Weeds are unwanted and invasive plants characterized by their rapid growth and ability to compete with crops for essential resources such as space, water, nutrients, and sunlight. This competition has a negative impact on crop quality and productivity. To reduce the influence of weeds, precision weeding is used, which uses image sensors and computational algorithms to identify plants and classify weeds using digital images. This study used images of maize (Zea mays L.) to detect four types of weeds (Lolium rigidum, Sonchus oleraceus, Solanum nigrum, and Poa annua). For this purpose, YOLO (You Only Look Once) architectures, YOLOv8s, YOLOv9s, YOLOv10s, and YOLOv11s versions, were trained and compared, along with an architecture based on RT-DETR (Real-Time Detection Transformer), version RT-DETR-1. The YOLO architectures are noted for their real-time detection efficiency, and RT-DETR-l allows evaluation of the impact of an architecture that dispenses with Non-Maximum Suppression (NMS). The YOLOv9s model had the best overall performance, achieving a mAP@0.5 of 0.834 in 60 epochs and an F1-score of 0.78, which demonstrates a optimal balance between accuracy and recall, although with less confidence in its predictions. On the other hand, the RT-DETR-l model stood out for its efficiency in convergence, reaching a competitive performance in only 58 epochs with a mAP@0.5 of 0.828 and an F1-score of 0.80.

The penetration resistance (PR) is a soil attribute that allows identifies areas with restrictions due to compaction, which results in mechanical impedance for root growth and reduced crop yield. The aim of this study was to characterize the PR of an agricultural soil by geostatistical and multivariate analysis. Sampling was done randomly in 90 points up to 0.60 m depth. It was determined spatial distribution models of PR, and defined areas with mechanical impedance for roots growth. The PR showed a random distribution to 0.55 and 0.60 m depth. PR in other depths analyzed showed spatial dependence, with adjustments to exponential and spherical models. The cluster analysis that considered sampling points allowed establishing areas with compaction problem identified in the maps by kriging interpolation. The analysis with main components identified three soil layers, where the middle layer showed the highest values of PR.

Growing demand for food exerts pressure on natural resources and may lead to the expansion of agricultural frontiers in developing countries. Most of this pressure appears in tropical zones, in native savannahs, with naturally infertile soils prone to degradation. Crop management in these regions is based on generalized estimates, leaving aside the inherent soil variability, leading to low production efficiency and high risk of environmental damage. This study aims at determining the spatial variability of some chemical properties, including organic carbon, pH, exchangeable acidity, exchangeable aluminum, P, Ca, Mg, K and Na for two Oxisols with different levels of agricultural intervention, in Puerto Lopez, Colombia, in order to identify guidelines for site-specific management. A forty-two point grid (25´ 25 m) was established for samplings at two depths: 0-100 and 100-200 mm. Descriptive statistics and geostatistics were used to analyze soil properties spatial dependence. Variogram models were obtained and from them maps of properties were drawn using ordinary punctual kriging. The results showed that spatial variability of the soil chemical properties depends upon the use of amendments, fertilizing methods, tillage and the inherent characteristics of each variable analyzed. A greater influence of the agricultural intervention on spatial variability was evident in the upper 100 mm of soil. Spatial dependence was found for most of the studied soil properties. However, K and Na presented variograms with pure nugget effects and/or very short ranges. The information generated is a base to derive guidelines for site-specific agriculture

