Using convolutional neural networks (CNNs) for image recognition is effective for early weed detection. However, the impact of training data curation, specifically concerning morphological changes during the early growth phases of weeds, on recognition robustness remains unclear. We focused on four weed species (giant ragweed [Ambrosia trifida L.], red morningglory [Ipomoea coccinea L.], pitted morningglory [Ipomoea lacunosa L.], and burcucumber [Sicyos angulatus L.]) with varying cotyledon and true leaf shapes. Creating 16 models in total, we employed four dataset patterns with different growth stage combinations, two image recognition algorithms (object detection: You Look Only Once [YOLO] v5 and image classification: Visual Geometry Group [VGG] 19), and two conditions regarding the number of species treated (four and two species). We evaluated the effects of growth stage training on weed recognition success using two datasets. One evaluation revealed superior results with a single class/species training dataset, achieving >90% average precision for detection and classification accuracy under most conditions. The other dataset revealed that merging different growth stages with different shapes as a class effectively prevented misrecognition among different species when using YOLOv5. Both results suggest that integrating different shapes in a plant species as a single class is effective for maintaining robust recognition success amid temporal morphological changes during the early growth stage. This finding not only enhances early detection of weed seedlings but also bolsters the robustness of general plant species identification.

Site-specific weed management (on the scale of a few meters or less) has the potential to greatly reduce pesticide use and its associated environmental and economic costs. A prerequisite for site-specific weed management is the availability of accurate maps of the weed population that can be generated quickly and cheaply. Improvements and cost reductions in unmanned aerial vehicles (UAVs) and camera technology mean these tools are now readily available for agricultural use. We used UAVs to collect aerial images captured in both RGB and multispectral formats of 12 cereal fields (wheat [Triticum aestivum L.] and barley [Hordeum vulgare L.]) across eastern England. These data were used to train machine learning models to generate prediction maps of locations of black-grass (Alopecurus myosuroides Huds.), a prolific weed in UK cereal fields. We tested machine learning and data set resampling methods to obtain the most accurate system for predicting the presence and absence of weeds in new out-of-sample fields. The accuracy of the system in predicting the absence of A. myosuroides is 69% and its presence above 5 g in weight with 77% accuracy in new out-of-sample fields. This system generates prediction maps that can be used by either agricultural machinery or autonomous robotic platforms for precision weed management. Improvements to the accuracy can be made by increasing the number of fields and samples in the data set and the length of time over which data are collected to gather data across the entire growing season.

Controlling weeds is a critically important task in sugarcane production systems. Weeds compete for light, nutrients, and water, and if they are not managed properly can negatively impact sugarcane yields. Accurate detection of weeds versus desired plants was assessed using hyperspectral and pigment analyses. Leaf samples were collected from four commercial Louisiana sugarcane varieties, and nine weed species commonly found in sugarcane fields. Hyperspectral leaf reflectance data (350 to 850 nm) were collected from all samples. Plant pigment (chlorophylls and carotenoids) levels were also determined using high-performance liquid chromatography, and concentrations were determined using authentic standards and leaf area. In all cases, leaf reflectance data successfully differentiated sugarcane from weeds using canonical discrimination analysis. Linear discriminant analysis showed that the accuracy of the classification varied from 67% to 100% for individual sugarcane varieties and weed species. In all cases, sugarcane was not misclassified as a weed. Plant pigment levels exhibited marked differences between sugarcane varieties and weed species with differences in chlorophyll and carotenoid explaining much of the observed variation in reflectance. The ratio of chlorophyll a to chlorophyll b showed significant differences between sugarcane and all weed species. The successful implementation of this technology as either an airborne system to scout and map weeds or a tractor-based system to identify and spray weeds in real-time would offer sugarcane growers a valuable tool for managing their crops. By accurately targeting weeds in sugarcane fields that are emerged and growing, the total amount of herbicide applied could be decreased, resulting in cost savings for the grower and reduced environmental impacts.

