CAUSES AND CONSEQUENCES OF FUNCTIONAL TRAIT VARIATIONS IN CROP POLLINATORS

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: AFRI COMPETITIVE GRANT
• Reporting Frequency: Annual
• Accession No.: 1015655
• Grant No.: 2018-67012-27978
• Cumulative Award Amt.: $165,000.00
• Proposal No.: 2017-07100
• Project Start Date: Apr 1, 2018
• Project End Date: Mar 31, 2020
• Grant Year: 2018
• Program Code: [A7201]- AFRI Post Doctoral Fellowships

Recipient Organization
CORNELL UNIVERSITY
(N/A)
ITHACA,NY 14853

Performing Department
Entomology

Non Technical Summary
Bees are essential for the production of our most high-value and nutrient dense foods. However, the supply of pollination services is threatened by declines in many important pollinator species driven primarily by loss of forage and nesting habitat. Simplification of agricultural landscapes can therefore undermine the very biodiversity and ecosystem services that would otherwise benefit crop production. Yet, while some bee species have experienced dramatic declines others are able to persist in simplified agricultural landscapes. Bees possess a diversity of functional traits (body size, sociality, diet breadth) that both mediate their responses to the environment and determine their efficiency as crop pollinators. These traits vary between and within species and our preliminary findings show that traits like body size are strongly influenced by landscape simplification. The proposed research will use bioinformatics approaches and natural history collections combined with field studies and laboratory experiments to examine how landscape simplification affects intraspecific body and its consequences for pollination services. The specific objectives are to: 1) characterize intraspecific body size variation and its relationship to landscape simplification across a broad range of bee species; 2) evaluate the potential mechanisms driving changes in size; and 3) assess the consequences for pollination services. An improved understanding of how bees respond to environmental stressors associated with landscape simplification will allow conservation practitioners to identify vulnerable species and appropriate interventions.

Animal Health Component

75%

Research Effort Categories

Basic

25%

Applied

75%

Developmental

(N/A)

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
211	3085	1070	100%

Knowledge Area
211 - Insects, Mites, and Other Arthropods Affecting Plants;

Subject Of Investigation
3085 - Other bees (non-Apis bees) - All bees except Apis cerana; includes Bombas, Alfalfa leaf-cutter;

Field Of Science
1070 - Ecology;

Keywords

bee health

functional trait

intra-specific variation

landscape

pollination

Goals / Objectives
The overall goal of this project is to explore the role of landscape simplification in mediating intraspecific trait variation in bees and the consequences for crop production.This broad goal will be addressed through the following three specific objectives:1) Characterize intraspecific variation in body size and its relationship to landscape simplification across a broad range of bee species. I will use specimens from contemporary landscape studies across cropping systems in addition to historical specimens and land use data for 30 bee species to conduct phylogenetically controlled analyses that examine the extent of intraspecific variation in body size.2) Evaluate the potential mechanisms responsible for intraspecific body size variation. I will conduct bioassays in the lab to examine possible factors such as diet and pesticide exposure. In addition reciprocal transplant experiments in the field will test for selection pressure on different body sizes in different landscape contexts.3) Assess the consequences of inter- and intraspecific body size variation for pollination services. I will conduct experiments in the field to test if the differences in body size observed in Objective 1 result in reductions in fruit production.

