Carrots as a Model for Defining Critical Period of Weed Management, Biofertilization, and Market Opportunities for Great Plains Vegetables Producers

Objective 1. Determine the length of the critical period of weed control (CPWC) in organically grown carrots  

We proposed to expand our 2021 pilot study (Fig 5), by conducting complementary studies during the 2022 and 2023 growing seasons at Townes Harvest Garden, Bozeman, MT, the MSU research and education farm (https://townesharvest.montana.edu/), and cooperators’ farms. Specifically, we are assessing the length of the CPWC, the period in which weed removal is necessary to avoid yield loss, in organically managed vegetable crops. At Townes Harvest Garden, the experiment followed a Randomized Block Design with four replications and four rows of carrots each.  Seedbed preparation followed the farm manager practices and included multiple tillage passes to achieve a fine seedbed to ensure carrot seed germination. Pre-emergence flame weeding was not carried out to better evaluate weed-crop competition.  Prior to seeding, soil compaction was measured using a soil penetrometer at four locations per plot and at four depth ranges (0–15, 15–30, 30–45 and 45–60 cm).  Soil nutrient content was evaluated with two randomly located soil core samples per rep. For each sample, soil was separated by depths (0-6 and 6-24) and analyzed for Olsen P, potassium, calcium, electrical conductivity, organic matter and pH.   

Experimental plots (2.5m x 4 rows) consisted of eight randomly applied weeding treatments, selected based on the 2021 pilot study: 1) weedy up to the 2-leaf stage, 2) weedy up to the 6-leaf stage, 3) weedy up to the 10-leaf stage, 4) weed free up to the 2-leaf stage, 5) weed free up to the 6-leaf stage, 6) weed free up to the 10-leaf stage, 7) season-long weedy, and 8) season-long weed free. To encompass a wide range of environmental conditions, two trials were carried out with different planting dates (June 28th and July 14th ) and in 2023 (May 30th and June 17th). To estimate labor demand and costs, the time spent weeding each plot was recorded. This information will be used to estimate labor needs associated with each weed management treatment. At harvest, carrot biomass, aboveground biomass, marketability, Brix sugar content, and beta-carotene levels were measured. Please note that the 2023 trials are currently being conducted. 

To fully understand the impact of any proposed weed management tactics, it is necessary to expand the research beyond yields and evaluate weed communities (1). This information is essential to assess the potential of undesired weed seeds inputs into the seedbank as well as difficult-to-manage shifts towards perennial weeds such as Canada thistle (Cirsium arvense) and field bindweed (Convolvulus arvensis), species that threaten carrot production through belowground competition, a research priority of organic growers in Montana (2,3).  To identify the species and functional traits of weed communities I assessed plant abundance by species in carrot stands at Towne Harvest Garden. To do this, a 10 meter by 0.2 meter (length, width) transect was set up along rows of carrots. All species and their abundance were recorded along the transect. Additionally, ten 0.5m x 20cm frames each were placed along the transect at the beginning of each meter. In each frame, species were identified and their abundance was recorded. In half of the frames were randomly selected and weed biomass was collected, separated by species, dried, and weighed. 

To assess the applicability of our results, a reduced version of this study was conducted by our farmer collaborators.  At each farm, treatments included weedy up to the 4-leaf stage, weed-free up to the 4-leaf stage, season-long weedy, season-long weed free, and weed management based on farmer’s practices. Having a farmer’s practice treatment allows us to compare the production and market benefits or shortcomings of using the CPWC as an approach to making management decisions. Like the experimental plots at Towne Harvest Garden, at harvest, carrot biomass, aboveground biomass, marketability, Brix sugar content, and beta-carotene levels were measured. 

Data analysis.  

Critical Period of Weed Control.  To generalize our results, prior to data analysis we correlated the carrot leaf stage with growing-degree days (GDD), accumulated from crop emergence,  Following (4), we used a four-parameter log-logistic regression analysis to assess the length of the CPWC:      

Y = [C + (D - C)] / {1 + exp[B(log X – log E)]}     (Equation 1)  

Where Y = relative response of the studied variable (percentage yield, Brix sugar, or marketability of the season-long weed-free response), C = lower limit, D = upper limit, X = time expressed in GDD, E = GDD giving a 50% response between the upper and lower limits, B = slope of the line at the inflection point.  All analysis is being conducted utilizing the dcr Package of the R statistical environment (http://cran.at.r-project.org). We use the plotrix, sciplot, and ggplot packages of R to graphically represent the impact of the length of crop-weed competition on yield, biomass, sugar content, and marketability. Simple linear regressions were carried out to evaluate the relationship between the different carrot measurements.  

