Enhancement of pollination by native bees in blueberries and cranberries

Final Report for SW08-056

Project Type: Research and Education
Funds awarded in 2008: $183,271.00
Projected End Date: 12/31/2011
Region: Western
State: Oregon
Principal Investigator:
Dr. Sujaya Rao
Oregon State University
Expand All

Project Information

Abstract:

Pollination is critical for production of blueberries and cranberries. In the Pacific Northwest, growers rent honey bee hives for pollination. However, honey bees do not perform as well as native bees in these crops due to prevailing weather conditions during bloom and due to their preference for other flowers in which nectar is easier to access. In contrast, native bees forage in wind, rain and on cloudy days. Also, economically, native bees are less expensive and invasive. Hence, the study was conducted to determine strategies for enhancing populations of native bees for increasing pollination and improving yield and grower profit.

Project Objectives:

Our long term goal was to enhance native bee populations in berry fields for augmenting pollination and thereby increasing yield and income for berry producers. In this project, two separate programs were integrated into one, with similar experiments conducted in blueberry fields in the Willamette Valley and cranberry fields on the coast. Our objectives were to:

1. Estimate native bee pollinator species diversity and abundance in berry fields.

2. Compare honey bee and bumble bee foraging behaviors in berry fields.

3. Evaluate the impacts of insecticide sprays on native bees.

4. Examine strategies for enhancing native bees in berry fields.

5. Build capacity in growers to identify, protect and enhance native bee pollinators in their fields for increasing berry production.

Introduction:

Pollination is a critical factor in production of blueberries and cranberries, two important ericaceous crops in the Pacific Northwest. Both crops require pollination by bees (Dogterom et al. 2000; Eck 1990, Brown 2006). Optimal pollination results in larger fruit, better fruit quality, consistency in the shape of the fruit and earlier ripening of berries. Hence, if pollination is maximized throughout bloom, growers profit in terms of quantity and quality of berries produced, earliness of harvest and greatest harvest at first picking.

Blueberry and cranberry growers typically rent one – four honey bee hives for pollination. Unfortunately parasitic mites and the recent colony collapse disorder have reduced the number and availability of honey bees and made their rental more expensive. In addition, honey bees do not perform well in both blueberries and cranberries due to cool weather and wind during bloom and due to preference for other flowers in which nectar at the base of the flower is easier to access (Free 1993). In contrast, native bees forage in wind, rain and on cloudy days. Often they are the first and last bees that are active during a day. Economically, native bees are less expensive and invasive as they are a natural part of the ecosystem.

Native bees, especially bumble bees, are also better pollinators of blueberries and cranberries because they are capable of buzz-pollination. Bumble bees hang on to the flower and buzz it by vibrating their muscles that control flight. The pollen in the flower is actively shaken loose and released onto the queen or worker bee, and the bee then grooms the pollen grains onto her hind legs. After visiting many flowers to collect pollen, she will have accumulated a large ball of pollen on each hind leg and will have cross-pollinated the flowers along the way. Honey bees are not able to buzz pollinate.

Worldwide there are reports of declining populations of native bees, especially bumble bees, due to diseases, pesticides and habitat changes (Biesmeijer et al. 2006). In the late fifties, the Pacific Northwest was reported to have a rich fauna of native bees (Stephen 1957). However, dramatic changes have been reported in recent years. On the west coast, the abundance of one of the key pollinator species, the bumble bee Bombus occidentalis, has dropped drastically since the late 1990s. Only seven specimens of this species have been recovered since 2003 (Rao and Stephen 2007). Other species may have declined, but their status remains largely unknown.

In addition, native bee populations are not consistent year to year, and their populations may be inadequate for the large numbers of flowers present in agricultural fields. Growers in majority of U.S. states can purchase commercial bumble bees for crop pollination. Growers in Oregon are at a disadvantage, as Oregon is one of three states where introduction of commercial non-native bumble bees is prohibited (http://www.oregon.gov/ODA/PLANT/IPPM/appr_insects.shtml), due to speculated simultaneous inadvertent introduction of diseases in the past. Berry farmers in Oregon are therefore dependent on naturally occurring populations which are, unfortunately, not consistent or predictable and low in numbers compared with the large abundance of flowers needing to be pollinated. To provide them a competitive edge in the market, there is an urgent need to enhance populations of native bees for increasing pollination and improving yield and grower profit.

Cooperators

Click linked name(s) to expand
  • Knute Anderson
  • Todd Anderson
  • George Bussman
  • Peter Dinsdale
  • Michael Helms
  • David and Marci Kranick
  • Robert Nelson
  • Kim Patten
  • William Stephen
  • Linda White

Research

Materials and methods:

Objective 1. Estimate native bee pollinator species diversity and abundance in berry fields.

