Source: MICHIGAN STATE UNIV submitted to NRP
CAN ATTRACTION OF PREDATORY BIRDS REDUCE FOOD SAFETY RISKS POSED BY PEST BIRDS IN ORCHARDS?
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
1026382
Grant No.
2021-67012-35133
Cumulative Award Amt.
$164,071.00
Proposal No.
2020-10478
Multistate No.
(N/A)
Project Start Date
Jun 15, 2021
Project End Date
Jun 14, 2024
Grant Year
2021
Program Code
[A7201]- AFRI Post Doctoral Fellowships
Recipient Organization
MICHIGAN STATE UNIV
(N/A)
EAST LANSING,MI 48824
Performing Department
INTEGRATIVE BIOLOGY
Non Technical Summary
Foodborne pathogens cause over 9 million illnesses each year in the United States. 46% of illnesses stem from produce consumption. This has led to industry regulations to reduce wildlife on farms, but monitoring for wildlife intrusion and contamination is costly for growers to adhere to. However, few data exist to assess the risk of wildlife to food safety, leaving uncertainty over best practices to reduce food safety risks. Wild birds are known to carry foodborne pathogens that could compromise food safety, but very few data document the risk birds pose to food safety. Additionally, deterrence of birds is a large challenge for growers because commercial bird deterrents are costly and generally ineffective. Recent work demonstrated that promotion of declining falcon populations in cherry orchards can decrease fruit-eating bird abundances. Therefore, kestrels may improve food safety by deterring other birds that can defecate on crops.This project has several complementary objectives designed to better understand the true risk wild birds pose to food safety and to manage to reduce fecal contamination of crops. We will first assess rates of fecal contamination on crops through visual inspection of fruit trees in paired kestrel/no kestrel orchards (Obj. 1). We will then elucidate which bird species defecated on crops using DNA barcoding (Obj. 2). Finally, we will estimate Campylobacter spp. prevalence (the most common foodborne pathogen in wild birds) in droppings (Obj. 3). Altogether, this proposal seeks to gain a greater understanding of the true risk wild birds pose to food safety and assess one eco-friendly management practice. The findings of this research will provide important information to develop food safety guidelines, and prior research suggests that growers will be willing to install kestrel nest boxes to provide multi-functional benefits, including improving food safety. More broadly, this proposal seeks to find solutions to co-manage biodiversity and human well-being.
Animal Health Component
90%
Research Effort Categories
Basic
10%
Applied
90%
Developmental
0%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
72208201070100%
Knowledge Area
722 - Zoonotic Diseases and Parasites Affecting Humans;

Subject Of Investigation
0820 - Wild birds;

Field Of Science
1070 - Ecology;
Goals / Objectives
Ipropose several complementaryobjectives to assess the risk birds pose to food safety in a sweet cherry system and to evaluate ifpromotion of kestrels may be one ecofriendly solution: Obj. 1: Conduct fecal contaminationsurveys to assess rates of fecal deposition on cherries inpaired kestrel/no-kestrel orchards;Obj. 2: Determine which species are responsible for defecating on trees; and Obj. 3: Estimateprevalence of Campylobacter spp. in bird feces on cherry trees. The proposed work issignificant because promoting raptors cansimultaneously reduce fruit damage and consumptionby pest birds (Plant Health and Production and Plant Products), potentially improvefoodsafety (Food safety, nutrition, and health), and support a rapidly declining species (Bioenergy,Natural Resources, andEnvironment).
Project Methods
Our project has several complementary objectives designed to better understand the true risk wild birds pose to food safety. This includes documenting that key steps occur in the foodborne pathogen spillover cycle from wild birds to humans (i.e., pathogens are present in wild birds and wild birds defecate on fruit).Obj. 1: Conduct fecal contamination surveys to assess rates of fecal deposition on cherries in paired kestrel/no-kestrel orchards. Bird fecal droppings must contact fruit to cause foodborne illness. Therefore, I will conduct fecal contamination surveys on sweet cherry trees. Approach: I will recruit 10 farms with no kestrel boxes and 10 farms with kestrel boxes. Shave et al. (2018) found a significant effect of kestrel nest box occupancy on pest bird abundances surveying 15 total orchards, so I anticipate sufficient power. I will then establish 100-m long transects of trees to estimate fecal contamination, one in the outer row of the orchard block that is next to natural habitat when available, and one in the center of the orchard block. I will use paired edge and interior transects because prior research has demonstrated that bird damage is greatest on the edge of plots and lowest in the interior, suggesting fecal contamination may follow similar trends (Kross et al., 2012). I will give priority to edges next to natural habitat because this has previously been linked to increased foraging, while edges adjacent to other crop fields likely do not have increased foraging activity (Boesing, Nichols, & Metzger, 2017; Heath, Soykan, Velas, Kelsey, & Kross, 2017; Lindell et al., 2016; Shave et al., 2018). In kestrel orchards, I will measure the distance to kestrel boxes for each point and place them as close as possible to active boxes within the set parameters. I will use ArcGIS to randomly assign transect start locations to avoid selection bias. I will conduct one visit to each farm in April 2021 to map the farm boundaries, confirm