O objetivo deste trabalho foi estudar o comportamento da variabilidade espacial da resistência mecânica à penetração (RP), da densidade do solo, da densidade de partículas do solo, da porosidade total e dos teores de areia, silte e argila, e as relações existentes entre as propriedades analisadas. O estudo foi desenvolvido em Puerto López (Meta-Colômbia). As amostras do solo foram coletadas em duas profundidades (0 - 0,10 e 0,10 - 0,20 m), em intervalos regulares de 25 m, em malha regular, compreendendo 42 pontos, sendo a RP medida até 0,42 m de profundidade. A RP não se correlacionou com nenhuma propriedade, na primeira profundidade. De 0,10 - 0,20 m de profundidade, a RP correlacionou-se com os teores de silte e areia. Os semivariogramas da RP não apresentaram correlação espacial nas duas profundidades, possivelmente devido ao sistema de preparo do solo, que modifica a dependência espacial na superfície, acarretando maior homogeneidade do solo. As propriedades ajustaram-se a modelos de semivariogramas exponenciais e esféricos. Para a profundidade de 0,10 - 0,20 m, as propriedades não se ajustaram a modelos de semivariogramas, apresentando efeito pepita puro, ou seja, não apresentaram dependência espacial, à exceção da densidade aparente. The objective of the present work was to determine the behavior of the spatial variability of the penetration resistance (RP), bulk density, soil density, total porosity and the sand, lime and clays contents, as well as the relation between these properties. The present study was conduced in Puerto López (Meta-Colombia). Soil samples were collected in a regular grid, comprising 42 points, spaced at separation distances of 25 m between points, for two depths (0,00 - 0,10 and 0,10 - 0,20 m). The RP was analyzed up to the depth of 0,42 m. The RP did not correlate with any property in the first depth. Between 0.10 - 0.20 m depth, the RP were correlated with the lime and sand contents. The RP semivariograms indicates no spatial correlation in both depths, possibly due to soil tillage practices, which modifies the superficial spatial structure, gradually inducing to soil homogeneity. The properties adjusted mainly to exponential and spherical semivariograms models. For 0,10 - 0,20 m depth, the properties presented nugget effect, without spatial correlation, except for the bulk density

El propósito de este estudio fue realizar la determinación de las propiedades granulométricas, funcionales y de color de la harina de quinua y de chontaduro. La quinua y el chontaduro poseen grandes atributos nutricionales, sensoriales y son substitutos potenciales del trigo en la elaboración de diversos alimentos. Sin embargo, existen pocas publicaciones sobre las propiedades funcionales y físicas de sus harinas. Sus propiedades fueron determinadas y a los datos se aplica estadística descriptiva, ANOVA y la prueba de rango múltiple Duncan (p<0,05). Los valores de las propiedades granulométricas, funcionales y de color son influenciados por el tipo de harina. La harina de chontaduro presenta propiedades funcionales más favorables que la harina de quinua para la elaboración de alimentos como salchichas, pastas, queso procesado y productos de panadería ya que presenta un mayor potencial de gelatinización de almidones, y mejores propiedades de color.

<![CDATA[Studies that involve soil attributes variability help the analysis of the behavior of hydrophysical attributes in space, to support the decision-making process within the conception of precision agriculture. The aim of the present paper was to characterize the spatial variability of the surface infiltration and its relation to some soil physical attributes. The sampling was done with a rectangular mesh of 64 points with perpendicular distances of 52 m x 45 m between sampling points. The attributes studied were bulk density, contents of sand, silt and clay, soil moisture, cumulative infiltration, basic infiltration, sorptivity, and saturated hydraulic conductivity (Ks). Data were analyzed using descriptive statistics and geostatistics, identifying the relationship between attributes, using the Pearson correlation and cluster analysis. Results suggest a behavior of parameters close to normal, attributes related to soil water movement had greater variability, with low basic infiltration (1.46 cm h-1) and Ks (0.80 cm h-1), which together with high values of bulk density (1.87 g cm-3) show limitations by compaction. Cumulative infiltration and Ks attributes did not fit to models of spatial distribution, finding ranges of 150 m for the other attributes, distance to optimize future sampling for the study of spatial variability. Pearson correlation and cluster analysis allowed establishing relationships between soil attributes; spatial relationships were corroborated in contour maps, where we observed the influence of clay and sand content in the behavior of bulk density, and the influence of these attributes in processes of water infiltration into the soil.]]>