Providing habitat for native pollinators and determination of native pollinator contribution to pollination of cucurbits and blueberries at farm sites

Final Report for GS08-077

Project Type: Graduate Student
Funds awarded in 2008: $10,000.00
Projected End Date: 12/31/2010
Grant Recipient: Univ. Tennessee
Region: Southern
State: Tennessee
Graduate Student:
Major Professor:
Dr. John Skinner
Univ. Tennessee
Expand All

Project Information


This study set out to document what groups of bees are providing visitation to insect pollinated crops in East Tennessee, and what bees will utilize a number of flowers used as bee food. In 2008-09, bee visitation data was collected on 12 farms and 10 different crops. On one of these farms, UT’s Organic Crops Unit, visitation data was collected for 26 different flowers that could be used in a bee food plot.


Native bees have been shown to provide significant levels of crop pollination in areas where natural habitat exists (Winfree et al 2007). While in areas of more intensive agriculture, such as found in California’s Central Valley, crop pollination is nearly entirely dependent on the non-native, commercially managed honey bee (Apis mellifera) (Kremen et al 2004). Regardless of whether we consider native bees or honey bees, bee populations in general are in decline. This fact threatens the ecology behind food production (Spivak et al . 2011). It is therefore important to document what bees are providing visitation to insect pollinated crops. To improve the health of bees, numerous strategies are underway including the enhancement of natural habitat for diverse and pesticide-free pollen and nectar sources through establishments of flower plots for bees (Spivak et al 2011). Regional planting guides for bee food plots can be found, however actual tests of the plants are not well developed.

Project Objectives:
  • Study the community composition of bees providing visitation to insect pollinated crops and flowers that could be used in bee food plots
    Determine the importance of native bees in crop pollination
    Determine if the flowers tested for bee food provide food to bees that provide crop visitation


Click linked name(s) to expand
  • Jerry Baird
  • Brian Baxter
  • Dan Eiser
  • Dennis Fox
  • John Skinner


Materials and methods:

In 2008-09, bee visitation data was collected on 12 farms and 10 different crops. On one of these farms, UT’s Organic Crops Unit, visitation data was collected for 26 different flowers that could be used in a bee food plot. At each farm, one or more transects were established within a given flower type, or bee food mix as appropriate. For each observation, a transect was walked and the number of bees visiting flowers recorded, noting the bee’s identification by sight within a classification scheme of 9 groups. A reference collection of specimens was developed by netting and trapping to better describe what species these 9 groups represent. This quick and efficient strategy allowed for 446 transect observations with 26,630 bees counted. Environmental data for each observation was recorded as the type of flower the bees occurred on, the date, time, location, if beehives were present, the size of the farm and plantings, if the farm was organic or conventional, and the number of flowers and sex of flowers when appropriate.

The data was analyzed in R by non-parametric, multivariate statistics and uni-variate means. For multivariate analysis, non-metric multidimensional scaling (NMDS) utilizing Bray-Curtis rank similarity was used to describe the community structure of bees at flowers. Permutation testing, weighted average confidence intervals, vector fitting, and thin plate regression splines, allowed for significance testing and graphics to describe differences in environmental factors on the NMDS ordination. Uni-variate, linear models were used to develop means of bee groups for individual flower types.

Research results and discussion:

Conclusions based on these analyses find that native bees are important visitors to crop flowers, but their importance and composition depend on the type of flower. Within a flower type, other environmental effects can shift the community structure. Plants selected for bee food plots can be chosen based on the similarity of the community of bees which utilize them as compared to crops. This study establishes a record of the importance of native bee visitation to crops within the study period and location. It also conceptualizes a method to develop a plant list for bee food plots specific to any location where the plants are tested.

Participation Summary

Educational & Outreach Activities

Participation Summary

Education/outreach description:

The publications are still in-press. This data will be presented in a masters thesis and two articles for peer review journals. Talking with co-operating farmers gave the opportunity to present the benefits they are seeing from native bees on their own farms.

Project Outcomes

Project outcomes:

Enough data was collected for a Master’s Thesis and 2 articles in peer review journals (still under development). This sets a good start to similar pollination ecology studies with the Entomology dept. at UTK. It also documents native bee benefits to crops in our area, underlining the importance of conservation and acts as a preliminary trial of several flowers that could be used in bee food plots.

Economic Analysis

Economically, this study shows a benefit in pollination from the surround habitat in places were no clear effort is being made to conserve or enhance the surrounding habitat. So, all pollination from native bees was provided free of charge. Considering that native bees often outnumbered honey bees, a greater amount of visitation was being provided by this free source then from rented honey bees.

Farmer Adoption

At least two of the farmers are planting food plots for bees. All of the farmers seemed to understand and appreciate the pollination services they were receiving by native bees. They also seemed to understand the need to still have honey bees available on their farms to supplement the native bee presence.


Areas needing additional study

Many of the plants tested for bee food plots did not attract many bees, even though these were listed in planting guides developed elsewhere. Additional trials are needed to test plants that are effective in providing food for bees in the southeast. Although the seed mixes contained many species of plants, few grew. Most were out competed by dominate flowers. A study is needed which examines the effectiveness of the flowers ability to grow and feed bees in various southeastern conditions.

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.