Soil Health Practices and Decision Drivers on Diversified Vegetable Farms in Minnesota

Abstract

Soil health is at the root of agricultural sustainability, and small-scale vegetable farmers are becoming an increasingly important part of the US food system. These farmers face unique challenges when it comes to managing soil on their farms. These challenges include reliance on intensive production practices, the use of primarily organic inputs with difficult to calculate nutrient concentrations, and lack of access to formal education tailored to their needs. We surveyed farmers at 100 small-scale vegetable farms in Minnesota to (1) develop a better baseline understanding of how small-scale vegetable farmers utilize key soil health practices including nutrient management, cover crops, and tillage; (2) explore how farm demographics influence the adoption of soil health practices; and (3) determine educational priorities to better support these growers. Here, we report a lack of understanding about the nutrient contributions of compost, which is often applied at very large volumes without guidance from soil test results, with implications for nutrient loading in the environment. Farmers in our study had high rates of cover crop adoption relative to other farmers in the region despite several barriers to using cover crops. More experienced farmers were more likely to utilize more tillage, with more use of deep tillage implements on larger farms. Overall, organic certification was correlated with higher adoption of soil health practices including utilization of soil tests and cover crop use, but it was not correlated with tillage. Other demographic variables including land access arrangement and race did not meaningfully influence soil health practices. Our findings suggest a need for more research, outreach, and education targeted to vegetable farmers about how to interpret laboratory soil test results, and how to responsibly utilize organic inputs including vegetative compost and composted manure at rates appropriate for crop production in a diversified farm setting. We also report a need to compensate farmers for their labor to incentive cover crop use on small farms, and a need for more research and support for farmers in the 3–50-acre range to utilize reduced tillage methods.

1. Introduction

Sustainable agriculture is defined as the ability of a crop production system to continuously produce food without environmental degradation [1]. A sustainable agricultural system integrates biological, physical, chemical, and ecological principles to utilize practices that are not harmful to the environment [2]. Soil is at the center of sustainable farming systems: sustaining soil health and fertility is key to maintaining food production as well as agricultural ecosystems [3].

Small-scale vegetable production is a growing industry in the Upper Midwest, but the soils and soil management strategies on these farms are understudied. With each recent census cycle, Midwest states such as Minnesota and Wisconsin lost thousands of farmers overall but gained hundreds of small-scale (<50 acres) fruit and vegetable farmers [4,5,6]. Due to lower barriers to entry than other types of farm enterprises, beginning farmers who do not come from farming backgrounds and who lack formal training often grow vegetables [7,8,9,10]. Increasingly, small-scale vegetable farms grow in both open fields and high tunnels. High tunnels are defined as plastic-covered enclosed growing structures without supplemental heating, and in-ground crop production [11]. To date, the United States Natural Resource Conservation Service has funded 26,216 high tunnels nationally, including 694 in Minnesota [12].

Studies of beginning small-scale vegetable farmers with limited agricultural backgrounds have characterized them as people with a desire to positively contribute to the natural environment while providing nutrient-dense fresh fruits and vegetables to their communities [13]. These farmers often lack both formal and informal training in production, stewardship, and business practices [9]. In Minnesota, these farmers are often referred to as “emerging farmers”, defined as “those who traditionally face barriers to the education and resources necessary to build profitable agricultural businesses, including immigrant farmers and farmers of color” [7]. These farms tend to be highly diverse, with farmers growing many, sometimes dozens of crops at a time [7,9,13].

The United States Natural Resource Conservation Service has determined key soil health principles to guide their work, which include minimizing disturbance, maximizing soil cover, maximizing biodiversity, and maximizing the presence of living roots [7]. The agency posits that healthy soils contribute to agricultural sustainability by producing high yields, reducing production costs, and therefore improving profits, protecting natural resources on and off the farm, reducing nutrient loading and sediment runoff, and sustaining habitat for wildlife and microorganisms [14]. Two key practices for achieving these goals are the use of reduced tillage and cover crops. The benefits of these practices are well documented: conservation tillage practices can increase soil microbial activity, soil moisture, organic matter, aggregate stability, cation exchange capacity, and crop yield [1,15,16,17,18].

While the benefits of reduced tillage and cover crops are well documented, small-scale vegetable farmers face unique barriers to implementing these practices. Many small-scale farmers do not inherit land from family and, therefore, must farm in a way that allows them to be profitable with limited land, infrastructure, and equipment [9,13]. Therefore, small-scale direct market vegetable production is typically characterized by intensive cultivation methods to produce high yields on a limited scale with tools that allow for season extension (e.g., high tunnels), succession planting to achieve multiple crop cycles per season, and reliance on hand tools or light machinery [19]. Due to the intensive nature of production, small-scale vegetable farms often have few opportunities for fallow periods (including cover crops), particularly in the context of high tunnels, which are most profitable when farmers take advantage of off-season production windows [20]. Despite the intensive nature of production, these farmers often prioritize soil health and reduced chemical inputs, and conceptualize their own practices as being directly in opposition to a more harmful paradigm of industrial agriculture [10,13]. For all of the above reasons, small-scale vegetable farmers represent a unique demographic of farmers with unmet educational needs. Understanding their practices and motivations can inform educational outreach programs, as well as research priorities and the development of best practices.