Site-specific weed management using open-source object detection algorithms could accurately detect weeds in cropping systems. We investigated the use of object detection algorithms to detect Palmer amaranth (Amaranthus palmeri S. Watson) in soybean [Glycine max (L.) Merr.]. The objectives were to (1) develop an annotated image database of A. palmeri and soybean to fine-tune object detection algorithms, (2) compare effectiveness of multiple open-source algorithms in detecting A. palmeri, and (3) evaluate the relationship between A. palmeri growth features and A. palmeri detection ability. Soybean field sites were established in Manhattan, KS, and Gypsum, KS, with natural populations of A. palmeri. A total of 1,108 and 392 images were taken aerially and at ground level, respectively, between May 27 and July 27, 2021. After image annotation, a total of 4,492 images were selected. Annotated images were used to fine-tune open-source faster regional convolutional (Faster R-CNN) and single-shot detector (SSD) algorithms using a Resnet backbone, as well as the “You Only Look Once” (YOLO) series algorithms. Results demonstrated that YOLO v. 5 achieved the highest mean average precision score of 0.77. For both A. palmeri and soybean detections within this algorithm, the highest F1 score was 0.72 when using a confidence threshold of 0.298. A lower confidence threshold of 0.15 increased the likelihood of species detection, but also increased the likelihood of false-positive detections. The trained YOLOv5 data set was used to identify A. palmeri in a data set paired with measured growth features. Linear regression models predicted that as A. palmeri densities increased and as A. palmeri height increased, precision, recall, and F1 scores of algorithms would decrease. We conclude that open-source algorithms such as YOLOv5 show great potential in detecting A. palmeri in soybean-cropping systems.

The widespread use of herbicides in cropping systems has led to the evolution of resistance in major weeds. The resultant loss of herbicide efficacy is compounded by a lack of new herbicide sites of action, driving demand for alternative weed control technologies. While there are many alternative methods for control, identifying the most appropriate method to pursue for commercial development has been hampered by the inability to compare techniques in a fair and equitable manner. Given that all currently available and alternative weed control methods share an intrinsic energy consumption, the aim of this review was to compare methods based on energy consumption. Energy consumption was compared for chemical, mechanical, and thermal weed control technologies when applied as broadcast (whole-field) and site-specific treatments. Tillage systems, such as flex-tine harrow (4.2 to 5.5 MJ ha−1), sweep cultivator (13 to 14 MJ ha−1), and rotary hoe (12 to 17 MJ ha−1) consumed the least energy of broadcast weed control treatments. Thermal-based approaches, including flaming (1,008 to 4,334 MJ ha−1) and infrared (2,000 to 3,887 MJ ha−1), are more appropriate for use in conservation cropping systems; however, their energy requirements are 100- to 1,000-fold greater than those of tillage treatments. The site-specific application of weed control treatments to control 2-leaf-stage broadleaf weeds at a density of 5 plants m−2 reduced energy consumption of herbicidal, thermal, and mechanical treatments by 97%, 99%, and 97%, respectively. Significantly, this site-specific approach resulted in similar energy requirements for current and alternative technologies (e.g., electrocution [15 to 19 MJ ha−1], laser pyrolysis [15 to 249 MJ ha−1], hoeing [17 MJ ha−1], and herbicides [15 MJ ha−1]). Using similar energy sources, a standardized energy comparison provides an opportunity for estimation of weed control costs, suggesting site-specific weed management is critical in the economically realistic implementation of alternative technologies.

Chemical weed control remains a widely used component of integrated weed management strategies because of its cost-effectiveness and rapid removal of crop pests. Additionally, dicamba-plus-glyphosate mixtures are a commonly recommended herbicide combination to combat herbicide resistance, specifically in recently commercially released dicamba-tolerant soybean and cotton. However, increased spray drift concerns and antagonistic interactions require that the application process be optimized to maximize biological efficacy while minimizing environmental contamination potential. Field research was conducted in 2016, 2017, and 2018 across three locations (Mississippi, Nebraska, and North Dakota) for a total of six site-years. The objectives were to characterize the efficacy of a range of droplet sizes [150 µm (Fine) to 900 µm (Ultra Coarse)] using a dicamba-plus-glyphosate mixture and to create novel weed management recommendations utilizing pulse-width modulation (PWM) sprayer technology. Results across pooled site-years indicated that a droplet size of 395 µm (Coarse) maximized weed mortality from a dicamba-plus-glyphosate mixture at 94 L ha–1. However, droplet size could be increased to 620 µm (Extremely Coarse) to maintain 90% of the maximum weed mortality while further mitigating particle drift potential. Although generalized droplet size recommendations could be created across site-years, optimum droplet sizes within each site-year varied considerably and may be dependent on weed species, geographic location, weather conditions, and herbicide resistance(s) present in the field. The precise, site-specific application of a dicamba-plus-glyphosate mixture using the results of this research will allow applicators to more effectively utilize PWM sprayers, reduce particle drift potential, maintain biological efficacy, and reduce the selection pressure for the evolution of herbicide-resistant weeds.