kestrel occupancy, and generate the random sampling pattern.I will conduct 2 fecal contamination surveys per orchard transect, coinciding with when fruit are ripe, with surveys on the same farm taking place 10 days apart. I will estimate fecal deposition rates on fruit by randomly selecting 10 cherry trees in each transect and mark them with flagging tape to survey the same trees for the 2 surveys. I will randomly select branches following the protocol of Lindell et al. (2016). For each tree, I will generate random numbers to delineate the horizontal and vertical components of the branch location. For the horizontal component, I will randomly select one of the eight half-winds of the compass (NNE, ENE, ESE, SSE, SSW, WSW, WNW, NNW). For the vertical component, I will measure the height of the tree and randomly select a number corresponding to 0.5-m intervals between the base of the plant's foliage and the top of the tree. I will then count the number of cherries on the branch with fecal droppings and the number without fecal droppings. Additionally, I will count the number of leaves with and without fecal droppings. Fecal droppings on leaves near fruit may still pose a risk to food safety as prior studies have demonstrated viable E. coli up to 5.33 ft from in-field fecal contamination (Jeamsripong, Chase, Jay-Russell, Buchana, & Atwill, 2019). The tree location will be GPS marked and the distance to a kestrel box recorded, if present. Analysis: I will use linear mixed effects models with a binomial distribution and logit link function to compare the number of cherries with and without fecal droppings between the kestrel and no-kestrel sites. I will also examine the impact of edge vs. interior sites and proximity to kestrel boxes on the number of cherries with fecal droppings. I will repeat the analyses examining the number of leaves with and without fecal droppings. Pitfalls and limitations: It is possible that we will find few droppings on fruit or leaves. In the event that fecal contamination is rare and statistical power is too low to determine a difference in kestrel/no kestrel or interior/edge sites, our study will remain useful because knowing that birds pose a low threat to food safety is critical missing information.Obj. 2: Determine which species are responsible for defecating on trees. Approach: During my fecal monitoring in Obj. 1, I will collect all fecal samples observed, up to 20 per farm per survey. I will place one-half of each sample in a cryotube with 200 proof ethanol, store samples on dry ice, and ship them to MSU, to be stored at -80ºC in collaborator Owen's lab until DNA extraction in the fall. I will place the other half in Cary Blair transport media, store on ice (Taff et al., 2016), and ship these sample subsets within 24 h to the MSU Veterinary Diagnostic Lab for Campylobacter spp. testing (see Obj. 3 below). With help from collaborator Owen in the fall, I will extract DNA from frozen and stored fecal samples using QIAamp® DNA stool mini kits (Qiagen, Hilden, Germany). I will then follow the protocol described in Joo & Park (2012) used in my dissertation work to identify bird species that defecated the samples by amplifying the COI gene using PCR and Sanger sequencing. I can then estimate Campylobacter spp. prevalence for individual bird species defecating directly on trees. Pitfalls and limitations: In my dissertation work, we were able to assign a species identity in 38.1% of samples (Smith, Edworthy et al., 2020). There did not appear to be a bias in ability to detect species in positive versus negative samples. However, there may be a bias in which species are able to be detected due to differential degradation of DNA in feces due to differences in fecal composition. Still, prior systematic bird surveys conducted in the same orchards by Shave et al. (2018) provide a list of which species use orchards and may defecate on crops, while my Obj. 2 and 3 will identify which of these birds are pathogen reservoirs. We will apply for additional grants from the Center for Produce Safety, Northcentral SARE, Michigan Horticulture Society, and MSU Project GREEEN to test additional samples and conduct mist-netting in the second year of the project to better estimate individual bird species pathogen prevalence. Finally, if insufficient feces are found on trees during my transect surveys, I will instead use fixed-time surveys where I search trees for 30 minutes until I meet my sample collection target. In my dissertation work, although feces were rare on plants (3.3%), I was able to find, label, and preserve 30-60 samples per farm within 30-60 minutes by covering large areas.Obj. 3: Estimate prevalence of Campylobacter spp. in bird feces on cherry trees. Regardless of fecal dropping presence, droppings must contain viable bacteria to cause human infection. Approach: As stated in Obj. 2, I will place one-half of each fecal sample in Cary Blair transport media in a cooler on ice (Taff et al., 2016). I will ship samples within 24 h to the MSU Veterinary Diagnostic Lab for Campylobacter spp. testing.Pitfalls and limitations: Similar to limitations noted in Obj. 2, fecal droppings on plants are often rare, and if I find few feces during my transect surveys, I will instead use fixed-time surveys. If feces and fecal contamination are rare, this project still provides crucial information to growers that wild birds likely do not pose a large food safety risk in this system, and growers may not need to use expensive and time-intensive practices to reduce risk. Our surveys can also identify which bird species are likely to be the largest issues to allow growers to target management more effectively.