Weed community composition.  Following (5), and using the R statistical environment, we will assess weed species richness and species diversity with appropriate linear mixed-effects models.  Changes in weed composition community composition will be evaluated with a permutational multivariate ANOVA (perMANOVA) in combination with unconstrained multivariate ordination analyses such as nonmetric multidimensional scaling (NMDS) or detrended correspondence analysis (DCA). \

Objective 2. Evaluate the impact of elevated CO2 on radish growth and weed competition 

Our initial Objective 2 "Evaluate the impact of on-farm generated biofertilizer as a soil amendment on carrot growth and weed competition" was modified due to a variety of issues The inability to track and account for the variability of input into the biodigesters created potential inconsistencies in the biofertilizer amendment. Therefore, pinpointing which aspects of the biofertilizer were specifically influencing weed-crop competition would have been difficult. Furthermore, some of the stored biofertilizer froze over the winter and its biotic components may have been altered, potentially contributing to further inconsistencies in the amendment. In addition to difficulties with the biofertilizer, due to the small size and sensitivity of carrot seeds, we were unable to achieve a successful, consistent emergence of the crop in a greenhouse setting. 

Given these challenges, we modified Objective 2. to "Evaluate the impact of elevated CO2 on radish growth and weed competition." This objective still follows the core topic of evaluating changes in weed-crop competition with the addition of fertilizer, given that the addition of CO2 can be considered a gas fertilizer amendment. Furthermore, this objective modification allows us to expand the evaluation of the length of the CPWC (Objective 1) by incorporating the impacts of climate change into the scope of the research project. More specifically, we evaluate how the CPWC will change with predicted CO2 levels. In the face of climate change, weed management must shift to be more adaptive (6) and studying how the competitive ability of weeds and crops will shift provides critical information to producers that enables them to take both adaptive and preventive steps in their management practices. Using growth chambers, we are comparing weed-crop competition in ambient and predicted elevated CO2 levels. Rather than carrots, we used radishes due to their ability to grow quickly and consistently under growth chambers conditions. For our weed, we use Bromus tectorum (downy brome, cheatgrass) and Fagopyrum esculentum (common buckwheat), due to their known ability to grow consistently in a growth chamber and their prevalence in the region. Like the field study on the CPWC (Obj. 1), for each weed species, we established six treatments of weedy up to or weedy after different growth stages as well as a weedy and weed-free treatment. Five replications of each treatment were carried out in both ambient (419ppm) and elevated (720ppm) CO2 conditions and the whole study is being repeated twice. Our proposed modified objective gives us clearer and more dependable results while still maintaining the initial goal of studying weed-crop competition with the addition of fertilizer.  

Data analysis. 

Plant emergence. For each species, we estimated the mean time of emergence (MTE) as:   

MTE = ∑nidi  / ∑nI      (Equation 2)  

where ni is the number of seedlings at timei and di is the number of days from the beginning of the experiment to timei.  

Percentage emergence inhibition (EI%) will be calculated as:  

EI%  = [(Eambient – Eelevated)/ Eelevated] * 100     (Equation 3)     

where Eambient and E elevated are the number of seedlings that emerged with ambient and elevated CO2, respectively (7). 

Differences in emergence dynamics belowground biomass allocations will be evaluated with a mixed model analysis of variance (ANOVA) with species as a fixed variable and trial as a random variable.   

Critical Period of Weed Control.  The analysis follows ther methods described in Obj. 1. To generalize our results, radish leaf stage will be correlated with growing-degree days (GDD), accumulated from crop emergence prior to data analysis.  Briefly, following (4), a four-parameter log-logistic regression analysis will be used to assess the length the critical period of weed control as:      

Y = [C + (D - C)] / {1 + exp[B(log X – log E)]}     (Equation 1)  

Where Y = relative response of the studied variable (percentage yield, Brix sugar, or marketability of the season-long weed-free response), C = lower limit, D = upper limit, X = time expressed in GDD, E = GDD giving a 50% response between the upper and lower limits, B = slope of the line at the inflection point.  All analysis will be conducted utilizing the dcr Package of the R statistical environment (http://cran.at.r-project.org). The plotrix, sciplot, and ggplot packages of R will be used to graphically represent the impact of the length of crop-weed competition on yield. 

