Location: Food Systems Research Unit
2025 Annual Report
Objectives
Objective 1: Characterize the current and potential capacity of regional food systems to support healthier diets and improved sustainability outcomes.
This objective investigates the current and potential capacities of regional food systems through two sub-objectives. Sub-objective 1a addresses improved sustainability outcomes. Subobjective 1b addresses healthier diets.
Sub-objective 1.A: Identify if farmer attitudes, motivations, experiences, and beliefs regarding soil health and related practices predict measured soil health metrics in New England.
Sub-objective 1.B: Quantify the capacity of field- to regional-scale food production systems to provision human diets.
Objective 2: Identify pathways and evaluate risks of transformations of regional food systems in response to changing climates, diets, and markets.
Sub-Objective 2.A: Identify and evaluate social and environmental resources under changing climates: food systems impact and resilience.
Sub-Objective 2.B: Adapting dairy to an unstable climate: Identifying and evaluating dairy forage production strategies to improve soil health and mitigate climate risks.
Objective 3: Enhance the long-term sustainability of regional food systems by building a research strategy designed around coproduction of knowledge with stakeholders.
Sub-objective 3.A: Compare models of experimentation, data integration, and knowledge co-production for long-term food systems research.
Approach
We combine modeling, field research, qualitative interviews, and secondary data analysis to ask, “How can regional food systems enhance environmental, economic, and social sustainability and improve human nutrition?” We focus on the Northeastern US. In Objective 1, we characterize the current and potential capacity of regional food systems to support healthier diets and improved sustainability. We link farmers' attitudes, beliefs, and motivations with soil health (Sub-Objective 1a) using surveys of farmers and on-farm sampling of soil health. Farmer-participants will be identified throughout the region with external stakeholders. We also estimate regional food production systems’ contributions to diets in historical and future scenarios (Sub-Objective 1b). We approach this challenge at the regional and field scales. Regionally, we will estimate the historical and potential future self-reliance of the Northeastern US. At the field scale, we will compile and assess contrasting conceptualizations of agricultural productivity and explore their implications for valuing different agricultural production systems. Both scales will rely extensively on public data or the reanalysis of pre-existing samples. In Objective 2, we identify pathways and evaluate the risks of transformations of regional food systems in response to changing climates, diets, and markets. We identify sociopolitical barriers to improving water access in Northeastern food systems. Here, too much and too little water are increasingly common, as the 2023 Vermont flooding catastrophe exemplifies (Sub-Objective 2a). We will identify the key climate change impacts to water systems of concern for farmers, processors, distributors, and consumers in the Northeastern U.S., the formal and informal institutions that exist to support farm system and food system resilience, and strategies these groups are taking to cope with recent and repeated flooding disasters. We approach this through a combination of literature review, a case study with stakeholders across the food system, and combining farmer interviews with historical water use data and hydrological models. We then evaluate adaptation strategies at the farm scale to erratic rainfall patterns by comparing forage production systems for their ability to improve soil health and reliably support productive dairies (Sub-Objective 2b). This field study will contrast standard practices, such as corn silage or improved hay, with emerging practices like winter-summer annual double crops, to assess tradeoffs between forage production, susceptibility to extreme weather, and resilience capacity via soil health. Finally, food systems is a new research area for the USDA-ARS, and programmatic innovation is essential to meet stakeholder needs. In Objective 3 and in close collaboration with the University of Vermont, we build a research strategy designed around the coproduction of knowledge with stakeholders to enhance the long-term sustainability of regional food systems. This coproduction research strategy will hasten efforts to develop local and regional food systems that are more productive, sustainable, and healthy.
Progress Report
Sub-objective 1a.
Substantial progress has been made on Sub-objective 1a, to identify if farmer attitudes, motivations, experiences, and beliefs regarding soil health and related practices predict measured soil health metrics in New England.
