MOVEMENT AND POPULATION DYNAMICS OF GREAT LAKES MALLARDS By Benjamin Zachariah Luukkonen A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Fisheries and Wildlife – Doctor of Philosophy 2024 ABSTRACT Mallard (Anas platyrhynchos) abundance in the Great Lakes states declined by 30% from 2000 to 2023 based on spring aerial surveys in Michigan, Minnesota, and Wisconsin. Concern among management agencies and uncertainty in the factors contributing to Great Lakes mallard decline led to a regional partnership which initiated this research project. The research objectives were to estimate Great Lakes mallard demographic parameters, determine which parameters had driven declines in abundance, and reduce uncertainty about the ecological and anthropogenic factors affecting movement and population dynamics. We captured, collected genetic and feather samples from, and marked 592 female mallards with GPS-GSM transmitters in Michigan, Wisconsin, Ohio, Indiana, and Illinois, USA from 2021–2023. We also used 32 years of banding, band recovery, and aerial survey data collected from mallards in Michigan and Wisconsin during 1991–2022 to develop an Integrated Population Model (IPM). Genetic ancestry analysis of GPS-marked female mallards via ddRAD sequencing revealed 44% were wild mallards and 56% were wild x domestic game-farm mallard hybrids. Hybrid mallards had shorter daily movement distances, were less likely to engage in autumn migration, and had higher use and selection of urban developed land cover than did wild mallards. Survival of female GPS-marked mallards was positively related to the proportion of locations in urban developed land cover, regardless of individual genotype, suggesting urban land cover use could be a source of individual heterogeneity in survival. Female mallards with greater domestic ancestry primarily used urban developed land cover, raising questions about their ability to survive in rural habitat types primarily used by wild mallards. Increasing proportion of domestic ancestry was associated with significantly lower probability of initiating nest incubation, indicating early generational hybrids had low productivity. Sedentary behavior, use and selection of urban areas, and low incubation incidence related to domestic ancestry raises concern regarding hybridization between wild and domestic game-farm mallards. Molting and natal origins estimated from stable hydrogen (δ2H) isotopes predicted 72%– 84% of adult females molted and 59%–77% of juvenile females hatched at the latitudes of the Great Lakes region. Emigration likely contributed little to population decline as 98% of surviving female Great Lakes mallards remained in or returned to the Great Lakes region in subsequent breeding periods. The IPM identified declining productivity and increasing natural mortality in adult and juvenile female mallards as the primary demographic drivers of population decline. Productivity was lowest at urban developed banding sites in the southern Great Lakes region, where prevalence of hybrid mallards was greatest. Productivity declined with loss of area enrolled in the Conservation Reserve Program within Michigan and Wisconsin during 2000– 2022. Natural mortality was 3.5–6.7 times and 1.3–4.2 times greater than harvest mortality for adult and juvenile female mallards, respectively, suggesting environmental factors during spring and summer, not harvest, drove annual mortality for female mallards. Attempts to increase or maintain Great Lakes mallard abundance should consider regional quantity and quality of nesting and brood-rearing habitat types and population genetics. Copyright by BENJAMIN ZACHARIAH LUUKKONEN 2024 To David Richard Luukkonen, whose life and philosophy were my greatest inspiration. v ACKNOWLEDGEMENTS The Great Lakes mallard project has been a substantial undertaking and its success relied on the contributions and teamwork of nearly countless people. My guidance committee provided key expertise, advice, and direction throughout the project. Dr. Scott Winterstein was always available to discuss whatever issue I was contemplating. His many stories served as useful examples to guide academics, research, and life. Dr. Dan Hayes’ wisdom encouraged deep thinking and to always ask “why”. His love for teaching is contagious and I learned much from his engagement with students; I would be lucky to someday be half the educator he is. Dr. Drew Fowler expanded my knowledge of waterfowl life history, population processes, and emerging techniques such as stable isotope analysis. He remained committed to the project from day one despite changing jobs and taking on new responsibilities, and I am grateful for his dedication. Dr. Jerry Urquhart strengthened my ability to consider the big picture in ecology, including the multitude of factors which can interact to affect wildlife populations. I was reminded that before mallards, and not too long ago, American black ducks were the preeminent nesting duck in the Great Lakes region. Dr. Barb Avers has been a longtime mentor and source of encouragement. I am thankful for her in providing many opportunities to discuss findings with wildlife managers and stakeholders. Sharing applied research with practitioners has been one of the most rewarding aspects of the project. Beginning at its inception and lasting throughout, support for the project was overwhelming. The following organizations provided vital support: Ducks Unlimited, Forbes Biological Station and Illinois Natural History Survey, Franklin College, Indiana Department of Natural Resources, Illinois Department of Natural Resources, Michigan Department of Natural Resources, Michigan State University, United States Fish and Wildlife Service, Upper vi Mississippi/Great Lakes Joint Venture, University of Texas El Paso and the Lavretsky Lab, Winous Point Marsh