Objective 3: Investigate value-added products for vegetables in the Great Plains, specifically those impacted by weed competition  

Assessing possible value-added products of rejected vegetables can contribute both to the economic and environmental sustainability of farms by increasing profitability and reducing waste. This is especially relevant to crops that may not be market-ready due to aesthetic concerns. Producing long, straight, and smooth carrots can be difficult, and our preliminary study found that those carrots that had higher weed competition scored a lower qualitative marketability, rating given their smaller size and often misshapen appearance (Fig. 1). Despite these “ugly” carrots being inadequate for market, they have potential to be incorporated into value-added products such as juices, jams, and pickles instead of being tossed away. To investigate this possibility, we are evaluating the potential market values of otherwise unmarketable products in collaboration with two locally owned food processing companies: Farmented Foods and Root Kitchen and Cannery . 

Figure 1. Example of “ugly carrots” that are likely to have low marketability making candidates for value-added products 

Farmented Foods (https://www.farmented.com/) is committed to reducing food waste through the fermentation of local, “ugly”, vegetables, and is evaluating the potential processing of weed-impacted carrots and estimating their marketability within value-added products.  Root Kitchen and Cannery (https://www.rootskitchencannery.com/) was founded in 2012 with the mission of crafting a range of canned goods using locally sourced fruit and produce.  They will explore crafting curried carrot pickle and spicy carrot chips with non-marketable carrots.  Both companies provided an estimate of the labor cost and potential value-added of the products they will produce, as well as the potential price they are willing to pay the farmer for the raw vegetables. This information will be incorporated into the enterprise budgets (Objective 4).  

Objective 4. Estimate enterprise budgets for carrot production that include the different weed management scenarios  

Enterprise budgets help producers make informed decisions when allocating limited resources such as land, labor, and equipment.  We will adapt the protocols develop by Iowa State University (8) to develop enterprise budget estimates for carrots produced under three management scenarios: 1) following famers’ typical weed management practices, 2) weed management based on the critical period of weed control, and 3) weed-free conditions. This analysis will help growers explore the financial impact of weed management tactics, including potential income from value-added products (Objective 3).

Data collection: Fixed and variable production costs including land and irrigation, labor, capital, machinery and equipment, and capital will be based on information provided by local vegetable producers and the Montana Department of Agriculture.  The returns will be based on common prices and yields received by farmers. Labor will be valued using the USDA NASS seasonal “Farm Labor” reports. Yields will be obtained from field trials (Objective 1), and organic carrot values will be quantified using seasonal weighted average retail price data available bi-weekly from USDA Agricultural Market News Service, “Weekly Advertised Fruit & Vegetable Prices” and farmers inputs. All estimates will be developed on a 4-foot by 100-foot bed basis to better represent small farm production.  Marketing and/or transaction costs will not be included in the enterprise budget as they vary based on whether products are distributed through a community-supported agriculture share, wholesaler, or farmers’ market outlet.  While the estimated budgets may not reflect the specific environmental and labor of individual farms, we will use them to train farmers on approaches to maximize net returns based on market goals, weed pressure, and production costs (see Education plan, below).

References – Research Methods and Analyses 

1. Neve, P., J. Barney, Y. Buckley, et al. 2018.  Reviewing research priorities in weed ecology, evolution and management: a horizon scan.   Weed Research 58: 250-258.
2. OAEC, 2013a. A survey report on the research and educational needs of organic grain producers in Montana. http://www.oaecmt.org/reports.html.
3. OAEC, 2013b. A survey report of the research and educational needs of organic vegetable and herb producers in Montana. http://www.oaecmt.org/reports.html
4. Knezevic, S. and A. Datta. 2015.  The critical period of weed control: revisiting the data.  Weed Science Special Issue: 188-202.
5. Dupre, M.E., T. Seipel, M. Bourgault, D. Boss and F. Menalled. 2022.  Predicted climate conditions, cover crop composition, and cover crop termination method modify weed communities in semiarid agroecosystems.  Weed Research 62: 38-48 (Special Issue on Cover Crops and Weed Management).  
6. Prato, T. 2008. Conceptual framework for assessment and management of ecosystem impacts of climate change. Ecological Complexity. 5(4): 329–338.
7. Menalled, F., D. Buhler, and M. Liebman.  2005.  Composted swine manure effects on germination and early growth of crop and weed species under greenhouse conditions.  Weed Technology 19: 784-789.