Which Wildflower is Best at the Recruitment of Native Bees into Agricultural Areas? A comparison of perennial vs. annual wildflowers

Final report for FS17-296

Project Type: Farmer/Rancher
Funds awarded in 2017: $10,000.00
Projected End Date: 03/14/2018
Region: Southern
State: Georgia
Principal Investigator:
Expand All

Project Information

Abstract:

Which Wildflower is Best at the Recruitment of Native Bees into Agricultural Areas: A Comparison of Perennial Verses Annual Wildflowers?

Each year, honeybees contribute $15-20 billion in pollination services to U.S. commercial agriculture. Some crop yields decrease by more than 90% without honeybee pollination. Honeybees are in decline from Colony Collapse Disorder (CCD), which puts the global food supply at risk. Today, honeybee colonies are down by 40% compared to colonies available in the 1970’s.

The reduced availability of honeybee colonies has increased food production costs and reduced potential crop yields. In order to ensure their crops are fully pollinated, nearly all fruit, nut, and vegetable farmers rent honeybee hives. The shortage of honeybee colonies has resulted in a rapid increase in the cost of renting honeybee hives.

One possible solution to the loss of the honeybees is the native bees already present in the environment. With nearly 3500 species in North America alone, the diversity of different forms, pollen-strategies, and behaviors of native bees provide a wide range of use for agricultural operations. It is estimated that native bees already annually contribute $3 billion to U.S. agriculture.

Methods need to be developed that will recruit native bees to farms or orchards. In the past decade, studies have begun to examine native bee pollination behaviors and even how to boost bee abundances. The two most common approaches suggested are to provide additional nesting habitat and to provide additional food resources. The use of wildflowers (food resources) on the edges of target agricultural areas has recently received a fair amount of attention. However, wildflower systems have not been optimized. Since each region of the U.S. has its own unique mix of native bees, regional studies will be needed to determine the optimal mix and types of wildflowers needed to boost bee abundances.

Experiments were designed for the 2017 growing season at Mountain View Orchard, to compare and contrast annual wildflower with perennial wildflower pollinator recruitment. The three wildflower plots (100×100 ft2) installed in March 2016 were mowed down to a few inches above the soil in December 2016. This upkeep method insured ideal conditions for perennial wildflowers to bloom. Additional wildflower seeds were added to the plots in February.

In order to measure the number of native bees attracted to the wild-flower patches, bees were sampled each week from April to October (30 weeks). Each flower plot had 12 pan-trap sets (1 set = 1 blue bowl, 1 yellow bowl, and 1 white bowl). Four bowl sets were placed on the out-side edge, 4 bowl sets were placed from 25 feet from the plot center, and 4 bowl sets were placed 5 feet from the plot center. This allowed us to measure differences in: (a) insect location in the plot (edge vs. center), (b) insect color preferences, (c) the time of year insect groups are present, and (d) insect abundance and diversity.

During the 2017 field season, both annual and perennial wildflower species were present, and 4051 bees were collected at Mountain View Orchard. 2188 bees were collected in the three wildflower plots (plot 1=618, plot 2 =741, and plot 3 = 829) during the 27 collection days (March-September). In the adjacent apple orchard, 1982 bees were collected during 10 collection days (March-September). The majority of bees were collected during the apple bloom in April. The most common bee groups collected were similar to 2016, except for a lack of bumblebees. The collection’s most abundant groups included the small sweat bees (Lasioglossum species) making up 41% of the total collection, mining bees (Andrena species) – 21%, large sweat bees (Halictus species) – 9%, green sweat bees (Augochorella species) -6%, and Carpenter bees (Xylocopa species)– 5%.

In 2016, 3 wildflower plots were added adjacent to the Mountain View Apple Orchard. The average harvest of apples prior to the addition of the wildflower plots was 2225 bushels per season (based on 2014 and 2015 data). After the addition of the wildflower plots, the average apple production at the orchard surged to 3,325 bushels of apples (based on 2016 and 2017 data). The wildflower plots and the additional pollinators they recruited appear to have had a substantial impact on apple production.

By establishing a strong network of native bees in Georgia agriculture, we can make Georgia Agriculture more secure and sustainable. In addition, an increase in reliance on native bees means that farmers will spend less money on pollination services (e.g. renting honey bee hives), thus increasing farmer profits and potentially reducing food costs for the general public.

 

Project Objectives:

Main Objectives

The main objective of the proposed study is to assess the effects of perennial wildflowers in recruiting native bees, particularly the mining bees (Genus Andrena). Mining bees are likely to be the best group of native bees to pollinate early blooming crops such as apples and blueberries.

The secondary objective of the proposed study is to assess whether wildflower plots can substantially increase crop yields. It appears so in 2016. Will it occur again in 2017?

In this proposed project, we will measure the benefits of adding three wildflower plots adjacent to Mountain View Orchard in terms of (1) native bee abundance, and (2) apple production. Native bee abundance and diversity has been measured the past seven years at this orchard, providing a detailed understanding of the typical native bee biodiversity. Using past data, we will be able to measure changes in native bee abundance and diversity, and apple production due to the three wildflower plots.