Conservancy, and Wisconsin Department of Natural Resources. From fieldwork and logistics to funding and analysis, Don Avers, John Coluccy, Taylor Finger, Auriel Fournier, Drew Fowler, Doug Gorby, Phil Lavretsky, Ben O’Neal, Josh Osborn, Amy Shipley, Brendan Shirkey, John Simpson, Greg Soulliere, and Jay Winiarski have been integral to the project’s success. Major thanks to Phil Lavretsky, Vergi Musni, and the Lavretsky Lab for processing and analyzing genetic samples, and Kim Sparks and the Cornell Stable Isotope Laboratory for processing and analyzing feather samples. Funding was provided by grant GG 751210000001883 from the Great Lakes Fish and Wildlife Restoration Act, grant WLD2101 from the Michigan Department of Natural Resources Wildlife Division, and through contribution from Ducks Unlimited, Michigan State University, Winous Point Marsh Conservancy, Franklin College and Indiana Department of Natural Resources, Upper Mississippi/Great Lakes Joint Venture, and by the Joeseph Laurence Maison scholarship and Joseph G. Schotthoefer Memorial Student Award by Safari Club International, Michigan Involvement Committee. Dr. Scott Winterstein’s participation in this research was supported in part by the intramural research program of the U.S. Department of Agriculture and National Institute of Food and Agriculture through MSU AgBioResearch Project MICL02588 (Modeling Wildlife Population Dynamics). This project would have been significantly less impactful without an unbelievably generous gift to Ducks Unlimited from Dave and Denise Bunning and the Bunning Family Foundation. They understand that research inferences can depend on sample size and their support enabled us to mark a powerful sample of mallards with GPS transmitters. It is deeply vii encouraging to know that people like the Bunnings care so much about ducks. Thank you to the Bunnings, John Coluccy, and Ducks Unlimited for this amazing support. I’d like to thank all current and past waterfowl banders, whose work has contributed to making mallards one of the most highly monitored wildlife species in the world. We would never have been able to deploy 592 transmitters across five states without the help of many banders. I’d like to acknowledge the people who trapped mallards, identified opportunities to capture nesting mallards, helped deploy transmitters, and assisted with other aspects of fieldwork: Don Avers, Taylor Finger, Auriel Fournier, Drew Fowler, Ben O’Neal, Josh Osborn, Amy Shipley, Brendan Shirkey, John Simpson, Jay Winiarski, Jozalynn Boucher, Bailey Marston, Shannon Stemaly, Sierra Avendt, Corey VanStratt, Brandon Blasius, Maddison Allen, Nathan Steelman, Zosha Kuiper, John Darling, Katie Farinosi, John Niewoonder, Tim Riley, Barry Sova, Ron Sting, Rob Hamilton, Tammy Giroux, Nate Levitte, Joe Robison, Brad Johnson, Caleb Eckloff, Brian Roell, Colter Lubben, Karen Sexton, Mike Richardson, Terry McFadden, Randy Knapik, Kaitlyn Barnes, Brandy Dybas-Berger, Pat Brickel, Cameron Dole, Chad Krumnauer, Adam Shook, Zach Cooley, Jeremiah Heise, Amber Frye, Melissa Nichols, Vern Richardson, Trey McClinton, Jessica Jaworski, Josh Martinez, Eric Kroening, Andrew Greenawalt, Austin Ficher, Chelsea Kross, Steven Gurney, Rachel Correia, Samantha Courtney, Marty Johnson, Cheyenne Beach, Andy Gilbert, Chad Creamer, Bryan Woodbury, Tom Cooley, Drew Hawley, Paul Samerdyke, Brenda Kelly, Ben Williams, Brendan Woodall, Kali Rush, Chuck Farrell, Jes Rees Lohr, Jeffrey Edwards, Andrea Spurck, Nate Stott, Alex Hanrahan, Alissa Kakatsch, Allicyn Nelson, Penelope Murphy, Joe Vermeulen, Aaron Wright, Steve Burns, Thomas Carlson, Steven Easterly, Derek Christians, Coleman Nadeau, and Kyle Vandyke. Additionally, Jake Nave, Brady Hannah, and Dan Nelson of USDA APHIS Wildlife Services assisted with transmitter viii deployments and graciously allowed me to mark several birds despite reducing progress towards their disease sampling quota. Don Avers went above and beyond to ensure the project was a success and I’ve learned much from his expertise. It is an understatement to say the project would have been unsuccessful without his help. Many people have also assisted with recovering transmitters and carcasses from mallard mortalities despite challenging or unpleasant conditions. Special thanks to staff at the Veterinary Diagnostic Laboratory for performing mallard necropsies, including Tom Cooley, Kelly Straka, Julie Melotti, and Katie Farinosi. Thanks to Paul Link for sharing his insights and experience with transmitter attachment techniques. Pam Garrettson graciously provided preliminary estimates of mallard band reporting probability for chapter 5. Thank you to all the waterfowl hunters that report banded birds and contribute to one of the largest and longest running citizen science projects in the world. Without your contributions we would have far less data and the analyses in Chapter 5 would not have been possible. Further, thank you to the hunters who reported harvesting and the individuals who reported encountering GPS-marked mallards. Return of transmitters allowed us to substantially increase our sample size of marked mallards. Numerous private landowners and public land managers gave us permission to trap and band ducks and search for mallard mortalities. Special thanks to Kraig Rasche for his commitment to wetland habitat conservation and for providing us with access to a special place. I would not be where I am today, professionally or personally, without my family. Thank you to Cassidy for her love and support which has made this endeavor possible. We discovered that in pursuing our dreams, we could be committed both to our careers and to one another. I’m extremely grateful for her understanding and perseverance through my mental and physical ix absences. Although it’s been far from easy, we have succeeded and built something to be proud of. I thank my Dad for his tremendous positive influence on my life. He always believed in me. I learned so much from him. His passing meant I simultaneously lost my greatest mentor, best friend, and hunting and fishing partner. The time since has been some of the most challenging. However, he will continue to inspire and guide me. I am so grateful for the opportunity to be part of this research and this work is dedicated to him. x TABLE OF CONTENTS CHAPTER 1: INTRODUCTION ................................................................................................... 