INTEGRATING ADVANCED CULTURAL AND MECHANICAL STRATEGIES FOR IMPROVED WEED MANAGEMENT IN ORGANIC VEGETABLES

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: OTHER GRANTS
• Reporting Frequency: Annual
• Accession No.: 1016587
• Grant No.: 2018-51300-28426
• Cumulative Award Amt.: $2,086,519.00
• Proposal No.: 2018-02869
• Project Start Date: Sep 1, 2018
• Project End Date: Oct 31, 2023
• Grant Year: 2024
• Program Code: [113.A]- Organic Agriculture Research & Extension Initiative

Recipient Organization
UNIVERSITY OF MAINE
(N/A)
ORONO,ME 04469

Performing Department
School of Food and Agriculture

Non Technical Summary
Weeds are consistently ranked as one of the top challenges to profitable production of organic vegetables. Organic farmers cite cultivation, crop rotation and cover cropping as their foundational weed management tools, but many continue to rely on outdated equipment and do not deploy disturbance and crop diversity in ways that reliably stress and reduce weed populations over time. Low and variable cultivation efficacy are combined with abundant weed seed rain, resulting in stable, or more commonly, increasing weed densities over time. Simulation models and farmers' experience suggests that individual advancements in cultivation or crop diversity strategies alone cannot change this trajectory. However, we contend that cultural weed management focused on weed seedbank-depleting practices, combined with advanced, highly effective and reliable cultivation tools, can result in declining weed populations over time. Integration of seed- and seedling-focused management will exploit a positive feedback loop: reduced germinable seeds -- reduced weed seedlings -- high levels of seedling mortality -- lower density of mature weeds - reduced weed seed rain, and thus, continued reduction in the germinable seedbank. There are proven weed seedbank management strategies applicable across all farm scales, and others that are more scale-dependent. Timely disturbance and cover cropping can encourage germination and seedling recruitment to deplete the germinable seedbank. Timely termination of cash and cover crops can preempt weed seed rain. Solarization or tarping has been enthusiastically adopted by small-scale farmers as a strategy to smother winter annual and perennial weeds, as well as to deplete the surface germinable seedbank, creating a stale seedbed. Highly effective weed seedling control, supported by supplemental hand weeding both maximize "debits" to the seedbank, while minimizing "credits." Likewise, there is a large cultivation toolbox to draw upon, both between- and in-row tools, and evidence that "stacking" tools can improve mean efficacy while reducing variability, but farmer adoption remains low. Many of the tools are manufactured in Europe, and thus have historically been difficult to source. Farmers are sometimes skeptical regarding tool performance on their farms, with varying slopes, soil textures, stones, and residue. Low adoption is due in part to farmers, researchers and extension personnel alike who too often perpetuate the belief that cultivation is an "art," mastered only by selected farmers featured in videos, bulletins or speaking at winter meetings. Lastly, a lack of reliable, evidence-based research on optimal use of these tools under diverse soil, crop and weed conditions inhibits adoption by risk averse growers of high value crops. Our aim is to work with innovative farmers, equipment manufacturers and researchers to address these barriers to adoption. Integration of weed seedbank management (Gallandt, 2014) with robust, highly effective advanced cultivation tools (Brown and Gallandt, In Press; Gallandt et al., 2018) will establish a positive feedback loop of improving weed control. Specifically, we will demonstrate that reduced seedbank and seedling densities will lower weed survival and reduce weed seed rain thereby improving weed management outcomes over time. To this end we have established four interrelated objectives:(1) Characterize and optimize efficacy and selectivity of advanced cultivation tools, including "stacked tools," in controlled environments and field conditions;(2) Optimize organic weed management through integration of mechanical tools with preventive and cultural approaches;(3) Conduct field experiments on small-, medium-, and large-scale organic farms, comparing farmers' standard weed management practices to those which integrate preventive and cultural approaches with advanced and optimized cultivation tools; and(4) Develop a network of engaged farmer experts, equipment manufacturers and suppliers, and researchers who use and share data and experiences to improve weed management outcomes.

Animal Health Component

60%

Research Effort Categories

Basic

30%

Applied

60%

Developmental

10%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
213	2300	1140	50%
216	1499	1140	50%

Knowledge Area
216 - Integrated Pest Management Systems; 213 - Weeds Affecting Plants;

Subject Of Investigation
1499 - Vegetables, general/other; 2300 - Weeds;

Field Of Science
1140 - Weed science;

Keywords

on-farm research

physical weed control

weed management

weed seedbank

Goals / Objectives
Project GoalsTo develop and demonstrate advanced weed management practices integrating proven weed seedbank management and stacked cultivation resulting in a positive feedback loop evidenced by decreasing weed populations and increasing profitability. To facilitate a high level of stakeholder engagement with on-farm research, data-driven decision-making, and farmer-to-farmer exchange of knowledge and experiences.Managing for a Positive Feedback Loop We contend that it is possible to "solve"--or at least greatly reduce--the weed problem, evidenced by declining weed populations over time, with a management philosophy that embraces seedbank management and "stacked" cultivation strategies. Integration of seed- and seedling-focused management will exploit a positive feedback loop: declining seedbank -- fewer seedlings -- high levels of seedling mortality -- lower density of mature weeds - reduced rain, and thus, continued reduction in the germinable weed seedbank. Most organic vegetable farmers have a weed management philosophy focused on the "critical weed-free period" (Brown and Gallandt, 2017b; Knezevic et al., 2002); i.e., they cultivate to maintain an early crop size advantage to minimize yield loss, but then weeds are neglected. Due to famously abundant seed rain, weeds are, unsurprisingly, a stable or increasing problem. We are not passing judgement; many successful organic farmers share this philosophy, but there are alternative strategies, for example, "zero seed rain," and "mulching," each with distinct trade-offs (Brown and Gallandt, 2017b). Management that encourages high levels of weed seedbank "debiting" while reducing or eliminating seed rain or "credits," (Forcella, 2003) have been promoted by researchers for many years (Gallandt, 2006; Gallandt, 2014; Norris, 1999; Riemens et al., 2007), and used by some farmers to solve their weed problem (Nordell and Nordell, 2009). However, most organic farmers have not adopted this seedbank-focused management philosophy, partly because of deeply held beliefs that it can't work (Jabbour et al., 2013); many farmers believe that extraordinary seed longevity and multiple immigration vectors make it a "fool's errand." While partly true, the fact is, most important agricultural weeds are relatively short-lived in the seedbank, with half-lives of one year or less (Van Acker, 2009). A single season of high recruitment and seedling mortality (e.g., with effective false seedbeds), and a short-duration crop harvested prior to weed seed rain, can cause a large reduction in the germinable seedbank. It is also true that weeds may be introduced to a farm or field by many natural and human vectored mechanisms (Petit et al., 2012), but for most farms, these immigrants are a concern only for introduction of pernicious new species. Immigrants, for example in manure, represent a very small fraction of the existing weed seedbank (Mt. Pleasant and Schlather, 2009).Project ObjectivesIntegration of weed seedbank management (Gallandt, 2014) with robust, highly effective advanced cultivation tools (Brown and Gallandt, In Press; Gallandt et al., 2018) will establish a positive feedback loop of improving weed control. Specifically, we will demonstrate that reduced seedbank and seedling densities will lower weed survival and reduce weed seed rain thereby improving weed management outcomes over time. To this end we have established four interrelated objectives:(1) Characterize and optimize efficacy and selectivity of advanced cultivation tools, including "stacked tools," in controlled environments and field conditions;(2) Optimize organic weed management through integration of mechanical tools with preventive and cultural approaches;(3) Conduct field experiments on small-, medium-, and large-scale organic farms, comparing farmers' standard weed management practices to those which integrate preventive and cultural approaches with advanced and optimized cultivation tools; and(4) Develop a network of engaged farmer experts, equipment manufacturers and suppliers, and researchers who use and share data and experiences to improve weed management outcomes.