University Extension programs across the country have begun to invest in local foods programs with dedicated local foods educators and researchers, urban agriculture programming, and small farms content [21,22,23]. However, small-scale direct market vegetable farmers have historically been underrepresented in university outreach programs. In Minnesota, three needs assessments shed light on the lack of contact between university and extension programs and this growing sector of the farming population. A 2019 needs assessment of mostly white fruit and vegetable farmers in Minnesota found that only 19% sought information from Extension or a university entity about farming practices [24]. Another survey of Hmong and Hispanic farmers in Minnesota found that only 5% would turn to Extension or a university entity about farming practices [25]. Finally, the 2020 emerging farmers report to the Minnesota state legislature cited that while Extension is a key resource, small-scale farmers felt that Extension needs more educators who could better support nontraditional farms [7]. Rather than seeking support from traditional “experts” or support people, farmers are likely to turn to each other as sources of information [24]. They also often turn to “master classes”, YouTube, and podcasts led by “celebrity” farmers who have an outsized influence on small-scale vegetable production, as evidenced by subscriptions to these channels. A study of beginning farmers in the Southeastern United States found that YouTube videos, inspirational speakers, and books were some of the most common sources of inspiration for first-generation farmers, both to start farming and to influence their practices [13]. Another study of farmers in Hungary and the UK found that farmers increasingly trust one another over “experts”, and that farmer influencers are becoming a more important source of information and influence [26]. As of March 2024, The No-Till Growers YouTube channel had 315K subscribers, Neversink Farms channel had 74.1k subscribers, and JM Fortier’s Market Gardener Institute’s channel had 59.8k subscribers. A common theme tying these channels and celebrity farmers together is vegetable production systems based on reduced or no tillage, intensive production on a small scale, and heavy inputs of compost.

One practice in particular, deep compost mulch, has emerged as a unique soil management strategy among many influential farmers. Deep compost mulch involves using large volumes of compost to quickly increase soil organic matter, suppress weeds, and improve soil moisture retention [27]. This system was first popularized by books like Regenerative Agriculture: A Practical Whole Systems Guide to Making Small Farms Work [28] and The Living Soil Handbook: The No-Till Grower’s Guide to Ecological Market Gardening [29]. While this system is aesthetically pleasing and can result in many soil health benefits (e.g., increased organic matter, increased soil available water, improved nutrient retention, and increased biodiversity such as more earthworms), researchers have begun to identify concerns about the accumulation of nutrients, particularly nitrates and risks to surface and groundwater from this practice [27].

A 2019 needs assessment of 315 fruit and vegetable farmers in Minnesota identified soil health and fertility as the top educational priority [24]. Despite relatively low rates of organic certification, the use of organic practices is common among emerging farmers in Minnesota [7] and among fruit vegetable farmers in Minnesota [24]. Soil fertility can be particularly challenging for farmers relying primarily on organic practices due to nutrient imbalances between organic inputs and crop needs [30] and because macronutrient concentrations are highly variable across compost and manure products, and, thus, more expensive and time-consuming to calculate [31]. A recent survey and focus groups with emerging farmers in Minnesota identified significant confusion about inputs. The term “compost” was used interchangeably among participants to describe composted yard waste and food scraps, composted manure, and commercial fertilizer products containing composted animal products [32]. Additionally, Extension educators at the University of Minnesota reported working with farmers who, between 2021 and 2022 experienced challenges after applying large volumes of compost on their farms. Examples included two urban farms that applied multiple cubic meters of compost to plots less than ¼ acre in size to mitigate compaction, resulting in high soluble salts that killed their plants. In another situation, a farmer added >6 tons of composted poultry manure to a single high tunnel attempting to use a deep compost mulch system, resulting in excessively high nutrient concentrations and high soluble salts.

Small-scale vegetable farmers, therefore, face a series of unique challenges when it comes to managing the soil on their farms: the influx of new farmers means this group may have more educational needs than a typical farmer. Reliance on intensive production due to space constraints limits opportunities for cover crops and fallow periods. The tendency to use organic inputs reduces opportunities to employ reduced tillage strategies and makes nutrient management more complicated. Finally, the lack of formal outreach and education for these farmers is often replaced by previously mentioned “celebrity” or otherwise influential farmers and “master classes”, which may or may not be based on best management practices and research-based information.

Based on these experiences and identified needs, a project was developed to study the soils at 100 diversified vegetable farms growing their crops in open fields or under high tunnels. The objectives of this project were the following:

• To develop a better baseline understanding of how small-scale vegetable farmers utilize key soil health practices including cover crops, reduced tillage, and nutrient management (including how the use of organic inputs like vegetative compost and composted manure factor into nutrient management decision-making).
• To explore how farm demographics such as size, organic status, experience, land ownership, race and ethnicity, and production system (open fields vs. high tunnels) influence soil health practices.
• To determine educational priorities for Extension and other farmer-focused education programs.