Researchers began implementing the social science survey to farmers in four states and are collecting three soil samples for each participating farmer’s land. In the first sampling season, researchers surveyed 130+ farmers and collected nearly 500 soil samples. Soil samples have been processed and analyzed as they have been collected. Researchers reviewed the initial coupled dataset to determine a targeted sampling approach for the second and final field season that captures the important farm systems of the region in the study. Sampling for the second field season is underway, with new survey and soil data being collected.
Sub-objective 1b.
Significant progress was made in collecting data for Goal 1 of Sub-objective 1b, which was conducting an analysis of historical self-reliance. Researchers at Burlington, Vermont, have compiled data on fruit production for the 12 Northeast States (CT, DE, MA, MD, ME, NH, NJ, NY, PA, RI, VT, and WV) from the Census of Agriculture. The dataset includes all Census years from 1900 to 2022, covering a longer time period than initially anticipated. This period covers a critical transition period in American agriculture in which diversified, subsistence farming was largely replaced by more specialized, commercial operations. Data collection is underway for Northeast vegetable crop production, which unlike fruits, is not reported directly but must be derived from estimates of vegetable acreage. This historical dataset will enable ARS scientists in Burlington, Vermont, to assess the past and present capacity of regional food systems to produce fruit and vegetable crops, critical to supplying the foods needed for healthier diets.
In support of Goal 2 of Sub-objective 1b, which was to investigate alternative definitions of crop productivity with an emphasis on crop contributions to improved human nutrition, ARS researchers elicited input from an international community of academic, industry, farmer, and nonprofit experts on this topic through three public presentations. As part of this engagement, researchers re-analyzed existing, long-term data to convert yield to three, alternative definitions of productivity: total protein, gross farm revenue, and contribution to American food independence. These alternative definitions substantially altered perceptions of the value and reliability of various grain cropping systems, while reaffirming the importance of extending crop rotations to improve agricultural productivity and reliability.
Sub-objective 2a:
Substantial progress has been made on Sub-objective 2A, “Identify and evaluate social and environmental resources under changing climates: food systems impact and resilience.” In support of Goal 1, researchers are modelling future hydrologic metrics of interest to agriculture across a six-state region in the Northeast. Model specification is nearing completion for the first state, Maine, and will shortly begin on the second state, Massachusetts. Researchers are designing a data dashboard for agricultural stakeholders and other interested public to access the data when it is complete.
Data collection was initiated toward Goal 2 of Sub-objective 2A: investigating the impacts of a major flooding event on the food systems of Vermont in 2023. This includes developing a map of losses associated with this flooding event.
Sub-objective 2b
Substantial process was made in identifying and evaluating dairy forage production strategies to improve soil health and resilience to extreme weather. Researchers hired critical personnel and secured long-term access to field space. Building on these successes, researchers have planted all study treatments to condition soil for intensive data collection in subsequent years. Researchers also collected baseline soil samples to enable tracking changes in soil health and other critical soil properties.
Objective 3
ARS scientists in Burlington, Vermont, collaborated with partners at the University of Vermont to compare the Long-Term Agroecosystem Research (LTAR) model and the Living Labs model of long-term agricultural research. A group of 24 scientists and staff members held six meetings between September 2024 and June 2025 to compare models and consider relevant questions for long-term research on Northeastern agriculture and food systems. This core team of ARS and university scientists determined that the two models of research are
complementary. The LTAR model provides continuity for conducting controlled experiments over multiple decades, while the Living Labs model tests the performance of current and alternative agricultural systems in on-farm settings. The core team has begun drafting a proposal for creating an LTAR site based in Vermont following a “hub and spokes” approach. A core set of research farms form the Hub of the site. The spokes will be formed by partnerships with commercial farms, and other food businesses, that are willing to test one or more alternative systems on their operations to assess the performance of these systems in real world conditions.