Site-specific weed management recommendations require knowledge of weed species, density, and location in the field. This study compared several sampling techniques to estimate weed density and distribution in two 65-ha no-till Zea mays–Glycine max rotation fields in eastern South Dakota. The most common weeds (Setaria viridis, Setaria glauca, Cirsium arvense, Ambrosia artemisiifolia, and Polygonum pensylvanicum) were counted by species in 0.1-m2 areas on a 15- by 30-m (1,352 points in each field) or 30- by 30-m (676 points in each field) grid pattern, and points were georeferenced and data spatially analyzed. Using different sampling approaches, weed populations were estimated by resampling the original data set. The average density for each technique was calculated and compared with the average field density calculated from the all-point data. All weeds had skewed population distributions with more than 60% of sampling points lacking the specific weed, but very high densities (i.e., > 100 plants m−2) were also observed. More than 300 random samples were required to estimate densities within 20% of the all-point means about 60% of the time. Sampling requirement increased as average density decreased. The W pattern produced average species densities that often were similar to the field averages, but information on patch location was absent. Weed counts taken on the 15- by 30-m grid were dependent spatially and weed contour maps were developed. Kriged maps presented both density and location of weed patches and could be used to establish management zones. However, grid-sampling production fields on a small enough scale to obtain spatially dependent data may have limited usefulness because of time, cost, and labor constraints.

The objective of this research was to determine the potential use of commercially available multispectral images to detect weeds at low densities during the critical period of weed control. Common lambsquarters seedlings were transplanted into plots of glyphosate-resistant corn at 0, 1, 2, and 4 plants/m2 at two sites, Agronomy Center for Research and Extension (ACRE) and Meig's Horticultural Research Farm at the Throckmorton–Purdue Agricultural Center (TPAC), in Indiana. Aerial multispectral images (12 to 16 cm pixel resolution) were taken 18 and 32 days after planting (DAP) at ACRE and 19 and 32 DAP at TPAC. Corn and common lambsquarters could not be reliably detected and differentiated at either site when weeds were 9 cm or less in height. However, economic threshold densities (2 and 4 plants/m2) of common lambsquarters could be distinguished from weed-free plots at TPAC when weeds were 17 cm in height. At this height, common lambsquarters plants were beyond the optimal height for glyphosate application, but could still be readily controlled with higher rates. Results from this study indicate that commercially available multispectral aerial imagery at current spatial resolutions does not provide consistently reliable data for detection of early season weeds in glyphosate-resistant corn cropping systems. Additional refinement in sensor spatial and spectral resolution is necessary to increase our ability to successfully detect early season weed infestations.

Weeds are often spatially aggregated in maize fields, and the level of aggregation varies across and within fields. Several annual weed species are present in maize fields before postemergence herbicide application, and herbicides applied will control several species at a time. The goal of this study was to assess the spatial distribution of multispecies weed infestation in maize fields. Ground-based imagery was used to map weed infestations in rain-fed maize fields. Image segmentation was used to extract weed cover information from geocoded images, and an expert-based threshold of 0.102% weed cover was used to generate maps of weed presence/absence. From 19 site-years, 13 (68%) demonstrated a random spatial distribution, whereas six site-years demonstrated an aggregated spatial pattern of either monocotyledons, dicotyledons, or both groups. The results of this study indicated that monocotyledonous and dicotyledonous weed groups were not spatially segregated, but discriminating these weed groups slightly increased the chances of detecting an aggregated pattern. It was concluded that weeds were not always spatially aggregated in maize fields. These findings emphasize the need for techniques allowing the assessment of weed aggregation prior to conducting site-specific weed management.

