Evaluating flowering plant selection for pollinator habitat enhancement: Open-pollinated natives vs. native cultivars

Project Overview

Project Type: Partnership
Funds awarded in 2012: $14,850.00
Projected End Date: 12/31/2014
Region: Northeast
State: Vermont
Project Leader:
Dr. Leonard Perry
University of Vermont

Annual Reports


  • Additional Plants: native plants, ornamentals


  • Education and Training: on-farm/ranch research
  • Farm Business Management: whole farm planning
  • Natural Resources/Environment: biodiversity, habitat enhancement, wildlife
  • Production Systems: agroecosystems, holistic management, organic agriculture, permaculture

    Proposal abstract:

    Ecosystem sustainability and the success of many agricultural crops in the northeast rely upon the mutualistic relationship between plants and pollinators. This research seeks to improve flowering plant selection for pollinator habitat enhancement by comparing “true” native plants (open-pollinated) to native cultivars (human-bred) in terms of their ability to attract and support native pollinators.

    Thirty-five percent of food crops worldwide require animal-mediated pollination (Klein et al. 2007), making sustainable pollination services integral to global food supply. Like most of America, the northeast relies heavily on the services of a single domesticated species, the European honey bee (Apis mellifera). Bee keepers have struggled in recent years to maintain healthy populations of honey bees, given their susceptibility to parasitic mites and colony collapse disorder. Declining honey bee populations and rising costs for employing their pollination services have led farmers to reevaluate the potential role of native bees, butterflies and hummingbirds for pollinating their crops. The USDA and NRCS are actively promoting the restoration of pollinator habitat to both agricultural and human-dominated landscapes.

    Restoring pollinator habitat with plants indigenous to the region is recommended because native plants are evolutionarily adapted to coexist with native pollinators (Comba et al. 1999). However, no previous research has focused on the ecological differences between open-pollinated native plants and native cultivars, which are more readily available for purchase at garden centers. This research will identify the potential tradeoffs of using native cultivars in place of open-pollinated “true” natives for the restoration of pollinator habitat.

    We will accomplish outreach by a demonstration garden, a specific project website, and presentations at NRCS regional meetings and others.

    Project objectives from proposal:

    The project will begin with an informal survey of plant nurseries in Vermont to assess the current interest and commitment to growing and selling native plants; issues with native plant availability; prominence of true native (open-pollinated, local genotype) plants versus native cultivars; and which species are most commonly grown and sold. A minimum of ten native herbaceous perennial species will be selected for the study and will be compared to common ‘improved’ cultivars of the same species.

    Cultivars will be chosen for the study based upon their popularity in the retail garden market, bloom period (early, mid, and late-season bloomers), and method in which the cultivar was established. When possible, one cultivar will originate from a natural selection and the second cultivar will originate from a controlled cross.

    Plants will be obtained in quart or gallon-sized pots from plant suppliers. Efforts will be made to obtain open-pollinated native plants with genotypes most local to the study sites in northwestern Vermont.

    Using the selected species, pollinator gardens will be designed for three study sites. Two of the study sites will be situated on the properties of cooperating farmers, River Berry Farm in Fairfax, Vermont and Full Circle Gardens in Essex, Vermont, and a third study site will be located on the University of Vermont campus. The study sites were chosen for the diversity in their immediate surroundings (fruit and vegetable crops, plant nursery, and urban campus), diversity in the greater surroundings (agricultural, residential/forested, and urban/suburban), and their ability to support the educational objectives of the study. The diversity in sites for the pollinator gardens will allow for comparisons not only between open-pollinated natives and native cultivars (and the method of cultivar selection), but also between site variables such as competing nectar/pollen availability and landscape surroundings.

    To control for environmental variables, each pollinator garden will contain both the true native and the native cultivars. The design of each garden will vary slightly based upon the site constraints and opportunities to display the plant material to the public.

    Each pollinator garden will have a total of 270 plants. Ten native herbaceous flowering perennial genera will be represented—one open-pollinated and two cultivars. Nine plants of each plant type will be grouped into three replicates of three plants each, in a randomized complete block experimental design for each genus at each site.