2. Materials and Methods

2.1. Recruitment

A team of 16 University of Minnesota Extension educators visited 100 small-scale (<50 acres) diversified vegetable farms between April and June 2023 to conduct soil tests and a soil management survey. In each region of the state, testing was completed as soon as fields were dry enough to sample following spring snowmelt. The county-based educators on the team first reached out to farmers in their local areas through local newsletters and personal contact. Following this local outreach, statewide staff posted a recruitment flier in the “University of Minnesota Fruit and Vegetable News” (1757 subscribers). Farmers were accepted into the trial until a total of 100 participants joined. Recruitment was carried out in this way versus a more random approach to ensure the geographic distribution of farm sites and to match educator capacity for conducting soil tests. A total of 200 vegetable sites (100 open fields, and 100 high tunnels) were identified on the participating farms. Participants with both high tunnel and field-grown vegetables were prioritized and made up the majority of trial sites (83 farms). On these 83 farms, one open field and at least one adjacent high tunnel were sampled, reducing variability by sampling one of each treatment group on the same soil type and location. The additional 17 farms did not have high tunnels, so a second high tunnel from one of the nearby 83 farms was included. Figure 1 illustrates the geographic location of the sites used in this study.

Extension educators collected soil and water for testing at each site, and during each visit, participants completed a survey collecting information on agricultural practices and land use history, administered via Qualtrics. Participants were not directly compensated for completing the project. Rather, they received free soil testing and consultation.

2.2. Survey

A 20-question survey questionnaire was developed using Qualtrics to learn about soil health and nutrient management practices on small-scale vegetable farms in Minnesota. The first draft of the survey was developed by the PI, and then it was reviewed for technical content and functionality by three University of Minnesota soil scientist colleagues, as well as four local Extension educators. The questionnaire was approved for distribution by the University of Minnesota Institutional Review Board (study #00018630).

The survey was divided into five sections: demographics and site history, fertility decisions and practices, soil health practices, compost use, and land tenure. It included a mix of multiple-choice and fillable text questions.

Participants completed the survey online via smartphone while the educator visiting their farm completed soil testing. In areas with limited internet access, they were given the link via email and encouraged to complete the survey when they had access to the internet. Participants completed a survey for each site sampled at their farm (e.g., one survey for their field and another for their high tunnel). The survey had an 88% response rate. Survey data were cleaned by the PI, which included de-identifying participants with a code for each farm and formatting the data for analysis in R. Unanswered questions were left blank in the dataset, and later filtered during statistical analysis.

2.3. Soil Tests

University of Minnesota Extension educators collected soil tests at each site between 7 April and 5 June 2023. Soil samples were collected with a standard soil probe from 0 to 15 cm, moving away any mulch present. Between 15 and 20 cores were collected from each site and aggregated into a single composite sample. The composite samples from each location were sent to the University of Minnesota Research Analytical Laboratory (RAL) to quantify Bray-P1 phosphorus, exchangeable potassium, and nitrate [33,34]. Additional soil testing was completed, which will be reported in a future publication.

2.4. Tillage Score

The tillage score was calculated based on survey results and an equation developed by the project team based loosely on the STIR rating [35], which attempts to assign a quantitative value to tillage intensity rather than classifying farms into vague qualitative groups like “no till”, “low till”, or “conventional till”. Following the approach of Büchi and colleagues [36], a weight was assigned to different types of tillage, and participants were asked to report the average frequency of each type of tillage. They reported on the number of deep (rototiller or moldboard plow), shallow (e.g., harrow or tilther), and manual (broadfork) tillage passes per year. The equation used was as follows:

Tillage score = (# deep tillage passes per year × 2) + (# shallow tillage passes
per year × 1) + (# manual tillage passes per year × 0.5)

(1)

To further assess tillage differences, we created a farm size variable. Sites were grouped by size and production type: high tunnels, ≤1-acre fields, 1.1–3-acre fields, 3.1–5-acre fields, and >5-acre fields. These categories were chosen based on equipment suitability at each scale (based on author experience and observations): one can feasibly manage a 1-acre farm using only hand tools, whereas two-wheel tractors and other motorized equipment such as rototillers become more common at the 1–3-acre scale and larger equipment such as tractors become more common around the 5-acre scale.

2.5. Statistics

Survey and soil data were analyzed using R v4.2.2 [37]. Graphs were generated with the ggplot2 package, and summary statistics were generated using basic commands within the dplyr package. A basic chi-square test was used to assess differences between non-ordered categorical variables. For ordered categorical variables, we also used the Spearman correlation coefficient. Variable levels were assigned a rank in R (e.g., for organic certification, conventional = 1, using mostly organic practices = 2, using exclusively organic practices = 3, and certified organic = 4). Finally, Kruskal–Wallis was used to compare categorical values with continuous variables.

3. Results

3.1. Farm Demographics

Participating farms had an average of five acres in production, with fields in vegetable production for an average of seven years. A total of 23% of participants had less than 5 years of experience farming, 25% had 5–10 years of experience, and 49% had more than 10 years of experience. Three percent did not share their level of experience. Participants self-identified as belonging to the following racial and ethnic groups: white (75%), American Indian or Alaska Native (6%), Asian, Native Hawaiian, or other Pacific Islander (4%), Black or African American (5%), and Hispanic or Latino (1%). Nine percent chose not to answer. While only 14% of participants were certified organic, 44% claimed to use exclusively organic practices, and an additional 18% used mostly organic practices. In total, 11% of participants used conventional practices, while 13% chose not to specify.

