Site-specific Herbicide Applications Research Articles

Time-Efficient Low-Resolution RGB Aerial Imaging for Precision Mapping of Weed Types in Site-Specific Herbicide Application

An efficient method for weed detection and the precise generation of spraying maps is crucial to optimising weed management strategies and minimising the costs associated with herbicide use. This study presents a method that leverages low-resolution UAV images to accurately detect and map weeds in sugarcane fields. The proposed approach integrates RGB images, vegetation indices and the HSV colour space, enhancing segmentation through histogram equalization (HE) and object-based image analysis (OBIA). The models developed in this study demonstrate exceptional performance in weed detection, with the most suitable dataset, achieving a detection capability of 95.78% for Broad-leaved Weeds (BLW) and the highest accuracy of 94.45% for Narrow-leaved Weeds (NLW). When considering multiple targets such as BLW, NLW, soil and sugarcane, the models exhibit a detection accuracy of 89.69%. Furthermore, the precision spraying maps generated by the coverage model method (CM) demonstrate remarkable accuracy, reaching 97.50% for weed control using agricultural drones. This method offers an efficient, cost-effective, and timely solution for precise weed detection, leading to improved weed control outcomes by enabling the selection of appropriate chemical substances tailored to each weed species. It reduces repetitive spraying costs and minimises chemical usage through spot spraying.

Smart sprayer a technology for site-specific herbicide application

Reducing the use of chemical-synthetic herbicides with site-specific application technology will play an important role in agricultural research and practice in the coming years and decades. This paper presents a sprayer for site-specific weed control in row crops using weed detection technology based on image analysis and spot application using weed detection data. Two small plot trials in sugar beet and maize in 2021 and seven On-farm field trials in sugar beet, sunflower and maize in 2022 are presented. A pair-wise comparison of broadcast application, untreated control and spotting application were realized. A zero-tolerance strategy for weeds was used as a spotting application in all experiments. Herbicide savings from 10% to 55% were realized without a significant decrease in efficacy in weed control for the spotting application compared to the broadcast application. The weed control efficacy for broadcast application varied from 70 to 100%, while for the spotting application 72–99%. Weed control between broadcast and spotting did not differ significantly within a trial, except one experiment. Application mistakes caused the partly observed weed control of under 80%. Yield tends to be higher for all spotting treatments compared to broadcast application. The presented system provides a robust spotting technology for effective weed management with lower environmental risk and a positive effect on biodiversity.

Detection of Weeds Growing in Alfalfa Using Convolutional Neural Networks

Alfalfa (Medicago sativa L.) is used as a high-nutrient feed for animals. Weeds are a significant challenge that affects alfalfa production. Although weeds are unevenly distributed, herbicides are broadcast-applied in alfalfa fields. In this research, object detection convolutional neural networks, including Faster R-CNN, VarifocalNet (VFNet), and You Only Look Once Version 3 (YOLOv3), were used to indiscriminately detect all weed species (1-class) and discriminately detect between broadleaves and grasses (2-class). YOLOv3 outperformed other object detection networks in detecting grass weeds. The performances of using image classification networks (GoogLeNet and VGGNet) and object detection networks (Faster R-CNN and YOLOv3) for detecting broadleaves and grasses were compared. GoogLeNet and VGGNet (F1 scores ≥ 0.98) outperformed Faster R-CNN and YOLOv3 (F1 scores ≤ 0.92). Classifying and training various broadleaf and grass weeds did not improve the performance of the neural networks for weed detection. VGGNet was the most effective neural network (F1 scores ≥ 0.99) tested to detect broadleaf and grass weeds growing in alfalfa. Future research will integrate the VGGNet into the machine vision subsystem of smart sprayers for site-specific herbicide applications.

OpenWeedLocator (OWL): an open-source, low-cost device for fallow weed detection

The use of a fallow phase is an important tool for maximizing crop yield potential in moisture limited agricultural environments, with a focus on removing weeds to optimize fallow efficiency. Repeated whole field herbicide treatments to control low-density weed populations is expensive and wasteful. Site-specific herbicide applications to low-density fallow weed populations is currently facilitated by proprietary, sensor-based spray booms. The use of image analysis for fallow weed detection is an opportunity to develop a system with potential for in-crop weed recognition. Here we present OpenWeedLocator (OWL), an open-source, low-cost and image-based device for fallow weed detection that improves accessibility to this technology for the weed control community. A comprehensive GitHub repository was developed, promoting community engagement with site-specific weed control methods. Validation of OWL as a low-cost tool was achieved using four, existing colour-based algorithms over seven fallow fields in New South Wales, Australia. The four algorithms were similarly effective in detecting weeds with average precision of 79% and recall of 52%. In individual transects up to 92% precision and 74% recall indicate the performance potential of OWL in fallow fields. OWL represents an opportunity to redefine the approach to weed detection by enabling community-driven technology development in agriculture.