Project Methods
Objective1. Characterize intraspecific body size variation across a broad range of bee species. In order to understand the extent of intraspecific body size variation across various cropping systems and bee clades, I will measure the specimens collected in recent landscape scale studies of pollinator communities. In addition to these contemporary collections, I will evaluate historical specimens from the 10 most common pollinators. I will relate my body size measures of museum specimens to digitized historical aerial photographs dating back to 1936. For each species, I will measure inter-tegular distance (ITD) as an estimate of body size.In addition, I will compile existing sequence data available on Genbank for each species and construct a phylogeny using Bayesian methods. Because bee species that are more closely related may be expected to show similar variation in body size, this will allow me to control for relatedness when testing for the effects of landscape simplification on body size.Data Analysis & Interpretation: To test whether landscape simplification has altered intraspecific body size across bee species, I will use a phylogenetically controlled regression analysis as well as linear and non-linear models implemented in the R packages picante (45) and nlme (46). I will also test for phylogenetic signal in body size variation to determine if changes in body size are more similar among closely related species.Objective 2. Evaluate the potential mechanisms responsible for intraspecific body size variation. My preliminary data suggest that both diet and pesticides may play a role in driving intraspecific changes in body size in response to landscape. Therefore, I will conduct bioassays using Osmia cornifrons, a managed crop pollinator that also occurs in wild population across the region and regularly colonizes trap nests. Trap nests were deployed on 19 farms in the spring of 2017. Offspring from these nests will be reared in the lab in 96-well microtiter plates. Pollen composition will be altered by substituting with mono-floral pollen mixes collected by honey bees and re-constituted to the consistency of a normal provision with 30% sucrose solution. In the first year of the study, I will individually test the role of diet composition (low, med, high diversity based on studies of pollen use in Danforth Lab at Cornell), provision quantity (representing 25th. 50th and 75th percentile of provision weights for females) and pesticides (imidacloprid v. control at concentrations found in field collected provisions) on 48 larvae per treatment group (24 larvae from simple and 24 from complex landscapes). Larvae will be allowed to develop to adulthood after which I will assess mortality and body size (ITD). In the second year, I will use a factorial design to test for interactions between factors that significantly affected body size in the first year of the study.Although both diet and pesticides may influence body size through plastic phenotypic responses, I cannot rule out that the patterns observed in my preliminary data do not represent adaptation of phenotypes to local optima. To address this question, I will conduct a reciprocal transplant experiment in year two. Pre-emergent adults will be collected while overwintering in the trap nests at each site. Overwintering adults will be extracted from their nests, identified, measured (ITD) and individually marked. For each species, both males and females from sites in simple landscapes will be introduced to nests in complex landscapes. Bees from sites in complex landscapes will also be introduced to nests at sites in simple landscapes. A subset of bees from each species will remain at the site from which their nest was collected. Nest establishment by marked females will be monitored every 3 days following introduction and at the end of the season nests will be collected. For each nest, I will record the number of offspring produced, sex ratio and size of offspring (ITD).Data Analysis & Interpretation: To test the effect of diet composition, quantity and pesticides on body size, I will use a mixed effects model with a Gaussian error structure and family group (bees from the same nest) as a random effect to explore interactions between source landscapes (simple vs. complex) and these factors. I will use similar analytical methods to test for interactions between factors in year two. To test whether body size changes represent local adaptation to landscape structure, I will compare fitness measures (number of offspring, sex ratio) and offspring body size between transplanted and local groups and in a series of generalized linear mixed models (GLMMs) (49). Population, Location and Population X Location will be included as a fixed effect and species will be included as a random effect. To evaluate the significance of fixed and random effects, alternative models without the variable of interest were compared to the full model using likelihood ratio tests.Objective 3. Consequences of inter- and intraspecific body size variation for pollination services. In order to compare the relative contribution of inter- vs. intraspecific body size variation for pollination services, I will conduct pollination experiments in the field which will allow me to exploit the body size variation present in natural pollinator communities from many species for which it is not possible to maintain adults in captivity. During bloom at each of these sites, small groupings of flowers will be observed until a bee visits one of two paired experimental flowers. Behavior observations will be recorded including approach orientation, visit duration pollen vs. nectar collecting etc. A minimum of 10 visits per species for up to 6 species per site will be observed. When ripe, the fruits from visited flowers along with the paired unvisited fruit from the same inflorescence were collected and weighed. Each bee's contribution to pollination services will be calculated by the weight of the resulting fruit to that of the other unvisited flower.Data Analysis & Interpretation: To test whether pollination services to fruit are influenced by the body size distribution of the pollinator community, I use linear mixed effects models to relate fruit productionto both average body size measures per species and intraspecific body size.Efforts designed to deliver knowledge generated during the course of the project include:Publication of three peer-reviewed research articles based on project findingsMentoring undergraduate researcher by the PD as well as mentoring of the PD by the faculty project mentorOrganization of a symposium on intra-specific variation and ecosystem services at a national society meetingEVALUATION PLANAt three month intervals, I will complete performance evaluations with Dr. Poveda to assess progress towards my project plan, training/career development plan, and mentoring plan. The project plan will be evaluated based on the major milestones for each research objective described in the timeline above. The conclusion of work related to each objective will be signified by dissemination of the results through three different means: 1) scientific publications, 2) conference presentations, and 3) extension presentations. We will also evaluate progress towards each of my training/career development objectives described above: developing collaborations within and across disciplines, gaining experience in teaching and mentoring, advancing skills in spatial modeling and analysis. To specifically assess my teaching and mentoring progress, I will obtain evaluations from students and faculty instructors for each education module I teach (one per semester) and annual evaluations from all undergraduate mentees involved in the project. These evaluations will serve as indicators of progress over the course of the fellowship.