Project Methods
Objective 1. Characterize and optimize efficacy and selectivity of advanced cultivation tools, including "stacked tools," in controlled environments and field conditions.Hypotheses: Improved knowledge of the mechanisms by which cultivation tool systems (design, settings and speed) kill plants, coupled with an understanding of plant tolerance to those mechanisms, will facilitate selection and design of optimal tool systems as well and cultivar choices for improved efficacy and selectivity under diverse conditions.General approach. Studies for this objective will be conducted in protected soil bins with actual crops and weeds, as well as crop mimics and surrogate weeds designed to efficiently understand and narrow down the large range of possible tool systems for field testing in our weed management systems studies (Objective 2) and on-farm research (Objective 3). These soil bin systems will also play an important role in tool demonstrations and trainings for farmers and farm interns (Objective 3) and will be used during our field days and workshops (Objective 4).Objective 2. Optimize organic weed management through integration of mechanical tools with preventive and cultural approaches.Hypothesis: Individually, top-performing cultivation treatments and proven seedbank-depleting practices will maintain initial weed seedbank densities, but in combination, seedbank densities will decline over time as a consequence of a positive feedback loop: reduced germinable seedbank, lower seedling densities, high and uniform seedling mortality, minimal weed seed rain, and a further reduced germinable seedbank.Weed management systems experiments. We will establish four-year cropping systems experiments to evaluate short and long-term effects of integrated weed seedbank and "stacked" cultivation for improved weed management. The cropping system field experiments will be established in the spring of 2019 and will continue through the 2022 field season. Experiments will be established in a randomized complete block, 2 by 2 factorial design, with four replicates. Weed seedbank management will serve as one factor, cultivation will serve as the second factor, also with two levels: (i) a reference treatment including sequential use of individual tools, e.g., tine-harrow and sweeps; and (ii) stacked intra-row tools including sweeps, torsion-, tine- and finger-weeders on a steerable or camera-guided toolbar.Objective 3. Conduct field experiments on small-, medium-, and large-scale organic farms, comparing farmers' standard weed management practices to those which integrate preventive and cultural approaches with advanced and optimized cultivation tools.Hypothesis: On-farm research will capture known variability in weed seedbanks, soils, crops, and weather, as well as farmers' knowledge and skills. General approach. During each year of this four-year project we will evaluate advanced cultivation tools in fields of "low" and "high" weed pressure on 14 farms in northern New England (primarily Maine, New Hampshire and Vermont), and in the Midwest (primarily Michigan and Wisconsin). Specifically, in each region, and each year, we will work with three small farms (< 2 acres; mostly hand tools), three medium farms (2-10 acres; hand tools and a small tractor), and one large farm (10 to 100 acres; mostly tractor-based). Our on-farm effort is ambitious: 7 farms per year x 2 regions x 4 years = 56 total farms over the life of the project.On-farm research interns. Participating Farmers will receive stipends to support travel and labor costs for a resident employee that will serve as the on-farm Research Intern. PDs, working with the Research Technician and Participating Farmers will identify Research Interns during January through March each year. In the Midwest, at least 3 of the 7 interns will be selected by farmers each year from MSU's Organic Farmer Training Program. This program consists of hands-on and classroom learning each Monday during the growing season, coupled with on farm internships four or more days of the week.Core experiments. The foundational treatments in our on-farm trials repeat those used in our research farm systems experiments (Objective 2, above): a 2 x 2 factorial combination of weed seedbank density (low vs. high) and cultivation tools (farmer's reference tools vs. stacked, intra- and inter-row tools).On-Farm Cultural Weed Management. In addition to the core on-farm experimental treatments occurring at all sites, a subset of at least 4 participating farms in each region will evaluate one of the following cultural practices within the low weed seedbank treatment: 1) pre-cultivation tarping; or 2) use of a cultivation tolerant variety (identified through soil bin testing in Objective 1). In both cases, subplots (± tarp or cultivation tolerant vs standard variety) within each seedbank main plot will be established.Objective 4. Develop a network of engaged farmer experts, equipment manufacturers and suppliers, and researchers who use and share data and experiences to improve weed management outcomes.Hypothesis: Recognizing that farmers prefer to learn from other farmers, our Research Interns will engage our Participating Farmers and their employees, and our Project Dashboard will provide them with compelling data and an online environment to question and share experiences among project participants and a larger online community. On-farm Research Interns. Applied researchers know all too well that it is easy to find farmers "interested in on-farm research," but for various reasons, they often fail to remain engaged in the research. We think that having "boots on the ground" at each farm will both improve the quality of our on-farm dataset, and as important, serve to engage the Participating Farmers, their employees, and neighbors.On-line Project Dashboard. Data should be fundamental to farmers' decision-making, but it often remains in the background. Researchers tend to focus their data communication efforts on other researchers; extension personnel know it is better to show photos of people and equipment than boring charts and graphs. Yet, data can engage, it can sell products, it can change behavior. Consider the small "Reviews" bar graphs and stars associated with on-line shopping, or "Energy Star" rankings on a new appliance. We aim to engage our Participating Farmers, and a larger on-line audience, with a Project Dashboard that will aggregate and share project data in real time. Moreover, participant data will be posted next to selected "benchmark" statistics. Participating Farmers and their employees will immediately be able to see how their weed seedbank compares to regional and project averages.Local field days and cultivation workshops. Field Days will be held on a selected small-, medium- and large-scale farm each summer, in both New England and the Midwest. Equipment demonstrations, hands-on activities evaluating cultivation efficacy, and review of the farm's data relative to the region and project totals will provide an engaging experience for local farmers. We will also develop and host a series of Cultivation Workshops at the University of Maine Rogers Farm, and the Michigan State University Horticulture Farm.Cultivation tool lending programs. At the conclusion of the project, we will coordinate with local and regional organic farming organizations (Maine Organic Farmers and Gardeners Association; Northeast Organic Farming Association; Michigan Organic Food and Farm Alliance), to establish tool lending programs to encourage continued grower engagement and experimentation with novel cultivation tools.

