DSpace Communidade: Embrapa Agricultura Digital (CNPTIA)

Advancing coffee management mapping through multisensor data and multistep ensemble learning.

2026-04-22T19:49:56Z

Título: Advancing coffee management mapping through multisensor data and multistep ensemble learning. Autoria: PARREIRAS, T. C.; BOLFE, E. L.; FURUYA, D. E. G. Conteúdo: Despite the advances, accurately identifying recently renovated and skeletonized coffee areas remains a challenge, as their altered canopy structure and reduced vigor produce spectral signatures similar to those of fallow or non-coffee areas. To address these limitations, upcoming research will focus on leveraging a space-time hybrid approach with deep learning and surface phenology modeling. Specifically, we plan to implement a workflow combining the spatial detail of Sentinel-2 with the temporal continuity of HLS.

From satellites to smartphones: opportunities for digital agriculture in apple orchards.

2026-04-22T18:48:57Z

Título: From satellites to smartphones: opportunities for digital agriculture in apple orchards. Autoria: FURUYA, D. E. G.; BOLFE, E. L.; PARREIRAS, T. C. Conteúdo: Mobile phones images of Fuji and Gala apples are being used to evaluate the performance of neural networks for fruit detection, which can serve as the basis for developing future mobile applications not only for fruit detection but also for the assessment of different characteristics and parameters, such as size, color, or quality. Such applications may support farmers and technicians by offering practical and low-cost tools for monitoring orchard conditions.

Estimating Arabica coffee productivity using Planetscope images and artificial neural networks.

2026-04-18T13:52:43Z

Título: Estimating Arabica coffee productivity using Planetscope images and artificial neural networks. Autoria: FAGUNDES, R. B.; KAYSER, L. P.; BENEDETTI, A. C. P.; AMARAL, L. de P.; BOLFE, E. L.; LEMOS, O. L. Conteúdo: Research on orbital remote sensing and artificial neural networks (ANN) for estimating coffee productivity has advanced in Brazil, but applications remain scarce in states such as Bahia. This study aimed to develop and validate a model to estimate Arabica coffee (Coffea arabica L.) yield using PlanetScope images and ANN, based on two commercial areas with contrasting production systems (rainfed and irrigated). Georeferenced productivity samples (2 per hectare) were collected during the 2024 harvest, and PlanetScope images were analyzed from August 2023 to May 2024. In R software, vegetation indices related to vigor, nutritional status, and water stress were calculated, and their minimum, mean, and maximum values were extracted. Most variables followed a normal distribution (Shapiro–Wilk test, p < 0.05), enabling Pearson’s correlation analysis. The best correlation was observed in Santa Vera, where the CCCI index of 08/20/2023 showed a moderate positive correlation with yield (r = 0.66; p = 0.001). After ANN training in SPSS software, the model achieved very high predictive performance in both areas, with R² and adjusted R² above 0.99, low errors (MAE = 0.031–0.072; RMSE = 0.042–0.087; relative RMSE = 0.029–0.043), and almost no bias (BIAS = 0.001; relative BIAS = 0). Among the vegetation indices, CCCI was the most relevant, followed by NDRE, NDVI, NDWI, and EVI. Overall, the findings demonstrate that combining PlanetScope imagery with ANN provides a highly accurate approach for estimating Arabica coffee productivity under different production systems in Bahia.

Calibration and evaluation of a carbon net primary productivity module based on the SIMPLE model.

2026-04-12T01:35:07Z

Título: Calibration and evaluation of a carbon net primary productivity module based on the SIMPLE model. Autoria: COLMANETTI, M. A. A.; OLDONI, H.; ANNIBAL, L.; WOLFF, W.; REJAILI, R. P. A.; BARIONI, L. G.; ALMEIDA, I. R. de; EWING, R. P.; CUADRA, S. V. Conteúdo: This study presents the carbon net primary productivity (CNPP) module, an imple-mentation of the simple generic crop model (SIMPLE model) that we have extendedto predict aboveground and belowground biomass production and carbon assim-ilation. The goal for this module is to predict biomass inputs at and below thesoil surface, during and at the end of crop cycles, while integrated into a broaderenvironmental carbon simulation framework, ProCarbon-Soil. CNPP was initiallyparameterized for soybean [Glycine max (L.) Merr.] and maize (Zea mays L.) usingmicrometeorological data, and for wheat (Triticum aestivum L.), bean (Phaseolusvulgaris L.), and perennial forage [Urochloa (syn. Brachiaria) brizantha (Hochstex A. Rich.) Stapf cv. Marandu] using agrometeorological experimental data and/ordata from the literature. Subsequently, calibrations for reference cultivars were per-formed, grouping cultivars by crop phenological characteristics and edaphoclimaticregions using farm-level data. CNPP accurately simulated leaf area index, evapotran-spiration, and biomass dry matter production and allocation for soybean and maizewhen evaluated at the sites with micrometeorological data (R2 > 0.76, Nash–Sutcliffeefficiency > 0.56, and relative root mean square error < 38% for all variables).Simulations for wheat, bean, and perennial forage exhibited lower performanceowing to lower availability of yield data. Nonetheless, the resulting statistics supportthis module’s efficacy in predicting crop productivity in major Brazilian agricul-tural areas. By employing a reduced and efficient parameter set, the CNPP moduleachieves enhanced performance and enables robust calibration across diverse crops,genotypes, and management schemes in multiple regions.