Accomplishments
1. Report on New England’s food self-reliance spurs stakeholder discussion and planning related to regional food systems. USDA-ARS scientists in Burlington, Vermont, contributed to the New England Feeding New England (NEFNE) Project, a research collaboration initiated to answer the question, “What would it take for the region to supply 30 percent of its food needs?” Released in June of 2023, the final project report has sparked conversation around the region. The stakeholder group, New England Food System Planners Partnership, has presented the report’s findings in over 100 in-person and on-line meetings with a wide range of stakeholders, including farmer organizations, cooperative extension, food business organizations, food and agriculture non-profits, and policy makers. The report has informed New Hampshire’s food and agriculture plan, Rhode Island’s food and agriculture plan, a Massachusetts’ emergency preparedness plan, and a Vermont plan for food security. In January of 2025, a USDA-ARS scientist in Burlington, Vermont, was contacted by staff from Team Pennsylvania, who are replicating the approach used to study New England for Pennsylvania.
2. Improved access and relevance of soil health assessments for northeastern U.S. farmers. Improved soil health has well-documented benefits and numerous programs exist to help farmers improve soil health. Yet, adoption of soil health-promoting practices is increasing only slowly. To reduce barriers to improving soil health, USDA-ARS and University of Vermont researchers discovered that internal decision-making factors, namely holistic systems thinking, underpins farmers’ perspectives on the importance of soil health and that currently available tests could better meet regional needs. To meet these needs, USDA-ARS researchers in Burlington, Vermont, consulted with University of Vermont scientists to establish a new, public-serving Soil Health Research and Extension Center (SHREC). Tests offered were selected based on farmers’ preferred assessment metrics and uniquely include holistic measures of biological and physical health. This center provides soil health tests and educational materials to farmers, consultants, and other stakeholders throughout northern New England and northern New York. Research results and SHREC are anticipated to help continue improving soil quality and farm profitability. SHREC meets farmer demands for soil health information that is relevant to their region and their unique systems.
3. Grain-legume intercrops increase feed grain production and reduce input needs. Grain-legume intercrops increase feed grain production and reduce input needs. Cool season grains like oat and field pea are low-cost, yet potentially valuable feed components and growing them improves the yields of corn and soybeans in rotation. However, cool season grains tend to be lower in feed value and yield than corn. Researchers in USDA-ARS locations of Brookings, South Dakota and Burlington, Vermont, discovered that intercropping oats and field pea could reduce these challenges while reducing susceptibility to weather, weeds, and pests. Relative to growing oat and field pea alone in monoculture, intercropping increased land-use efficiency by up to 5% while providing a mixed grain with better-balanced nutritional content as dairy feed. Critically, and unlike monocultures, both the quantity and the quality of intercropped feed were stable across different weather conditions. This provides farmers with certainty in feed production. The intercrop also showed significant potential cost savings: It was as effective at suppressing weeds as herbicides, and it increased the activity of predatory arthropods by 150%. Small and medium-sized animal farms often produce their own feed and will benefit most from this research, which highlights how such farms may better use a constrained land base and combat pests and weeds without costly synthetic inputs.
4. Toward profitable and verified carbon markets that are fair to farmers. Payments for specific farming practices or outcomes have been an important revenue source for American farmers – for example, by certifying as organic to access higher-value markets. An emerging opportunity for farmers is soil carbon markets, whereby farmers are compensated for implementing certain soil management practices. There are questions about if and how these new markets might resolve technical and market uncertainties—in particular, estimating how much carbon is sequestered in a given location—that can create challenges for farm operators and investors. USDA-ARS researchers in Burlington, Vermont, and collaborators found that as constructed, voluntary carbon markets are unlikely to internally resolve issues of credit measurement and uncertainty in resource access. Resolving these technical and market uncertainties in the emerging voluntary soil carbon market for U.S. producers is essential to improve investor confidence and thus ensure farmers are fairly compensated.