Progress 06/15/21 to 06/14/24

Outputs
Target Audience:Participating farmers: We created reports concerning all data collected for each farm in the first year of the study and distributed reports to their respective farmers. Farmers interested in managing birds on farms: We presented results for several extension and outreach venues. PD Smith and mentor Catherine Lindell talked about the planned work on the Michigan State University Ag and Wildlife Podcast in 2021.PD Smith also presented results on the Michigan State University AgBites podcast in 2022.PD Smith presented resultsas part of two courses for farmers and other agricultural professionals put on by the nonprofit Wild Farm Alliance. One was the"Role of Birds on Farms Course" in 2022, and one was the "Support Avian Pest Control Services California Continuing Education Course" (2023). PD Smith and mentor Lindell presented the results at theGreat Lakes Fruit, Vegetable & Farm Market EXPO in two sessions in 2022.Good Fruit Grower Magazine published a news article on the project in 2022. Mentor Lindell presented results for farmers at the Southeast Fruit and Vegetable Conference and Ontario Fruit and Vegetable Convention. PD Smith presented the results as part of larger talks for birders on co-managing farmlands for bird conservation, crop production, and food safety for the Capital Area Audubon Society and W. K. Kellogg Biological Station Bird Sanctuary in 2024. Academics: PD Smith gave a 15 minute talk on the results at the Michigan State University Ecology, Evolution, and Behavior Program colloquium in 2021. PD Smith discussed some of the results as part of larger seminars on how birds impact agriculture at University of Illinois at Urbana-Champaign (2023), Penn State University (2023), Michigan State University (2024), and Siena Heights University (2024). PD Smith and mentor Lindell also presented the work for visiting researchers from Jordan and Georgia (country) in a field day in 2023. The field day was part of a one-week training on avian disease ecology that the researchers were participating in.A manuscript from the study is currently drafted and nearing submission to a peer-reviewed journal. General public: PD Smith has regularly communicated research activities through social media platforms included Instagram, Twitter, and Facebook. Changes/Problems:We encountered fewer feces on trees than we had desired in year 1 (total 159). Additionally, the feces had lower Campylobacter spp. prevalence than expected based on prior research. Therefore, in 2022, we captured birds using mist-nets to collect fecal fresh samples. The added benefit of this approach was to identify (1) which specific bird species were shedding Campylobacter bacteria and (2) that we could estimate the prevalence of Campylobacter spp. in fresh bird feces and compare that to old feces to assess bacterial survival. Our proposal stated that we would follow this course of action upon low fecal encounter. Our research was conducted with the approval of Michigan State University's Institutional Animal Care and Use Committee (IACUC ID: PROTO202100100). Mist-netting was conducted on the federal bird banding permit #22766 permitted to Catherine Lindell. We requested a one-year no-cost extension on the project to allow more time for laboratory assays to be finished. We brought on three collaborators to conduct this work who experienced unexpected delays. The delays did not impact total expenditures. The laboratory assays have been completed, and we have drafted a publication of the results to be submitted for peer review soon. We also brought in a collaborator to conduct whole genome sequencing on our Campylobacter isolates, which allows us to determine if the birds are likely exchanging pathogens with humans and if they carry antibiotic resistance genes. This additional component allows us to better understand how likely birds are to make people sick and how bad the particular strains of Campylobacter they carry are. What opportunities for training and professional development has the project provided?Training activities: PD Smith gained invaluable experience in designing, coordinating, conducting, and leading this research project. PD Smith mentored a graduate student, Olivia Utley, on the project. Olivia Utley assisted with the field work. Two undergraduate students (Ashley de Borchgrave d'Altena and Talia Swartout) helped culture Campylobacter in feces as part of their Veterinary Diagnostic Lab internships and are included as coauthors on the publication. Another graduate student (Sofia Varriano) was invited to collaborate on the project under the guidance of postdoctoral researcher Pedro Rodrigues and gained additional experience in molecular biology methods. Additionally, two postdoctoral researchers (Karla Vasco and Jose Rodrigues) were invited to conduct whole genome sequencing on Campylobacter isolates and be coauthors on the publication. PD Smith and mentor Lindell also presented the work for visiting researchers from Jordan and Georgia (country) in a field day in 2023. The field day was part of a one-week training on avian disease ecology that the researchers were participating in. Professional developmentactivities: PD Smith was able to present results at multiple conferences and seminars. PD Smith gave a 15 minute talk on the resultsat the Michigan State University Ecology, Evolution, and Behavior Program colloquium in 2021. PD Smith discussed some of the results as part of larger seminars on how birds impact agriculture at University of Illinois at Urbana-Champaign (2023), Penn State University (2023), Michigan State University (2024), and Siena Heights University (2024). How have the results been disseminated to communities of interest?We presented results for several extension and outreach venues.PD Smith and mentor Catherine Lindell talked about the planned work on the Michigan State University Ag and Wildlife Podcast in 2021.PD Smith also presented results on the Michigan State University AgBites podcast in 2022.PD Smith presented resultsas part of two courses for farmers and other agricultural professionals put on by the nonprofit Wild Farm Alliance. One was the"Role of Birds on Farms Course" in 2022, and one was the "Support Avian Pest Control Services California Continuing Education Course" (2023).PD Smith and mentor Lindell presented the results at theGreat Lakes Fruit, Vegetable & Farm Market EXPO in two sessions in 2022.Good Fruit Grower Magazine published a news article on the project in 2022.Mentor Lindell presented results for farmers at the Southeast Fruit and Vegetable Conference and Ontario Fruit and Vegetable Convention. PD Smith presented the results as part of larger talks for birders on co-managing farmlands for bird conservation, crop production, and food safety for the Capital Area Audubon Society and W. K. Kellogg Biological Station Bird Sanctuary in 2024. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Ipropose several complementaryobjectives to assess the risk birds pose to food safety in a sweet cherry system and to evaluate ifpromotion of kestrels may be one ecofriendly solution: Obj. 1: Conduct fecal contaminationsurveys to assess rates of fecal deposition on cherries inpaired kestrel/no-kestrel orchards. 1) Major activities completed / experiments conducted; We conducted fecal contamination surveys for two summers (2021-2022) in 16 sweet cherry orchards (8 with kestrels and 8 without kestrels). 2) Data collected; We examined 1240 sweet cherry tree branches for bird feces and noted the number of feces found. 3) Summary statistics and discussion of results; We found 96 bird droppings across all 1240 focal cherry branch sections, equating to an average of 0.08 feces per 1-m of branch across sites. We found feces on 4.68% of the 1240 branches examined across surveys. However, feces on the actual cherries were rare (4/15,890 [0.025%]). Sites with active nest boxes had fewer branches with feces than control sites (2.33% vs. 6.88%; β = -1.18 ± 0.42 (SE), P = 0.005). The odds of finding feces on branches were 3.26 times higher at control sites. At kestrel sites, the trees further from nest boxes had increased probabilities of having bird feces on branches (β = 0.0038, P = 0.050): with every 1-m increase in distance from the box, the odds of finding bird feces increased by 1.004. Altogether, our findings suggest that attracting American kestrels to cherry orchards may be an effective way to reduce fecal contamination on sweet cherry trees. 4) Key outcomes or other accomplishments realized. Our study is the first to show that birds of prey can be used to improve food safety. We have presented our results for farmers and researchers at several venues, and the feedback we have received suggests our findings will lead to more farmers installing kestrel nest boxes to improve food safety. We expect this will improve food safety while conserving a species of conservation concern. Obj. 2: Determine which species are responsible for defecating on trees; 1) Major activities completed / experiments conducted; We conducted observational bird transect surveys in 2021 to determine which bird species use sweet cherry orchards. We also collected fecal samples from sweet cherry trees in 2021 and 2022, then used molecular barcoding to identify which species were responsible for depositing feces on trees. 2) Data collected; We collected data on bird abundances and collected fecal samples. We extracted DNA from the fecal samples, amplified bird DNA using PCR, then sequenced the DNA to identify the source of the sample. 3) Summary statistics and discussion of results; We observed 236 individual birds from 28 species during transect surveys. The most commonly observed species were American robin (Turdus migratorius; n = 58), common grackle (Quiscalus quiscula; n = 30), and black-capped chickadee (Poecile atricapillus, n = 17). Sites with active kestrel nest boxes had fewer birds detected in orchard transects than control sites (β = -0.89 ± 0.42 (SE), P = 0.033). We traced the feces found on crops back to the bird species responsible in 132/310 (42.6%) samples and were able to identify the genus responsible for one more. Feces were most often traced back to American robin (n = 48), American crow (n = 41; Corvus brachyrhynchos), and common grackle (n = 23). Other species were responsible for five or fewer samples each. Feces came from at least 12 species. It appears that American robin, American crow, and common grackles are species of higher concern for farmers in our study region due to their high densities and the frequency with which they defecate on crops. 4) Key outcomes or other accomplishments realized. Our study is the first on avian-mediated food safety risks in produce conducted in the Midwest region and for tree fruit. Thus, prior work couldn't elucidate which species were of importance to these farmers. Farmers can use this information to assess if the species on their farms are likely to be an issue and if they should take action against them. Knowing which species are of concern is an important step for guiding future research on how to deter the key pests. Obj. 3: Estimateprevalence of Campylobacter spp. in bird feces on cherry trees. 1) Major activities completed / experiments conducted; We collected 310 bird fecal samples from cherry tree surfaces across2021 and 2022. Half of each sample was cultured for Campylobacter. The other half was used for PCR assays for Campylobacter spp. 2) Data collected; We had 8 samples that were culturally positive for Campylobacter. For these, we identified the species of Campylobacter using aMALDI-TOF MS Microflex LT Biotyper (Bruker Daltonics, Germany). Our PCR assays included primers for 5 species of Campylobacter. 3) Summary statistics and discussion of results; Viable Campylobacter was cultured from 3/310 (0.97%) fecal samples collected from cherry trees.All came from non-kestrelsites and were identified as C. jejuni. Using molecular barcoding, we traced two samples with C. jejuni back to American crow (n = 1) and American robin (n = 1), while one source remained unidentified. Campylobacter detection rates were again higher using PCR than culture. The Campylobacter 23S rRNA gene was detected in 97/310 (31.29%) samples; however, only 29 of these 97 samples could be characterized using species-specific primers from either PCR protocol. Our species-specific PCR assays detected C. jejuni in 9/310 (2.9%) samples, C. coli in 6/310 (1.94%), C. lari in 9/310 (2.9%), C. upsaliensis in 9/310 (2.9%), and C. fetus subsp. fetus in 10/310 (3.23%). One or more specific Campylobacter species were detected using culturing and/or PCR in 33/310 samples (10.65%). We detected Campylobacter in 10/91 (11.0%) samples from kestrel sites and 21/219 (9.59%) samples from non-kestrelsites (the difference was not significant; β = -0.041 ± 0.60 (SE), P = 0.95). Overall, we detected little viable Campylobacter in bird feces fromcherry trees. 4) Key outcomes or other accomplishments realized. Our bacterial testing suggests that viable Campylobacter is uncommon in bird feces, so birds are not a huge food safety issue in cherry orchards in Michigan.