Progress 04/01/18 to 03/31/20

Outputs
Target Audience:1. Agricultural and Conservation Scientists 2. Extension Agents and Crop Consultants 3. Growers and Land Managers Changes/Problems:A majorchange in approach was made for Objective 2 (Evaluate the potential mechanisms responsible for intraspecific body size variation). We had originally plannedto conduct bioassays and reciprocal transplants using field collectedOsmiaspecies from sites across a land use gradient. Due to poor establishemnt and the release of non-local populations of Osmiawithin the vicinityof our field sites, the approach was changedto instead incorporate a modeling basedassessment of drivers of size variation within five of the mostcommon strawberry pollinators. These species were readily available at sites and were already measured as a result of Objective 1. Additionally, this change in approach allowed for a new collaboration among researchers at Penn State University and Dickinson College. Thisapproach was supplementedwith a review of the literature on mechanistic studies that have evaluated drivers of size variation in bees using experimental approaches. What opportunities for training and professional development has the project provided?The project has provided training opportunities to the PI Heather Grab through one-on-one mentoring from K. Poveda. Additionally, Heather Grab provided mentoring to multiple technicians and undergraduate researchers who assisted in aspects of the project. The project also provided professional development for Heather Grab through attendance at the Entomological Society of America National Meeting, participation in professional development workshops at Cornell University and participation in week discussion seminars on plant-insect interactions and pollination biology. How have the results been disseminated to communities of interest?Our results have been disseminated to target audiences and stakeholder groups primarily through peer-reviewed publications as well as a publically available thesis. We have presented results to grower and extension educator audiences through presentations and published proceedings. Additionally we have highlighted out work by delivering press releases, engaging with social media and publishing research updates on relevant websites. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Over the project period Ihave accomplished myoverall goal of exploring the role of landscape simplification in mediating intraspecific trait variation in bees and the consequesces for crop production. Through myon-farm research activities Ihave increased grower awareness of how farming practices can influence variation within species and how this variation can in turn impact the delivery of pollination services. Ihave contributed to knowledge change among researchers and practitioners by publishing these results in peer-reviewed academic journal and collaborating with media outlets to publicize results. Accomplishments relevant to each of our three objectives are outlined below. 1) To characterize intraspecific size variation and its relationship to landscape simplification in a broad range of bee species, I collected bees visiting the flowers of crops including strawberry, raspberry, hemp and tomato from more than 30 farms over the course of two growing seasons. Farms varied from 5-85% agricultural land in the surrounding landscape allowing us to evaluate how variation in contemporary land use impacts variation in size at both the species and intraspecific level. Additionally, Iinitiated a collaboration with researchers in the Isaacs lab at Michigan State University to explore trends across regions,evaluate whether local resource enhnacement can offset landscape scale habitat loss and evaluate differences in trends among species. Finally, over 3,000 specimens from Cornell University Insect Collection were measured to estimate trends in body size changes across time focusing on phylogentically controlled contrasts between closely related species that have been reported to be experinecing population declines or stable populations over the past century. These research activities have generate both datasets as well as specimens that have been or will be deposited in appropriate repositories to ensure their utility to future researchers. The results of these research activities show that many, but not all, bee species are experiencing downward shifts in body size in association with landscape simplification driven by agricultural land covers. These effects are also operating at the community level often resulting in pollinator communities comprised of both smaller species and smaller individuals within species. At the intraspecific level, the tendancy the decline in body size in simplified landscapes is strongest among both small-bodied and large-bodiedspecies while medium-sized species tend not to show size changes across the land use gradient. In both New York and Michigan, local scale habitat enhnacements near crop fields appear to buffer against body size declines, as individuals collected from farms in simplified landscapes but with a wildlfower enhancement tended to be the same size, or in some cases larger, than their counterparts in landscapes with abundant natural habitat. These results suggest that lack floral resources are one of the main drivers of size declines, however, availabiliy of suitable nesting habitat may also be important. Key findings related to this objective have been published or are in preparation for publication in peer-reviewed journals. 2) To evaluate the potential mechanisms responsible for intraspecific body size variation, I initiated a collaboration with researchers at CornellUniversity, Penn State University and Dickinson University and mentored an undergraduate honors student in a project to integrateremotely sensed models of pesticide exposure risk as well a remotely sensed floral resource estimates as predictors of body size decline among the five most common strawberry pollinators. Additionally, I conducted a literature review synthesizing patterns of body size variation over the past centrury and in contemporarystudues across land use and climate gradients. The review highlights results from mechanistic studies, suggesting drivers of observed size changes and outlining future research needs. The results of these research activities indicate that raw land use change aswell as landscape scale indices of pesticide exposure and floral resource availability are all good preictors of intraspecific size variation. Among these predictors, pesticide exposure consistently explained the greatest variation in body size across farms. Although the review effort identified several mechanistic studies that have evaulated the impact of pesticides on body size variation within species, the results among studies were mixed and the few studies that have evaluated the impact of both pesticides and habitat loss have found habitat loss to better explain size variation. These results suggest that the importance of different drivers is likely to be context dependent (moderated by bee species, habitat type and potnetially pesticide class). The finding from our modeling study have been published as an honors Thesis and the literature review will be submitted for publication in the journalGlobal Change Biology. 3) To assess the consequences of inter- and intraspecific body size variation for pollination services, I conducted controlled single visit pollination trials in strawberry, raspberry, squash and tomato. In these experiments, virgin flowers wereexposed to the natural pollinator community until they recieved a single visit. Following the visit, the flower was bagged along with an unvisited control and the visitor was captured, identified to species and measured to estimate body size. Additionally, data on visit duration and flower size were collected when possible. To determine whether these effects can scae up to crop pollination under more natural conditiones we analyzed the community size data from strawberry collected in Objective 1 against estimates of crop yield at each of the strawberry fields.As a result of these research activities we demonstrate that individual visitor size within a speciesis a good predictor of pollination services in two of the four crops (strawberry and squash). Larger species were not more efficient than small species in any crop with the possible exception of tomato which was visited nearly exclusivelyby the large-bodied speciesBombus impatiensbut also recieved a small number of less efficient visits from bees in thesmall-bodied genusLasioglossumspp. Our community analysis in the strawberry system suggests that the shift in community mean size distribution due to both changes in the relative abundance of small vs large species and the intraspecific reduction in body size can have negative effects on pollination services. Results of these studies have been submitted for publication in the journal Nature Communications.