Land use history (how the site was managed prior to growing vegetables, as reported by study participants) varied substantially. The largest category of previous land use was row crop farming (32%), followed by “other” (16%), fallow (13%), grazing animals (9%), specialty crop production under a different manager or owner (6%), woodland (5%), prairie (5%), and former industrial site (1%). The remaining respondents did not specify (13%). Write-in responses included hay or alfalfa, lawn, mixed use, and fruit production (7, 4, 3, and 2% of the total responses, respectively).

3.2. Demographics and Soil Testing

Farmers in our study were equally likely to test their soil in high tunnels versus fields, and the decision to collect soil for testing was not impacted by the farmer’s race or ethnicity (

Table 1. Impacts of various demographic factors of survey participants on how often they test their soil. For each variable, the chi-square test is reported. For ordered variables (e.g., experience), the Spearman correlation coefficient is also reported. Responses (% of total response) for each category are listed for variables that were significantly correlated with soil test frequency. Variables with significant correlations are shaded in gray.

Soil Test Frequency	Never	Once	4+ Years	2–3 Years	Every Year
Tunnel vs. field: χ2 = 0.609, p = 0.9620
Race or ethnicity: χ2 = 31.51, p = 0.1396
Organic certification χ2 = 32.063, p = 0.0014; rs = 0.26, p = <0.001
Conventional	12%	24%	24%	29%	12%
Mostly organic	34%	14%	6%	40%	6%
Exclusively organic, not certified	20%	16%	22%	31%	11%
Certified organic	7%	0%	7%	52%	34%
Experience: χ2 = 27.823, p = 0.0005; rs = 0.087, p = 0.265
<5 years	41%	16%	0%	38%	6%
5–10 years	16%	11%	11%	36%	27%
>10 years	11%	15%	24%	38%	11%
Land access: χ2 = 47.843, p = 0.03556
Informal lease	27%	18%	0%	18%	36%
Formal lease	30%	0%	0%	50%	20%
Own	21%	14%	17%	36%	12%
Land trust or community farm	0%	33%	0%	33%	33%
Tribal land	0%	33%	0%	67%	0%

). Across all testing sites (including fields and high tunnels), organic certification, farmer experience, and land access situation were all significantly correlated with the frequency of soil testing (Table 1). In general, certified organic farmers were the most likely to test their soil at regular intervals. Farmers who claimed to use organic practices without certification were not more likely than those who self-identified as conventional farmers to do regular soil tests. Less experienced farmers were also less likely to have tested their soil than those with at least five years of experience. While land access was significantly correlated with soil testing frequency, it did not follow a clear pattern, and having more stable land access did not necessarily make someone more or less likely to test their soil (Table 1).

3.3. Nutrient Management and Input Decision Drivers

Survey participants reported using soil tests as one of their top three decision-making factors more than any other source of information when deciding which fertility products to use (and how much) (61% in high tunnels, 55% in fields). Soil tests were followed by observations of previous crop performance (55% in high tunnels, 52% in fields). Farmers were more likely to seek advice from other farmers about fertility (40% in high tunnels, 41% in fields) than they were to ask private consultants (20% in high tunnels, 22% in fields), Extension (14% in high tunnels, 16% in fields), co-ops or input sellers (10% in high tunnels, 9% in fields), or nonprofit support organizations (5% in high tunnels, 18% in fields). A total of 14% of participants selected “other” in high tunnels and 18% selected it in fields as one of their top decision-making factors. Write-in responses for “other” included applying based on observation of plant performance (n = 3), performing the same practices they have used before (n = 2), not applying any fertilizer (n = 4), based on university or agronomist recommendations (n = 4), personal or family knowledge (n = 5), “plant needs” (n = 3), personal research based on books or online articles (n = 4), and applying as much as is available n = 1) or as much as the participant could afford (n = 1).

3.3.1. Soil Testing

Since participants ranked their soil tests as the primary driver of nutrient management decisions, we created a new binary variable based on whether a survey respondent indicated soil tests as a factor in their decisions, then did basic t-tests to determine whether using a soil test for fertility decisions impacted the amount of each macronutrient in the soil. There were no significant differences in soil phosphorus (p = 0.1378), potassium (p = 0.1284), or nitrate (p = 0.1285) between farmers who claimed to use soil tests as a basis for fertility decisions and those who did not. This is consistent with the findings above, indicating that the use of soil tests is not significantly correlated with soil macronutrient values, except for people who had never completed a soil test in their high tunnels.

We also looked at whether soil test frequency influenced soil nutrient concentrations. If someone had never taken a soil test in their high tunnel, their soil nitrate concentrations were likely to be higher than someone who had taken a soil test, but farmers who tested more frequently had similar nitrate concentrations to those who tested less frequently (Figure 2). This same dynamic was true for phosphorus, but only in high tunnels (Figure 3). Using the Bray-PI test for phosphorus, 41 ppm is considered “very high” for most vegetable crops, and no additional phosphorus is recommended after 50 [38]. Of the 100 tunnels and 100 fields we sampled, 87% of tunnels and 84% of fields exceeded the “very high” threshold for soil phosphorus. Even on farms that tested their soil every year, soil phosphorus levels were on average well above the “very high” threshold at which no additional added phosphorus is recommended (Figure 3). Potassium concentrations were similar across all groups, including those who had never tested their soil (Figure 4).

