Source: UNIVERSITY OF CALIFORNIA, RIVERSIDE submitted to NRP
BUMBLE BEE WORKER BODY SIZE: A NEXUS OF SOCIALITY AND GLOBAL CHANGE
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
ACTIVE
Funding Source
Reporting Frequency
Annual
Accession No.
1030666
Grant No.
2023-67014-39986
Cumulative Award Amt.
$748,797.00
Proposal No.
2022-08489
Multistate No.
(N/A)
Project Start Date
Jul 1, 2023
Project End Date
Jun 30, 2027
Grant Year
2023
Program Code
[A1113]- Pollinator Health: Research and Application
Recipient Organization
UNIVERSITY OF CALIFORNIA, RIVERSIDE
(N/A)
RIVERSIDE,CA 92521
Performing Department
(N/A)
Non Technical Summary
Bumble bees are important pollinators in natural and agricultural ecosystems. One of the most unique, but still poorly-understood, features of bumble bee biology is the high level of size variation observed among workers in nests. Worker body size influences colony function and also has broader consequences for plant pollination. Yet, there many unresolved questions about worker size, including about how it is generated, its relationship to colony fitness, and how it is being impacted by global change. This project focuses on these aspects of worker body size variation, and includes lab work with the managed species Bombus impatiens that combines thermal imaging and video observation to study how brood feeding and incubation interact to shape size. The project also uses a semi-field experiment (with B. vosnesenskii) to study how landscape characteristics (wildfire and the presence of honey bees) and food resource acquisition influence size variation within nests, coupled with a similar but more controlled experiment in B. impatiens. Work with wild, free-foraging bumble bees (and lab-reared colonies) in California and Alaska will examine relationships between landscape characteristics, body size, and plant pollination, including in apple, a growing crop system in Alaska. Together, this work will shed new light on the factors that impact bumble bee body size and how this might be altering pollination outcomes, including for a set of economically and ecologically important plant species. ?
Animal Health Component
(N/A)
Research Effort Categories
Basic
100%
Applied
(N/A)
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
21130851070100%
Goals / Objectives
The proposed project explores bumble bee worker size variation at multiple levels of biological organization. Part 1 (Social dynamics) investigates how size variation is generated within nests, what its consequences are for colonies, and how it changes in response to dynamic environmental conditions. Part 2 (Ecosystem dynamics) examines relationships between (i) real-world landscapes and body size distributions; (ii) body size and foraging distance; and (iii) size and pollination outcomes. This part of the project involves fieldwork in two unique regions (California and southern Alaska). Our experiments include the following:Experiment 1.1 will explore the drivers and consequences of size variation in free-foraging bumble bee nests. The experiment will use lab-initiated, field-deployed colonies of B. vosnesenskii.We will deploy colonies in both burned and unburned areas to better relate data from this experiment to Part 2 (described below). We will then track colonies across the season and collect data on worker size distributions, floral resource acquisition (using RFID), and ultimately queen production as a proxy for colony fitness.Experiment 1.2 will examine how food resource availability influences colony size distributions in further detail, under highly controlled conditions. This experiment will use lab-reared B. impatiens.Experiment 1.2 will yield a comprehensive understanding of how body size is shaped by social interactions (feeding and incubation) in bumble bee nests. This will be generated by estimating the relative amounts of feeding and incubation that brood receive, and quantifying the resulting growth patterns of individual developing workers in the nest.Experiment 1.3will test the hypothesis that queens produce smaller-bodied offspring (relative to when larvae are reared by workers) because queens cannot match the amount or quality of brood care that workers can provide, when they assume brood care later in the nest cycle.In part 2.1, we will repeatedly sample workers at focal field sites to track body size variation in wild bumble bees across the season. Sites will vary in wildfire history and presence of honey bees. As a companion to our wild bee collections, we will also monitor worker size distributions in sets of captively-reared colonies produced under controlled, invariable conditions.Part 2.2 will study the relationship between size and foraging distance. We will sample intensively along a 20x20 km grids, use genetic data for sibship analysis to assign bees to natal colonies, estimate the nest location, then use regression analysis to examine the relationships between size and foraging distance, both within and across natal colonies.In part 2.3, we will integrate environmental variation, size variation, and pollination outcomes, to better understand how anthropogenic stressors influence pollination services indirectly through their effects on body size.We will test how wildfire (in California) and honey bees (in California and Alaska) ultimately impact pollination outcomes for a set of focal plants in both regions. In Alaska we will study a pair of agriculturally (apple; Malus domestica) and ecologically (fireweed; Chamaenerion angustifolium) important species.
Project Methods
Experiment 1.1: We will capture and rear wild queensthen deploy their nests to sitesthat either burned in 2024 or have not burned in the last 10+ years. A subset of nests will be affixed to an RFID system that automatically records bees exiting and entering the nest. We will use weekly photographs of nests to estimate worker sizes; some nests will be collected mid-season (to get data on all worker sizes) and others will be allowed to develop until the end of the season to count queen number.From the data generated from this experiment, we can explore how treatment group (burned versus unburned) impacts pollen collection, worker size, and colony reproductive success.Experiment 1.2: Young queens will be obtained frommature colonies of Bombus impatiens from Koppert Biological Systems. Their nests (n = 96) will be split equally into two nest types: early queens (with eggs but no larvae yet; to be used to study queen decision-making) and mature nests (~35 workers; to study worker decision-making). Within each of these nest types, a set of controlcolonies (n = 12) will