Both blueberries and cranberries are known to be well pollinated by bumble bees. In Oregon there is considerable diversity in native bees, especially bumble bees (Stephen 1957). However, there is little information on which species are present during berry crop bloom. Hence, two separate studies were conducted to determine the species diversity and abundance of native bees during blueberry and cranberry bloom in Oregon. Each study was conducted over two years.

In both studies, bees were monitored using blue funnel traps that have proven to be effective sampling devices in earlier studies (Stephen and Rao 2005). The traps consist of a clear plastic collecting jar, 15 cm diameter x 15 cm high, fitted with a polypropylene screw cap funnel into which two 24 x13 cm (3 mm thick) polypropylene semitransparent blue cross vanes are inserted (SpringStar™ LLC, Woodinville, WA).

Blueberry: Three blueberry farms (A,B,C) in the Willamette Valley, a key blueberry production area in western Oregon, were selected for the study. Farm A is conventionally managed, Farm B was conventionally farmed in 2009 and began the transition to organic practices for certification in 2010, and Farm C is certified organic. At each site, approximately 10-ha were included in the survey which represented the entirety of Farms A and C and a subsample of Farm B. Each farm included an on-site residence and was bordered by agriculture, residential development and roads. Each farm stocked commercial honeybee colonies in April and May during blueberry bloom. Farm B stocked eight hives per acre; Farm A stocked two-three hives per acre; and Farm C stocked three-four hives per acre.

To cover periods of bloom and post bloom, each site was monitored over two years between May and September at approximately three-week intervals for a total of six sampling events per year (May 17, June 3, June 24, July 16, August 13, September 6 in 2009, and May 9, June 3, June 24, July 16, August 6, September 2 in 2010). Sampling dates were shifted as necessary to ensure nominally favorable weather and avoid any effects from large-scale or local field disturbances such as storms, spraying or harvest. For each sampling event, the traps were hung for a 48-hour period, then collected and transported to the laboratory where bee specimens were frozen, pinned, sorted and identified.

Cranberry: The cranberry-growing region of southern coastal Oregon extends from just north of Bandon to just north of Port Orford, a distance of more than 40 km. Because no previous research has been done on the native bee composition of this region, sites were chosen to cover the north-south gradient. Four farm sites (Sites 1-4) were selected based on availability, spanning a distance of 22 km. Absolute distances between the sites ranged from 4.2 km to 9.8 km. Cranberry acreage at each site varied from 4.5 hectares to over 65 hectares. The composition of the area surrounding each site also varied from roadside and pastureland to completely forested.

Blue vane traps were hung at each site on rebar posts between two cranberry beds and the surrounding environment. Two traps were placed at each farm site at dawn and collected at dusk (12 hours later) for a total of eight traps per sampling event. Sampling events occurred once every other week before and after cranberry bloom, and once per week during bloom. Samples were collected from mid-May through late July in 2008 and 2009. The bees caught in the traps were transferred to paper cups with lids and frozen; they were later thawed, pinned, labeled and identified.

2. Compare honey bee and bumble bee foraging behaviors in berry fields.

Both blueberry and cranberry growers rent honey bee hives for pollination. However, honey bees are not considered to be very effective in these crops. In contrast, bumble bees are considered to be excellent pollinators for both crops, but their efficiencies under Oregon conditions are not known. Hence, studies were conducted to examine the foraging behaviors of honey bees and bumble bees in blueberries and cranberries in Oregon.

Blueberries: To determine whether honey bees from hives placed in blueberry fields were foraging on the crop, pollen traps were placed in honey bee hives. The traps were placed for two days in three blueberry orchards in 2009 and 2010, one-three times during bloom. A subsample of 50 pollen loads from each trap was subjected to acetolysis, which burns internal parts so that structures on the external pollen shell are preserved for identification.

To compare foraging behaviors of honey bees with bumble bees, in a separate study, pollen loads were examined.

Cranberries: A similar pollen trap study was conducted to examine honey bee foraging behavior. In 2011, over six weeks during cranberry bloom, six honey bee hives were placed on a cranberry farm near Langlois on the south Oregon coast. In addition, due to availability of bumble bee colonies, six bumble bee colonies were also placed in the same area. Pollen was collected from the honey bee colonies using Sundance pollen traps, and from the bumble bee colonies using direct observation and collection in vials. One hundred pollen grains were counted from 1,647 pollen loads individually analyzed using acetolysis, and pollen loads were identified via light microscopy using a reference collection prepared from the research site. For the reference collection, anthers were collected from flowers observed blooming around cranberry beds, and the pollen processed as described above. From each load, 200 pollen grains were identified. Because ericaceous pollen grains are similar in morphology, but differ in size, size was used to separate cranberry pollen (12.5 – 15.0 ?m across) from other ericaceous pollen (10.0-12.0 ?m).