Progress 09/01/18 to 08/31/23

Outputs
Target Audience:Weeds have long been a production challenge for organic vegetable farmers, a problem currently exacerbated by climate change and a shrinking labor pool. Weeding tools are used to reduce weed density and thus minimize handing weeding, while improving harvest, crop yield, and quality. The weeding tools are not very effective, often killing only half the weeds that are present, and selectivity is equally poor, resulting in high levels of crop injury or mortality. In this project we worked with northeast and Midwest farmers, introducing them to a new generation of weeding tools, worked to optimize their performance, and tested likely benefits of combining innovative tools with ecologically based strategies to reduce their weed seedbanks. If organic farmers started with fewer weeds, and used better tools, weed management outcomes will improve. We developed new methods for testing weeding tools, looked for cultivation-tolerant crop varieties, evaluated novel tools to improve efficacy and selectivity, and worked with 15 farms in northern New England, and 16 farms in the Midwest to test novel tools on their farms. Field experiments at our research stations followed the combined effects of seedbank management plus stacked weeding tools. Field days and online educational materials aimed to bring our teams experiences to a larger audience. ? Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?This project provided basic, applied and on-farm research experiences for 11 undergraduate students, one of whom presented their research on the development of a next generation artificial weed at the 2022 Weed Science Society of America Annual Meeting in Arlington, VA. Undergraduate students also contributed presentations at our field day events. Five B.S. and one M.S. level Project Research Technicians participated in laboratory, field and on-farm research projects, and presented results at field days and Project Advisory Meetings. We trained six graduate students during the project (5 M.S. and 1 Ph.D.); they are all currently employed, working in research and various aspects of the food and agriculture system. Lastly, we provided training to two post-doctoral Research Associates. How have the results been disseminated to communities of interest?Our primary target audience for this research was organic vegetable farmers in New England and the Midwest. We emphasized in-person events, hosting large regional field day events in Maine in 2019 (100 attending) and 2023 (65 attending), and in Michigan in 2022 (165 attending). In Maine we also hosted 4 smaller field day events (145 attending), while in Michigan we hosted 9 similar events (225 attending). To bring results and participating farmers' experiences to a national audience we developed and posted 3 videos to the UMaine zeroseedrain YouTube Channel, and 12 videos to the MSU Mechanical Weed Control website (combined views as of November 2023 = 35k). Our project website (https://weedmanagement.umaine.edu) launched in year 1 aimed to expand the reach of our project, but was an initiative that was not enthusiastically received by our participants, and momentum from year 1 was halted by the COVID pandemic shutdowns. During this period our Project Advisory Team suggested we consider a more open forum type website to reach a national audience, resulting in the development and launch of the Physical Weed Control Forum (https://forum.physicalweedcontrol.org), which averages 1,000 page views monthly, with slow but steady growth in content and activity. We worked closely with 15 farms in northern New England, and 16 farms in the Midwest, testing novel weeding tools in comparison to farmers' standard practice. These project participants and their respective neighbors and circles of friends gained valuable insights into weed management tools and strategies. Feedback from our participating farmers was very supportive: "The on-farm mechanical weed control research project was an excellent way for our farm to try out a new and innovative tool for managing weeds in vegetable crops. By using this new tool alongside the tools we normally use, we were able to assess the pros and cons of each one.We truly value this type of research, which is geared towards smaller scale, organic vegetable farmers like us, and hope to see much more of it in the future." David Levinson, Beech Hill Farm, Mt. Desert, Maine. "We had a gap in our cultivation equipment that needed to be filled. We needed to get really close to very young carrot, beet, bean and sweet corn seedlings. By using the HAK steerable cultivator that the university provided, myself and one employee were able to effectively cultivate acres of crops at various stages of growth with very little adjustment in just a couple hours. The only hand weeding we had to do all season was pull large lambs quarter from our carrots. And that only took a few man hours. Going forward we will either be purchasing one of these units or shop building one on the farm. Either way we go we'll know the investment is solid because we've already trialed the equipment successfully on the farm." Mike Bahner, Bahner Farm, Belmont, Maine. "Tarping has helped to reduce weed pressure, improve crop growth, and reduce my overall stress (because everything has to be ready well in advance, so there's less last-minute pressure). I now make regular use of the Terrateck wheel hoe to cultivate very newly germinated crops; it is small and precise enough to do a great job cleaning up weeds very early in their growth cycle. The addition of tarping, plus the very early season cultivation afforded by the Terrateck, weed control from many of my crops has been reduced almost to an afterthought. This is a most welcome change!" Clayton Carter, Fail Better Farm, Dixmont, Maine. "Controlling and minimizing the detrimental impacts of weeds on our farm has been a constant struggle for us ever since the beginning. So it has been very helpful to us when the folks at Michigan State organize demonstrations, and field days, and experiments and trials, like they did at our farm a couple of years ago. We can then see from side by side treatments which equipment and management practices are most effective in battling weeds. They were even helpful to us by loaning us a European tool that has added some new twists to a weeding tool that has been around for a while. (We ended up investing in the same tool ourselves shortly afterwards)." Peter Seely, Springdale Farm, Plymouth, WI "Partnering on your organic weed management project has been a great experience for our team at UPREC and collaborators at Little Parsley Farm! The project allowed us to investigate new tools for weed management that were scaled to our unique operations with very little risk or capital investment. This process resulted in quantitative data and experiential knowledge to inform future decision making, and has encouraged us to reassess our entire production system for carrots from cover crops to planting, irrigation and beyond." James DeDecker, UPREC Farm, Chatham, MI. "Our farm is doing a better job controlling weeds because of our involvement in your project. Both because of the specific tools we used in our field but also due to the exposure to ideas and experience of others. This was a very worthwhile project and we are glad to live with the lasting improvements to weed control on our farm." Fred Monroe, Monroe Family Organics , Alma, MI We were very pleased by the broad interest in our project, evidenced by invited talks at regional (2), national (10), and international (2) farmer-focused events. While our primary goal was to improve weed management outcomes for farmers, we also aimed to improve understanding of physical and ecological weed management by the scientific community. To this end we presented research results at the national Weed Science Society of America Annual Conference as well as the American Society for Horticultural Science. Lastly, we also had broad national and international reach through our highly engaged Project Advisory Team consisting of farmers (4), farm advisors (1), researchers (4), and equipment manufacturers (7). Online meetings were held in 2018, 2021 and 2023, with in-person meetings in 2019 and 2023. Participation and engagement of this group was remarkable, with only two of our original Team dropping out over the course of the project and over 90% attendance at each of our meetings. The depth and breadth of expertise represented by this group captured most of the world's most accomplished researchers and representative experienced farmers. These meetings were valued by all participants and were especially useful for our group. ? What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Weeding tools: Basic research Experienced farmers suggest that there is an "art" to effectively use weeding tools, requiring years of trial and error. Modeling and mechanistic theory will contribute to a more systematic study of weeding tools, defining tool selection and adjustment to bring early success to farmers new to farming. We adapted the anchorage force model of Kurstjens et al (2004) to include predictions of weeding tool selectivity (killing weeds but not the crop) based on early weed and crop growth characteristics. When applied to carrot-weed combinations, our model predicts that annual grass weeds are most effectively managed with hilling (burial) shortly after planting, while broadleaf species are most susceptible to uprooting at this time.Simulation also supports the idea that stacking weeding tools may improve selectivity of in-row weeds depending on their height and anchorage force relative to the crop. To conduct rapid and repeatable tool testing, we built a soil bin in the controlled environment of a greenhouse. A mobile carriage can carry weeding tools at precise speeds from 0.4 to 19.0 km hr-1, and "artificial weeds" (golf tees) are then scored to reflect "uprooting" or burial, providing insights into weeding tool mechanisms. A more realistic artificial weed with silicone "roots" and plastic aquarium plant "shoots," was better anchored in the soil and responded to a tine weeder much like real plants they aimed to mimic. The soil bin system allows high-throughput testing of weeding tools, with easy adjustment of depth, angles, spacing or speed. We were able to routinely conduct experiments that would be nearly impossible in the field while capturing slow-motion video from multiple angles to observe tool-soil-weed dynamics. Changing the angle of a tine rake, for example, shifted the weeding mechanism from uprooting to burial. Speed was important for tines set deep, but not for shallow depth settings. A finger weeder angled away from the direction of travel had a "hilling" action that resulted in greater artificial weed burial, information farmers can use to improve weed control near and within crop rows. These mechanistic insights will allow farmers to select and adjust their weeding tools to more precisely target any particular weed problem. We took our basic research related to weeding tools to the field as well, developing a simple "flour assay" to visualize soil surface disturbance caused by different weeding tools. The flour assay starts with a 0.25 m-2 template over which a known amount of all-purpose baking flour is sifted. A light box is placed over the area to capture a "pre-cultivation" image, the area cultivated with a tool of interest, and a "post-cultivation" image taken of the same area. Software allows analysis of the soil disturbance relative to the crop row, a variable well correlated with the weed control observed in the field. Farmers found the flour assay very engaging as they observed various weeding tools at field day events. LIDAR and 3D camera imagery were used to characterize soil microtopography before and after weeding events. Analysis of hilling disks revealed both variation in tool alignment relative to the crop-line, and variation in the vertical movement of soil relative to the tool affected efficacy. Thus, efforts aimed at improving steering may have equal or greater potential for improving efficacy than changes in tool design targeting depth control. Two years of greenhouse and field studies confirmed that crops such as condiment mustard are suitable surrogates for real weeds in physical weed control research. Surrogate weeds provide a high density of even aged seedlings to ensure robust data sets related to weeding tool efficacy. Weeding tools: Applied research Farmers relying on physical weed control would benefit from "cultivation tolerant" crop varieties that are better able to survive aggressive in-row disturbance of early weeding events. We demonstrated that commercially available beet varieties varied in their height and anchorage force at the time of early cultivation as well as their tolerance to mechanical cultivation in some cases. Carrot farmers in particular would benefit from cultivation tolerant cultivars, but early growth (1- to 3-leaf) of carrot genotypes were mostly similar despite differences in mid-season growth. Alternatively, using only the largest seed-size fraction of commercially available carrot cultivars resulted in taller and better anchored seedlings at the time of first cultivation, but unfortunately this did not translate into detectable improvements in crop tolerance to hilling or finger weeding. These results suggest that for most crops, weeding tool design and optimized settings are more promising areas for improving crop/weed selectivity than is cultivation tolerance of available varieties. Stacking two tools on the Terrateck Double Wheel Hoe from France improved weed control efficacy a small amount but it did not reduce variability. However, two particular tools, the sweep + torsion weeder, in either sequence, outperformed all other tool combinations. Thus, stacking must also include consideration of tool design. Certain tools really stood out as consistent performers: the hoe ridger following a sweep achieved remarkably high near- and in-row efficacy while the Japanese Q-hoe, a unique torsion weeder, offered impressive in-row weed control and crop selectivity without the challenging adjustment of conventional torsion weeders. These results will directly benefit farmers aiming to improve their weed control, prioritizing use of particular tools in a sequence that reliably offer improved efficacy and selectivity. Combining ecological weed management with advanced weeding tools We evaluated weeding tools and seed bank management over three years in Maine and Michigan, using bush bean, table beet, and sweet corn as test crops. Contrary to our previous work, tool stacking was an advantage in fewer than half of all weeding events. Of note, when finger weeders and hoe ridgers were used, weed mortality was consistently greater than the single-tool treatment, achieving an average efficacy of 95%. Thus, tool selection may have a greater impact on weed management outcomes than simply stacking available tools. In Maine, seedbank management decreased seed density from 3,300 m-2 in year one to 1,600 m-2 in year three; in plots receiving no seedbank management density increased from 3,100 seeds m-2 in year one to 9,000 seeds m-2 in year two. However, our results demonstrate that the effects of seedbank management vary based on species vulnerability to particular seedbank management practices. For example, at the Michigan site, removal of weeds prior to crop harvest reduced seed production and seedbank densities of common purslane which has rapid reproductive potential, but had no effect on common lambsquarters that is typically killed prior to maturitiy within short-duration vegetable crops of snap beans and beet. Crop yields were positively affected by seedbank management in two out of three years in Maine, and in one out of three years in Michigan. Contrary to expectations, we did not observe a positive feedback loop of decreasing weeds from stacking tools combined with seedbank management. These results unfortunately suggest that our management effects are not sufficiently robust to overcome variability in space, time and species inherent in organic vegetable cropping systems. Lastly, on-farm research remains an important but challenging goal. Despite efforts to support farmers with paid interns, and aims to motivate them with an innovative online data visualization space, larger farm-level challenges often relegate research to a lower priority. The on-farm data, however, seemingly were less important to our participating farmers than were their personal experiences testing various tools, and observations and peer interactions at field day events.