not be manipulated. The other sets of colonies will be subjected to one of the following three treatments (n = 12/treatment): episodic-brief food limitation (each week, will go three days without fresh food; all pollen removed but nectar left in nest), episodic-long food limitation (one week in the middle of the experiment without fresh food; all pollen removed but nectar left in nest), and chronic food limitation (fresh food supplied daily but half the amount calculated from Heinrich 2004). We will also track focal brood to quantify how the amounts of feeding and incubation influence developmental rate and body size.To quantify feeding, we will affix cameras (VIGICA Peashooter QD520) to the tops of nests. We will also quantify incubation using a FLIR T650sc camera.The development time (days from egg to eclosion), final body size, and dry mass will be recorded for each focal bee.Experiment 1.3: After queens have laid their first five egg cups,we will either transfer eggs to a new queen of the same age (treated similarly to the mother-queen) or will give them to a nest with one, five, 10, or 20 newly-eclosed workers (all from the same natal colony, which will differ from the natal colony of the queen).Video recordings will be used to estimate the amount of brood feeding and incubation, which will be related to body size and growth rate.Alaska fieldwork and site selection for Part 2: Alaska fieldwork will take place in March through May. Sampling will take place in or around a set of apple orchards. We will select a set of eight focal orchards, primarily based on distributing the sites geographically across the growing region; spacing them >10 km apart (which is also beyond the typical flight range of honey bees; Visscher & Seeley 1982); and use of honey bee hives (yes/no categorical variable).California fieldwork and site selection for Part 2: We willidentify a set of locations in CA that have just burned the previous year, or burned two or five years prior. Sampling will take place in each of these areas and also an adjacent area (<10 km distant) that has not burned in the past 10+ years (referred to as unburned), to have pairs of burned and unburned sites. Within each given burn history site type, we will also locate sites that are either protected (National Park) or not.Part 2.1: For the lab-reared colony component in Alaska, in early March, queens of multiple species (n = 20 per species) will be collected and housed in a rearing facility at UAA. Queens and their nests will be provided with food ad libitum as described above. Workers will be periodically paint-marked to designate age cohorts. At the end of May, whole nests will be collected into a -80C freezer, subsequently dissected, and worker body sizes will be measured. For California, queens will be collected and reared similarly, but at the Insectary and Quarantine Facility at UCR.In our wild bee collections, we will use a paired-site design to sample within a focal site (e.g., an orchard in Alaska) and then in a natural area that is within a 10 km buffer around the site. Bees (n = 50/site/visit) will be hand-netted haphazardly and the marginal cell length for both wings and the IT distance will be measured. We will record the total number of honey bees observed while walking transects and will collect pollen loads from bumble bees and a set of 50 honey bees/site/visit to keep for potential future use.Part 2.2: We will focus on B. vosnesenskii, the most abundant and widespread bumble bee in California. Fieldwork will take place in a large, continuous 20x20 km area in California. The specific location will be determined at the start of the project because it will depend on wildfires in the previous years. The location selected will be unburned within the last 10+ years; with the same land cover type across the site (based on the National Land Cover Database); devoid of agricultural, urbanization, and large bodies of water; and with a < 500 m elevational gradient across the site. The site will be divided into a grid with transects separated by 2 km, with a total of 121 collection locales within the grid (including the margins). Collections will occur over a two- to three-day period during early summer with the exact dates depending on the site. We will generate low coverage (5X) genomes from these samples, for a total of 726 bees from California for sibship analysis. DNA extractions and indexed library preparation will take place in the Woodard lab and will follow standard llumina Novaseq submission protocols. Sequencing will be done at UC Berkely's Sequencing Core Facility. Sequencing reads will be mapped to a high-quality B. vosnesenskii genome that the lab recently generated through the California Conservation Genomics Consortium project. We will use the program COLONY 2.0 to assign bees to natal colonies, then estimate nest locations based on the centroids of sister- foragers.From putative nest locations, we can estimate foraging distances for each worker, then use regression to identify how they vary as a function of body size, within and across colonies.Part 2.3: We will use bee visitation experiments focusing on floral handling and pollen transfer. Data of each type will be collected from 10 visits per plant species, per site. In Alaska we will collect data from each of the eight apple orchards and nearby (< 1 km) locations where fireweed is observed. For California, we will collect information about dominant plant species at field sites in year two, then collect data in year three focused on a subset of these species. Briefly, unopened flowers will be covered in a gauze bag to prevent insect access, then unbagged after anthesis. Virgin flowers will be presented to foraging bumble bees, then the bee and the flower will be immediately collected after a single visit to measure body size (as described above). To estimate pollen removal, anthers will be removed from flowers, separated into dehisced versus undehisced, then the latter will be placed in a vial until dehiscence. Pollen grains will be counted on each anther types. The number of pollen grains removed will be calculated as the difference between the number of pollen grains in dehisced versus undehisced anthers. To estimate pollen deposition, emasculated flowers will be presented to bees, then after a single visit the stigma will be removed, stained with fuschin, and the number of pollen grains will be counted (Kearns and Inouye 1993). The lengths of visits will also be calculated. Linear regression will be used to assess relationships between worker body size and pollen removal, pollen deposition, and visit duration.