During cranberry bloom, it is typically cold and windy on the Oregon coast. Hence, a second study on honey bees and bumble bees was conducted to determine the impacts of temperature and wind on foraging behaviors. The numbers of foraging honey bees and bumble bees were estimated during two-minute counts of visual observations along a 1 m-wide transect in one cranberry bed. Before every count, wind speed and temperature were measured with an anemometer. In addition, bees were hand-collected by placing a vial over a flower, chilled, identified, and their pollen loads removed with a paintbrush. Subsequently, the bees were released. The labeled pollen was chilled in the field and subsequently dried, weighed and identified using acetolysis. This study documented the composition of pollen on bees foraging in the crop, while the study described earlier documented foraging of bees in colonies placed near the crop.

Data on the numbers of foraging honeybees and bumble bees were analyzed using R (R Development Core Team, 2010) to create binomial logistic regression models for bee foraging behavior. Single-variable regressions were used to determine the correlation between wind speed and foragers and temperature and foragers. A multivariate regression model was created to examine the relationship between wind, temperature and foragers together. Pollen load weights of bumble bees and honeybees were compared using a Wilcoxen rank-sum test. All tests were analyzed with ? = 0.05.

Objective 3. Evaluate the impacts of insecticide sprays on native bees.

Studies on impacts of pesticides on bees have largely focused on honey bees, which differ considerably from bumble bees in life cycle and behavior. The few studies that have been conducted with bumble bees have examined pesticide impacts on workers. However, during blueberry bloom in western Oregon, queens rather than workers dominate. Also, most studies have examined effects of insecticides, and there is little information on fungicide effects.

In this study, the toxicity of pesticides (two reduced risk insecticides and one fungicide) typically applied to blueberry foliage was assessed. Highbush blueberry plant material was collected from an unsprayed plot at the North Willamette Research and Extension Center (NWREC) of Oregon State University (OSU) in Aurora, OR. The upper 15 cm of material (flowers, stems and leaves) was removed and used in bioassays on the day of collection. The following pesticides registered for foliar application during blueberry bloom periods were evaluated:

1. imidacloprid (Admire® 2 (21.4% a.i.), Bayer CropScience Inc., Research Triangle Park, NC) – neonicotinoid insecticide registered for suppression of aphid pests (DeFrancesco and Bell 2008).

2. spinosad (Success® (22.8% a.i.), Dow AgroSciences, LLC, Indianapolis, IN) – spinosyn registered for management of lepidopteran pests (DeFrancesco and Bell 2008).

3. pyraclostrobin/boscalid (Pristine® (12.8/25.2% a.i.), BASF Corporation, Florham Park, NJ) – fungicide registered for management of botrytis grey mold (Botrytis cinerea) and Anthracnose fruit rot (Colletotrichum spp.) (Pscheidt 2008).

Pesticide residues were generated using a Potter Precision Laboratory Spray Tower. Treatments included an untreated control and three pesticide rates: the minimum, maximum and twice the maximum rate approved for field application of each chemical in each crop. For each pesticide, 2 ml was applied in each cage, the equivalent of 100 gallons per acre. After pesticide application, residues were allowed to dry on plant material for a period of one-two hours prior to use in the experiments.

Queens of the bumble bee, Bombus vosnesenskii were collected from unsprayed blueberry plants, placed in plastic vials, cooled, transported to the lab and exposed to test materials on the day of collection. Bees were exposed to the pesticide treated foliage in cylindrical cages consisting of a 15.00 cm plastic petri dish top and bottom surrounded with a 45.70 cm x 5.10 cm strip of metal screen that formed a circular insert to provide ample room for bumble bees to fly and defecate. Plant material was condensed to 6.50 cm x 12.50 cm, and 15.50 g were distributed evenly within cages. Two queens were randomly assigned to treatments. During the experiment, bees were kept under controlled environmental conditions of humidity (50-60%), temperature (28oC ± 2oC) and photoperiod (D7:L17). Mortality was assessed after 24, 48 and 72 hours by recording the number of dead bees within each cage. The experiment was set up as a randomized block design with six replications. Queens that were alive after 72 hours were released at their respective collection sites.

Data analysis: The data were subjected to an analysis of variance (ANOVA) using S-Plus version 8.0. To meet the assumptions of normal variance, all values were arcsine, square root transformed prior to analysis. When significant differences were present, treatment comparisons were made using Tukey’s multiple means comparison. All tests were performed at a significance level of ? = 0.05.

Objective 4. Examine strategies for enhancing native bees in berry fields.

While native bees are diverse and abundant during bloom in cranberry and blueberry crops in Oregon, their populations fluctuate year to year. In addition, worldwide there are reports of declining populations (Beismeijer et al. 2006; Rao and Stephen 2007). Hence strategies are required for enhancing native bee populations in berry crops.

In separate studies, we examined the effects of hedgerows on native bee populations in blueberries and of the addition of attractive blue vanes in cranberries.

Blueberries: Bloom in blueberry crops lasts around three weeks while native bee life cycles last several months. Thus, native bees need alternative foraging resources before and after crop bloom. Hence, a study was conducted to determine if bee abundance or richness was correlated with the availability of floral resources for forage.