Publications

• Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: H. Grab & M. Lopez-Uribe. Managed and Wild Pollinators of Strawberry. Mid Atlantic Fruit and Vegetable Growers Conference, Hershey PA. January 2019
• Type: Journal Articles Status: Published Year Published: 2019 Citation: Grab, H., Brokaw, J., Anderson, E., Gedlinski, L., Gibbs, J., Wilson, J., Loeb, G., Isaacs, R., & K. Poveda (2019) Habitat enhancements rescue bee body size from the negative effects of landscape simplification. Journal of Applied Ecology 56 (9), 2144-2154
• Type: Journal Articles Status: Under Review Year Published: 2020 Citation: Grab, H., Perez, R., Bergmann, A., Loeb, G., & K. Poveda (in review) Bees in simplified landscapes are smaller, less effective pollinators. Nature Communications.
• Type: Journal Articles Status: Under Review Year Published: 2020 Citation: Schroeder, H., Grab, H ., Kessler, A., Poveda, K. (in review) Land use change as agents of selection on plant traits mediated by changes in the arthropod community.
• Type: Journal Articles Status: Other Year Published: 2020 Citation: H. Grab & K. Poveda (in prep) Causes and consequences of body size shifts within bee specie. Global Change Biology
• Type: Journal Articles Status: Under Review Year Published: 2020 Citation: RICARDO PEREZ-ALVAREZ1, HEATHER GRAB, ANTHONY POLYAKOV, AND KATJA POVEDA (in revision) Landscape composition mediates the relationship between predator body size and pest control
• Type: Theses/Dissertations Status: Published Year Published: 2019 Citation: Osicky, A. Land Use Change, Floral Resource Availability and Pesticide Application Impact Bee Body Size. Undergraduate Honors Thesis Under the supervision of Dr. Heather Grab and Dr. Katja Poveda. Cornell University Department of Entomology