Weed management decisions based on weed threshold models offer the opportunity to reduce herbicide use by allowing the possibility of forgoing treatment or lowering rates. Weed thresholds based on a relative leaf-cover model were tested during a 4-yr period at two locations. Two 1.62-ha fields, planted to conventional and glyphosate-resistant corn (2004, 2005, 2007) or soybean (2006), were divided in 900 m2 sections. Herbicides were applied postemergence to each of these sections with either variable rates based on weed thresholds, or constant full rates. Variable herbicide rates included: no application, half rate, or full rate. Relative weed cover values of 0.2 and 0.4 (corn) or 0.1 and 0.3 (soybean) served as thresholds for incremental rates. Digital images were used to evaluate the relative weed cover. Weed density was assessed before and after herbicide application. Weed seed production was estimated for two species in 2004 and 2005. No difference in crop yield, relative weed cover, weed density, or weed seed production was observed between conventional and glyphosate-resistant cropping systems. During the first year, herbicide use reduction was obtained (−85.4%) with marginal crop yield loss (5 to 15%). In the subsequent 3 yr, preherbicide weed densities increased and concomitant increases in relative weed cover values did not allow more than a 10% overall reduction in herbicide use. This threshold model designed to maintain crop yields within a given year did not allow significant reduction in herbicide use during the following 3 yr. Residual weed populations most likely replenished the seed bank to levels that allowed weed densities to increase afterward. Increased weed density over time in plots treated with full rates of herbicide every year also indicated that a single postemergence herbicide treatment was not sufficient to contain weed populations at low levels every year in this corn–soybean rotation.

Managing weed infestations in a spatially precise manner requires accurate and cost-effective weed identification techniques. The goal of our research was to quantify the accuracy of continuous weed presence–absence maps and assess how management based on those maps may affect producer net returns. Each continuous sampled map covered the entire field and contained vector polygons labeled as either wild oat presence or wild oat absence. The accuracy of the continuous wild oat maps at each sampling time was determined from georeferenced quadrats of wild oat densities. The accuracy of the continuous wild oat seedling maps ranged from 48.3 to 87.1% among the six site-years. The accuracy of the wild oat seedling maps improved by at least 8% when a 10-m buffer was included around areas mapped as wild oat presence. The accuracy of continuous wild oat panicle maps from the combine at harvest ranged from 65.8 to 90.9% among the six site-years. The variation in accuracy for the wild oat seedling maps among sites was greater than the accuracy of the panicle maps. Net returns ($ ha−1) for four site-years were calculated and compared for four possible weed management approaches on each field. A site-specific herbicide application to areas mapped as wild oat presence always generated higher net returns than a herbicide application over the entire field for four sites. A site-specific herbicide application to areas mapped as wild oat presence plus a surrounding 10-m buffer area only resulted in a higher net return in one of the 12 site-years compared with a site-specific herbicide application without the 10-m buffer. This site had the lowest (48.3%) wild oat seedling map accuracy, and uncontrolled wild oat had a high-yield effect. This research indicates that using a continuous weed sampling method based on presence or absence for site-specific herbicide application can be profitable over a herbicide application to the entire field, even with the associated technology cost and seedling map errors.

The implementation of site-specific weed management requires information about weed cover and decision support systems to determine weed cover thresholds and concomitant herbicide rates. Although it is possible to create accurate weed cover maps over large areas, weed cover thresholds have generally been evaluated using tedious weed density counts. To bridge this gap between weed cover obtained by machine vision and the concept of economic threshold, crop advisers specializing in weed scouting were asked to evaluate over 2,500 weed cover images (2 m by 3 m) and determine if a given image would require herbicide application or not. Using the area under the “receiver operating characteristic” curve method, an optimal weed cover threshold was established. The derived economic thresholds ranged from 0.06 to 0.31% weed cover contingent on the level of tolerance of the expert adviser. Although this threshold seems low, it is comparable with economic threshold values based on weed density.

Field research was conducted to determine the potential of hyperspectral and multispectral imagery for late-season discrimination and mapping of grass weed infestations in wheat. Differences in reflectance between weed-free wheat and wild oat, canarygrass, and ryegrass were statistically significant in most 25-nm-wide wavebands in the 400- and 900-nm spectrum, mainly due to their differential maturation. Visible (blue, B; green, G; red, R) and near infrared (NIR) wavebands and five vegetation indices: Normalized Difference Vegetation Index (NDVI), Ratio Vegetation Index (RVI), R/B, NIR-R and (R − G)/(R + G), showed potential for discriminating grass weeds and wheat. The efficiency of these wavebands and indices were studied by using color and color-infrared aerial images taken over three naturally infested fields. In StaCruz, areas infested with wild oat and canarygrass patches were discriminated using the indices R, NIR, and NDVI with overall accuracies (OA) of 0.85 to 0.90. In Florida–West, areas infested with wild oat, canarygrass, and ryegrass were discriminated with OA from 0.85 to 0.89. In Florida–East, for the discrimination of the areas infested with wild oat patches, visible wavebands and several vegetation indices provided OA of 0.87 to 0.96. Estimated grass weed area ranged from 56 to 71%, 43 to 47%, and 69 to 80% of the field in the three locations, respectively, with per-class accuracies from 0.87 to 0.94. NDVI was the most efficient vegetation index, with a highly accurate performance in all locations. Our results suggest that mapping grass weed patches in wheat is feasible with high-resolution satellite imagery or aerial photography acquired 2 to 3 wk before crop senescence.