To determine the presence of floral resources in the landscape, the landscape surrounding berry fields was scored, and the plants that are present in hedgerows and in adjacent habitats were recorded. Prior to beginning the study, an aerial map of each farm site was overlaid with a grid of numbered 100-m2 plots. A list of random numbers, with replacement, was generated for each site to provide randomized plot selections for each of the 12 46 sampling events (six in each year). At each sampling event, three plots were selected, separated into quadrants and marked with flags. Each quadrant was walked in a serpentine pattern, and the bee-foragable, melittophilous plant species in bloom were recorded. Plants were identified using Kozloff (2005) and Burrill et al. (1996). The quadrants were scored for percentage of area with plants in bloom that could provide forage for bees. Bare ground and plants that were either not in bloom or were in bloom but are not utilized by bees (grasses, sedges and microflora not observed to be visited by bees, e.g., Microsotis) were not included in the score. Floral resource scores were simple percentages of cover of each quadrant in bloom. The scores for all four quadrants of each plot were averaged into a single plot score, then the three scores per site were averaged to a site-level floral resource score. This single site floral resource score was the measure used for statistical analyses.

Data analysis: Analyses of the relationships between total bee abundance, species richness and floral resource availability were performed using correlations, multivariate analysis of variance (MANOVA) and least-squares linear regression in SPSS (2010).

Cranberry: In 2009, we conducted an experiment to determine if a large display of blue vanes would attract more pollinators to a cranberry bed than the number attracted without this large display. Four Oregon cranberry growers in Coos and Curry counties allowed us to set up the blue vane displays. The largest farm of 65.0 hectares allowed us to erect two displays separated by approximately one mile. The other three farms each had one display.

On May 27, 2009, at approximately 10% bloom, twelve blue vane traps were hung from a line stretched between two, 2m tall posts for a distance of approximately 4m. Each display contained two of these lines at right angles to each other to allow for maximum visibility from multiple directions. Displays remained in the field for the duration of cranberry bloom in 2009 (to July 22).

Each week, we performed timed walking counts close to (=Near) and away (=Far) from the blue vane displays to determine the numbers of bumble bees, honey bees and other smaller cranberry flower visitors (not identified, except as “other pollinator”). Areas designated as “Far” were, at minimum, two cranberry beds from a blue vane display.

Objective 5. Build capacity in growers to identify, protect and enhance native bee pollinators in their fields for increasing berry production.

Blueberry and cranberry growers in Oregon see many native bees in their crops, especially bumble bees, but they are unable to tell them apart. To build capacity in growers to recognize various bees, separate workshops were organized for growers of each crop. Instruction was provided by Sujaya Rao, OSU (Oregon State University) Professor, and W. P. Stephen, world renowned OSU Emeritus Professor, who has > 50 years experience with native bee research.

Blueberry: A four-hour workshop on native bee identification was organized in March 2011 at the OSU North Willamette Research and Extension Center (NWREC) in Aurora, OR. The workshop was organized in collaboration with Wei Yang, OSU extension agent, who assisted with advertisement of the workshop, selection of the site and local arrangements, including acquisition of microscopes.

The workshop consisted of an initial short presentation, followed by hands-on exercises in bee identification. In the initial presentation, background information was provided on pollination by honey bees and native bees, life cycles of bumble bees, issues related with declines in their populations and conservations efforts that can facilitate their preservation and build up. Subsequently, attendees were engaged in hands on identification. Attendees were provided with color pictorial keys developed for blueberry growers, microscopes and bee specimens. First they learned to separate various groups such as honey bees, bumble bees, solitary bees, yellow jackets and flies that resemble bees. Next they learned to separate various common bee families such as Apidae (bumble bees), Halictidae (sweat bees) and Megachilidae (and leaf cutter bees). Finally, they were exposed to differences between bumble bee species that are common in blueberry fields in western Oregon.

Cranberry: A similar workshop for cranberry growers was organized in conjunction with the annual cranberry growers’ meeting in February 2010. Local county agent Linda White assisted with advertisement of the workshops, selection of a site and local arrangements, including acquiring microscopes.

At the end of both workshops, attendees were engaged in discussions on native bees and pollination of berry crops. Topics covered included the life cycle and foraging behaviors of honey bees and bumble bees, colony build up in social bumble bees and the critical need for the continued presence of adequate nectar and pollen resources,not just during berry bloom but throughout the life cycle of the colony. Attendees participated in discussions on strategies for protection and preservation of native bee pollinators. These included timing of pesticide sprays and addition of flowering plants that are in bloom before and after the period when crop is in bloom. Growers were encouraged to consider implementation of options at the landscape level; in particular, they were encouraged to maintain plant diversity around their fields for sustainability of native bees during periods when their crop is not in bloom.