Publications

• Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: McCollough, M.R., R. Champagne, E.R. Gallandt, and D.C. Brainard. Coupling seedbank management and cultivation tool stacking for improved weed control in organic vegetables. Weed Science Society of America Annual Meeting, Arlington, VA (1/31/2023, Abstract 207).
• Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Parks, J.W., M.R. McCollough, and E.R. Gallandt. Speed, angle and spacing effects on finger weeder performance. Weed Science Society of America Annual Meeting, Arlington, VA (1/31/2023, Abstract 242).
• Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Gallandt, E.R. and D.C. Brainard. Advancing physical weed control: Farmers love field days, but on-farm research is very challenging. Weed Science Society of America Annual Meeting, Arlington, VA (2/2/2023, Abstract 449).
• Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Ortiz, S., E.R. Gallandt, B. Khoda, M.R. McCollough, and D.C. Brainard. Rational design of an artificial weed for high-throughput soil bin physical weed control tool testing. Weed Science Society of America Annual Meeting, Arlington, VA (1/31/2023, Abstract 130).
• Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Codega, A.J., R. Clements, J.W. Parks, R. Champagne, E.R. Gallandt. Chlorophyll content and fluorescence fail to detect crop injury from physical weed control tools. Weed Science Society of America Annual Meeting, Arlington, VA (1/31/2023, Abstract 113).
• Type: Journal Articles Status: Published Year Published: 2023 Citation: Parks, J. W. and E. R. Gallandt (2023). "Testing a tine weeder with artificial weeds in the controlled conditions of a soil bin." Weed Research 63: 328-337.
• Type: Theses/Dissertations Status: Published Year Published: 2023 Citation: Parks, Jordan W., "Deconstructing the Art of Physical Weed Control" (2023). Electronic Theses and Dissertations. 3765. https://digitalcommons.library.umaine.edu/etd/3765
• Type: Theses/Dissertations Status: Published Year Published: 2023 Citation: Champagne, Rebecca J. Evaluating Physical and Cultural Methods to Improve Weed Management in Organic Vegetables. The University of Maine, 2022.
• Type: Conference Papers and Presentations Status: Published Year Published: 2021 Citation: Champagne R, Gallandt E. Combining Weed Seedbank Management with Improved Physical Weed Control to Reduce Weed Densities in Organic Vegetables. In ASA, CSSA, SSSA International Annual Meeting 2021 Nov 9. ASA-CSSA-SSSA.
• Type: Theses/Dissertations Status: Published Year Published: 2021 Citation: Priddy, Daniel Murphey. Soil surface characteristics and cultivar choice affect mechanical weed control efficacy in organic vegetables. Michigan State University, 2021.
• Type: Theses/Dissertations Status: Published Year Published: 2021 Citation: Sanchez, Johnny J. Evaluating methods for research in physical weed control and farm asset tracking. The University of Maine, 2021.
• Type: Websites Status: Published Year Published: 2021 Citation: Physical Weed Control Forum (https://forum.physicalweedcontrol.org)
• Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Birthisel, S.K., and E. Gallandt. Looking for Rare weeds of agronomic concern in a changing climate. Weed Science Society of America Annual Meeting, New Orleans, LA (2/13/2019, Abstract 350).
• Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Gallandt, E. and L. Pedrosa. Artificial and surrogate weeds for physical weed control research. Weed Science Society of America Annual Meeting, New Orleans, LA (2/13/2019, Abstract 338).
• Type: Theses/Dissertations Status: Published Year Published: 2022 Citation: Connors, Noelle Anne. Addressing Weed and Soil Management Trade-Offs in Vegetables Through Integrated Cultural and Mechanical Strategies. Michigan State University, 2022.
• Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Gallandt, E.R. Weed management in organic grains: Competition, harrowing and hoeing (Invited). American Society of Agronomy National Conference, San Antonio, TX (11/9/2019).
• Type: Websites Status: Published Year Published: 2019 Citation: Organic Weed Management (https://weedmanagement.umaine.edu)
• Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Brown, B., D.C. Brainard, P. Marshall. Using "Stacked" Cultivation Implements to Improve Weed Control Selectivity. Weed Science Society of America Annual Meeting, Arlington, VA (1/31/2023, Abstract 351).
• Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Brainard, D.C. and N.A. Connors. To Bury or Uproot? Modelling Selectivity of Physical Weed Control in Sensitive Crops. Weed Science Society of America Annual Meeting, Arlington, VA (1/31/2023, Abstract 179)
• Type: Conference Papers and Presentations Status: Published Year Published: 2022 Citation: Connors, N.A. and D.C. Brainard. 2022. Optimizing in-row mechanical cultivation tools in carrots and onions: a modeling approach. American Society for Horticultural Science Annual Conference, Chicago, IL. 7/31/22. Paper 37792 https://ashs.confex.com/ashs/2022/meetingapp.cgi/Paper/37792
• Type: Conference Papers and Presentations Status: Published Year Published: 2022 Citation: Connors, N.A. and D.C. Brainard. 2022. Improving in-row cultivation efficacy in carrots through seed selection and timing optimization. 10th International Integrated Pest Management Symposium, Denver, CO. 3/1/2022. Poster 20
• Type: Conference Papers and Presentations Status: Published Year Published: 2021 Citation: D.M. Priddy, D.C. Brainard, Z.D. Hayden and M. Hemker. Integrated Physical and Cultural Weed Management in Table Beets. Weed Science Society of America Annual Meeting (Virtual) (2/17/2021; Abstract 162).
• Type: Conference Papers and Presentations Status: Published Year Published: 2021 Citation: D.M. Priddy, D.C. Brainard, Z.D. Hayden and M. Hemker. Soil Surface Effects on Finger and Flextine Cultivation Efficacy in Vegetables. Weed Science Society of America Annual Meeting (Virtual) (2/16/2021; Abstract 247).
• Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Benzle, M.M. and D.C. Brainard. Improving selectivity of physical weed control in winter squash: Cultivation tolerant varieties and traits. Weed Science Society of America Annual Meeting, New Orleans, LA (2/13/2019, Abstract 284).

Progress 09/01/21 to 08/31/22

Outputs
Target Audience: Nothing Reported Changes/Problems:Engaging small-scale farmers seems to be increasingly difficult, possibly due to continued economic stresses and farm labor shortages. We plan to start advertising to our farm groups in late 2022 to recruit our final cohort of farmers for the 2023 field season. Our cooperating farmers all seem to enjoy and really value the opportunity to test various physical weed control tools on their farms but remain less interested in following through on the additional effort to collect data on weed control efficacy and crop injury. We will continue to build our on-farm dataset where possible but will likely have to rely more on participating farmers' experience and testimonials than we intended as we wrap up the project and evaluate impact. What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest? Fennimore. 2021. Californian Perspectives on Camera Guidance and Robotic Cultivation in Veggies Mechanical Cultivation Session. Great Lakes Fruit, Vegetable, and Farm Market Expo. Grand Rapids, MI. December. Oral Presentation and Discussion (Session organized by Brainard and Oschwald Tilton; Moderated by Oschwald-Tilton) Bishop, H. and J. Oomen 2021. Farmer Perspectives on Mechanical Cultivation at Various Scales. Great Lakes Fruit, Vegetable, and Farm Market Expo. Grand Rapids, MI. December. Oral Presentation (Session organized by Brainard and Oschwald Tilton; Moderated by Oschwald-Tilton) Priddy, D. B. Scott; L. Arriaga and D.C. Brainard. 2022. Cultivation tool demonstration. Organic Farmer Training Program; MSU Student Organic Farm. Holt, MI. June. Brainard, D.C. and D. Priddy. 2022. Stale seed bed and stacked cultivation tool demonstration. Weed Day; MSU Horticulture Teaching Research and Extension Station, June. Hendrickson, J. and D.C. Brainard. 2022. Mechanical weeding demonstration with whisker weeders from Japan. Midwest Mechanical Weed Control Field Day; South West Michigan Research and Extension Center, Benton Harbor, MI. Sept Priddy, D and D.C. Brainard. 2022. Mechanical weeding with Oz robot demonstration. Midwest Mechanical Weed Control Field Day; South West Michigan Research and Extension Center, Benton Harbor, MI. Sept What do you plan to do during the next reporting period to accomplish the goals?In 2023, we will continue field evaluations of cultural and mechanical tools both on research farms and commercial farms, but shift our emphasis towards publication of journal articles, bulletins and fact sheets based on information we have collected over the past three years. We will continue to post information on our Physical Weed Control Forum website and YouTube Channels, and present information at sessions at the Mechanical Cultivation Session at the GLEXPO. In this final year of this project, we will also reconvene our Advisory Group to report on and evaluate progress and discuss future opportunities for continued collaboration. We continue our effort to develop an artificial weed for use in PWC tool assays in controlled conditions of the soil bin system. It has been challenging to identify materials that are both realistic in their friction and elasticity to mimic plant roots, that are also inexpensive and easy to handle to allow construction of hundreds of test subjects. We have a candidate artificial weed constructed with silicon tubing for "roots" that has an anchorage force profile similar to brassica seedlings. In this final year of our project, we will design and produce an advanced artificial weed using 3D printing and surface modification. We will measure the uprooting resistance of the advanced artificial weed and compare its anchorage force profile to those of established surrogates and alternative artificial weeds. Lastly, we will test the advanced artificial weed with a tine harrow to compare the mean cultivation efficacy rate and its variance to that of established surrogate and artificial weeds.