 

Cooperators

Click linked name(s) to expand
  • Prof. Mark Schlueter (Educator and Researcher)
  • x x

Research

Materials and methods:

The main goal of the study is to determine which wildflower species are best at the recruitment of native bees into agricultural areas. In particular, we want to compare perennial wildflowers and annual wildflowers. Of particular interest is to determine whether perennial wildflowers (because they bloom early in the season) will be able to recruit mining and mason bees. It is hypothesized that early blooming wildflowers (perennials) will be the only wildflowers that have a chance to recruit these bees since these bees emerge in March (early in the season). Last year, annual wildflowers that bloom a little later (May) failed to recruit these groups of native bees.

Experiments were designed for the 2017 growing season at Mountain View Orchard, to compare and contrast annual wildflower with perennial wildflower pollinator recruitment. The three wildflower plots (100×100 ft2) installed in March 2016 were mowed down to a few inches above the soil in December 2016. This upkeep method insured ideal conditions for perennial wildflowers to bloom. Additional wildflower seeds were added to the plots in February.

In order to measure the number of native bees attracted to the wild-flower patches, bees were sampled each week from April to October (30 weeks). Each flower plot had 12 pan-trap sets (1 set = 1 blue bowl, 1 yellow bowl, and 1 white bowl). Four bowl sets were placed on the out-side edge, 4 bowl sets were placed from 25 feet from the plot center, and 4 bowl sets were placed 5 feet from the plot center. This allowed us to measure differences in: (a) insect location in the plot (edge vs. center), (b) insect color preferences, (c) the time of year insect groups are present, and (d) insect abundance and diversity.

The comprehensive sample routine allowed us to measure the success of the wildflower patches in recruiting pollinator groups over the whole growing season. The wildflower seed mix chosen has a mix of early spring blooming plants, summer blooming plants, and fall blooming plants, providing a continuous food source for the pollinators.

During the past 7 years, we have gathered comprehensive bee data at Mountain View Orchard. Using older data, we will be able to determine the recruitment effectiveness of the three wild flower plots.

In addition, by using the apple production records, we will be able to assess any changes in apple production in 2017. While many factors (rain, temperature, etc.) may impact apple production, we will be able to compare changes at Mountain View Orchard from 2010-2015 (no wildflower patches), the 2016 season (wildflower patches – only annual wild flowers) and the 2017 season (wildflower patches – annual and perennial wild flowers).

Research results and discussion:

  Figure-1.-Graph1                  Figure-2-Graph-2             Figure-3-Graph3      

Bee Abundance and Diversity Results

During the 2017 field season, both annual and perennial wildflower species were present, and 4051 bees were collected at Mountain View Orchard. 2188 bees were collected in the three wildflower plots (plot 1=618, plot 2 =741, and plot 3 = 829) during the 27 collection days (March-September). In the adjacent apple orchard, 1982 bees were collected during the 10 collection days (March-September). The majority of bees were collected during the apple bloom in April. The most common bee groups collected were similar to 2016, except for a lack of bumblebees. The collection’s most abundant groups included the small sweat bees (Lasioglossum species) making up 41% of the total collection, mining bees (Andrena species) – 21%, large sweat bees (Halictus species) – 9%, green sweat bees (Augochorella species) -6%, and Carpenter bees (Xylocopa species)– 5% (See Figure 1).

In comparison, during the 2016 field season when only annual wildflower species were present, 3779 bees were collected at Mountain View Orchard. 1797 bees were collected in the three wildflower plots (plot 1=646, plot 2 =490, and plot 3 = 661) during the 30 collection days (March-October). In the adjacent apple orchard, 1982 bees were collected during the 14 collection days (March-October). The majority of bees were collected during the apple bloom in April. The most common bee groups collected were the small sweat bees (Lasioglossum species) making up 32% of the total collection, mining bees (Andrena species) – 30%, Bumblebees (Bombus species) – 12%, and large sweat bees (Halictus species) – 10% (See Figure 2).

The wildflower plots (2016 and 2017) significantly increased native bee numbers from the pre-wild flower plots years (2010-2015).   However, the shifting from Annual wildflower species in 2016 to Perennial and Annual wildflower species in 2017 did not make a huge difference in overall bee abundance. However, the types of bees recruited did change.

The small sweat bee (Lasioglossum species) abundance significantly increased. Mining bees (Andrena species) and large sweat bees (Halictus species) did not significantly change. Bumble bee abundance significantly dropped; while the large size Carpenter Bee abundance significantly increase from less than 1% of the 2016 collection to over 5% of the 2017 collection.