Research results and discussion:

Objective 1. Estimate native bee pollinator species diversity and abundance in berry fields.

Blueberries: Native bees belonging to five families, 14 genera and 51 species were collected from the three farms across the two years of the study (Table 1; Bergh 2011). Mean bee abundance between sampling events was significantly lower in September, when catches were far lower than in the other months (P=0.03). Apidae was the most abundant family, with individuals belonging to 15 species, including seven species of Bombus, which comprised 28% of the total bee abundance. Halictidae was the most diverse family, with individuals belonging to 26 species and morphospecies. Halictus and Lasioglossum spp. represented 15% each of total bee abundance. There were 12 genera in five families collected in the native bee guild present in May, of which seven genera in Apidae and Halictidae persisted through the season: Bombus, Ceratina, Synhalonia, Agapostemon, Lasioglossum (Dialictus), Lasioglossum (s.str.) and Halictus. All of these genera were seen to forage and collect pollen on blueberry flowers during bloom in April and May. Of the bumble bee species observed in the study, five out of seven were present in the May sample events.

A reference collection of native bees associated with blueberries was prepared. It is being displayed at a fruit stand at Farm A for increasing public awareness of bees associated with blueberry pollination in western Oregon.

Cranberries: Over the two growing seasons, bees representing five families, thirteen genera and around 30 species (Table 2). Several genera were trapped in higher abundance than honey bees; these included bumble bees (Bombus), metallic sweat bees (Agapostemon) and small sweat bees (Lasioglossum).

Bumble bees were observed in high numbers; Bombus comprised 25.1% of all bees captured. Five species of bumble bees were captured, including the closely related Bombus caliginosus and Bombus vosnesenskii, B. mixtus, B. melanopygus and B. californicus. The most abundant was the western yellow-faced bee, B. vosnesenskii, which constituted 50.1% of all bumble bees.

Two species of Agapostemon were trapped: Agapostemon texanus and A. virescens. Of these, A. texanus was particularly abundant, comprising 24.6% of all bees captured and 98.5% of all Agapostemon.

Agapostemon texanus, B. vosnesenskii and the Lasioglossum complex were all present when cranberries were blooming on the Oregon coast. As the Lasioglossum complex contains at least six species which are difficult to separate, it was impossible to determine if any one of them has a life cycle synchronized with cranberry bloom.

A reference collection of native bees associated with cranberries was prepared and donated to the Coos County Extension office.

2. Compare honey bee and bumble bee foraging behaviors in berry fields.

Blueberries: Out of 850 loads processed over two years, on average, 0.38% of the pollen loads in 2010 alone consisted of blueberry pollen (Fig. 1).

Out of 65 honey bees collected on flowers, only 3% had pollen loads. In comparison, 23% of bumble bees collected at the same time had pollen.

Cranberries: Results: Overall, there was a greater proportion of cranberry pollen in pollen loads on bumble bees compared with honey bees (Fig.2,3). For honey bees, on average, < 50% of the pollen loads comprised cranberry pollen while for bumble bees, while there was considerable variation, on average, > 65% of the pollen loads comprised of cranberry pollen except in the 6th week, when bloom was limited.

During 67 two-minute counts made of bees foraging in cranberry beds over the two year study, honeybees (68.1%) and bumble bees (31.6%) were the most abundant. The average number of honeybees observed was 4.88 ± 0.60, and the average number of bumble bees was 0.40 ± 0.25. Of the 96 bumble bees observed, the majority comprised of the B. vosnesenskii / B. caliginosus complex (70.8%), followed by B. mixtus (22.9%) and B. melanopygus (6.3%). A single halictid bee was observed, representing 0.3% of the total foragers.

Temperature was highly correlated with both honeybee and bumble bee foragers (P<0.001; Fig. 4). The interquartile range of bumble bee foraging was 18.3 – 22.2°C, while that of honeybees was 21.1 – 26.7°C. There was no correlation between the number of foragers and either the minimum (P=0.067) or maximum (P=0.104) wind speed. When considered together in a multivariate analysis, both temperature (P<0.001) and wind speed (P=0.001) were significant predictors of bee foragers (Table 3). The average temperature observed during the visual counts was 19.5°C. The highest numbers of bumble bees were observed foraging between 18 and 22°C, while highest numbers of numbers of honeybees were recorded when temperatures exceeded 24°C, which occurred during less than 20% of observations over the study period.

Of the foraging bees collected for pollen analysis, 62.8% of honeybees and 88.7% of bumble bees had pollen loads. Bees collected without pollen loads were observed foraging for nectar rather than pollen. Honeybee loads (2.0 ± 3.6 mg) were significantly smaller than bumble bee loads (6.8 ± 12.9 mg; P<0.001). On average, honeybee loads contained 89.4% cranberry pollen, with the remainder comprised of one-three other pollen types, while bumble bee loads contained 82.0% cranberry pollen with one-five other pollen types. For both bee species, the most common non-cranberry pollen was from fabaceous plants (7.1% of all pollen), while pollen from plants belonging to Asteraceae (2.4%), Ranunculaceae (1.2%) and Rhamnaceae (0.9%) were also recorded. No statistical difference (P=0.300) in pollen composition was observed between the two bee genera.

