Exploring the efficacy of prairie strips as a soil health promoting practice

Our study will use a mixed methods approach to understand the efficacy of prairie strips in building soil health and optimizing economic profitability within row-crop agriculture. The research component includes experiments on-station and on-farm. The education and outreach component includes an interactive decision support tool to help farmers decide when and where it is economically advantageous to plant prairie strips, as well as programming to revise and disseminate this tool. 

Objective 1: Assess how prairie strips impact soil health and yield. 

The extent to which prairie strips can enhance soil health and yield will be examined using the newly established KBS LTAR. The proposed work is the first experiment of its type to explore how prairie strip placement in low fertility and low yielding areas improves soil health and crop yield. The experiment will be conducted in the LTAR scale-up plots, which are designed to mimic farmer fields in size (500 m by 500 m). The experiment is a randomized complete block design with two replicated blocks. Treatments include: 1) a ‘business as usual’ treatment consisting of a tilled corn-soybean rotation, 2) an ‘aspirational’ treatment, managed as a no-till corn-soybean-wheat-canola-forage, with rye and clover cover crops planted after corn and soybeans, respectively. Each phase of the rotation is present every year. Prairie strips are only found within the aspirational fields and have been incorporated at four locations per field. Prairie strips were planted in locations that were known to be low yielding areas based on a decade’s worth of yield stability and profitability maps. 

The prairie strips were established in April of 2022 in 9.1 m rows. The prairie strips were sown using a 22-species seed mix consisting of 4 grass species and 18 forbs, purchased from Native Connections (Kalamazoo, MI). The mix is designed to have species in bloom throughout the course of the growing season.

Soil cores will be collected at designated sampling stations. The predetermined sampling stations capture variability in field topography (depression, edge, slope, and summit). In the aspirational fields where prairie strips exist, sampling will occur at the four topographically designated sampling stations within the strip and four topographically designated sampling stations in the adjacent row cropped area. The prairie and row-cropped sampling stations are located 5 m apart. At each sampling station, ten 0-10 cm soil cores will be collected and composited.

Soil health analyses will assess soil food web structure and ecosystem function: Soil Biological Health measurements (beneficial nematodes, lipid-based fungal and bacterial community composition, microbial biomass carbon, extracellular enzyme assays, soil respiration rates, soil protein, permanganate oxidizable carbon); Soil Chemical Health measurements (soil organic matter, total carbon and nitrogen, total organic carbon and organic nitrogen); and Soil Physical Health (aggregate stability, infiltration, texture, bulk density). 

Yield monitoring: Crop yield and aboveground plant biomass will be determined for each crop and forage at the field scale. Yield data will be collected using a combine. This will allow us to compare large-scale field assessments in the aspirational versus ‘business as usual’ treatments to determine to what extent the presence of prairie strips influences yield at scale. Above ground plant biomass will be collected using quadrants at sampling stations. 

Sampling will occur in years 1-3 of the study. Moreover, since each phase of the rotation is present every year, we will be able to assess how prairie strips are influencing soil health and yield in diversified crop rotations through time. 

Data analysis: Treatment effects of  cropping system; soil source (prairie strip or adjacent cropland), and topographical location, will be assessed each year and over time for the three year project period. For data collected each year, we will assess the effects of treatments on all univariate soil health metrics using a blocked design random linear effects model. To assess changes during the 3-year project period, we will incorporate  a repeated measures term into the model. For lipid analyses, we will examine total lipid biomass and fungal-to-bacterial lipid ratios as univariate measurements, but we will also examine the multivariate lipid-based communities using ordination approaches and permutational analysis of variance.

Objective 2: Establish on-farm chronosequence trial and prairie strip demonstrations

We will partner with three farmers to explore how prairie strip implementation influences soil health and yield on farms across Michigan. Our partner farmers have a wide range of prairie strip experience – some are in the initial phases of implementing prairie strips, while others have managed prairie strips for multiple years (see letters of support). We will conduct sampling across a chronosequence of prairies to assess how the effects of prairie strips on soil and crop yields could be predicted to change over time.  For this portion of the project, we will collect soils from prairie strips that have only recently been established (year 1), strips that are nearing maturity (years 3-4), and strips that have been in place for almost ten years (Edward Lowe Foundation (ELF) Farm). In conversations with farmers, one of their biggest questions they ask about incorporating prairie strips is how long it will take to see the benefits of using this conservation practice. Our chronosequence approach will allow us to demonstrate soil health benefits in new (on-farm) vs. young (on-farm and KBS) vs. established (ELF) prairie strips.  

At each farm, farmers will select locations for soil sampling. At these locations we will collect soil samples along a transect. Four soil samples within each prairie strip will be collected per year, as well as four samples in the adjacent row-cropped area, 5 meters from the prairie strip. The same soil health metrics conducted for Objective 1 will be analyzed to address Objective 2. We will ask farmers to share yield data, which we will then compare to soil health metrics. This will provide information on how proximity to prairie strips influences yields. 