3.3.2. Inputs

Inputs were similar across high tunnels and fields. Composted manure was the most frequently used input in both systems, with over half of respondents using it every year or more than once per year. This was closely followed by “organic supplemental fertilizers like bone meal, fish meal, blood meal, feather meal, etc.” (Figure 5 and Figure 6). Because educators have noticed confusion about the use of the term “compost”, and the fact that it is often used interchangeably to describe composted manure and vegetable-based compost, we asked participants to share where they source their compost, and what type of compost they most often use. A total of 29% of respondents said they use animal-based compost, 16% said they use plant-based compost, and 54% said they use a mix of the two. Most compost was generated on the farm. Off-farm compost primarily came from commercial facilities. Farmer participants were more likely to purchase compost from a cooperative or input store for use in their high tunnel, whereas, for use in fields, they were more likely to source it from neighbors (

Table 2. Survey responses to questions about compost sourcing in both high tunnel and field environments at 100 Minnesota vegetable farms. Percentages reflect the percentage of participants who selected each option on a multiple-choice survey.

	High Tunnels	Fields
Source of compost (multiple responses allowed)
Generated on farm	62%	62%
Commercial compost facility	35%	41%
Neighbors	10%	15%
County or city delivery program	7%	8%
Agricultural cooperative or input store	9%	4%
Yard waste site	4%	3%
Reasons for adding compost (multiple responses allowed)
To add fertility (nutrients)	83%	81%
To improve soil structure	70%	79%
To add organic matter	72%	76%
To lessen soil compaction	32%	32%
To bury weeds	7%	8%
To fill beds	10%	6%
To remediate contamination	0%	0%
Other	10%	9%
How much compost do you apply each year?
A set amount (e.g., 1 inch per bed)	41%	34%
As much as I can get	24%	34%
Based on soil tests	16%	16%
Enough to bury weeds	3%	4%
Other	16%	12%

).

The highest-ranked motivation for using compost among survey participants was to add fertility to the soil, indicating that many farmers do consider compost to be a source of nutrients. This was closely followed by the desire to improve soil structure and increase organic matter, with fewer participants using compost for compaction remediation, to bury weeds, or to fill beds. No one reported using compost to remediate contamination (Table 2). All of the “Other” responses were related to increasing crop yields, except for one write-in response from an open field indicating that the producer adds compost to improve soil biodiversity.

While participants ranked soil tests as their primary motivating factor in making fertility decisions, soil tests were less important than other factors when it came to deciding how much compost to apply. According to survey responses, participants were most likely to apply a set amount of compost each year (41% in tunnels, 34% in fields), or to apply as much as they could source (24% in tunnels, 34% in fields) (Table 2). In both tunnels and fields, only 16% of respondents applied compost based on soil test results. Write-in responses for “Other” included comments about the limited experience and still figuring out compost applications, amounts based on equipment (e.g., “one spreader full” or one truckload), purchasing as much as people could afford, and as much compost as is generated in the respondent’s chicken coop.

3.4. Cover Crop Use

Overall, farmers were significantly more likely to use cover crops in open fields than in high tunnels (χ² = 17.208 p = 0.0006). While 72% of the farmer participants had planted a cover crop in their fields, only 41.5% had done so in their high tunnel (

Table 3. Survey responses to questions about cover crop use in high tunnels and fields at 100 Minnesota vegetable farms, percentages reflect the percentage of participants who selected each option on a multiple-choice survey.

	High Tunnels	Fields
Cover crop use frequency
Never	58.50%	28%
Occasionally	19.50%	26%
Every 2–3 years	10%	21%
Every year	12%	26%
Cover crop barriers
Cost	5%	10%
Limited space	15%	15%
Questions about logistics	16%	11%
Limited time	23%	29%
Concern about performance of next crop	6%	2%
Other	35%	33%

).

Despite these differences in cover crop adoption between high tunnels and open fields, barriers were nearly identical across environments (Table 3). In fields, farmer participants were slightly more likely to cite costs and time constraints as a barriers. In high tunnels, they were slightly more likely to cite performance about the next crop and questions about logistics as barriers. The choice most frequently selected by the farmers for both fields and high tunnels when asked about barriers to cover crop use was “Other”, suggesting that a key factor (or factors) was missed in the survey questionnaire. There were 52 write-in responses, some of which included multiple barriers: 28 responses related to a lack of time, not necessarily in terms of labor, but in terms of being able to fit cover crops into existing crop rotations. Of these 28 responses, 12 were related to fields, and 16 were related to high tunnels. Two people mentioned struggles incorporating cover crops into their perennial plantings, and seven mentioned weather as a barrier, specifically a lack of rainfall at the right time. Seven people mentioned lacking experience or being unsure of where to find seed. Two people mentioned labor and cost, and five discussed equipment barriers. Among the equipment barriers, two felt that the equipment needed to plant and terminate cover crops was too costly, and the other three shared that cover crops do not work with their system due to the use of plastic mulch or landscape fabric.