Objective 3. Evaluate the impacts of insecticide sprays on native bees.

Responses of queen bumble bees varied based on pesticide, rate and period of exposure (Table 4).

Imidacloprid: There were no significant differences in queen mortality among the four treatments 24 hours after exposure (P = 0.50). By 48 hours, over twice as many queens were dead compared to the control, but the difference was not statistically significant (P = 0.30). However, by 72 hours, queen mortality was three times greater at the 2x maximum rate compared with the control, and this difference was statistically significant (P = 0.04).

Spinosad: There was no statistical difference in mortality among the treatments 24 hours after exposure (P = 0.77). However, at both 48 and 72 hours after exposure, queen mortality was highest at the minimum dose compared with the other doses and the control, but the difference in mortality was significant only at 72 hours (P < 0.01).

Pyraclostrobin/boscalid: After 24 hours, over six times as many queens were killed with the 2x maximum dose, but the difference was not statistically significant when compared with the untreated control. This was due to high control mortality, for reasons unknown, as there were significant differences when the 2x treatment was compared with the minimum and maximum dose treatments (P = 0.01). However, there were significant differences in mortality of the queens at the three rates tested, as well as the control when examined 48 or 72 hours after initial exposure to pesticide residues (P = 0.09; P = 0.26).

In summary, though high control mortality likely masked some effects, all the pesticides tested did have negative effects on queen bumble bees. Of the reduced risk insecticides tested, imidacloprid was observed to be toxic while a reverse dose response was recorded for spinosad. The higher mortality at low doses compared to high doses is speculated to be due to a trigger of detoxifying enzymes, which increases the metabolism of pesticides, reducing mortality at high doses.

Objective 4. Examine strategies for enhancing native bees in berry fields.

Blueberries: Floral resource scores included 35 melittophilous plant species in 12 families and 31 genera. Plants included the blueberries themselves and five other cultivated species and varieties, eight native species and 23 introduced species. Phenological observations of these floral resources revealed 19 species that appeared in floral resource scores at least three times during each of the study years, which represents a bloom period of approximately two months. Additionally 16 species were identified as providing a “favored” floral resource for bees as evidenced by heavy foraging (Table 5; Bergh 2011). Mean floral resource site scores were similar from 2009 to 2010 and between sites and sample events (Fig 5). Bee abundance was highly correlated with floral resource score (R=0.565, P<0.001), as was species richness (R=0.602, P<0.001).

Cranberries: Fig. 6 shows the difference among bumble bees, honey bees and “other pollinators” observed per minute during walking counts on all samples days and at all locations in the study. There were no differences in numbers of each pollinator group, honey bees, bumble bees or other pollinators close to and away from the blue vane displays.

Based on the study, while blue vane traps are effective in drawing a great diversity and abundance of native bees for monitoring studies, the addition of a string of vanes hung on a line is not effective for drawing bees to a cranberry crop.

Objective 5. Build capacity in growers to identify, protect and enhance native bee pollinators in their fields for increasing berry production.

In all, there were 23 and 15 attendees in the workshops for blueberry and cranberry growers, respectively. Besides, growers, Master Gardeners and staff from state and federal agencies such as the Oregon Department of Agriculture and U.S. Fish and Wildlife also attended the workshops.

Reference Cited:

Bergh, J. E. 2011. Native Bee Diversity and Floral Resource Availability in Two Willamette Valley Oregon Ecosystems. MS thesis, Oregon State University.

Biesmeijer J.C., Roberts, S. P.M., Reemer, M., Ohlemüller, R., Edwards, M.,. Peeters, T., Schaffers, A. P., Potts, S.G., Kleukers, R., Thomas, C. D., Settele, J. and Kunin, W. E. 2006. Parallel Declines in Pollinators and Insect-Pollinated Plants in Britain and Netherlands. Science, 31 (5785): 351-354.

Brown, A.O., and McNeil, J.N., 2006. Fruit production in cranberry (Ericaceae: Vaccinium macrocarpon): a bet-hedging strategy to optimize reproductive effort. Amer. Jo. of Botany. 93:910-916.

Buchmann, S.L. 1983. Buzz pollination in angiosperms. In: C.E. Jones and J.R. Little (eds.). Handbook of experimental pollination biology. Van Nostrand Reinhold, New York. pp. 73-113.

Burrill, L.C., S.A. Dewey, D.W. Cudney, and B.E. Nelson. 1996. Weeds of the West. Western Society of Weed Science, Las Cruces. 630pp.
DeFrancesco, J. and Bell, N. 2008. Blueberry pests. In: Hollingsworth CS (ed) Pacific Northwest Insect Management Handbook. Oregon State University, Corvallis pp 81-82.