Data Analysis: Treatment effects (i.e. three prairie strip age categories) and soil source (prairie strip or adjacent cropland) will be assessed each year and over time for the three year project period. Analysis using a blocked random linear effects model will be used to test the effects of treatment and soil source on all univariate soil health metrics. To assess changes over time during the 3-year project period, we will use a repeated measures version of this model. Multivariate lipid-based communities will be examined usingordination approaches and permutational analysis of variance. 

By the end of year three, we will have a large dataset of soil parameters and crop yields from numerous farmer fields with different soil types and management histories. To integrate this information we will use exploratory factor analysis (EFA) to assess which aspects of the soil food web most influence soil C and overall prairie ecosystem functioning within each treatment and time point (Wade et al., 2022). EFA is a latent variable analysis that can be used to assess which underlying variables (i.e. nematode communities) best predict a given factor (i.e., SOM cycling). Understanding how the relationships between different parts of the soil food web shift through time and under different management will be compelling as we work to uncover the mechanisms that drive soil health on farmer fields. 

Objective 3: Launch decision support tool to help farmers understand when prairie strips are economically optimal. 

Economists at MSU have developed a prairie strip budget to calculate opportunity costs associated with prairie strip implementation. In theory, prairie strips create a monetary loss because they take land out of production. However, in practice, farmers can use yield maps and only add prairie strips in low yielding areas. Areas that yield 29% lower than the USDA average have zero opportunity costs, meaning that it is no longer worth planting or applying fertilizers in those areas (Basso et al., 2019). This means, at no net loss to the farmer, that land could be put under a conservation management practice (Gammans et al., in prep). MSU economists created a partial budget style framework to explore the associated gains (increased revenues) and associated losses (lost revenue) with prairie strip implementation. Revenue gains include payments from the CRP, since prairie strips are a new practice considered in this program as of 2018 (Agricultural Improvement Act, 2018).  Prairie strips also have the potential to qualify farmers for a Climate-Smart Commodities partnership project in the future. Decreased costs associated with planting, fertilizer, and harvest are also considered as monetary gains within the budget. The main potential source of lost revenue when prairie strips are implemented is from land taken out of production, reducing crop yield. Other losses are directly linked to increased costs associated with prairie strip planting. Site preparation including tillage and herbicide application could add to labor and input costs. Perhaps the largest cost considered is acquiring the native seed and continued management in subsequent years. The budget takes these gains and losses into consideration and can help farmers identify when it is most optimal to incorporate prairie strips within their farm operations. 

To date, this partial style budget only exists in a spreadsheet and is not publicly available to the farming community. A major objective of this study is to create and launch a decision support tool based on the calculations conducted by our economists (see letters of support). This tool can be altered to accommodate changes in the 2023 Farm Bill, longer-term soil C sequestration benefits and possibly extended to the Partnerships for Climate-Smart Commodities program goals once we know more about how prairie strips influence the soil health metrics we propose to measure. 

Our team will collect management surveys and economic data that will further aid in the development of the decision support tool. We will collaborate with economists to update calculations so that the tool can be used beyond corn-soybean and wheat growers so that the decision support tool can be used by a wider range of farmers across Michigan. Lastly, we will also work to assess how soil health information can be integrated into the decision support tool, which could further inform payments for future C credits. We will hold a webinar so that Extension Educators and farmers can learn how to use the tool. A recording of the webinar will remain on a MSU extension page along with the interactive decision support tool, so that farmers can easily access training when working with the tool. We will also hold a special workshop for Extension Educators, so that they can better understand the tool, ask researchers questions about the decision support tool and can then advertise the tool more broadly. Launching this web-based decision support tool will expose hundreds of farmers across the state to prairie strip Implementation and we will use this to expand our on-farm trials and demonstrations in years 2 and 3 of this study. 

References: 

Basso, B., Shuai, G., Zhang, J., Robertson, G.P., 2019. Yield stability analysis reveals sources of large-scale nitrogen loss from the US Midwest. Scientific Reports 9, 5774. 

Gammans, M. Laporte, J., Wilke, B., Schultheis, E., and Baziari, F.  In Prep. Prairie Strips Ecological Benefits - a partial style budget. Michigan State University Extension. 

H.R.2 - 115th Congress. Agricultural Improvement Act of 2018. 

Wade, J., Culman, S.W., Gasch, C.K., Lazcano, C., Maltais-Landry, G., Margenot, A.J., Martin, T.K., Potter, T.S., Roper, W.R., Ruark, M.D., Sprunger, C.D., Wallenstein, M.D., 2022. Rigorous, empirical, and quantitative: a proposed pipeline for soil health assessments. Soil Biology and Biochemistry. 170, 108710.