Progress 07/01/24 to 06/30/25

Outputs
Target Audience:A graduate student at UAApresented findings to the Alaska Pollinator Coordination Group and Alaska Pioneer Fruit Growers Association Changes/Problems:The biggest changes we have made are that we switched from captive rearing colonies of B. vosnesenskii to collecting wild bees and using genetic analysis and inferring sibship based on genetic data; this allowed us to expand our work across California although we will have less granular data on size distributions for entire colonies. We have also started our California pollination work, which focuses on Bombus crotchii andB. vosnesenskii visitation in Southern California. This work is also being supported by a small internal grant and we are developing a Southern California bumble bee-specific eDNA assay with the USGS eDNA lab. What opportunities for training and professional development has the project provided?Two graduate students (one at UCR, one at UAA) were trained through the project, as well as a researcher (Assistant Specialist) at UCR. The students received training in bee morphology, field data collection and analysis, genetic analyses, and laboratory experimental design and data analysis. How have the results been disseminated to communities of interest?The UAA graduate student gave a presentation to theAlaska Pollinator Coordination Group and Alaska Pioneer Fruit Growers Association. Other work is either still beingcompleted (see above) or is being prepared for publication. What do you plan to do during the next reporting period to accomplish the goals?We will return to Alaska in spring 2026 to rear queen bumble bees and study distributions. We will also complete our work with wild-caught California bees and begin our work on flower visitors in California.

Impacts
What was accomplished under these goals? Experiment 1.1: We have modified our plan and have collected wild B. vosnesenksii across California (n = 665 samples across the entire state); these samples will be measured for body size and used for genetic analysis to assign to colony. Experiment 1.3: This experiment has been run and we are watching videos of brood feeding now. Part 2.1: This work was published in the last reporting period (Costa et al. 2024). Part 2.2: We have completed these collections and are measuring bees for body size and preparing them for genetic analysis. Part 2.3: We have complete this data collection in Alaska with the exception that we are generating a DNA data set (for the community of apple visitors) and eDNA data set (from apple flowers) that is currently in progress.

Publications


    Progress 07/01/23 to 06/30/24

    Outputs
    Target Audience: Nothing Reported Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Dr. Claudineia Costa (project scientist) received mentorship and training for the project from PD Woodard and Co-PD Carlson, including in the use of GIS (a new skillset for her). A graduate student at UAA received mentorship from Co-PD Carlson for his Master's research. Additionally, four undergraduate researchers from UCR (supported through other funds) were able to travel to Alaska to participate in field research in association with the projects. These students learned about fieldwork, vegetation surveys, bee biology, and other topics through their participation. How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals?Thanks to our highly successful field season in summer 2024 we are spending most of 2025 processing samples, analyzing data, and preparing our findings for publication during the next reporting period. We are continuing to collect some laboratory data as well (for the lab experiment component). We will also engage in preparations for a 2026 field season in California that is framed around experiment 1.1 and the remaining work for Part 2 of the project.

    Impacts
    What was accomplished under these goals? Experiment 1.1: This work was performed with wild B. vosnesenskii (as opposed to captive colonies) in 2024 and has been accepted for publication in the journal Ecology and Evolution. In this article, we show thatB. vosnesenskiiworkers are larger in recently-burned areas of the southern Sierra Nevada Mountains. We plan to extend this work to wild-caught bees in summer 2026. Experiment 1.2: This experiment has been carried out and we are currently analyzing feeding and size data with the expectation of submitting for publication in late 2025. Experiment 1.3:This experiment has been carried out and we are currently analyzing video and size data with the expectation of submitting for publication in late 2025. Part 2.1: We carried out a successful field season in Alaska in summer 2024 and collected size (and other) data from 1000+ workers across the Anchorage area. We also carried out extensive vegetation surveys and these data are being analyzed together. We expect this component to be completed in 2026. Part 2.2: We successfully carried out genetic sampling in Alaska in summer 2024 andexpect this component to be completed in 2026. Part 2.3: We collected data on apple visitors in Alaskain summer 2024. This work is being combined with eDNA sampling and we expect this component to be completed in late 2025or early 2026.

    Publications