Impacts
What was accomplished under these goals? Objective 1: Characterize and optimize efficacy and selectivity of advanced cultivation tools, including "stacked tools," in controlled environments and field conditions Activity 1.1. Stacking tools to improve finger weeder efficacy and selectivity. Field studies in Michigan were continued to compare the effects of finger weeders used alone or in combination with either Kyoho whisker weeders, flextine cultivators or hilling disks. Results from these comparisons suggest that when finger efficacy is lower than 50%, additional tools often improve efficacy. Stacking finger weeders with whisker weeders appears to be a relatively low-cost approach to improving both efficacy and selectivity, particularly when soil crust is present. In Maine, we tested tools that specialize in breaking up the soil so finger weeders could work properly behind them. Activity 1.2. Assessing interactions between tools and soil surface conditions using LIDAR and 3-D cameras. To evaluate the potential for LIDAR and 3D cameras to facilitate understanding of physical weed control, we continue to generate and analyze topographical data using LIDAR and 3D imagery both before and after cultivation events with various tools. LIDAR and 3D scans hold promise for characterizing differences in soil surface roughness and clumping associated with different tools and soil conditions. Activity 1.3. Effects of carrot seed size on anchorage force, height and tool selectivity. Greenhouse and field studies were continued into their second year to evaluate whether seeds of the same cultivar (Bolero) varying in size differed in their early growth characteristics and tolerance to mechanical cultivation. Commercial seeds of the cultivar 'Bolero' were separated into small and large seed classes using a seed blower and evaluated for height, anchorage force and tolerance to finger weeding and hilling. Seeds from the larger seed class resulted in taller seedlings with greater anchorage forces at the time of early cultivation. However, these differences did not result in detectable differences in tolerance to finger weeders or hilling disks under field conditions. Activity 1.4. Characterizing dynamics of onion and weed tolerance to uprooting and burial tools. In both greenhouse and field experiments, the anchorage force and height of onions and various broadleaf and grass weed species were measured at 3-4 day intervals for 3-5 weeks. Comparisons of growth curves were used to predict the selective potential of cultivation tools that uproot or bury at various timings.Preliminary analysis suggests that height and anchorage force data are insufficient to adequately predict outcomes of these tools under field conditions. Activity 1.5 Evaluating Finger Weeders. Finger weeder angle and spacing was evaluated for the second year in beets. Three angles were used to test effects of scrubbing, hilling, and neutral mechanisms. Three different spacings were used to capture increasing aggressiveness from space between fingers, fingers touching, to overlap with the fingers. In 2021, the results showed the hilling tool angle to provide the best weed control in the intra and inter row zones. There were no detectable differences between the three different spacings or beet yield. Finger weeder speed effects on efficacy was tested for the second year in bush bean. Three different speeds were used to capture a range within the constraints of our equipment. In 2021, there was no effect of speed on efficacy between 3.2, 6.4, and 8.0 Km h-1. Activity 1.7. Soil Bin Testing. In Maine, finger weeder angle and spacing were also tested in the soil bin with artificial weeds and artificial crops in 2022. Results matched the tool angle effects shown in the field with the addition of a spacing effect. As the fingers were brought closer together, the more aggressive they were on the artificial crops. This experiment will be replicated again in the soil bin in October 2022. Objective 2: Optimize organic weed management through integration of mechanical tools with preventive and cultural approaches Activity 2.1. Cropping Systems Trial.We completed the final year of a cropping systems trial evaluating the interactive effects of cultivation tools (single tools vs stacked tools) and weed seedbank management intensity (low vs high) on cropping system performance indicators including crop yields, weed density and biomass, weed seedbank dynamics, soil quality and profitability. Rotational crops were snap beans in 2019, table beets in 2020 and sweet corn in 2021. High intensity weed seedbank management consisted of pre-planting tarping and/or flaming (to promote germination and subsequent death of seeds in the germination zone), coupled with rogue hand weeding to minimize weed seed production. Tool comparisons included either finger weeding alone, or finger weeding + hilling disks or whisker weeders depending on timing and crop. Results so far demonstrate reductions in both weed emergence and seed bank density in high seedbank management treatments, as well as equivalent or greater efficacy and selectivity of stacked vs single tools. In Michigan, seedbank management influenced crop yields in some cases, but tools did not. Seedbank management effects on yields varied by crop and appear to have been driven primarily by soil effects rather than weed competition. Objective 3: Conduct field experiments on small-, medium-, and large-scale organic farms, comparing farmers' standard weed management practices to those which integrate preventive and cultural approaches with advanced and optimized cultivation tools Activity 3.1. On farm tool and weed seedbank management studies. In 2022, in the Midwest, we conducted on-farm tool testing in collaboration with 5 farms in MI. On-farm experiments included evaluation of the effects of seedbank management (including tarping, cover cropping and flame weeding) and cultivation tools (standard versus advanced tools) on weed management efficacy, crop injury, and hand weeding time. Farmers and/or apprentices were provided half-day trainings and loaned tools across multiple scales including the Terrateck double wheel hoe, flextine cultivators (Treffler and Einbock), and tractor mounted steerable toolbars (Kult Kress and Stekete) equipped with various stackable tools including side knives, finger weeders and torsion weeders. Experiments were conducted in carrots, cabbage and broccoli. As expected, adoption of new techniques involved a learning curve, and frequent visits were required to help growers calibrate and test tools effectively. In several cases, growers reported improved efficacy and reduced hand-weeding requirements associated with alternative seedbank management and cultivation tools. In other cases, improvements in tillage and/or planting equipment and techniques, and/or untimely rainfall were identified as continued barriers to effective mechanical cultivation. The University of Maine group worked with four farms in 2022; three in Maine and one of New Brunswick, Canada. Farmers continue to be excited about adding finger weeders to their existing physical weed control toolkit, and with careful testing and adjustments, they are generally impressed with results. One farm tested the KULT Kress Duo units and finger weeders using a HAK 3-point hitch mounted steerable cultivator in mid-summer planted fall carrots with very good results. This farmer aims to fabricate a steerable tool bar over the winter. Objective 4: Develop a network of engaged farmer experts, equipment manufacturers and suppliers, and researchers who use and share data and experiences to improve weed management outcomes Activity 4.1. We will continue to build content and encourage activity in our new online forum (https://forum.physicalweedcontrol.org).

Publications

Progress 09/01/20 to 08/31/21

Outputs
Target Audience:We are continuing to work with organic vegetable farmers in both Maine and Michigan, conducting on-farm testing of weeding tools and weed seedbank management practices. Changes/Problems:Our central effort to engage and connect our group was the "Data Dashboard" that was launched during our first field season in 2019 (https://weedmanagement.umaine.edu). While it proved challenging to get our on-farm interns and farms to enter data in a timely manner, the data visualization and ability to search and select data by crops, tools and region worked as designed. However, the software platform for the site (Django), has proven to be technically challenging for our team, and we have not been able to contract with consultants to help with maintenance. An additional challenge was the COVID-19 related cancellation of on-farm work during the 2020 field season. We are exploring an alternative platform, a forum (https://forum.physicalweedcontrol.org) to engage both our participating farmers and the public with our project specifically, while broadly sharing information about physical weed control. The forum is visible to the public but requires membership to post or comment, settings designed to reduce inappropriate or unrelated content. The software (Discourse) is user friendly and seems to be a promising way to share curated, high-quality information about physical weed control tools and their use in a wide range of farming systems. Activity has been slow during our soft launch of the site, with 43 registered users and only 555 page views in the last month, but we are starting to advertise and are optimistic that activity will increase over the coming winter more widely What opportunities for training and professional development has the project provided?Farmer trainings included field day events listed below. How have the results been disseminated to communities of interest?1. Tilton, S.H. 2020. Nuts and Bolts: Selecting and calibrating cultivation tools for your farm. Mechanical Cultivation Session. Great Lakes Fruit, Vegetable, and Farm Market Expo. Grand Rapids, MI. December. Oral Presentation 2. Bishop, H. 2020. Mechanical cultivation at PrariErth Farm. Mechanical Cultivation Session. Great Lakes Fruit, Vegetable, and Farm Market Expo. Grand Rapids, MI. December. Oral Presentation 3. Courtens, J.P. 2020. Balancing weed control with soil health: A farmer's perspective. Mechanical Cultivation Session. Great Lakes Fruit, Vegetable, and Farm Market Expo. Grand Rapids, MI. December. Oral Presentation 4. Gallandt, E. 2020. Stacking tools for better weed management. Mechanical Cultivation Session. Great Lakes Fruit, Vegetable, and Farm Market Expo. Grand Rapids, MI. December. Oral Presentation 5. How does cultivation actually control weeds? Using this knowledge to improve efficacy. Gallandt, E.R. 2021 Empire State Producers Expo, Virtual Conference (1/14/2021). National 6. Improving weed control through stacking tools. Gallandt, E.R. Minnesota Fruit & Vegetable Growers Association Virtual Conference (1/15/2021). National 7. Ecological and Physical Tactics for Improved Weed Control in Organic Cereals. Gallandt, E.R. Midwest Organic and Sustainable Education Service (MOSES) Virtual Conference (2/24/2021). National 8. Managing for Weed Control in Organic Cereals. Gallandt, E.R. Culinary Breeding Network Grains Week Virtual Conference (5/6/2021). National 9. Mechanical cultivation tools for small scale vegetable production. University of Maine Rogers Farm Field Day, Stillwater, ME. June 17, 2021. 10. Priddy, D., N. Connors, and D.C. Brainard. 2021. In-row cultivation tool demonstration. Organic Farmer Training Program; MSU Student Organic Farm. Holt, MI. July. Demonstration. 11. Intra-row cultivation tools for organic corn and dry bean production. University of Maine Rogers Farm Field Day, Stillwater, ME. July 8, 2021. 12. Brainard, D.C. and D. Priddy. 2021. Mechanical cultivation tools for small scale vegetable production. MSU North Farm Field Day, Chatham, MI. August. Demonstration. https://vegetablegrowersnews.com/news/michigan-state-university-to-host-organic-vegetable-field-day-in-up/ 13. Stacking tools for better weed control in organic carrots. Field Day, Crystal Spring Farm, Brunswick, ME. September 1, 2021. What do you plan to do during the next reporting period to accomplish the goals?In 2022, we will continue field and greenhouse experiments related to crop injury and finger weeder optimization. Terrateck tool stacking will further be tested in the soil bin system, using artificial weeds to examine optimal tool settings. We developed a method to score our first-generation artificial weeds (golf tees) to address "uprooting" as well as "burial," and are planning soil-bin experiments to compare efficacy of these artificial weeds with mustard and millet surrogates. We will continue our internship program and on-farm tool testing. In addition, we are planning to increase dissemination of study results through expanded posting on our Physical Weed Control Forum website and through outreach events including the Midwest Mechanical Weed Control Field Day (Benton Harbor, MI Sept 2022) and the Mechanical Cultivation Session at the Great Lakes Expo Conference Site (GLEXPO). In our final year of this project, we will also reconvene our Advisory Group to report on and evaluate progress and discuss future opportunities for continued collaboration.