This study describes the distribution patterns of Johnsongrass populations present in 38 commercial corn fields located in three major corn growing regions of Spain. A total of 232 ha were visually assessed from the cabin of a combine during harvesting using a three-category ranking (high density, low density, no presence) and recording the georeferenced data in a tablet personal computer. On average, 10.3 and 3.9% of the surveyed area were infested with high and low density of Johnsongrass, respectively. Most of the infested area was concentrated in a few large patches with irregular shape. Small patches (less than 1,000 m2) represented only 27% of the infested area. Management factors could explain much of the spatial distribution of this weed in the studied fields. Tillage direction was the main factor explaining patch shape: the length width−1 ratio of the patches was greater than two in the tillage direction. In sprinkler irrigated fields, higher levels of infestation were generally observed close to the sprinkler lines. Areas close to the edges of the field had a higher risk of infestation than the areas in the middle of the fields: a negative relationship between distance from the edge and weed abundance was established. Because a few patches, located in some predictable parts of the field, such as field edges, represent most of the seriously infested area, site-specific treatments of these areas could reduce herbicide inputs, until more reliable, spatially precise and practical detection, mapping, and spraying systems are developed.

Weeds generally occur in patches in production fields. Are these patches spatially and temporally stable? Do management recommendations change on the basis of these data? The population density and location of annual grass weeds and common ragweed were examined in a 65-ha corn/soybean production field from 1995 to 2004. Yearly treatment recommendations were developed from field means, medians, and kriging grid cell densities, using the hyperbolic yield loss (YL) equation and published incremental YL values (I), maximum YL values (A), and YL limits of 5, 10, or 15%. Mean plant densities ranged from 12 to 131 annual grasses m−2 and < 1 to 37 common ragweed m−2. Median weed densities ranged from 0 to 40 annual grasses m−2 and were 0 for common ragweed. The grass I values used to estimate corn YL were 0.1 and 2% and treatment was recommended in only 1 yr when the high I value and either the mean or median density was used. The grass I values used for soybean were 0.7 and 10% and estimated YL was over 10% all years, regardless of I value. The common ragweed I values were 4.5 and 6% for corn and 5.1 and 15.6% for soybean. On the basis of mean densities, fieldwide treatment would have been recommended in 6 of 9 yr but in no years when the median density was used. Recommendations on the basis of grid cell weed density and kriging ranged from > 80% of the field treated for grass weeds in 3 of 4 yr in soybean to < 20% of the field treated for common ragweed in 2002 and 2004 (corn). Grass patches were more stable in time, space, and density than common ragweed patches. Population densities and spatial distribution generally were variable enough so that site-specific information within this field would improve weed management decisions.

Information on weed spatial distribution could improve weed management decisions. Herbicide use could be reduced if applied only to field zones with an infestation level higher than a specific economic threshold. In this study, we surveyed 31 winter barley fields in different regions of Spain to describe the spatial distribution of sterile oat and to analyze the relationship between sterile oat infestation level and landscape and crop yield attributes. Elevation and slope angle, crop yield and slope aspect were the main factors in order of importance in explaining the distribution of sterile oat. In general, greater infestation levels were observed in flat lowland and concave landscapes, with a low crop yield, and on northern exposures (when slope ≥ 2%). We could define field zones with a higher risk (> 6 times probability) of having sterile oat problems. High-probability zones, defined by topographic attributes alone, occupied 24% of the total area and contained 46% of the high infestation levels, whereas zones defined by topographic and crop yield attributes constituted 14% of the total area and contained 31% of the infestation.

