Resumo em Inglês:

ABSTRACT X-ray fluorescence (XRF) spectroscopy is a promising tool for laboratory soil analysis, and optimizing data acquisition setup enhances operational efficiency while maintaining accuracy. This study evaluated the impact of reducing scanning times and increasing replicates on XRF performance for predicting clay content, exchangeable (ex-) K, and ex-Ca using a dataset of 58 soil samples. Four scenarios were tested —30s+1r, 15s+2r, 10s+3r, and 6s+5r—each maintaining a total instrument time of 30 s. Models were calibrated using 41 samples and validated on the remaining 17. The 10s+3r scenario provided comparable predictive accuracy to longer scanning times, achieving R2 = 0.97 and RPIQ = 3.34 for clay, R2 = 0.55 and RPIQ = 2.09 for ex-Ca, and R2 = 0.67 and RPIQ = 0.81 for ex-K; thus, predictions were “very good” (RPIQ ≥ 2.00) for clay and ex-Ca, while ex-K remained poor across scenarios (RPIQ < 1.40). Despite a higher number of replicates, the 6s+5r scenario showed reduced precision due to increased noise. The 10s+3r configuration optimizes analysis time without compromising accuracy, making it a suitable option for high-throughput soil analysis in laboratory settings. In conclusion, these findings provide insights that can enhance laboratory workflows, reducing time per analysis while maintaining data reliability

Resumo em Inglês:

ABSTRACT In precision agriculture (PA), the evaluation of soil spatial variability to optimize crop management requires dense sampling. This costly activity often results in sparser sampling grids and may compromise both map quality and the return on variable rate fertilizer applications. This study evaluated whether the choice of interpolation method influences the quality of fertilizer prescription maps under sampling limitations. In two areas with different amounts of samples and degrees of spatial dependence, the interpolation of phosphorus (P) and potassium (K) contents was evaluated using deterministic methods (TPS, IDW) that consider mathematical functions based on distance, stochastic methods (OK, KED) that consider the spatial autocorrelation of the data, and machine learning methods (SVM, RFSI) that learn complex relationships. TPS demonstrated superior performance in predicting content, and the agreement of fertilization classes (Lin’s correlation coefficient and Kappa’s agreement index were higher). However, considering the recommendation classes, the differences between the methods were reduced. Thus, if this approach to creating recommendation maps is adopted, the interpolator can be selected based on the simplicity of the method, with TPS being a promising alternative.

Resumo em Inglês:

ABSTRACT Soil spatial variability in vineyards is often underestimated in management planning. This study assessed pedological heterogeneity in a high-altitude vineyard (0.4 ha) by sampling soils adjacent to 68 vines at two depths (0–20 and 20–40 cm) for texture, soil organic carbon, and exchangeable cations, in addition to profile description and classification. Three soil classes were identified (NITOSSOLO BRUNO, LATOSSOLO VERMELHO, and LATOSSOLO BRUNO) with short-range transitions and marked differences in clay, carbon, and fertility. Interpolated maps showed patterns consistent with topography. The results yield information layers useful for soil-management planning and may, in subsequent studies integrating plant and yield data, support the delineation of edaphically homogeneous zones and spatially differentiated recommendations. We integrated edaphic maps with per-vine production metrics and derived operational zones, configuring a proof-of-concept for precision management at the block scale.

Resumo em Inglês:

ABSTRACT Obtaining reliable yield data is a challenge. Optical remote sensing (RS) can be an alternative for yield prediction. However, one of its limitations lies in its wavelength, which captures only information from the top of the crop canopy. In contrast, synthetic aperture radar (SAR) has greater interaction with plants due to its distinct wavelength. Therefore, this study aimed to investigate whether the inclusion of SAR images in a dataset composed of optical vegetation indices (VIs) derived from different acquisition principles could enhance yield prediction performance. Using yield monitor data as reference for oat and sorghum crops, we employed four optical vegetation indices (EVI, TCG, PVI, and SFDVI) from Sentinel-2 and three SAR variables (VH, VV, and DPSVI) from Sentinel-1 at the peak vegetative stage of the crops. In addition, for SAR images, we tested three different backscatter normalizations: σ0, β0, and γ0. Correlation analysis, principal component analysis, and machine learning techniques were applied for prediction using the Random Forest algorithm under multiple scenarios, aiming to compare predictive performance with and without SAR data inclusion. When only one vegetation index was used, the addition of SAR data contributed to yield prediction. However, when multiple optical vegetation indices were employed together, SAR data no longer added predictive power to the model. Thus, for short-stature crops such as oat and sorghum, the inclusion of SAR data does not provide predictive gains; therefore, the use of optical vegetation indices derived from different acquisition principles is sufficient for yield prediction.

Resumo em Inglês:

ABSTRACT Assessing the spatial variability of soil chemical attributes and yield is fundamental to understanding crop responses to site-specific management practices. This study aimed to analyze the spatial structure of soil chemical attributes and carrot yield, as well as their spatial correlation, under conventional management. The experiment was conducted in an irrigated carrot field, where one hundred sampling points were established on a regular 20 × 20 m grid. At each point, carrot yield and soil chemical attributes were evaluated. Statistical and geostatistical analyses were applied to the parameters studied. Average variability was observed in carrot yield. Phosphorus, potassium, electrical conductivity, and potential cation exchange capacity exhibited spatial dependence, with no evidence of nutrient deficiency in the area. The remaining attributes showed no spatial dependence within the 20 m grid, indicating the need for alternative sampling strategies. Among the analyzed variables, potential cation exchange capacity had a direct influence on the spatial variability of carrot yield, showing a positive spatial correlation. These results highlight the importance of spatial characterization of soil chemical attributes as a key tool for supporting precision agriculture.

Resumo em Inglês:

ABSTRACT Animal communication through intranidal vibrations in stingless bees offers promising applications for precision meliponiculture. Monitoring vibrational patterns within nests provides reliable indicators of colony status and can enhance management practices. This study investigated anomalies in the vibrational patterns of a Scaptotrigona aff. depilis colony during an induced interspecific invasion. A piezoelectric sensor connected to an Arduino microcontroller was installed on the inner wall of a rational hive to record surface vibrations. The colony was monitored for two hours: one under normal conditions and another during a simulated invasion, with data collected at 100 readings per second. During the induced event, significant changes in vibrational amplitude and frequency were detected, corresponding to defensive behaviors against intruders. These findings may help meliponiculturists detect colony disturbances early and implement timely interventions to minimize losses. This research also contributes to future studies in biotremology of stingless bees, providing a practical and accessible method that combines scientific precision with field applicability.