Impacts
What was accomplished under these goals? Accomplishments by State and Objective 1) Characterize and optimize efficacy and selectivity of advanced cultivation tools, including "stacked tools," in controlled environments and field conditions Maine Activity 1.1. Evaluation of Weeding Tools. The Johnny's tine rake was evaluated to see how angle, speed, and depth effect efficacy based on assays with artificial weeds. Uprooting was affected by tool handle angle, as were burial rates. Slow-motion video indicated that the curvature of the tines at the shallow angle caused the sand substrate to build up and cover more artificial weeds. Speed did not affect efficacy, but increasing tine depth increased efficacy. Three field experiments were conducted to determine whether finger weeder efficacy could be optimized based on angle, speed and tool stacking. These experiments have not been analyzed to date. Tool stacking with the Terrateck Double Wheel Hoe was tested in both bush bean and table beet demonstrating that stacking can increase weed control efficacy in both crops, especially with tools such as L-sweeps, tine harrows, and torsion weeders. At a two-leaf growth stage, tool stacking did not cause greater crop mortality in bush bean but increased mortality in table beet. Activity 1.2. Soil Bin Testing. A soil bin was constructed at the University of Maine to evaluate physical weed control tools in a controlled environment. We developed artificial weed protocols and methods to prepare and maintain the sand substrate for routine testing. Activity 1.3. Effects of carrot seed size on anchorage force, height and tool selectivity. Greenhouse and field studies were initiated in 2021 to evaluate whether seeds of the same cultivar (Bolero) varying in size differed in their early growth characteristics and tolerance to mechanical cultivation. Seeds from the larger seed class resulted in taller seedlings with greater anchorage forces at the time of early cultivation but these differences were insufficient to result in practical improvement in selectivity with either finger weeders or hilling disks under field conditions. Michigan Activity 1.1. Stacking tools to improve finger weeder efficacy and selectivity. Field studies were continued to compare the effects of finger weeders used alone or in combination with either Kyoho whisker weeders, flextine cultivators or hilling disks. Results from these comparisons suggest that when finger efficacy is lower than 50%, additional tools often improve efficacy. Stacking finger weeders with whisker weeders appears to be a relatively low-cost approach to improving both efficacy and selectivity, particularly when soil crust is present. Activity 1.2. Assessing interactions between tools and soil surface conditions using LIDAR and 3-D cameras. To evaluate the potential for LIDAR and 3D cameras to facilitate understanding of physical weed control, we generated and analyzed topographical data using LIDAR and 3D imagery both before and after cultivation events with various tools including flextine cultivators, finger weeders and hilling disks in both soil bin and field studies. LIDAR and 3D scans were able to characterize differences in soil surface roughness and clumping associated with different tools and soil conditions. Soil surface conditions (roughness, hardness, crusting) were manipulated in various studies through bed preparation (rolled vs not), crusting induction (molasses application or not) and organic matter additions (historic compost use or none). Efficacy was influenced in different ways by roughness, hardness and crusting. For example, rolling a bed prior to flextine cultivation may improve flextine efficacy by reducing surface roughness, but can also reduce flextine efficacy by increasing hardness (penetrometer resistance) and crusting under some field conditions. Thus, excessive soil preparation may be counterproductive and expensive. Activity 1.3. Beet cultivar effects on selectivity of finger weeders and hilling disks. Field studies were continued into a second year to evaluate the effects of beet cultivar (Boro [B]; Chiogga Guardsmark [CG]; Moneta [M]; and Touchstone Gold [TG]) on characteristics which influence tolerance to cultivation with either finger weeders or hilling disks. Cultivars differ in their height and anchorage force at early growth stages, and that larger cultivars are generally more tolerant to both finger weeding and hilling. Activity 1.4. Characterizing dynamics of onion and weed tolerance to uprooting and burial tools. In both greenhouse and field experiments, the anchorage force and height of onions and various broadleaf and grass weed species were measured at 3-4 day intervals for 3-5 weeks. Comparisons of growth curves were used to predict the selective potential of cultivation tools that uproot or bury at various timings. The actual selectivity of finger weeders and hilling disks were compared to predicted selective potential at two timings in the field. Preliminary analysis suggests that height and anchorage force data are insufficient to adequately predict outcomes of these tools under field conditions. 2) Optimize organic weed management through integration of mechanical tools with preventive and cultural approaches Activity 2.1. Cropping Systems Trial. The third year of the cropping systems trial examining cultivation tool stacking and seedbank management was completed. Sweet corn was examined in 2021. Tools included a single reference tool using sweeps versus tool stacking using finger weeders followed by hoe ridgers. In Maine, tarping reduced early-season intra-row weed densities from 90 weeds m-2 to fewer than 20 weeds m-2. Tool stacking provided 95% weed control efficacy compared to 65% efficacy using sweeps alone, without an increase in crop mortality. Additionally, in systems where only physical weed control using sweeps was performed, the germinable weed seedbank increased from 4,000 seeds m-2 to 12,000 seeds m-2. In contrast, high seedbank management decreased the germinable weed seedbank from 4,500 seeds m-2 to 3,000 seeds m-2. In Michigan, seedbank management influenced crop yields, but tools did not. Seedbank management effects on yields varied by crop and appear to have been driven primarily by soil effects rather than weed competition. 3) Conduct field experiments on small-, medium-, and large-scale organic farms, comparing farmers' standard weed management practices to those which integrate preventive and cultural approaches with advanced and optimized cultivation tools Activity 3.1. On farm tool and weed seedbank management studies. In Maine, six farms participated in the 2021 on-farm weed management study, including one large-scale, two medium-scale, and three small-scale farms. Some farmers expressed interest in purchasing the new tools or incorporating a stacked tool cultivation practice on their farms due to their experiences thus far. In the Midwest, six farms in MI and IL participated in 2021, evaluating seedbank management (including tarping, cover cropping and flame weeding) and cultivation tools (standard versus advanced tools). ­­­In several cases, growers reported improved efficacy and reduced hand-weeding requirements associated with alternative seedbank management and cultivation tools. In other cases, improvements in tillage and/or planting equipment and technique were identified as barriers to effective mechanical cultivation. 4) Develop a network of engaged farmer experts, equipment manufacturers and suppliers, and researchers who use and share data and experiences to improve weed management outcomes Activity 4.1. We will continue to populate our original Dashboard with our 2021 on-farm data and will reevaluate our online engagement strategy with our Project Advisory Team at our January 2022 meeting. We will continue to build content and encourage activity in our new online forum (https://forum.physicalweedcontrol.org).

Publications

Progress 09/01/19 to 08/31/20

Outputs
Target Audience: Nothing Reported Changes/Problems:In Michigan, due to Covid-19 restrictions, we were unable to attain University approval to conduct on-farm cultivation trials in 2020. In addition, Covid-19 further delayed construction of our soil bin system. We completed construction of a 72 x 30' high tunnel in fall of 2019 to house the soil bin system and began construction of bins and tracking system in spring 2020, but Covid-19 slowed completion of the project. We have now constructed and begun testing our first soil bin prototype and hope to begin formal experiments for tool testing in Sept 2020. Likewise, in Maine, our on-farm research and outreach efforts were mostly cancelled. We had farms committed to the project in March 2020; we anticipate that these farms will remain interested in the project for the 2021 field season. While our field season was delayed in Maine due to required approvals, we initiated experiments at our University research farm in June and completed several critical field experiments related to this project, including the second year of our four-year cropping systems experiment. What opportunities for training and professional development has the project provided?Northeast Mechanical Weed Control Expo, described above. How have the results been disseminated to communities of interest?1. Brainard, D.C. 2020. The white thread: Early season weed management. Great Lakes Vegetable Producers Network. Podcast Interview. 2. Brainard, D.C. and S.H. Tilton. 2019. Mechanical cultivation and cover crops for weed control in carrots. Carrot Session. Great Lakes Fruit, Vegetable, and Farm Market Expo. Grand Rapids, MI. December. Oral Presentation 3. Brainard, D.C. 2019. Smart weed management for organic production. Beginning Organic Session. Great Lakes Fruit, Vegetable, and Farm Market Expo. Grand Rapids, MI. December. Oral Presentation 4. Seely, P. 2019. Demonstration of Treffler flextine weeder. Midwest Mechanical Weed Control Field Day. Demonstration and discussion of Treffler flextine weeder by farmer-participant in internship program. 5. Bishop, H. 2019. Demonstration of Stekete camera guidance system. Midwest Mechanical Weed Control Field Day. Waukesha, WI. Sept. Demonstration and discussion by farmer participant in internship program. 6. Brainard, D.C. 2019. Balancing the efficacy and selectivity of in row cultivation tools in vegetables. Northeast Mechanical Weed Control Expo. University of Maine, Orono, ME. Sept. Oral Presentation 7. Brainard, D.C. and S.H. Tilton. 2019. Stacked in-row cultivation tools for weed management in vegetables. Vegetable Weed Management Session. Ontario Fruit and Vegetable Conference. Niagara Falls, Ontario, Canada. February. Oral Presentation 8. Brainard, D.C. 2019. In-row cultivation tools demonstration. Oceana Vegetable Research Tour. Hart, MI. September. Oral Presentation and Demonstration 9. Brainard, D.C., K. Brandt and S. Geurkink. 2019. In-row cultivation tool demonstration. Student Organic Farm Farmer Field School. Green Wagon Farm, Ada, MI. July. Demonstration 10. Brainard, D.C., D.M. Priddy and M. Hemker. 2019. Cultivation tools for in-row weed management. Weed Science Field Day. Holt, MI. June . Demonstration 11. Ginakes, P.G., R.C. Champagne, J. Sanchez, and R. Sexton. Tools and weed management practices for the small farm. 2019 MOFGA Common Ground Fair in Unity, ME. 12. Ginakes presented on-farm overview and results at the annual 2019 Northeast Vegetable and Fruit Growers Conference held in Manchester, NH to an audience of approximately 300 farmers from the Northeast region. 13. Gallandt, E.R. Innovations in seedbank and seedling management for improved weed control. Western Colorado Food and Farm Forum, Montrose, CO (1/25/2020). National, Invited 14. Ginakes, P, E Gallandt, D Brainard, R Champagne, J Sanchez, R Clements, D Priddy. Mechanical Weed Management for Diversified Vegetable Farms: Research Update. Maine Agricultural Trades Show, Augusta, ME (1/14/2020). State 15. Gallandt, E.R. Advances in physical weed control. New England Vegetable and Fruit Conference, Manchester, NH (12/11/2019). Regional, Invited 16. Gallandt, E.R. Seedling, seed and system-level strategies to improve organic weed management. Albert Lee Seed Annual Conference, Albert Lee, MN (11/22/2019). National, Invited. What do you plan to do during the next reporting period to accomplish the goals?Laboratory, greenhouse and field research will continue related to Project Objective 1. Particular efforts will be placed on highly controlled studies using our soil-bin systems. We will evaluate our newly established "common assays" that were tested in several UMaine experiments this year. Specifically, all field tool evaluation experiments included new assays to measure soil disturbance, a standardized artificial weed, surrogate weeds, field measurements of anchorage force, and extracted plants characterized using the WinRhizo system. In 2021 we will have a renewed effort to engage farmers with on-farm testing and attempt to re-launch the project on-farm research dashboard which has languished over the past season due to lack of on-farm work. The Maine team will plan and host a second Northeast Mechanical Weed Control Field Day or will work to develop virtual versions depending on COVID-19 conditions.