Yellow colored wildflower species recruited the most native bees both in wide range of species attracted and numbers of individuals attracted. Yellow wildflowers should be emphasized in future wild flower planting. White wildflowers were more common in the Perennial wildflower species planted compared to the annual wildflower species that contained a large percentage of yellow flowers. Purple and blue wildflowers yielded poor results in attracting the native bees. The results indicated that flower color was a more important factor than whether the wildflower was Perennial or Annual. Overall, the results indicate the early blooming yellow wildflower species are best in attracting and maintaining native bees in the orchard.

The wildflower plots were designed to increase the numbers of native bees recruited to the area prior to the apple bloom. Compared to apple blooms in past years without wildflower plots, the number of native bee species did increase. During the whole season, 1863 bees were collected in the apple orchard; however, the 3 collections during apple bloom in April inside the orchard make up nearly 50% of the bees sampled. The most important “apple flower pollinating” native bee species were mining bees (Andrena species) – 63%, carpenter bees (Xylocopa species)– 20% and small sweat bees (Lasioglossum species) – 6%.   The Carpenter bees had replaced the Bumblebee as the large size “apple flower pollinating” native bee species.

For apple orchards, increasing the abundance of mining bees, Bumblebees, and Carpenter bees should be a targeted goal. However, it is possible that Bumblebee and Carpenter bees (both very large size bees) could interfere with each other’s abundances. Small sweat bees (Lasioglossum species) may also play an important role in apple pollination. However, observations over the past 5 years have shown population abundance surges in the mid-summer for these species, thus reducing their impact on apple pollination. These bees may be more important in the pollination of summer flowering vegetables.

Several other species also may play a minor role in apple pollination (they make up 1-3% of the bloom collections), these include the green sweat bees (Augochorella species), large sweat bees (Halictus species), and mason bees (Osmia species).

Apple Production/Harvest Results

In 2016, 3 wildflower plots (each 100 x 100 ft2 ) were added adjacent to the Mountain View Apple Orchard. The average harvest of apples prior to the addition of the wildflower plots was 2225 bushels per season (based on 2014 and 2015 data). After the addition of the wildflower plots, the average apple production at the orchard surged to 3,325 bushels of apples (based on 2016 and 2017 data).

In 2017, approximately 3500 bushels were harvested at Mountain View orchard. The 2017 harvest was even greater than the 2016 harvest (3150 bushels).   The wildflower plots and the additional pollinators they recruited appear to have had a substantial impact on apple production. Figure 3 illustrates the huge difference in apple production before and after the addition of wildflower plots.

Participation Summary
1 Farmer participating in research

Educational & Outreach Activities

10 Consultations
1 Curricula, factsheets or educational tools
12 On-farm demonstrations
2 Webinars / talks / presentations

Participation Summary

1 Farmers
1 Ag professionals participated
Education/outreach description:

Scientific Conference Presentation

(1) Local Presentations

Georgia Gwinnett College

 

(2) National Presentations

Entomological Society of America Annual Meeting

Learning Outcomes

1 Farmers reported changes in knowledge, attitudes, skills and/or awareness as a result of their participation
Key changes:
  • Pollinator Habitat Changes (Addition of wildflower plots next to the apple orchard)

Project Outcomes

1 Farmers changed or adopted a practice
3 New working collaborations
Project outcomes:

The project was successful in implementing a change (e.g. addition of wildflower plots) in the orchard that was successful in both (1) boosting the recruitment of native bee pollinators, and (2) increasing apple production throughout the orchard.

This new practice of adding wildflower plots adjacent to orchards or farms could boost crop production and lower production costs (farmers could rent fewer honeybee hives). In addition, wildflower plots could be a new method to recruit native pollinators in sufficient numbers to either replace or supplement the honeybee, which is in decline due to colony collapse disorder (CCD).

 

Potential Contributions

Why should the public care about boosting native bee populations in agriculture?

Increasing the number of native bee pollinators in commercial agriculture will result in: (1) lower production costs (e.g. fewer honey bee hives rented), (2) increased food production, and (3) lower food costs for the general public.

Food security will also be improved by recruiting native bees. One-third of the human food supply will no longer be completely dependent on the honeybee. We will have good pollination alternatives to the honeybee. This could be critical if honeybees continue to decline or die out.

 

Outreach

The study’s results and information about native bees was presented at scientific conferences (regional and national) and regional farm events/trade-shows.

In addition, the project’s results will be posted on the Bees of Georgia Website.

The website address is: http://native-bees-of-georgia.ggc.edu/. This website was created in 2015 to provide information on Georgia’s bees for farmers, researchers, and the general public. Acknowledgements to SARE are listed on the homepage.

 

Recommendations:

Wildflower plots clearly boost native bee abundance and increase apple production. We should repeat the experiment in order to determine if the results are similar each year. Moreover, additional experiments need to be performed to determine which wildflower species and wildflower combinations best recruit native bees.

In order to boost native bee recruitment in early Spring next season, several additional wildflower species will be added to the plots. It is hypothesized that early flowering (March) wildflowers will recruit even more bee species (mining and mason bees) to the apple orchard.

Wildflower plots that boost native bee pollinators are the next logical step in sustainable agriculture.

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.