Publications

  • Type: Journal Articles Status: Other Year Published: 2025 Citation: Smith, O.M., Rodrigues, P.A., Groendyk, S., Utley, O.J., dAltena, A., Carbonell, S., Davis, K.L., Rodrigues, J.A., Swartout, T., Varriano, S., Vasco, K.A., Kettler, N., Mani, R., Manning, S.D., Owen, J.C., Snyder, W.E., and Lindell, C.A. Falcons reduce food safety risks and damage posed by wild birds (in preparation for Journal of Applied Ecology).


Progress 06/15/22 to 06/14/23

Outputs
Target Audience:Farmers: PD Smith and mentor Lindell presented the results at theGreat Lakes Fruit, Vegetable & Farm Market EXPO in two sessions.Good Fruit Grower Magazine published a news article on the project. Mentor Lindell presented results for farmers at the Southeast Fruit and Vegetable Conference and Ontario Fruit and Vegetable Convention.PD Smith presented resultsas part of an onlinecoursefor farmers and other agricultural professionals put on by the nonprofit Wild Farm Alliance. Academics: PD Smith discussed some of the results as part of larger seminars on how birds impact agriculture at University of Illinois at Urbana-Champaignand Penn State University. General public: PD Smith has regularly communicated research activities through social media platforms included Instagram, Twitter, and Facebook. Changes/Problems:We requested a one-year no-cost extension on the project to allow more time for laboratory assays to be finished. We brought on three collaborators to conduct this work who experienced unexpected delays. The delays will not impact total expenditures. We also brought in a collaborator to conduct whole genome sequencing on our Campylobacter isolates, which will allowus to determine if the birds are likely exchanging pathogens with humans and if they carry antibiotic resistance genes. This additional component will allowus to better understand how likely birds are to make people sick and how bad the particular strains of Campylobacter they carry are. This will not increase project costs as another grant is paying for the sequencing. The collaborators are making use of isolates we collected in this study. What opportunities for training and professional development has the project provided?Training activities:PD Smith is gaininginvaluable experience in designing, coordinating, conducting, and leading this research project. PD Smith mentored a graduate student, Olivia Utley, on the project. Olivia Utley assisted with the field work. Two undergraduate students (Ashley de Borchgrave d'Altena and Talia Swartout) helped culture Campylobacter in feces as part of their Veterinary Diagnostic Lab internships and will beincluded as coauthors on the publication. Another graduate student (Sofia Varriano) was invited to collaborate on the project under the guidance of postdoctoral researcher Pedro Rodrigues and gained additional experience in molecular biology methods. Professional developmentactivities: PD Smith was able to present results at multiple conferences and seminars. PD Smith gave a 15 minute talk on the resultsat the Michigan State University Ecology, Evolution, and Behavior Program colloquium in 2021. PD Smith discussed some of the results as part of larger seminars on how birds impact agriculture at University of Illinois at Urbana-Champaign (2023) and Penn State University (2023). How have the results been disseminated to communities of interest?We presented results for several extension and outreach venues.PD Smith and mentor Catherine Lindell talked about the planned work on the Michigan State University Ag and Wildlife Podcast in 2021.PD Smith also presented results on the Michigan State University AgBites podcast in 2022.PD Smith presented resultsas part of two courses for farmers and other agricultural professionals put on by the nonprofit Wild Farm Alliance. One was the"Role of Birds on Farms Course" in 2022, and one was the "Support Avian Pest Control Services California Continuing Education Course" (2023).PD Smith and mentor Lindell presented the results at theGreat Lakes Fruit, Vegetable & Farm Market EXPO in two sessions in 2022.Good Fruit Grower Magazine published a news article on the project in 2022.Mentor Lindell presented results for farmers at the Southeast Fruit and Vegetable Conference and Ontario Fruit and Vegetable Convention. What do you plan to do during the next reporting period to accomplish the goals?Finish collecting data: We will continue our laboratory assays to determine what percent of bird fecal samples have Campylobacter determined through PCR methods. The PCR approach allows us to detect Campylobacter that is nonculturalable. We will also finish our molecular barcoding to determine what species were responsible for leaving feces on trees. Our collaborators at the University of Georgia are responsible for this component, so we will periodically check-in with them until the data are collected and objectives are achieved. Analyze data: Conduct final analyses on all data. Write a manuscript for peer review: We have begun drafting a manuscript but are awaiting the laboratory results to finish the manuscript. Once we get all data, the analyses, figures, and writing will be finalized.