Impacts
What was accomplished under these goals? Accomplishments by State and Objective 1)Characterize and optimize efficacy and selectivity of advanced cultivation tools, including "stacked tools," in controlled environments and field conditions Maine Activity 1.1. Stacking with Hand Tools. Stacking experiments with the Terrateck Double Wheel Hoe in both bush bean and table beet were repeated in 2020. In both crops, tool stacking has shown increased weed control efficacy but also greater crop mortality compared to single tool use. Activity 1.2. Evaluation of Surrogate Weeds. Because surrogate weeds are often used in controlled efficacy studies but their similarity to wild species have never been rigorously and quantitatively assessed, an early growth analysis of four commonly used surrogate weed species were grown in a controlled greenhouse environment, alongside one wild weed species, for the purpose of measuring characteristics that mightsuggest suitability as substitutes, such as anchorage force and root architecture. Activity 1.3. Evaluation of Tine Harrow Hand Tools. Six flex-tine harrows were evaluated in terms of efficacy and selectivity while their inherent design characteristics were assessed in order to determine which, if any, were directly related to their respective measures of effectiveness. Results from the first iteration of this study suggests that design characteristics such as the tine diameter and number of gangs on a tine harrow may be related to the efficacy of the tool. Activity 1.4. Field Measurements of Root Anchorage Force. Root anchorage force measurements were taken of surrogate and ambient weeds before cultivation events in order to identify any correlation between the force required to uproot a weed and cultivation tool efficacy. In addition, we measured the anchorage force of crops before and after cultivation events in order to help identify the effects of individual and stacked tools on crop injury. Activity 1.5. Measuring Soil Disturbance. Methods for measuring soil disturbance have been developed and used across numerous experiments. In brief, flour is spread across the soil surface before cultivation and images are taken before and after cultivation using a light box. Images are analyzed using the MIPAR image analysis software program and the difference in areas covered by flour in before and after cultivation images represent soil disturbed by specific tools. Activity 1.6. Soil Bin Testing. The UMaine soil bin testing system was completed in February 2020. Safety and operational protocols are now in place and we are developing testing methods using a tine harrow and golf tees as artificial weeds. Michigan Activity 1.1. Tool effects in table beets. The impacts of finger weeders, rakes, torsion weeders, and hilling disks used alone or in various combinations on efficacy (weed mortality) and selectivity (weed mortality/crop mortality) were evaluated in table beets at various crop and weed growth stages. Results suggest that: 1) at early beet growth stages (<3 lf), tool selectivity in the in-row zone (within 5 cm of crop row) is often poor, so one hand-weeding is generally required to avoid unacceptable beet stand loss; 2) at the 4-6 beet leaf stage (and < 2 lf weed stage), tool selectivity improves dramatically and tool combinations occasionally improved efficacy relative to single tools. In-row tool combinations which appear to show greatest promise include finger weeders + hilling disks at beet 4-6 leaf stage, and finger weeders + kyoho whisker weeders at later growth stages. Activity 1.2. Combining finger weeders with other tools in broccoli, turnips, beets and snap beans. Field studies were conducted to compare the effects of finger weeders used alone or in combination with either Kyoho whisker weeders, flextine cultivators or hilling disks. Results from these comparisons suggest that when finger efficacy is lower than 50%, additional tools often improve efficacy. Activity 1.3. Beet cultivar and planting depth effects on emergence, early growth and cultivation tool selectivity. Greenhouse and field studies were conducted to evaluate the effects of beet cultivar (Boro [B]; Chiogga Guardsmark [CG]; Moneta [M]; and Touchstone Gold [TG]) and planting depth (1, 2, 3 or 4 cm) on emergence and early growth. We hypothesized that deep placement of seeds might be a useful strategy for delaying emergence and hence facilitating pre-emergence flextine cultivation or flame weeding. Deeper seed placement can delay emergence of B and TG cultivars by 1-2 days without adversely affecting the total number of emerged seedlings. However, for M and CG, greater seeding depth either reduced emergence (M) or had little effect on the timing of emergence (CG). Activity 1.4. Effects of soil surface conditions on the efficacy of flextine and finger weeders in snap beans and sweet corn.A second year of field studies is underway to evaluate the impact of soil conditions on the efficacy and selectivity of flextine and finger weeders in snap beans and sweet corn. Soil surface conditions (roughness, hardness, crusting) were manipulated in various studies through bed preparation (rolled vs not), previous cultivation event (hilled vs fingered), crusting induction (mollasses application or not) and soil organic matter content (historic compost use or none). 2) Optimize organic weed management through integration of mechanical tools with preventive and cultural approaches Activity 2.1. Cropping Systems Trial. The second year of our 4 year cropping systems trial was initiated to evaluate the interactive effects of cultivation tools (single tools vs stacked tools) and weed seedbank management intensity (low vs high) on cropping system performance indicators including crop yields, weed density and biomass, weed seedbank dynamics, soil quality and profitability.Single versus stacked tools are being evaluated along with cultural approaches that aim to reduce the germinable weed seedbank. As with 2019, high intensity weed seedbank management consisted of pre-planting tarping and rogue hand weeding to minimize weed seed production. With beets as the chosen crop for this year, flame weeding approximately three days after planting was also implemented. 3)Conduct field experiments on small-, medium-, and large-scale organic farms, comparing farmers' standard weed management practices to those which integrate preventive and cultural approaches with advanced and optimized cultivation tools Activity 3.1. On farm tool and weed seedbank management studies. Farmer participants for the 2020 on-farm research season were contacted and confirmed by mid-March, but this portion of the project was then postponed for the year due to COVID-19.We have maintained stakeholder engagement. A farmer in VT is testing a Garford camera-guided cultivation system on his relatively large organic vegetable farm; he also field tested a set of the Kyoho whisker weeders. A farmer in southern Maine is testing a HAK steering cultivatorin preparation for participating in our project during the 2021 field season. 4) Develop a network of engaged farmer experts, equipment manufacturers and suppliers, and researchers who use and share data and experiences to improve weed management outcomes The Northeast Mechanical Weed Control Expo was held on September 12, 2019 in Old Town, Maine at Rogers Farm, the University of Maine research farm. More than 100 attendees heard five research presentations, saw tool demonstrations from six manufacturers, viewed nine research posters, and engaged in an end-of-day round table discussion. Evaluations were overwhelmingly positive and topic ideas for a 2021 expo were generated. Following the Expo, a two-day project meeting was hosted at the Schoodic Institute in Acadia National Park including 23 attendees in total: six UMaine and MSU researchers and students, and advisory team members including four farmers, nine tool manufacturer representatives, and four researchers.

