Enhancing sustainable pollination on urban farms using native plant conservation strips and outreach

The overall purpose of this project is to 1) better understand how native plant conservation strips influence the composition of bee communities on urban farms and their associated ecosystem services and 2) determine the best approach to increase awareness and implementation of pollinator management on urban farms and, 3) create outreach materials and conduct field days and workshop series with our partnering organizations to increase overall awareness of native bees and their pollination services on urban farms.

Objective 1: Improve in-farm habitat for pollinators by utilizing native plant conservation strips on urban farms.

To assess the effects of implementing native plant conservation strips in urban farms on bee communities in cucurbits, we will establish conservation strips on urban farms and community gardens in the Detroit and Lansing metropolitan areas of Michigan. We will compare bee communities in similarly sized and managed urban farms with and without conservation strips. We will only include farms that practice either certified organic or similar management practices. This project will initially involve growers with a strong interest in sustainable agriculture and committed to maintaining management practices like habitat enhancements on their land. We hypothesize that habitat enhancements will positively affect bee abundance and richness in urban farms, similarly to farms in rural settings. We also predict that farmers will be receptive to our outreach efforts and gain an understanding of how to help pollinators.

This project will be conducted on urban farms and gardens which typically grow a large diversity of different crops on their farm, however, we will focus on bee communities within cucurbit plantings due to their dependence on insect pollination. We will conduct our surveys at 20 sites that grow cucurbits in Detroit and Lansing; 10 where we will establish native plant conservation strips and 10 that will serve as control sites without plantings. Each conservation strip will be ~30 m² and placed as close as practically possible to the cucurbit planting. We will measure and account for that distance during statistical analysis and we will standardize the size of the cucurbit planting size and type as closely as possible.

We will prepare the growers and provide expectations for establishing and maintaining flower strips prior to planting them. Pre-season, our team will visit each site to identify appropriate locations to install the plantings. We will conduct site visits to work with each individual grower to ensure that the conservation strip will complement their current farm setup. To prepare growers we will host an orientation meeting in both Detroit and Lansing to provide outreach for pollinator management, explain the purpose of pollinator habitat enhancements, and give an overview of the program.

To establish the native plants, farmers will be required to clear a ~30 m² section of the farm. Since we will be using plant plugs rather than seeds, we expect a high establishment success rate and will only need to clear the land immediately prior to planting. Growers will be given the option to clear their land either using a rototiller or broad forking method, depending on their preference. To reduce weed competition and increase the aesthetics of plantings, growers will be supplied with mulch and landscape fabric to be placed around plants. Some study sites will be planted in Spring 2023 under the habitat kit program led by our collaborator Stefanie Steele (see letter of support). The remaining sites will be planted in Spring 2024 with funds received from NCR SARE. Based on prairie plant establishment recommendations, we will advise growers to mow plots with a mower twice per year for the first two years to decrease weed competition and increase sunlight exposure; once prior to plant emergence in the spring and once again when plants become dormant in the fall. Due to the heat and frequent droughts that occur in urban spaces, farmers will be advised to frequently water plantings for the first 2 years. The health and maintenance of each planting will be monitored by our team with monthly check-ins with the growers and at least two site visits per summer.

Plant species (Table 1) for the conservation strips were chosen to provide the best foraging resources for native bees while considering stressors and limitations that are common in urban environments. We will obtain plants and receive consultation from Bill Schneider at Wildtype Native Plant Nursery (see letter of support) to ensure that we are including native plants that are attractive to pollinators, have at least one species that is consistently blooming throughout the growing season, are resistant to the unique stressors of urban environments (heat, drought), and do not exceed maximum plant heights ordered by city ordinances (under 4 ft). These species are also readily available for our team to source locally. We used multiple sources to determine that these species were the best fit for our project including technical reports, research datasets, and communications with local pollinator experts, including the Xerces Society’s Great Lakes Pollinator Resource Center (native plant lists, habitat installation guides), USDA PLANTS Database, Bill Schneider with Wildtype Native Plant Nursery, and MSU’s Michigan Pollinator Initiative.