Progress 04/01/18 to 03/31/19

Outputs
Target Audience: Agricultural and Conservation Scientists Extension Agents and Crop Consultants Growers and Land Managers Changes/Problems:Several of the sites where we had previously has well established local wild populations of Osmia in our trap nests did not survive the winter due to a fungal pathogen outbreak. Additionally, multiple sites had to be discarded from the study when another researcher released the same species of Osmia that had been purchased online at these sites. We made the decision to eliminate them because it could not be determined whether all the bees in our nests had originated at the site or it some had come from the other researchers stock. We are currently establishingnests a new sites and expect to conduct lab bioassays in spring 2020. What opportunities for training and professional development has the project provided?So far, the project has provided multiple avenues for professional development and training. Firstly, I have established collaborations with researchers at other institutions both inside and outside the U.S. in order to obtain specimens for the project. I have also had substantial opportunities to develop my teaching and mentoring capacity by contributing to courses in statistics, field biology, and applied entomology, as well as mentoring four undergraduate researchers. How have the results been disseminated to communities of interest?Results of the project to date have been communicated to the scientific community primarily through two peer-reviewed publications (one accepted, one under review) and to the broader agricultural community through talks at regional extension meetings and in trade publications. I have communicated our work to the public through various outreach events and to the grower community through project newsletters and in direct communications with our grower collaborators. What do you plan to do during the next reporting period to accomplish the goals?During the next funding period, I plan to conduct the historical analyses of land use change in relation to body size variation. We will continue our experimental work with trap nesting Osmia and we will repeat the study of body size effects on pollination in up to threeadditional crops (raspberry, tomato and squash).

Impacts
What was accomplished under these goals? Progress towards the overall goal of exploringthe role of landscape simplification in mediating intraspecific trait variation in bees and the consequences for crop production have been made in each of the three objective areas. Body size variation has been characterized in greatdepth for 10 species and more shallowly for an additional 25 species, which has allowed us to 1) partition statistically the community weighted body size variation across land use gradients into the intra-and inter-specific components; and 2) begin to determine species level traits that predict body size declines. We have made advances in multipleareas to determine the potential mechanisms responsible for intraspecific body size variation. First, we have used large-scale remotely sensed databases of floral resource availability and pesticide risk to explore the potential for these two predictors to associate with intra-specific body size. Secondly, we have developed methods to analyze pesticide residues in individually collected bee specimens across a range of species, which we have related to individual body size measures giving a much finer scale perspective on predictors of variation.Lastly, we have maintained and developed existing populations of a wild trap nesting species for use in lab experiments. To assess the consequences of inter and intraspecific body size variation on pollination we have completed both a broad scale correlational studyand a more controlled experimentusing measures of inter and intraspecific size variation in relation to strawberry pollination which has been submitted toNature Communicationsfor publication.

Publications

• Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: H. Grab & M. Lopez-Uribe. Managed and Wild Pollinators of Strawberry. Mid Atlantic Fruit and Vegetable Growers Conference, Hershey PA. January 2019
• Type: Journal Articles Status: Awaiting Publication Year Published: 2019 Citation: Grab, H., Brokaw, J., Anderson, E., Gedlinski, L., Gibbs, J., Wilson, J., Loeb, G., Isaacs, R., & K. Poveda (in press) Habitat enhancements rescue bee body size from the negative effects of landscape simplification. Journal of Applied Ecology.
• Type: Journal Articles Status: Submitted Year Published: 2020 Citation: Grab, H., Perez, R., Bergmann, A., Loeb, G., & K. Poveda (in review) Bees in simplified landscapes are smaller, less effective pollinators. Nature Communications.