Spectral differentiation of sugarcane from weeds

Site-specific application of herbicides is highly desirable for optimising its usage and reducing environmental damages. Thus, developing techniques for identification and mapping of weeds is necessary for a proper precision agriculture adoption. Such weed identification for site-specific management is difficult when the main crop is already established in the field. This study shows the possibility of differentiating sugarcane plants from weeds by the spectral behaviour of the leaves. The performance of two modelling methods, SIMCA (soft independent modelling by class analogy) and the RF algorithm (random forest) was compared. The simplification of the Vis-NIR spectrum into only four bands of interest (500–550 nm; 650–750 nm; 1300–1450 nm; and 1800–1900 nm) was verified by demonstrating they had the same differentiation ability as the full visible-near infra-red spectrum. Thus, it was shown that performing the proper band selection and local calibration using a spectral classification approach may allow weed mapping and facilitate localised herbicide application.

Multi-Temporal Site-Specific Weed Control of Cirsium arvense (L.) Scop. and Rumex crispus L. in Maize and Sugar Beet Using Unmanned Aerial Vehicle Based Mapping

Sensor-based weed mapping in arable fields is a key element for site-specific herbicide management strategies. In this study, we investigated the generation of application maps based on Unmanned Aerial Vehicle imagery and present a site-specific herbicide application using those maps. Field trials for site-specific herbicide applications and multi-temporal image flights were carried out in maize (Zea mays L.) and sugar beet (Beta vulgaris L.) in southern Germany. Real-time kinematic Global Positioning System precision planting information provided the input for determining plant rows in the geocoded aerial images. Vegetation indices combined with generated plant height data were used to detect the patches containing creeping thistle (Cirsium arvense (L.) Scop.) and curled dock (Rumex crispus L.). The computed weed maps showed the presence or absence of the aforementioned weeds on the fields, clustered to 9 m × 9 m grid cells. The precision of the correct classification varied from 96% in maize to 80% in the last sugar beet treatment. The computational underestimation of manual mapped C. arvense and R. cripus patches varied from 1% to 10% respectively. Overall, the developed algorithm performed well, identifying tall perennial weeds for the computation of large-scale herbicide application maps.

A True-Color Sensor and Suitable Evaluation Algorithm for Plant Recognition.

Plant-specific herbicide application requires sensor systems for plant recognition and differentiation. A literature review reveals a lack of sensor systems capable of recognizing small weeds in early stages of development (in the two- or four-leaf stage) and crop plants, of making spraying decisions in real time and, in addition, are that are inexpensive and ready for practical use in sprayers. The system described in this work is based on free cascadable and programmable true-color sensors for real-time recognition and identification of individual weed and crop plants. The application of this type of sensor is suitable for municipal areas and farmland with and without crops to perform the site-specific application of herbicides. Initially, databases with reflection properties of plants, natural and artificial backgrounds were created. Crop and weed plants should be recognized by the use of mathematical algorithms and decision models based on these data. They include the characteristic color spectrum, as well as the reflectance characteristics of unvegetated areas and areas with organic material. The CIE-Lab color-space was chosen for color matching because it contains information not only about coloration (a- and b-channel), but also about luminance (L-channel), thus increasing accuracy. Four different decision making algorithms based on different parameters are explained: (i) color similarity (ΔE); (ii) color similarity split in ΔL, Δa and Δb; (iii) a virtual channel ‘d’ and (iv) statistical distribution of the differences of reflection backgrounds and plants. Afterwards, the detection success of the recognition system is described. Furthermore, the minimum weed/plant coverage of the measuring spot was calculated by a mathematical model. Plants with a size of 1–5% of the spot can be recognized, and weeds in the two-leaf stage can be identified with a measuring spot size of 5 cm. By choosing a decision model previously, the detection quality can be increased. Depending on the characteristics of the background, different models are suitable. Finally, the results of field trials on municipal areas (with models of plants), winter wheat fields (with artificial plants) and grassland (with dock) are shown. In each experimental variant, objects and weeds could be recognized.