Impacts
What was accomplished under these goals? Ipropose several complementaryobjectives to assess the risk birds pose to food safety in a sweet cherry system and to evaluate ifpromotion of kestrels may be one ecofriendly solution: Obj. 1: Conduct fecal contaminationsurveys to assess rates of fecal deposition on cherries inpaired kestrel/no-kestrel orchards. 1) Major activities completed / experiments conducted; We conducted fecal contamination surveys for two summers (2021-2022) in 16 sweet cherry orchards (8 with kestrels and 8 without kestrels). 2) Data collected; We examined 1240 sweet cherry tree branches for bird feces and noted the number of feces found. 3) Summary statistics and discussion of results; We found 96 bird droppings across all 1240 focal cherry branch sections. We found feces on 4.68% of the 1240 branches examined across surveys. However, feces on the actual cherries were rare (4/15,890 [0.025%]). Sites with active nest boxes had fewer branches with feces than control sites (2.33% vs. 6.88%). Altogether, our findings suggest that attracting American kestrels to cherry orchards may be an effective way to reduce fecal contamination on sweet cherry trees. 4) Key outcomes or other accomplishments realized. Our study is the first to show that birds of prey can be used to improve food safety. We have presented our results for farmers and researchers at several venues, and the feedback we have received suggests our findings will lead to more farmers installing kestrel nest boxes to improve food safety. We expect this will improve food safety while conserving a species of conservation concern. Obj. 2: Determine which species are responsible for defecating on trees; 1) Major activities completed / experiments conducted; We conducted observational bird transect surveys in 2021 to determine which bird species use sweet cherry orchards. We also collected fecal samples from sweet cherry trees in 2021 and 2022, and are now usingmolecular barcoding to identify which species were responsible for depositing feces on trees. 2) Data collected; We collected data on bird abundances and collected fecal samples. We extracted DNA from the fecal samples and are working on amplifying bird DNA using PCR. DNA will be sent forsequencingto identify the source of the sample. 3) Summary statistics and discussion of results; We observed 236 individual birds from 28 species during transect surveys. The most commonly observed species were American robin (Turdus migratorius;n= 58), common grackle (Quiscalusquiscula;n= 30), and black-capped chickadee (Poecile atricapillus,n= 17). Sites with active kestrel nest boxes had fewer birds detected in orchard transects than control sites. We are currently working on identifying what species defecated on trees using molecular barcoding methods. 4) Key outcomes or other accomplishments realized. Our study is the first on avian-mediated food safety risks in produce conducted in the Midwest region and for tree fruit. Thus, prior work couldn't elucidate which species were of importance to these farmers. Farmers can use this information to assess if the species on their farms are likely to be an issue and if they should take action against them. Knowing which species are of concern is an important step for guiding future research on how to deter the key pests. Obj. 3: Estimateprevalence of Campylobacter spp. in bird feces on cherry trees. 1) Major activities completed / experiments conducted; We collected 310 bird fecal samples from cherry tree surfaces across2021 and 2022. Half of each sample was cultured for Campylobacter. The other half is beingused for PCR assays for Campylobacter spp. 2) Data collected; We had 8 samples that were culturally positive for Campylobacter. For these, we identified the species of Campylobacter using aMALDI-TOF MS Microflex LT Biotyper (Bruker Daltonics, Germany). Our PCR assays will includeprimers for 5 species of Campylobacter. 3) Summary statistics and discussion of results; ViableCampylobacterwas cultured from 3/310 (0.97%) fecal samples collected from cherry trees.All came from non-kestrelsites and were identified asC. jejuni. 4) Key outcomes or other accomplishments realized. Our bacterial testing suggests that viable Campylobacter is uncommon in bird feces, so birds are not a huge food safety issue in cherry orchards in Michigan.