Dogterom, M.H., Winston, M.L.and Mukai, A. 2000. Effect of pollen load size and source (self, outcross) on seed and fruit production in highbush blueberry cultivar ‘Bluecrop’ (Vaccinium corymbosum; Ericaceae). Amer. J. Bot. 87: 1584-1591.

Eck, P. 1990. The American Cranberry. Rutgers University Press, New Brunswick.

Free, J. B. 1993. Insect pollination of crops. Academic Press, London, UK.

Kozloff, E.N. 2005. Plants of Western Oregon, Washington & British Columbia. Timber Press, Portland. 608pp.

Pscheidt, J. W. 2008. Blueberry plant disease index. In: Pscheidt JW and Ocamb CM (ed) Pacific Northwest Plant Disease Management Handbook. Oregon State University, Corvallis, pp 128-135.

R Development Core Team. 2010. R: A language and environment for statistical computing v 2.11.1. R Foundation for Statistical Computing, Vienna, Austria.

Rao, S. and Stephen , W. P. 2007. Bombus occidentalis (Hymenoptera: Apiformes): In decline or recovery? Pan-Pacific Entomologist 83(4): 360-362.

Rao, S. and W.P. Stephen. 2010. Abundance and diversity of native bumblebees associated with agricultural crops: The Willamette Valley Experience. Psyche 2010: 9 pages. [online] URL: http://www.hindawi.com/journals/psyche/2010/354072.html.

SPSS for Mac, Release 19.0.0. 2010. Chicago: SPSS Inc.
Stephen, W. P. 1957. Bumble bees of western America. OSU Technical Bulletin 40. 162 pp.

Stephen, W. P. and Rao, S. 2005. Unscented color traps for non-Apis bees (Hymenoptera: Apiformes). Journal of the Kansas Entomological Society, 78: 373-380.

Research conclusions:

Objective 1. Estimate native bee pollinator species diversity and abundance in berry fields.

The two studies documented that native bees are diverse and abundant during bloom in blueberry and cranberry fields in western Oregon. In particular, five species of bumble bees are present in each crop, and these likely contribute considerably to pollination of the crop. This could potentially enable blueberry and cranberry growers to reduce their dependence on honey bees for crop pollination.

2. Compare honey bee and bumble bee foraging behaviors in berry fields.

The studies confirmed that honey bees do not forage extensively for pollen in blueberries and cranberries in western Oregon. While they may contribute to pollination in both crops via pollen collected on other parts of their body, it may benefit producers if more attention is paid to enhancing bumble bee pollination.

Objective 3. Evaluate the impacts of insecticide sprays on native bees.

Bumble bee queens that are potentially key pollinators of crops such as blueberries are at risk to pesticides used in Willamette Valley crops, and hence caution is critical when pesticides are applied when queens are in flight. Also, impacts of pesticides observed in the study differed from impacts on honey bees documented from other studies, and this underscores the need for toxicity studies conducted on bumble bees prior to registration of new compounds.

Objective 4. Examine strategies for enhancing native bees in berry fields.

The study highlights the value of the presence of hedgerows for maintenance of native bees in the area beyond the period of crop bloom. This is critical as the number of native bees present at the end of the year affects the size of the population available for crop pollination the following year.

Objective 5. Build capacity in growers to identify, protect and enhance native bee pollinators in their fields for increasing berry production.

Feedback from the attendees indicated that they found the workshop to be very useful and informative; they appreciated the hands-on exercises and instruction provided by the world renowned bee expert. In an email, one grower attendee indicated the following: “The workshop was very interesting for ..(names removed). On Monday I saw a very large Bombus melanopygus and a smaller Bombus vosnesenskii in my yard on some heather. I was very proud of the fact that I actually knew what they were specifically, instead of saying a kind of black bumble bee and a kind of orange one. Also interesting to know about their life cycles and realize that these were both queens”.

Participation Summary

Educational & Outreach Activities

Participation Summary

Education/outreach description:

Presentations made related to blueberry and cranberry pollination research:

Stephen, W. P. and Rao, S. Evaluating pollination by bees for enhancing blueberry production, Blueberry Growers’ Meeting, Corvallis, OR, January 2009.

McKenney, M., Rao, S. and Stephen, W. P. Native bees and honey bees associated with cranberry production, Annual Cranberry Growers’ Meeting, Bandon, OR, January 2009.

McKenney, M., Rao, S. and Stephen, W. P. Honey bees and bumble bees in Oregon cranberry bogs, Oregon State University 1st Annual Student Research in Entomology Symposium, Corvallis, OR, February 2009.

McKenney, M., Rao, S. and Stephen, W. P. Native bee and honey bee foraging in Oregon cranberry, Entomological Society of America 57th Annual Meeting, Indianapolis, IN, December 2009.