Integration of Aerial Imaging and Variable-Rate Technology for Site-Specific Aerial Herbicide Application

Abstract. As remote sensing and variable-rate technology are becoming more available for aerial applicators, practical methodologies for effective integration of these technologies are needed for site-specific aerial applications of crop production and protection materials. The objectives of this study were to demonstrate how to integrate an airborne multispectral imaging system and an IntelliStar variable-rate aerial application system for site-specific management of the winter weed henbit. The airborne imaging system was used to acquire natural color and near-infrared (NIR) images of a fallow field near College Station, Texas, for mapping the infestation of henbit shortly before herbicide application. The images were then rectified, classified, and converted to a binary prescription map with the full application rate on infested areas and the zero rate on non-infested areas. The variable-rate aerial application system mounted on an agricultural aircraft was used to apply glyphosate over the field based on the prescription map. Airborne imagery was collected 14 days after the herbicide application. The as-applied map along with the prescription map and the post-application imagery were used to assess the performance of the site-specific application. Spatial and statistical analysis results showed that the imaging system was effective for mapping henbit infestations and for assessing the performance of site-specific herbicide application, and that the variable-rate system accurately delivered the product at the desired rate to the prescribed areas for effective control of the weed. The methodology and results from this study will be useful for aerial applicators to incorporate airborne imaging and variable-rate application systems into their aerial application business to increase their capabilities and profits. Keywords: Aerial application, As-applied map, Henbit, Imaging system, Prescription map, Variable-rate technology.

Route planning for agricultural tasks: A general approach for fleets of autonomous vehicles in site-specific herbicide applications

Route planning in agricultural fields is a major challenge closely related to the amount of inputs consumed and the associated soil compaction. Current approaches primarily focus on reducing the travelled distances (i.e., the trajectories that vehicles have to cover to carry out the task) and generally do not consider other optimization criteria such as input costs (e.g., fuel, herbicides, labor). Furthermore, although few approaches consider more than one vehicle, none of them takes into consideration vehicles with different characteristics, such as different speeds or different turning radii, and some variabilities of the field such as the weed distribution have not been studied yet. All these factors affect the cost of routes to be followed to accomplish agricultural tasks such as site-specific treatments. In this context, this study proposes a very general approach to optimize the routes that considers: (1) different criteria such as the travelled distance, the time required to perform the task and the input costs, even simultaneously, (2) vehicles with different features (e.g., working speeds, both intra and inter-crop, turning radii, fuel consumptions, tank capacities and spraying costs), (3) the variability of the field and (4) the possibility of tank refilling.The proposed approach has special relevance for route planning in site-specific herbicide applications. This case requires a tank on board the vehicle to store an agrochemical product, and its capacity must be considered because it affects the routes to be followed, specifically in those cases in which the tank capacity may not be sufficient to treat the entire field even when working in cooperation with other vehicles. In such cases, refilling (i.e., a round trip to the refilling depot) may be essential despite the extra cost involved in this operation.The proposed approach was validated by solving several illustrative problems. The results showed that the proposed route planner covers a broad range of agricultural situations and that the optimal routes may vary considerably depending on the features of the fleet vehicles, the variability of the field and the optimization criteria selected. Finally, a comparative study against other well-known agricultural planners was carried out, yielding routes that improved those produced by the reference approaches.

Autonomous systems for precise spraying – Evaluation of a robotised patch sprayer

Advances in different technologies, such as global navigation satellite systems, geographic information systems, high-resolution vision systems, innovative sensors and embedded computing systems, are finding direct application in agriculture. These advances allow researchers and engineers to automate and robotise agricultural tasks no matter the inherent difficulties of the natural, semi-structured environment in which these tasks are performed. Following this current trend, this article aims to describe the development and assessment of a robotised patch spraty -20ying system that was devised for site-specific herbicide application in agricultural crops and is capable of working in groups or fleets of autonomous robots. The robotised patch sprayer consisted of an autonomous mobile robot based on a commercial agricultural vehicle chassis and a direct-injection spraying boom that was tailor-made to interact with the mobile robot. There were diverse sources (on-board and remote sensors) that can supply the weed data for the treatment. The main features of both the mobile robot and the sprayer are presented along with the controller that harmonised the behaviour of both main subsystems. Laboratory characterisation and field tests demonstrated that the system was reliable and accurate enough to accomplish the treatment over 99.5% of the detected weeds and treatment of the crop with no weed treated was insignificant; approximately 0.5% with respect to the total weed patches area, achieving a significant herbicide savings.