At the field scale, weeds generally appear aggregated rather than randomly distributed, and this aggregation is linked to the spatial heterogeneity of biotic and abiotic factors. Crop management practices shape the spatial pattern of weed infestations by modifying certain factors having an impact on weed emergence and growth. Although crop seeding is often the last in-field disturbance before crop and weed emergence, its effect on the distribution of weeds has received little attention in the literature. The purpose of this study was to assess the influence of the planting operation on weed cover and presence in corn fields using digital images to investigate the possibility of sensing the interrow to infer the presence or absence of weeds on the corn row. A total of 18 site-years under conventional tillage treated with a single POST application of herbicide were selected across seven locations. Image analysis, at the V2 to V4 growth stage of corn, was used to compare the weed cover in three zones: the undisturbed interrows, the corn rows, and the interrows compacted by tractor wheel traffic. For 61% of site-years, there was no significant difference among the zones. When there was a significant difference compared with the other two zones, the undisturbed interrow was usually less infested. Point-to-point comparisons of weed presence or absence (based on a threshold of five pixels) between the interrow and the corn row revealed 70 or 73% correspondence, depending on the type of interrow (undisturbed or tracked). However the error of inference of the corn row weed cover generated by sensing only adjacent interrows may be too high for efficient commercial weed control.

Nascent research on the profitability of site-specific weed management has focused on reduced herbicide use, ignoring significant information costs for scouting, making treatment maps, and patch herbicide application. Including these information costs results in few, if any, studies, fully covering added costs with herbicide savings. Real-time, sensor-driven site-specific herbicide management promises to overcome many of the scouting and map-making costs so long as weed species recognition accuracy continues to improve and commercial capital costs to decline. Two means to increase revenues are possible from site-specific weed management: value-added crop products (e.g., herbicide-free, nontransgenic crops) and environmental stewardship payments from government programs. Site-specific ecological weed management offers a more fruitful means to garner both higher prices from “green” consumers and revenues for sound environmental stewardship. It entails using spatial tools to enhance habitat for weed predators and diseases and their vectors, as well as to grow crops that are better able to compete with weeds. The most economically promising areas for future weed management will draw upon both spatial and developmental management tools.

One approach to site-specific weed control is to map weeds within a field and then divide the field area into smaller grid units. The decision to apply a herbicide to individual grid units, or decision units, is made by using yield loss models to establish an economic threshold level. However, decision units often contain weed populations with aggregated distributions. Many yield loss models have not considered this because experiments dealing with weed–crop competition typically assume uniform weed distributions. Therefore, these models may overestimate yield losses. Field experiments conducted in 1999 and 2000 compared the effects of common ragweed having a uniform distribution vs. an aggregated distribution on soybean seed yield, moisture content, and dockage. Field experiment data were used to calculate and compare economic thresholds for both distributions. Economic thresholds that considered drying costs and dockage also were compared. There was no significant difference in I parameters (yield loss as density approaches zero) between the two ragweed distributions in either year. Seed moisture content and dockage increased with increasing common ragweed densities, but increases were not significant at the break-even yield loss level. Economic threshold values were similar for both distributions with differences between aggregated and uniform of 0.14 and 0.01 plants m−2 in 1999 and 2000, respectively. The economic threshold values were reduced by 0.01 to 0.06 plants m−2 when drying costs and dockage were considered.

The overall objective of this study was to identify common patterns in the spatial distribution of the major weed species present in the corn-growing region of central Spain, exploring the scale dependence of these patterns and the possible associations or dissociations between individual species. Weed density was assessed in 16 commercial fields using digital images acquired in a 9-m by 9-m sampling grid. A set of six species was found in all the fields: black nightshade, common cocklebur, fierce thornapple, johnsongrass, purple nutsedge, and velvetleaf. Spatial analysis by distance indices and inverse distance weighting interpolation methods were performed to create weed distribution maps. The results showed aggregated spatial distribution patterns for all individual species regardless their life cycle, annual or perennial. Some associations and dissociations among species were found in the analysis of interactions. Nevertheless, the spatial patterns of co-occurrence of weed species were field-specific and therefore cannot be considered general patterns of weed co-occurrence. In order to explore the scale dependence of these results, an additional study was conducted in an experimental field located in the same area using a 1.0-m by 0.75-m sampling grid. Although this resolution allowed for a better definition of the positions of the weed patches and weed-free gaps, the results obtained revealed similar patterns to those observed with a coarser sampling resolution.