    The pollinator gardens will serve to provide valuable data and will also serve to educate visitors to the sites. The farmstand and greenhouse at River Berry Farms, the retail plant nursery at Full Circle Gardens, and the University of Vermont campus are open to the public and have regular visitors, who will be able to tour the perimeter of the pollinator gardens and learn about the value of native plants and pollinators from interpretive signage. Each plant species and cultivar will be clearly labeled and additional educational signage will be designed and installed.

    During the first growing season, ongoing maintenance of the study gardens, including watering and weeding as needed will be performed. Preliminary field surveys will be conducted in year one to assess species richness and abundance of native pollinators including bees, butterflies, and hummingbirds. This preliminary data will be used to identify the predominant pollinator species in the research area. In consultation with UVM entomologists and statisticians, the most rigorous method for grouping and quantifying these pollinators will be employed. For example, due to the large abundance of native bee species and the difficulty of identifying them in the field to the species level, it may be deemed appropriate to group native bees into four sub-groups based upon morphology, field-recognizability and pollination efficiency. Such groups could include bumble bees, other large bees, green bees, and small bees, similar to methods utilized by Winfree et al. 2007.

    During the second growing season, field data will be collected by the project manager under the guidance of the principal investigator. Each garden will be visited weekly throughout the blooming season (early-June through late September). Each visit will rotate between morning, midday, and evening. Pollinator visitation rates will be measured for native and native cultivar flowers as pollinator visits per flower per unit time. Pollinator visits to as many flowers as can be viewed simultaneously within each approximately one by one meter plant block will be recorded during 60 second scans. Each plant block will be scanned three times per garden visit.

    At the visually-observed peak of each plant’s bloom period, nectar volume and sugar concentration will be measured using a microcapillary tube and a hand-held refractometer. Pollen mass will be measured manually. A minimum of 8 samples will be taken per plant type at each site, totaling 24 nectar/pollen samples per plant type.

    The data will be compared using analysis of variance and mean separation to determine if pollinator preference differs among plant types for each genus at each site. If there is insignificant variation among sites, then the sites will be analyzed together. This is a two factor statistical analysis where plant type is one factor and site is the second. Analyses will be will performed for each pollinator group.

    Collaborating entomologists, plant physiologists, and statistical consultants from the University of Vermont will review the proposed data collection methods and provide guidance throughout the study to ensure the research methods are rigorous and yield statistically valid results.

    The dissemination of information will begin early in the research process and will continue throughout the conclusion of the project. The pollinator gardens designed and installed for the purpose of this research will also serve as educational sites. Customers visiting the nursery at Full Circle Gardens, the farm stand at River Berry Farm, and the University of Vermont Campus will be able to view the pollinator gardens, identify individual plants, read interpretive signage about the research, and take away an educational pamphlet on the importance of native pollinator habitat. A project website will be created and maintained by the Project Manager to blog about all phases of the research including the design, implementation, data collection, analysis, and results. The website will also serve as a portal for information on native plants and pollinator habitat enhancement in the northeast. The results of this research will be provided in a final project report to Northeast SARE. Information will also be tailored for inclusion in the Northeast SARE newsletter. Project findings and recommendations will be presented directly to regional NRCS programs that are involved in promoting pollinator habitat restoration in agricultural landscapes. The results will also be disseminated to the regional nursery and landscape industry through the Vermont Nursery and Landscape Association newsletter. Written articles will be tailored to specific audiences (academia, farmers, conservationists, growers, retailers, designers and home gardeners) and submitted for publication. These may include HortScience for the academic audience, American Vegetable Grower for vegetable farmers, Native Plant Journal for conservationists, American Nurseryman for the plant industry, and fine gardening for home gardeners. A minimum of three articles will be written and submitted for publication. Three or more public speaking engagements will be sought to present the research findings to a diverse audience. Speaking opportunities may include the New England Vegetable & Fruit Conference, Vermont Nursery & Landscape Association annual meetings, New England Nursery Association meetings and conferences, master gardener groups and local farm and garden workshops.

    Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.