Derivation of weed intensity maps in wheat (Triticum aestivum) using spatial data with GIS in the central districts of Punjab, India

Remote sensing technologies are playing an increasingly important role in agricultural production. Because of their potential for high spatial and spectral resolution, satellite and aircraft images can contain detailed site specific information about conditions in agricultural fields. It can be used for monitoring crop growth, yield potential, soil conditions, weed intensity etc. For the commercial extension of site-specific herbicide application technology, rapid and cost effective methods for creating accurate weed maps are required. An experiment was conducted to generate weed prescription maps using spatial data with GIS and remote sensing in the central districts of Punjab, India 2007-08 and 2008-2009. The objective of this research was to demonstrate the potential of optical airborne remote sensing in the detection of some specific weeds and their densities in wheat (Triticum aestivum L.) crop in the central districts of Punjab, India. The analysis of spectral and ground measurement was done to select wave bands (wavelength regions) suitable for distinguishing weed-infested and weed-free crop areas. The results of this study showed that weed prescription mapping can be used for forecasting the infestations, on the basis of which farmers can be take the preventive control measures for weed control, which ultimately helps them in preventing yield losses and these weed maps can also be used as an input in yield forecasting models in future.

Herbicide savings and economic benefits of several strategies to control Sorghum halepense in maize crops

This study was conducted to assess the herbicide savings and the cost efficiency of site-specific herbicide application strategies in comparison with other strategies based on uniform application of herbicides throughout the whole field. The specific situation considered was Sorghum halepense infested maize fields in Spain. The results from a theoretical economic model were contrasted with the information derived from a S. halepense survey conducted in 37 commercial maize fields distributed over three Spanish maize production areas. Seven application strategies were simulated: 1) no herbicide; 2) overall full-rate; 3) overall half-rate; 4) and 5) site-specific spraying infested cells with full- and half-rate, respectively; 6) and 7) site-specific spraying infested cells plus adjacent buffer areas with full- and half-rate, respectively. The simulation results showed that site-specific weed management was the most profitable strategy when S. halepense infested area ranged between 6.5 and 18.7%. This scenario was present in 22% of the surveyed fields. In fields with less than 6.5% infestation (a situation present in 51% of the surveyed fields), yield losses were slight and the most profitable strategy was using no herbicide. When the infested area ranged between 18.7 and 40.8% (19% of the surveyed fields), no significant differences were observed between the net benefits of the various strategies. Full rate herbicide applications throughout the entire field resulted in the highest net returns in fields with more than 40.8% weed infestation (8% of the surveyed fields). In these cases, the added costs of weed detection, mapping and site-specific herbicide application were not justified by the herbicide savings obtained. Likewise, herbicide savings obtained with the various strategies depended on the proportion of the field infested. Considering the high difference between site-specific treatments and uniform application of herbicides when the infested area ranged between 18.7 and 40.8%, and that no significant differences in net returns were observed in this range, site-specific treatments should be preferred. Adding a buffer area to the site-specific treatments does not seem a suitable strategy due to the considerable increases in costs and reductions in herbicide savings.

Site-specific wild oat (Avena fatua L.) management

Beckie, H. J. and Shirriff, S. 2012. Site-specific wild oat ( Avena fatua L.) management. Can. J. Plant Sci. 92: 923–931. Variation in soil properties, such as soil moisture, across a hummocky landscape may influence wild oat emergence and growth. To evaluate wild oat emergence, growth, and management according to landscape position, a study was conducted from 2006 to 2010 in a hummocky field in the semiarid Moist Mixed Grassland ecoregion of Saskatchewan. The hypothesis tested was that wild oat emergence and growth would be greater in lower than upper slope positions under normal or dry early growing season conditions. Three herbicide treatments were imposed on the same plots each year of a 2-yr canola (Brassica napus L.) – wheat (Triticum aestivum L.) sequence: (1) nontreated (weedy) control; (2) herbicide application to upper and lower slope positions (i.e., full or blanket application); and (3) herbicide application to lower slope position only. Slope position affected crop and weed densities before in-crop herbicide application in years with dry spring growing conditions. Site-specific wild oat herbicide application in hummocky fields in semiarid regions may be justified based on results of wild oat control averaged across slope position. In year 2 of the crop sequence (wheat), overall (i.e., lower and upper slope) wild oat control based on density, biomass, and dockage (i.e., seed return) was similar between site-specific and full herbicide treatment in 2 of 3 yr. Because economic thresholds have not been widely adopted by growers in managing wild oat, site-specific treatment in years when conditions warrant may be an appropriate compromise between no application and blanket herbicide application.