Publications


    Progress 06/15/21 to 06/14/22

    Outputs
    Target Audience:Participating farmers: We created reports concerning all data collected for each farm in the first year of the study and distributed reports to their respective farmers Farmers interested in managing birds on farms: We presented some preliminary results as part of Wild Farm Alliance's Role of Birds on Farms Course. This course provides certification for participants. PD Smith presented as part of Lesson 6:Co-Existing With Birds and Making Farms Safe Academics: PD Smith gave a 15 minute talk on the results from the year of data collection at the Michigan State University Ecology, Evolution, and Behavior Program colloquium General public: PD Smith has regularly commicated research activities through social media platforms included Instagram, Twitter, and Facebook. Changes/Problems:We encountered fewer feces on trees than we had desired in year 1 (total 159). Additionally, the feces had lower Campylobacterspp. prevalence than expected based on prior research. Therefore, in year 2, we conducted mist-netting to capture birds to collect fresh fecal samples.Netting was conducted between 20 June and 31 July 2022 between 6:10 and 11:45 AM. We netted each site once then revisited the four sites expected to have the best captures to maximize our sample size. We used six polyester mist-nets that were 2.6 m high and 12-m long. Half had 30-mm mesh and half had 38-mm mesh. At each site, we placed one net at the start and one net at the end of the transect expected to have higher captures based on transect surveys. We placed four nets at the edge of that transect, preferably adjacent to seminatural edge if present. Each net had an effort of approximately 4 hours.We placed birds extracted from mist-nets in new brown paper lunch bags to defecate. Fecal samples were preserved the same way as the environmental fecal samples. Our proposal stated that we would follow this course of action upon low fecal encounter.Our research was conducted with the approval of Michigan State University's Institutional Animal Care and Use Committee (IACUC ID: PROTO202100100). Mist-netting was conducted on the federal bird banding permit #22766 permitted to Catherine Lindell. What opportunities for training and professional development has the project provided?Undergraduate students: two undergraduates were provided training opportunities because they were employed by the Veterinary Diagnostic Lab at Michigan State University to cultureCampylobacterspp. Both undergraduates will be coauthors on the publication to provide additional experience and professional development opportunities. Graduate student: one graduate student assisted with field work in 2021 and was mentored by PD Smith. She was asked to help with methodology, farmer coordination, nest box monitoring, and data collection. She will be a coauthor on the publication. Postdoctoral Fellow: PD Smith gained additional training and professional development opportunities by directing this project and leading the research. PD Smith also has gained professional development opportunities by mentoring the graduate student assistant and presenting the results. How have the results been disseminated to communities of interest?Participating farmers: We created reports concerning all data collected for each farm in the first year of the study and distributed reports to their respective farmers Farmers interested in managing birds on farms: We presented some preliminary results as part of Wild Farm Alliance's Role of Birds on Farms Course. This course provides certification for participants. PD Smith presented as part of Lesson 6:Co-Existing With Birds and Making Farms Safe Academics: PD Smith gave a 15 minute talk on the results from the year of data collection at the Michigan State University Ecology, Evolution, and Behavior Program colloquium Social media: PD Smith has regularly commicated research methods and results through social media platforms included Instagram, Twitter, and Facebook. What do you plan to do during the next reporting period to accomplish the goals?Statistical analyses:We plan to analyze all field data in fall 2022. Molecular work: Collaborator Shannon Manning's lab has extracted DNA from all 8Campylobacterspp. isolates. The one collected in 2021 has already been sequenced, and the remaining 7 will be shortly. We are preparing to extract DNA from fecal samples to identify what species defecated them. Additionally, we intent to test forCampylobacterspp. in fecal samples using PCR in case bacteria died prior to culturing. Manuscript preparation: We are currently preparing our manuscript. We plan to submit it to a peer reviewed journal in spring 2023. Presentations/outreach: Farmer: We will present our results at 3 invited sessions of the Great Lakes Expo, an annual conference for fruit and vegetable growers. Additionally, PD Smith was invited to participate in an upcoming video outreach series by Wild Farm Alliance. Finally, we will share full data summaries with participating farms in fall 2022. Academic: We will present our results at the American Ornithological Society conference.

    Impacts
    What was accomplished under these goals? Recent foodborne illness outbreaks have heighted pressures on produce growers to improve food safety, including possible threats from wildlife. Wild birds pose a unique challenge because they are difficult to deter, and efforts to deter birds from farms can jeopardize conservation and biological control efforts. Yet, the probability that foodborne pathogens spillover from birds to humans remains uncertain, and it remains unclear what individual farmers can do to reduce risk. Promotion of predatory birds with nest boxes may be one feasible solution to improve food safety because predatory birds can reduce the abundance of birds in crop fields. In this project, we are evaluatingif the American kestrel (Falco sparvarius), a predatory bird, can reduce food safety risks from birds in sweet cherry orchards in Michigan, USA using a barriers-to-spillover approach. In orchards with and without active kestrel nest boxes, we first conducted avian transect surveys. Second, we collected fecal samples from captured birds alongside fecal samples collected from sweet cherry trees then tested feces for the most common foodborne pathogen in birds,Campylobacterspp. We then estimated the percent of branches and cherries with feces. We are currently working onsequencingtheCampylobacterisolates found in birds to compare those to isolates in human illnesses. Our