approach was compared with four soil attribute mapping methods: conventional (single composite sample), high-density uniform grid (one sample per hectare), rectangular cells (one composite sample per cell of 5 to 10 hectares), and random cells (one composite sample per cell of varying sizes). Leave-one-out cross-validation evaluated the performance of each sampling method. Results showed that combining VV/VH index and autoencoders for MZs delineation provided more accurate soil attribute estimates, outperforming the conventional, random cells, and often the rectangular cell method.