Weed mapping using a machine vision system

Weed mapping is a useful tool for site-specific herbicide applications. The objectives of this study were (1) to determine the percentage of land area covered by weeds in no-till and conventionally tilled fields of common bean using digital image processing and geostatistics, and (2) to compare two types of cameras. Two digital cameras (color and infrared) and a differential GPS were affixed to a center pivot structure for image acquisition. Sample field images were acquired in a regular grid pattern, and the images were processed to estimate the percentage of weed cover. After calculating the georeferenced weed percentage values, maps were constructed using geostatistical techniques. Based on the results, color images are recommended for mapping the percentage of weed cover in no-till systems, while infrared images are recommended for weed mapping in conventional tillage systems.

Spectral discrimination of wild oat and canary grass in wheat fields for less herbicide application

Wheat, Triticum durum L, is a major cereal crop in Spain with over five million ha grown annually. Wild oat, Avena sterilis L., and canary grass, Phalaris spp., are distributed only in patches in wheat fields but herbicides are applied over entire fields, thus leading to over-application and unnecessary pollution. To reduce herbicide application, site-specific management techniques based on weed maps are being developed to treat only weed patches. Intensive weed scouting from the ground is time-consuming and expensive, and it relies on estimates of weeds at unsampled points. Remote sensing of weed canopies has been shown to be a more efficient alternative. The principle of weed remote sensing is that there are differences in the spectral reflectance between weeds and crops. To test this principle, we studied spectral signatures taken on the ground in the visible and near-infrared windows for discriminating wheat, wild oat and canary grass at their last phenological stages. Late-season phenological stages included initial seed maturation through advanced maturation for weeds, and initial senescence to senescent for wheat. Spectral signatures were collected on eight sampling dates from April 28 through May 26 using a handheld field spectroradiometer. A stepwise discriminant analysis was used to detect differences in reflectance and to determine the accuracy performance for a species classification as affected by their phenological stage. Four scenarios or classification sets were considered: wheat-wild oat-canary grass, with each species represented by a different group of spectra; wheat and grass weeds, combining the two weed species into one spectral group; wheat and wild oat with each represented as a single group, and finally, wheat and canary grass. Our analysis achieved 100% classification accuracy at the phenological stages of initial seed maturation, and green and advanced seed maturation and partly green for weeds and wheat, respectively, between the dates of April 28 and May 6. Furthermore, we reduced the number of hyperspectral wavelengths to thirteen out of 50. Multispectral analysis also showed that broad wavebands corresponding to those of QuickBird satellite imagery discriminated wild oat, canary grass and wheat at the same phenological stages and dates. Our findings are very useful for determining the timeframe during which future multispectral QuickBird satellite images will be obtained and the concrete wavelengths that should be used in case of using airborne hyperspectral imaging. Accurate and timely mapping of the spatial distribution of weeds is a key element in achieving site-specific herbicide applications for reducing spraying volume of herbicides and costs.

Precision farming for weed management: techniques

Site-specific weed control techniques have gained interest in the precision farming community over the last years. Managing weeds on a subfield level requires measuring the varying density of weeds within a field. Decision models aid in the selection and adjustment of the treatments, depending on the weed infestation. The weed control can be done either with herbicides or mechanically. A site-specific herbicide application technology can save large amounts of herbicides. Mechanical weed control techniques adapting to the weed situation in the field are applicable to a wide spectrum of crops. Site-specific techniques for the detection and management of weeds are presented. A system for the discrimination of different weed species and crops from images is described, which generates weed maps automatically. Models for the yield effect of weeds are developed and applied in on-farm-research experimental setups. Economic weed thresholds are derived and used for a herbicide application with a patch sprayer.