Publications

• Type: Journal Articles Status: Published Year Published: 2020 Citation: Sanchez, J. and E. Gallandt. (2020). Functionality and efficacy of Franklin Robotics' Tertill" robotic weeder. Weed Technol. Published online: 24 August 2020.
• Type: Other Status: Published Year Published: 2020 Citation: Brainard, D.C. 2020. Finger weeders provide an affordable option for in-row weeds. Organic Grower Magazine. May Issue. https://organicgrower.info/article/finger-weeders-provide-an-affordable-option-for-in-row-weeds/
• Type: Other Status: Published Year Published: 2020 Citation: Brainard, D.C. and D. Priddy. 2020. Tips for planting table beets. Vegetable Notes. Vegetable Grower News Magazine. June Issue.
• Type: Journal Articles Status: Published Year Published: 2019 Citation: Brainard, D.C. and D. Priddy. 2019. Tips for successful stale seed bedding. Vegetable Notes. Vegetable Grower News Magazine. November Issue

Progress 09/01/18 to 08/31/19

Outputs
Target Audience:Small- to large-scale, diversified and specialized organic vegetable farmers in the midwestern and northeastern U.S. Changes/Problems:We have had significant delays in construction of our controlled environment soil bin systems for conducting high throughput experiments on the optimal use of tools under different, weed, crop and soil conditions. In Michigan, the original site planned for construction proved unsuitable and the expense of constructing a new facility exceeded our budget, so we are seeking additional departmental and college funding to support construction of an alternative structure. In Maine, the University engineering group contracted to design and build the soil bin has encountered problems with the cable drive system that continues to delay installation. The on-farm activities required more time and site-visits by senior project personnel than originally planned, a consequence of the unique challenges and learning curve associated with mounting and setting up new cultivation tools. While the on-farm Research Interns were largely successful in data collection and entry, this can be improved. Our intern training day will be held later in the season next year, as several months passed between the training and first cultivation events for most farms; we will also work to simplify and streamline methods for interns, and plan to have project personnel on-farm for initial layout of experiments and execution of the first cultivation events. What opportunities for training and professional development has the project provided?MSU Intern Training Workshop and Tool Demo. Seven on-farm interns and one MSU extension educator attended a full-day workshop at the Horticulture Research Teaching and Education Center in Holt, MI. Activities included hands-on demonstration of cultivation tools; tarping demonstration; weed identification; and data collection protocols. One MS student and four undergraduate research aides were also trained through the various activities of this grant, including participation in the intern training workshop, calibration and operation of tools for research farm experiments, weed id and weed seedbank community evaluation. In Maine, a training day for on-farm research interns was held on May 13, 2019. Five interns attended along with one farmer and two Johnny's Selected Seeds technicians. Interns were given necessary tools and equipment for data collection, and practiced all the steps of the on-farm work, including uploading data to the data dashboard. How have the results been disseminated to communities of interest?Green Wagon Farm Weed Field Day, Ada, MI We organized and executed a full-day workshop at one of our apprentice farms as part of the Farmer Field School Workshop series hosted by MSU's Student Organic Farm. Approximately 25 farmers attended the workshop which featured demonstrations of weeding tools for various scales. Michigan State University Weed Day, Holt, MI. Approximately 40 participants attended an approximately 30 minute demonstration of cultivation equipment in snap beans as part of a larger field day featuring weed science projects Oceana Research Tour, Hart, MI. Approximately 50 participants attended 30 minute presentation featuring display of several options for "tool stacking" for weed control in vegetable crops. Artificial and Surrogate Weeds for Physical Weed Control Research. E. Gallandt and L. Pedrosa. (338) 2019 Weed Science Society of America Annual Meeting in New Orleans, LA. Seedling and seedbank management practices for improved weed control. E. Gallandt. New Mexico Organic Agriculture Conference, Albuquerque, NM (2/15/2019; invited). What do you plan to do during the next reporting period to accomplish the goals?In 2020, we will repeat field and greenhouse experiments summarized above. In addition, both Michigan and Maine teams hope to complete construction of soil bin systems. Brainard will recruit a PhD student to begin more detailed mechanistic studies outlined in our proposal. The Maine team will focus on methods for quantifying soil disruption and weed seedling mortality using image analysis tools. Surveys will be used to assess the effectiveness of the novel on-farm research methods, and farmers' attitudes toward weed control strategies used in the research trials.

Impacts
What was accomplished under these goals? Our four project objectives focus on physical weed control tools, integrating these tools with proven weed seedbank management practices, engaging small-, medium- and large-scale farmers to evaluate promising equipment and strategies over a wide range of environments, while building a team of researchers and farmers to help prioritize and evaluate project activities. Year 1 activities included personnel hiring, equipment acquisition, website and methods development, as well as laboratory, greenhouse, research station and on-farm experiments. Objective 1. Characterize and optimize efficacy and selectivity of advanced cultivation tools, including "stacked tools," in controlled environments and field conditions Activity 1.1. Tool effects in table beets. The impacts of finger weeders, rakes, torsion weeders, and hilling disks used alone or in various combinations on efficacy (weed mortality) and selectivity (weed mortality/crop mortality) were evaluated in table beets at various crop and weed growth stages. At early beet growth stages (<3 lf), tool selectivity in the in-row zone (within 5 cm of crop row) was generally low, requiring hand-weeding to avoid unacceptable beet stand loss; 2) at the 4-6 beet leaf stage (and < 2 lf weed stage), tool selectivity improved dramatically and tool combinations occasionally improved efficacy relative to single tools. In-row tool combinations which appear to show greatest promise include finger weeders + hilling disks at beet 4-6 leaf stage, and finger weeders + kyoho torsion weeders at later growth stages. Activity 1.2. Combining finger weeders with Kyoho torsion weeders in broccoli, turnips and snap beans. Field studies were conducted to compare the effects of finger weeders used alone, with finger weeders used in combination with a Kyoho torsion weeder. Activity 1.3. Beet cultivar and planting depth effects on emergence and early growth. Greenhouse and field studies were conducted to evaluate the effects of beet cultivar (Boro [B]; Chiogga Guardsmark [CG]; Moneta [M]; and Touchstone Gold [TG]) and planting depth (1, 2, 3 or 4 cm) on emergence and early growth. We hypothesized that deep placement of seeds might be a useful strategy for delaying emergence and hence facilitating pre-emergence flextine cultivation or flame weeding. Preliminary analysis of greenhouse evaluations suggest that deeper seed placement can delay emergence of B and TG cultivars by 1-2 days without adversely affecting the total number of emerged seedlings. However, for M and CG, greater seeding depth either reduced emergence (M) or had little effect on the timing of emergence (CG). Planting depth effects on early growth, root-shoot partitioning and anchorage force have not yet been analyzed. Activity 1.4. Effects of soil surface conditions on the efficacy of flextine cultivation in beans. A field study was conducted to evaluate the impact of soil bed preparation (rolled or not), historic compost application (3 dryT/yr vs none) and soil surface moisture conditions (moist vs dry) on the efficacy and selectivity of flextine cultivation in snap beans at the early post-emergence stage. Few if any detectable effects of historic compost application or soil surface moisture on flextine efficacy. Rolling the bed prior to planting reduced soil surface roughness but did not improve weed control efficacy of the flextine cultivator. Surprisingly, the mortality of mustard surrogate weeds (but not ambient purslane weeds) in the crop row was lower when beds were rolled. Activity 1.5. Field evaluation of tine-harrows. Six flex-tine harrows (five hand-tools, and one tractor-mounted) were evaluated in terms of efficacy and selectivity while their inherent design characteristics were assessed in order to determine which, if any, were directly related to their respective measures of effectiveness. These tines were evaluated in two different crops (beets and beans) to represent the two ends of the cultivation tolerance spectrum. Results from the first iteration of this study suggest that design characteristics such as the tine diameter and number of gangs on a tine harrow may be related to the efficacy of the tool. Activity 1.6. Comparing surrogates to real weed species. We examined early growth of four Brassica species commonly used as "surrogate weeds," and wild radish, Raphanus raphanistrum. Early growth characteristics were similar between the surrogates and the real weed suggesting they are suitable as substitutes for research purposes. Furthermore, in a field study, the surrogate and real weed species did not differ in their response to tine-harrow cultivation. Objective 2. Optimize organic weed management through integration of mechanical tools with preventive and cultural approaches Cropping Systems Trial. In both Michigan and Maine, 4 year cropping systems trials were initiated to evaluate the interactive effects of cultivation tools (single tools vs stacked tools) and weed seedbank management intensity (low vs high) on cropping system performance indicators including crop yields, weed density and biomass, weed seedbank dynamics, soil quality and profitability. In 2019, high intensity weed seedbank management consisted of pre-planting tarping (to promote germination and subsequent death of seeds in the germination zone), coupled with rogue hand weeding to minimize weed seed production. Tools included early flextine cultivation in all treatments, followed by either finger weeding alone, or finger weeding + hilling disks or torsion weeders depending on timing. Objective 3. Conduct field experiments on small-, medium-, and large-scale organic farms, comparing farmers' standard weed management practices to those which integrate preventive and cultural approaches with advanced and optimized cultivation tools On-farm tool and weed seedbank management studies. Farm apprentices, working in collaboration with seven farms in the Midwest and in Maine evaluated the impact of weed seedbank management approaches (including tarping, cover cropping, flame weeding, solarization and planting depth strategies) and standard versus advanced tools or tool combinations on weed management efficacy, crop injury, and hand weeding time. Trials are in progress and data has not yet been analyzed. As of August 2019, Maine researchers spent over 50 person-hours on farms reviewing the research plan, delivering and setting up new tools, and assisting with cultivation events. Objective 4. Develop a network of engaged farmer experts, equipment manufacturers and suppliers, and researchers who use and share data and experiences to improve weed management outcomes Virtual Advisory committee meetings, phone calls, farm visits, on farm trials, and field days all contributed to the development of our growing network. We have a diverse and engaged Project Advisory Team comprised of fifteen experienced researchers, farmers and industry representatives. All participated in our first virtual meeting on December 12, 2018, and two-thirds of the group will attend our first in-person meeting on September 13, 2019 in Maine. We built an interactive website to post videos and results from our on farm trials. The "data dashboard" was created in spring of 2019 (weedmanagement.umaine.edu), where on-farm interns can enter data from their trials. Thus far, four Northeastern farms have reported results from their experiments into the data dashboard, and all farmers have visited the dashboard to view posted data. We also initiated a MailChimp account for the project to provide regular emails to project advisers, equipment manufacturers, researchers, and collaborating farmers and interns (direct email list as of August 2019 includes 46 people total).

Publications