preliminary results indicate that there were fewer birds present in crop fields with active American kestrel nest boxes and thatthis translated into fewer branches with fecal contamination (6.56%vs. 2.03%). However, contamination of individual cherries was rare (4/15,786 [0.025%]).Campylobacterspp. were more 4.2 times more often detected in fresh feces taken from captured birds (5/122 [4.10%]) than in samples that were found on crops (3/310 [0.97%]). This suggests that bacteria do not survive long in the environment. Food safety and wildlife conservation are often thought to be in conflict, and produce growers have few tools to effectively manage birds. However, our preliminary findings suggest that promotion of predatory birds using nest boxes may be a way growers can effectively conserve a declining species, reduce crop damage, and improve food safety. Obj. 1: Conduct fecal contaminationsurveys to assess rates of fecal deposition on cherries inpaired kestrel/no-kestrel orchards We conducted surveys from 2021-2022 in 16 sweet cherry orchards located in Leelanau County, MI. Half of the sites included active kestrel nest boxes that were established between 2012-2016. In 2021, our study included 8 sweet cherry orchards with active kestrel nest boxes and 8 without boxes (controls). In 2022, one site with a nest box was unoccupied, so we surveyed 7 sites with active boxes plus the 8 control sites.We established two 100-m long transects in each sweet cherry orchard where we conducted bird transect surveys, fecal contamination surveys, and deployed nets for mist-netting. For our fecal contamination surveys, we sampled 10 trees in each of the 32 transects between 28 June-27 July 2021 and 17 June-19 July 2022. Each transect was surveyed twice per year, coinciding with when fruit were ripe and ready for harvest.Within each transect, we surveyed the first tree coinciding with the transect start point then selected the other 9 focal trees spaced at 10-m intervals along the transect. We alternated between the left and right-side rows in a zig-zag pattern to minimize microclimate effects. For each plant, we randomly selected a branch following Lindell et al. (2016). We surveyed the outermost 1-m of each focal branch. We counted the total number of cherries along the 1-m-section of the branch and noted the number of cherries with bird feces. Because we rarely observed feces on cherries (4/15,786 [0.025%]) and pathogens can be mechanically vectored or moved by water, we also examined leaves for feces with the assumption that feces on branches may still represent a food safety risk. We also counted the number of cherries with bird damage. We have completed fecal contaminationsurveys to assess rates of fecal deposition on cherries inpaired kestrel/no-kestrel orchards.Preliminarily, in 2021, 1.88% (6/320) of cherry tree branchesin kestrel sites had bird feces. In 2022, 2.50% (7/280) of branches in kestrel sites had bird feces. In 2021, 8.13% (26/320) of branchesin controlsites had bird feces. In 2022, 5.00% (16/320) of branches in control sites had bird feces. We ran binomial generalized linear mixed effects models in fall 2021 to test if kestrels reduced the percent of branches with bird feces in the first year, and they did.We will soon be starting our semi-final models using both years of data.If kestrels do significantly reduce the percent of branches with bird feces, then we can recommend them as a means to improve food safety in orchards. Obj. 2: Determine which species are responsible for defecating on trees First, we conducted three bird density surveys along each of our 100-m transects between 28 June and 22 July 2021. These surveys aimed to elucidate bird species identity and density in orchards during the late ripening and harvest period. We used a fixed-width of 10-m and standard methods to maximize detectability of birds. Each survey lasted 10-min and transects were travelled at a pace of 10-m per minute.All birds seen or heard were recorded. We also collected all fecal samples observed on focal branches during fecal contamination surveys described above. Due to low fecal encounter rates, we collected up to 10 feces per tree within the transect per survey, prioritizing all feces on focal branches first. One halfof each sample was stored in a cryotube in 200-proof ethanol, stored on dry ice, shipped back to Michigan State University, and stored at -80ºC awaiting DNA extraction. We plan toextract bacterial DNA from avian feces this fall. Extracted DNA will bestored at -20ºC until COI testing. We willfollowthe protocol described in Joo & Park (2012) to identify bird species that defecated the samples collected from fields using the COI gene and Sanger sequencing, except that we willnot include the plasmid insertion and cloning steps (Smith et al. 2020a). We observed 235 individual birds from 27 species during transect surveys. The most commonly observed species were American robin (Turdus migratorius; n = 58), common grackle (Quiscalus quiscula; n = 30), and black-capped chickadee (Poecile atricapillus, n = 17). We collected 310 avian fecal samples from trees across the two years that will be used to identify which species defecated on trees. Obj. 3: Estimateprevalence of Campylobacter spp. in bird feces on cherry trees. We tested the samples that we collected from branches (described above) forCampylobacterspp.One-half of each fecal sample was preserved in 1.25 ml Cary Blair media, stored on wet ice, and shipped back to the Michigan State University Veterinary Diagnostic Lab to be cultured for Campylobacter spp. within 24-h of collection.Both direct and enrichment culture methods were used for isolating Campylobacter spp. from fecal samples. We detected Campylobacter spp. in 3/310 (0.97%) fecal samples from trees. All came from control sites and were identified as Camplyobacter jejuni. Due to low fecal encounter rates on trees in 2021(159 samples), we alsoconducted mist-netting in 2022 to capture birds to test their fresh feces forCampylobacterspp. We processed samples using the same methods described above for feces from trees. In total, we collected 122 fresh fecal samples from 20 species of birds.Of these, we detected C. jejuniin 5 samples (4.10%) total, which included samples from American robin (n = 3/18 [16.7%]), common grackle (n = 1/6 [16.7%]), and northern cardinal (Cardinalis cardinalis; n = 1/5 [20.0%]). We will conduct additional analyses this fall comparingCampylobacter recovery in fresh vs. old samples and further classifyisolates using whole genome sequencing.

    Publications