An economic evaluation of site-specific herbicide application

The objective of this study was to determine if site-specific application of postemergence herbicide was economically viable with current technologies. This objective was accomplished by: developing an algorithm that determined the economic optimal postemergence herbicide rate; creating models to determine the impact that postemergence herbicide rate has on yield; and determining whether site-specific application of postemergence herbicide has greater net returns than those from a uniform application of postemergence herbicide. Weed species identification and population counts were done on a regular grid in five fields across Kansas. A decision algorithm was developed to determine the economic optimal rate of postemergence herbicide for each grid cell. The site-specific herbicide rate and four standard herbicide rates [0, 0.5, 0.75, and full (1×) label rate] were applied according to a split-plot design. Weed population observations made three weeks after application showed that the site-specific treatment controlled the weeds present in the fields. Production functions developed to determine whether postemergence herbicide rate had an impact on yield showed that it had a positive, yet statistically insignificant, effect on yield. The difference in estimated net returns between applications of site-specific rate and uniform full-label rate covered all of the costs associated with site-specific application of postemergence herbicide. The margin between the estimated net returns for site-specific and uniform application of the economic optimal rate covered only a portion of the costs associated with site-specific application of postemergence herbicide.

Patchy weed distribution and site-specific weed control in winter cereals

Site-specific weed control in winter cereals was performed on the same fields every year over a 5-year period (1999–2003). The most common weeds (Apera spica-venti, Galium aparine, Veronica hederifolia, Viola arvensis) were counted by species, at grid points which were georeferenced and the data were analysed spatially. For weed control, weeds were grouped into three classes: grass, broad-leaved weeds (without Galium aparine), and Galium aparine. Based on weed distribution maps generated by the spatial analyses, herbicide application maps were created and site-specific herbicide application was carried out for grouped and or single weed species. This resulted in a significant reduction in herbicide use. Averaging the results for all fields and years, the total field area treated with herbicides was 39% for grass weeds, 44% for broad-leaved weeds (without Galium aparine) and 49% for Galium aparine. Therefore, site-specific weed control has the potential to reduce herbicide use compared to broadcast application, thus giving environmental and economic benefits.

Site-specific herbicide applications based on weed maps provide effective control

More-effective weed control in agricultural fields can be achieved by utilizing information about the spatial distribution of the previous year's mature weeds. In our study, variable-rate herbicide applications based on weed infestation maps developed just before the previous year's harvest provided effective weed control. The results showed that when information about the spatial distribution of the previous year's weed seedlings or mature weeds was used, weed control was comparable to uniform, one-rate, herbicide applications, while the total amount of herbicide applied decreased. Herbicide use was reduced an estimated 39% for the seedling map and 24% for the mature map approach. However, incorporating the weed-seed redistribution from harvest to application time into the treatment maps could further improve weed control.

Evaluating the Potential for Site-Specific Herbicide Application in Soybean

Field experiments were conducted on two North Carolina research stations in 1999, 2000, and 2001; on-farm in Lenoir, Wayne, and Wilson counties, NC, in 2002; and on-farm in Port Royal, VA, in 2000, 2001, and 2002 to evaluate possible gains from site-specific herbicide applications at these locations. Fields were scouted for weed populations using custom software on a handheld computer linked to a Global Positioning System. Scouts generated field-specific sampling grids and recorded weed density information for each grid cell. The decision aid HADSS™ (Herbicide Application Decision Support System) was used to estimate expected net return and yield loss remaining after treatment in each sample grid of every field under differing assumptions of weed size and soil moisture conditions, assuming the field was planted with either conventional or glyphosate-resistant (GR) soybean. The optimal whole-field treatment (that treatment with the highest expected net return summed across all grid cells within a field) resulted in average theoretical net returns of $79/ha (U.S. dollars) and $139/ha for conventional and GR soybean, respectively. When the most economical treatment for each grid cell was used in site-specific weed management, theoretical net returns increased by $13/ha (conventional) and $4.50/ha (GR), and expected yield loss after treatment was reduced by 10.5 and 4%, respectively, compared with the whole-field optimal treatment. When the most effective treatment for each grid cell was used in site-specific weed management, theoretical net returns decreased by $18/ha (conventional) and $4/ha (GR), and expected yield loss after treatment was reduced by 27 and 19%, respectively, compared with the whole-field optimal treatment. Site-specific herbicide applications could have reduced the volume of herbicides sprayed by as much as 70% in some situations but increased herbicide amounts in others. On average, the whole-field treatment was optimal in terms of net return for only 35% (conventional) and 57% (GR) of grid cells.