5. Predicting biomass and nitrogen content of green manures with a smartphone. Green manures, especially those composed of legumes like clover, can improve soil fertility and provide the myriad other benefits of cover crops. Estimating how much they can improve soil fertility, and by extension reduce fertilizer expenses for cash crops, is a persistent barrier to cover crop adoption. Researchers with USDA-ARS in Burlington, Vermont and the University of Minnesota identified an algorithm that utilizes a smartphone camera to predict nitrogen content and quantity of a medium red clover green manure. This algorithm was robust to poor lighting and field conditions and thus required minimal user training. Because smartphones are ubiquitous globally, this method could be used by farmers and researchers to evaluate potential fertility improvements from green manures regardless of their location and financial resources.
6. Reference soils and data diagnostics are necessary to validate soil microbial community sequence results. Soil biology is an important factor driving the productivity and sustainability of soils. To understand the diversity and function of soil microbes, DNA sequencing methods are now commonly employed, but reproducibility and the potential to introduce error throughout the workflow steps are ongoing challenges. In response, eleven ARS researchers in 12 separate locations conducted a cross-laboratory study to evaluate sources of variability in soil community sequencing experiments and develop diagnostics to detect errors. One of the seven sequencing runs in the experiment produced results that were clearly incorrect based on comparison to the other runs, yet the internal control (microbial mock community) and standard diagnostics did not suggest errors. The researcher in Burlington, Vermont, generated one of these sequencing runs and contributed to data analyses and editing of the report. This study provides new simple bioinformatic diagnostics to identify erroneous sequence results and recommends reference soils rather than simple “mock” communities to validate sequence runs. Researchers and commercial facilities following these guidelines can have greater confidence in the accuracy and reproducibility of soil community sequence results, which is essential to a holistic comparison of agricultural soils across regions and management practices.
Review Publications
Rosen, L.B., Ewing, P.M., Runck, B. 2024. RGB-based indices for estimating cover crop biomass, nitrogen content, and carbon:nitrogen ratio. Agronomy Journal. 116(6):3070-3080. https://doi.org/10.1002/agj2.21657.
Neupane, D., Osborne, S.L., Roeder, K.A., Knoll, A.E., Ewing, P.M. 2025. Feed grain polycultures mitigate weather risk, support arthropods, and suppress weeds in the Western Corn Belt. Agriculture, Ecosystems & Environment. 393. Article 109773. https://doi.org/10.1016/j.agee.2025.109773.
Manter, D.K., Reardon, C.L., Ashworth, A.J., Ibekwe, A.M., Lehman, R.M., Maul, J.E., Miller, D.N., Creed, T.B., Ewing, P.M., Park, S., Ducey, T.F., Tyler, H.L., Veum, K.S., Weyers, S.L., Knaebel, D.B. 2024. Unveiling errors in soil microbial community sequencing: A case for reference soils and improved diagnostics for nanopore sequencing. Communications Biology. 7. Article e913. https://doi.org/10.1038/s42003-024-06594-8.
Hammond Wagner, C.R., White, A., Darby, H., Ewing, P.M., Faulkner, J., Fisher, B., Galford, G., Horner, C., Jones, W.V., Neher, D., Von Wettberg, E., Zeraatpisheh, M. 2025. Holistic systems thinking underpins Vermont farmers and practitioners soil health preferences and beliefs. Soil Security. 19. Article 100186. https://doi.org/10.1016/j.soisec.2025.100186.
Kantar, M.B., Ewing, P.M., Bancic, J., Blair, H., Garba, I., Jamshidi, S., Jannink, J., Jha, P., Jungers, J., Pathak, H., Paul, S., Raghavan, B., Runck, B.C., Singh, J., Subedi, S., Joshi, V.R., Wang, D.R. 2025. Computational design for more engaged, impactful, and dynamic agricultural research.. Crop Science. Article e70034. https://doi.org/10.1002/csc2.70034.
Lehman, R.M., Osborne, S.L., Ewing, P.M. 2024. When are you measuring soil ß-glucosidase activities in cropping systems? Agricultural & Environmental Letters. 9(2). Article e70002. https://doi.org/10.1002/ael2.70002.