Of the demographic variables surveyed, only organic certification was correlated with cover crop use (χ² = 21.149, p = 0.012). Overall, certified organic farmers were the most likely group to plant a cover crop every year in both environments, and the least likely group to have never planted a cover crop (

Table 4. Frequency of cover crop use across environments according to farmer’s organic certification status based on farmer survey responses at 100 Minnesota vegetable farms. Percentages reflect the percentage of participants who selected each option on a multiple-choice survey.

	Never	Occasionally	Every 2–3 Years	Every Year
High tunnels
Conventional (n = 7)	86%	14%	0%	0%
Using mostly organic practices (n = 17)	71%	6%	24%	0%
Using exclusively organic practices but not certified (n = 43)	58%	26%	2%	14%
Certified organic (n = 15)	33%	20%	20%	27%
Fields
Conventional (n = 11)	27%	9%	45%	18%
Using mostly organic practices (n = 17)	35%	18%	18%	29%
Using exclusively organic practices but not certified (n = 44)	32%	34%	16%	18%
Certified organic (n = 14)	7%	21%	21%	50%

). In high tunnels, farmers claiming to use mostly or exclusively organic practices without certification were slightly more likely to plant a cover crop than farmers who identified as conventional, though, in fields, the results were more mixed (Table 4).

Farmer experience (χ² = 5.287, p = 0.5075), land access arrangement (χ² = 28.194, p = 0.2519), and race (χ² = 24.046, p = 0.1535) did not significantly impact cover crop use. In fields, total acreage was not correlated with cover crop use (χ² = 34.345, p = 0.356).

3.5. Tillage Practices

Overall, tillage intensity, as determined by tillage score (Equation (1)), was slightly lower in high tunnels than in fields (Figure 7). This difference was borderline significant according to a chi-square test (χ² = 3.1817, p = 0.0745).

Organic certification and land access did not significantly correlate to tillage score (χ² = 6.5276, p = 0.1631 and, χ² = 8.9553, p = 0.3461, respectively) but more experience was significantly correlated with tillage score (χ² = 8.8471, p = 0.0120). In general, experienced farmers were more likely to use more intensive tillage than their less experienced counterparts (Figure 8). More experienced farmers did, on average, have larger farms than those with less experience (Figure 9).

When we looked at types of tillage rather than the overall tillage score, we saw differences between farms of different scales. The smallest farms used less frequent deep tillage (rototillers or moldboard plows) and less shallow tillage (harrows, tilthers) than larger farms, but used manual tillage techniques like broadforks much more frequently. Farmers with 1–3 acres were more likely to use shallow tillage techniques, whereas farmers with more than 3 acres were more likely to use more deep tillage techniques. Tillage techniques in high tunnels fell somewhere in the middle (Figure 10).

There was no significant association between cover crop use and tillage (χ² = 0.1795, p = 0.9808).

4. Discussion

4.1. Nutrient Management Decision-Making

More experienced farmers and certified organic farmers in our study were more likely to utilize soil testing than their less experienced or non-certified organic peers. The National Organic Program requires that certified organic producers manage inputs in a manner that does not contribute to contamination of crops, soil, or water by plant nutrients [39]. Rule §205.601(j) prevents farmers from adding micronutrients to their soil without first documenting a deficiency by testing the soil [39]. These rules, along with support provided by certifiers likely contribute to the higher rates of soil testing among certified organic farmers. While there is limited research exploring the links between organic certification and soil testing, organic certification has been linked to improved soil health in a variety of studies and contexts [40,41,42].

This finding suggests a need for more outreach and education targeted to beginning farmers about how to correctly collect soil samples for soil testing and also how to interpret laboratory soil test results, particularly for farmers who are not certified organic or pursuing certification. Extension agents have been identified as crucial for promoting organic farming practices and other agroecological practices [42]. As such, Extension programs should invest in promoting soil testing and soil test interpretation with this audience and also consider promoting organic certification to connect farmers with additional support for sustainable nutrient management practices.

Even when farmer participants tested their soil regularly, this did not translate to meaningful differences in soil nutrient concentrations, and most fields and high tunnels in the study had excessive soil phosphorus. Excess phosphorus in agricultural soils is correlated with runoff and leaching, resulting in eutrophication of freshwater ecosystems [43,44]. A 2022 soil health-related needs assessment of emerging and beginning farmers in Minnesota identified that even when farmers test their soil, they often struggle to interpret soil test results and make nutrient management decisions accordingly [32]. These results support the finding that soil testing does not always translate into nutrient management decision-making and further emphasizes the need for additional education about soil test interpretation among small-scale vegetable farmers. This is consistent with farmers in both Michigan and Ghana, whose observations of physical and biological attributes of soil generally aligned well with soil health assays, but whose perceptions of chemical properties of soil did not consistently align with soil test results [45,46].