Skyrm, S., Rao, S. and Fisher, G. Impact of pesticide residues on a native bumble bee pollinator, Bombus vosnesenskii (Hymenoptera: Apidae). Annual Pacific Northwest Insect Management Conference 69th Annual Meeting, Portland, OR. January 2010.

Rao, S., McKenney, M. and Phillips, K. Native Bees and Honey Bees Associated with Cranberry Pollination, 2010 Oregon Cranberry School, Bandon, OR. February 2010.

Phillips, K. A comparative study of pollination in Oregon cranberries: bumble bees versus honey bees, 2010 Oregon Cranberry School, Bandon, OR. February 2010.

Phillips, K., Rao, S., Stephen, W. P. and White, L. Pollination efficiency by bumble bees and honey bees in Oregon cranberries. Oregon State University 2nd Annual Student Research in Entomology Symposium, Corvallis, OR, February 2011.

Phillips, K., Rao, S., Stephen, W. P. and White, L. Bumble bees versus honey bees: a comparison of pollination success in Oregon cranberries. Annual Meeting of the Pacific Branch Entomological Society of America, Boise, ID, March 2010.

Phillips, K., Rao, S., Stephen, W. P. and White, L. Comparison of pollination efficiency by bumble bees and honey bees in Oregon cranberries. Insect Explorer Series, Entomology Program, Oregon State University, April 2010.

McKenney, M. Rao, and Stephen, S. Pollination by Native Bees and Honey Bees in Cranberry. Celebrating Undergraduate Excellence, Oregon State University, May 2010.

Phillips, K. Oregon cranberry pollination research updates. Cranberry Field Day, Oregon State University Extension Service and Oregon Cranberry Growers Association, June 2010.

Phillips, K., Rao, S., Stephen, W. P. and White, L. Bumble bees versus honey bees: a comparison of pollination success in Oregon cranberries. Annual Meeting, Entomological Society of America, San Diego, CA, December 2010.

Phillips, K. Foraging preferences of honey bees and bumble bees in Oregon cranberries, 2011 Oregon Cranberry School, Bandon, OR. February 2011.

Phillips, K., Rao, S., Stephen, W. P. and White, L. Foraging preferences of honey bees and bumble bees in Oregon cranberries. Oregon State University 3rd Annual Student Research in Entomology Symposium, Corvallis, OR, February 2011.

Publications related to blueberry and cranberry pollination research:

McKenney, M., Rao, S. and Stephen, W.P. 2009. Pollination in Cranberries on the South Oregon Coast: Honeybees and Native Bees. South Coast Grower News. 3(1):3-5.

Phillips, K., White, L., Rao, S., and Stephen, W. P. 2010. Bumble bees versus honey bees: a comparison of pollination success in Oregon cranberries. South Coast Grower News. 4(2): 4-5.

Skyrm, K. and Rao, S. 2010. Impact of pesticide residues on a native bumble bee pollinator, Bombus vosnesenskii (Hymenoptera: Apidae). In Proceedings of the 69th Annual Pacific Northwest Insect Management Conference. Portland, OR. pg. 15-18.

Rao, S. and Stephen, W. P. 2010. Abundance and diversity of bumble bees associated with agricultural crops: The Willamette Valley Experience. Psyche doi:10.1155/2010/ 354072.

Broussard, M., Rao, S., Stephen, W. P. and White, L. 2011. Native bees, honey bees and pollination in Oregon cranberries. HortScience. 46: 885-888.

Bergh, J. E. 2011. Native Bee Diversity and Floral Resource Availability in Two Willamette Valley Oregon Ecosystems. MS thesis, Oregon State University.

Project Outcomes

Project outcomes:

Currently, growers rent honey bee hives for pollination of blueberry and cranberry crops. Due to the adverse weather conditions during bloom in both crops, investments in honey bee rentals are high so as to maximize pollination during the few days when weather conditions favor foraging by honey bees. However, honey bees are not as efficient as bumble bees in pollination of both crops. Bumble bee populations are high during bloom in both crops. Hence, crops can reduce their dependence on honey bees and growers can raise their crops more economically.

Farmer Adoption

Blueberry and cranberry growers have approached us for implementation of strategies for sustaining and building bumble bee populations in their fields. They have sought our input on plants that they should add to their farms for increasing bloom during periods when their crops are not in bloom. One grower expressed an interest in rearing native bumble bee colonies, and we gave him supplies to get his started. We have been receiving more questions about timing of pesticide sprays so as not to negatively impact bumble bees.

Recommendations:

Areas needing additional study

Research is needed to determine if, indeed, adequate pollination can be achieved in blueberry and cranberry crops without investment in honey bee hives by depending on native populations or by investment in purchase of colonies of native bumble bees. Research is also needed on determining which plants are best suited for addition in hedgerows around farms of various crops, not just berry crops in different landscapes across the west coast of the U.S.

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.