From 2023-2025, sweep netting and visual observations will be used to survey bee communities at each site 1-2 times per season. We will visually observe 10 random cucurbit flowers (5 male, 5 female) for 1 min each and identify and count the number of bees that visit flowers (contact anthers or stigmas) including squash bees (Apidae: Eucera pruinosa), honey bees (Apidae: Apis mellifera), bumble bees (Apidae: Bombus spp.), and other wild bees. Other insect taxa (i.e., wasps, flies, beetles, butterflies, moths) will not be counted. We will conduct four 10-minute sweep netting sessions along 10 m x 2 m random transects within the flower strips at each site. All data collection will be done when weather conditions are sunny with no precipitation/wind. Wild bee species will be collected for later identification by bee taxonomist, Dr. Gibbs (see letter of support) at the University of Manitoba (CA). Bees that are collected will be placed in glass vials into a cooler to keep bees dormant and then placed into a freezer in the lab. To avoid overcollection and promote conservation, we will limit bee collection to those that we are unable to identify in the field. We will compare abundance and richness of bee taxa using data between farms with and without the conservation strips to determine whether increased floral resources impact native bee communities. Since cucurbits are an annual crop and many farmers practice crop rotation, we plan to investigate whether these perennial strips enhance overall pollinator habitat on urban farms. Urban farms are typically smaller in size (below 3 acres), so we hypothesize that incorporating floral resources will benefit bee communities based on their larger dispersal ranges.

To determine whether plant species composition, provided by conservation strips or other natural habitats, influence bee communities and pollination, we will also survey plant diversity within farms, and surrounding farms. At each field, we will place 10 random quadrats (1 m²) to count and identify blooming and non-blooming flowers to calculate bloom density. We will also conduct four 10m x 2m transects from the center of the perimeter of each side of the field, we will identify and count the number of flowering plant species and estimate bloom cover percentage.

To determine whether conservation strips have improved cucurbit pollination compared to sites without habitat enhancements, we will compare bee abundance from floral and transect observations in sites with and without conservation strips. To measure fruit set, we will tag 20 unopened squash blooms per site; we will place mesh bags over 10 blooms at each site to prevent insect pollination and leave the remaining 10 uncovered. Only female blooms will be tagged which can be identified by the presence of the ovary. We will then come back and remove mesh bags once the fruit begins to form and count the seeds within the fruit once it fully ripens (5-6 days later).

Objective 2: Understand urban farmer perceptions and decision-making tools used for making pollinator management decisions.

We will investigate how educational tools provided by universities and community-funded organizations are being utilized currently to inform pollinator management decisions. We will gather this information using (1) a survey that will be developed with the assistance of Dr. Liere (see letter of support) to survey farmers in Michigan and, (2) semi-structured interviews of farmers from the study sites. Consent forms will be submitted to the Institutional Review Board (IRB) for approval. We will include farmers from historically underserved backgrounds through extension educators and community contacts from the MSU Detroit Partnership in Food, Learning and Innovation (DPFLI), the Greater Lansing Garden Project, Michigan Food & Farming Systems (MIFFS), the Xerces Society for Invertebrate Conservation, and the USDA NRCS. We will use social survey methods to investigate the following metrics: their level of satisfaction with pollination, what pollinator management methods are currently being used if any, if growers are satisfied with these, their current knowledge of pollinator management methods, if they have access to educational resources, how they access resources currently, and whether differences in pollinator management on farms are influenced by sociodemographic factors (age, gender, education, experience, motivations).

We will lead a series of workshops in Detroit and Lansing educating urban farmers on the benefits and care of native flower plantings. Consequently, we will work with urban farmers to assist with preparing and planting the flower strips. Our proposed project will involve farmer participation from beginning to end; we have already conducted pre-surveys asking farmers about their concerns and willingness to implement pollination management methods. We plan to work with farmers individually as well as expand our outreach by holding workshops, field-days and presenting at regional meetings. We will work with and compensate farmers from the community to co-facilitate and host these workshops and field days. After data collection is complete, we will distribute farmer surveys created with the assistance of sociologists to determine whether our methods of outreach were effective with participating farmers and where we can improve. Based on the survey findings, we will generate outreach materials in collaboration with MSU Extension, MIFFS, and the Xerces Society, such as an urban pollination flip book, a squash bee and pollinator website, articles, and fact sheets. Ultimately, this project will build on existing knowledge of the role of native bees in sustainable crop production and will include many underserved communities and farmers.

Our analysis of our findings will focus on understanding the impacts of this project on the economic, environmental, social, and productivity of sustainable urban agriculture in these communities. We will analyze all data from our study using statistical methods with R Studio. To determine the effects of conservation strips on bee community abundance and richness, we will use generalized linear models to compare experimental sites and control sites. Bee community composition among sites will be compared using non-metric multidimensional scaling. To understand the effects of farmer educational tools and outreach on the implementation of pollinator management methods, we will use social-ecological modeling to analyze survey data by creating a global generalized linear model to compare outreach mediums and rate of implementation.