The farmers in our study based fertility decisions more on their own observations of previous crop performance and other farmers’ recommendations than on consultant or Extension recommendations. This is consistent with findings from studies on farmer decision-making across the region and the world, in which farmers prefer to learn from one another versus seeking advice from agricultural professionals [26,47,48,49]. By leaning into this finding and using train-the-trainer models, peer learning cohorts, and other methods to engage communities of farmers in learning together, educators can have more impact when sharing best practices for soil health and nutrient management.

4.2. Compost Confusion

This study highlights a particular need for compost-specific and composted-manure-specific education among diversified vegetable farmers. Composted manure was the most commonly used input among farmers in the study. They applied it primarily to add fertility and yet treated it very differently from other nutrient sources. They tended to apply as much as possible or a set amount each year rather than calculating compost and composted manure inputs based on soil test values. A previous qualitative study identified similar issues among beginning and emerging farmers in Minnesota, specifically, over-application of composts including composted manure, and conceptualizing composted products separately from nutrient sources [32]. Our findings here show that these challenges do not just apply to emerging farmers and that even experienced diversified vegetable farmers struggle to apply nutrient management principles to the use of composted manure and vegetative compost.

Compost applications can enhance the sustainability of a farm by improving root growth and nutrient uptake, improving crop yields and quality, and improving soil characteristics including porosity, aggregate stability, moisture and nutrient contents, and organic matter [50]. Mismanagement of synthetic fertilizers and pesticides has led to pollution and the degradation of soil on a global scale [51], and supplementing fertility programs with compost can mitigate some of these challenges. The use of compost to improve soil conditions has implications for the ability of a farm to remain sustainable, including the ability to withstand unpredictable weather conditions, manage pathogens, reduce soil erosion and nutrient runoff, and reduce the need for chemical fertilizers and irrigation [50]. Composting also improves the sustainability of global waste management and phosphorus management. A significant portion of human municipal waste is compostable, with estimates of 50% in Europe [52], and 40–70% globally [53]. This is particularly true in urban environments [54]. By diverting this waste from landfills and recycling it into high-quality compost, there are sustainability implications for greenhouse gas reductions and soil organic matter improvement at a global scale [52,53]. A 2011 analysis of the Minnesota Twin Cities Capitol Region Watershed estimated that food consumed and wasted by humans was the largest source of phosphorus inputs to the watershed and that diverting food and yard waste from landfills would reduce the storage of phosphorus in the watershed landscape by 66% [55].

Using compost to supplement soil fertility programs contributes to improving the sustainability of agriculture by improving nutrient use efficiency and productivity [56]. However, to promote more informed use of composts and mitigate externalities, there is a need for more research to develop accurate recommendations about the right volumes and types of compost needed for substituting commercial or synthetic fertilizers across different soils, crops, and growing conditions [53,56]. Compost is a highly variable product, referring to any organic matter that has been derived from plants, animals, or people and decomposed under aerobic conditions [50]. The final product depends on composting temperature, moisture, carbon-to-nitrogen ratio, source, and particle size of the feedstock, the presence of microorganisms, and the amount of oxygen and aeration available during the process [50].

If farmers are not using soil tests to determine compost application rates, but using compost and composted manure as primary fertility sources, soil tests have limited utility. This, combined with the excessively high soil nutrient concentrations on the farms in our study, represents a need for more targeted education about how to translate soil test results for farming systems primarily relying on composted vegetable scraps and manure for fertility.

4.3. Cover Crops

The majority of farmers in this study (72%) used cover crops at least occasionally in their fields, while the majority (58.5%) had never done so in their high tunnels. Adoption of cover crops was significantly higher in this group overall compared to estimates of the average Minnesota farmer. According to the 2022 Census of Agriculture, only 9% of Minnesota farms that participated in the census planted cover crops, representing about 3% of farmland represented in the census [6]. Of the demographic variables studied, only organic certification was significantly correlated with higher cover crop adoption. Farm size, experience, race, or land access were not significantly correlated with cover crop use. Other studies of small-scale vegetable farmers have identified that this group of farmers tends to be very conservation-minded, prioritizing soil health practices as a core motivation behind their approach to farming [10,13]. While a study of 541 organic fruit and vegetable farmers in the United States found that smaller farmers (<40 acres) were more likely to use cover crops and other conservation practices than their larger counterparts, almost all of the farmers in our study fell into the “small” farm category [41].

Despite the lower adoption of cover crops in high tunnels, we did not identify any barriers that were more significant in high tunnels than in open fields. Overall, the most significant barriers for cover crop use were a lack of time, and not being able to fit cover crops into crop rotations. This is consistent with the characterization that small-scale vegetable farms tend to plant very intensively, using season extension and succession planting to efficiently produce high yields on a small scale, leaving limited windows for cover crops and fallow periods [19].

The cost was one of the least significant barriers to cover crop adoption. This is important, as many of the conservation payment programs that incentivize cover crop use cover the cost of seed, but not of farmers’ time. A 3-year study on economic tradeoffs for high tunnel cover crops in Minnesota, Kansas, and Kentucky found that seed made up only 1% or less of the cost of planting cover crops, and that labor accounted for the vast majority of the cost [57]. Our results support the finding that incentive programs may be more effective if they can offset the costs of labor related to planting cover crops versus simply covering the cost of seed [57].

Only 16% of high tunnel participants and 11% of field participants identified educational barriers (questions about logistics). This suggests that the primary barriers to cover crop use on small-scale vegetable farms are not educational in nature. Nonetheless, targeted educational outreach could still be beneficial for improving cover crop adoption among the farmers who identified it as a barrier.

4.4. Tillage

More experienced farmers in our study were more likely to use more intensive tillage than their less experienced counterparts. While we do not have enough data to investigate the reasons behind this trend, this finding is consistent with a study of 96 organic (including those using organic practices but not certified) field crop and vegetable producers in Michigan. The survey found that farmers with fewer years of experience were more likely to be interested in reduced tillage methods and that vegetable farmers were more likely to be interested in reduced tillage methods than field crop farmers [58]. While this study did not report on why, they cited that smaller, newer farms generally have less capital to invest in equipment like reduced tillage machinery and that the return on investment for equipment may be lower on highly diversified vegetable farms where equipment may only be suitable for a subset of the crops grown. Small beginning farmers are also likely to be debt-averse [10]. Thus, smaller vegetable farms may be more motivated to find alternatives [58]. Our breakdown of tillage types among different farm scales supports this hypothesis. Our results also show that while reduced tillage methods are common among the smallest-scale vegetable farmers (1 acre or less), there is a need for education, as well as research into methods for reducing tillage at the 3–50-acre scale. This is especially important for organic systems; conventional no-till approaches rely heavily on herbicides [59,60,61].

While organic certification was significantly correlated with soil testing and cover crop use, it was not significantly correlated with tillage score.

Finally, a commonly cited tradeoff of using cover crops, particularly in systems that do not use herbicides, is that tillage is typically required to terminate them [62]. We did not find a correlation between the frequency of cover crop use and tillage score to support this. This may be partially explained by the fact that if well timed, the weed suppression benefits of cover crops may offset additional tillage passes for weed management [62].

4.5. Limitations and Future Research

Because of the number of participants in our study and the complexity of their agricultural systems (i.e., many crops per farm), our analysis lacks a detailed accounting of inputs. While our general categorization of input frequency (e.g., number of times per year participants applied a variety of inputs), the lack of quantification limits our ability to draw detailed conclusions about nutrient management practices. We can infer that the use of compost, particularly composted manure, is contributing to excessive nutrient accumulation based on soil nutrient levels and qualitative survey data, but without a full cost accounting of inputs, these conclusions are speculative in nature.

We also acknowledge the limited soil data presented in this paper. Due to the volume of data produced in this project, a follow-up paper will document more detailed soil test data from the 100 farms. Future research should address both agronomic challenges addressed here, including the development of better guidance for using compost products for fertility in a diversified agricultural system, as well as the social dynamics of how to effectively communicate soil health practices to farmers.

5. Conclusions

Small-scale vegetable farmers face unique challenges related to managing their soils due to intensive planting windows, labor demands, insufficient access to capital and equipment, and heavy reliance on inputs like compost and composted manure with hard to calculate nutrient concentrations.

Extension and other educational programs should develop targeted resources and education to support farmers with nutrient management in these systems, particularly related to the application of compost and avoiding over-application of nutrients like phosphorus as well as soil testing and test interpretation. Farmers in our study used composted manure and vegetative compost far more often than synthetic fertilizers (58% of farmers used composted manure at least once per year, 34% used vegetative compost compared to 18% using synthetic complete fertilizers, and 11% using synthetic nitrogen at least once per year in fields), and yet the majority (87% of high tunnels and 84% of fields) had soil phosphorus levels that were “very high”, compromising the sustainability of these systems. Farmer participants did not treat vegetative compost and composted manure like other fertility inputs: rather than relying on soil tests to make decisions about application rates, they tended to apply as much as possible or a set amount each year, likely contributing to the over-application of phosphorus. More research is needed to help farmers make informed decisions about organic inputs, and educators may be most effective by tapping into existing farmer networks and leveraging peer learning.

Cover crops are particularly challenging for this audience due to short planting windows, labor availability, and equipment requirements. Despite this, small-scale vegetable farmers in this study far surpassed the average Minnesota farmer in their adoption of cover crops with 72% of farmers in our study using cover crops at least occasionally, compared to 9% of Minnesota farmers in the 2022 agriculture census. Organic certification was the only significant driver of cover crop adoption; farm size, farmer experience, race, and land access were not significantly correlated with cover crop adoption. Conservation incentive programs that compensate labor in addition to seed costs in combination with more targeted educational outreach may significantly benefit small-scale vegetable farmers and other farmers.

The use of tillage was related to farm size (or at least land in vegetable production), with smaller areas (high tunnels and open fields < 1 acre) relying mostly on manual tillage and larger areas using mechanized tillage. Other demographic variables including organic certification, farmer experience, race, and land access were not significantly correlated with tillage score. More investment in reduced tillage practices and methods for small-scale growers in the 3–50-acre range is critical for reducing tillage among small-scale vegetable farmers.

Overall, farmers in our study were likely to use practices widely considered sustainable including the use of organic inputs, cover crops, and reduced tillage. However, they need more targeted information and educational resources to support their continued sustainability.