1 MIDCONTINENT MALLARD MANAGEMENT ................................................................... 1 GREAT LAKES MALLARDS .................................................................................................. 2 RESEARCH QUESTIONS, GOAL, AND OBJECTIVES ........................................................ 4 DISSERTATION CONTENT .................................................................................................... 7 FIGURES .................................................................................................................................... 8 LITERATURE CITED ............................................................................................................. 10 CHAPTER 2: GENETICS AFFECT GREAT LAKES MALLARD MOVEMENT AND RESOURCE SELECTION ........................................................................................................... 15 ABSTRACT .............................................................................................................................. 15 INTRODUCTION .................................................................................................................... 16 STUDY AREA ......................................................................................................................... 19 METHODS ............................................................................................................................... 20 RESULTS ................................................................................................................................. 29 DISCUSSION ........................................................................................................................... 32 MANAGEMENT IMPLICATIONS ........................................................................................ 37 TABLES ................................................................................................................................... 39 FIGURES .................................................................................................................................. 43 LITERATURE CITED ............................................................................................................. 52 CHAPTER 3: GREAT LAKES MALLARD SURVIVAL .......................................................... 60 ABSTRACT .............................................................................................................................. 60 INTRODUCTION .................................................................................................................... 61 STUDY AREA ......................................................................................................................... 63 METHODS ............................................................................................................................... 64 RESULTS ................................................................................................................................. 70 DISCUSSION ........................................................................................................................... 74 MANAGEMENT IMPLICATIONS ........................................................................................ 81 TABLES ................................................................................................................................... 82 FIGURES .................................................................................................................................. 84 LITERATURE CITED ............................................................................................................. 89 APPENDIX A: SURVIVAL MODELING SUPPORTING INFORMATION........................ 97 CHAPTER 4: GREAT LAKES MALLARD NESTING ECOLOGY AND FIDELITY .......... 103 ABSTRACT ............................................................................................................................ 103 INTRODUCTION .................................................................................................................. 104 STUDY AREA ....................................................................................................................... 107 METHODS ............................................................................................................................. 108 RESULTS ............................................................................................................................... 118 DISCUSSION ......................................................................................................................... 121 MANAGEMENT IMPLICATIONS ...................................................................................... 130 TABLES ................................................................................................................................. 132 FIGURES ................................................................................................................................ 134 xi LITERATURE CITED ........................................................................................................... 139 APPENDIX B: GLOBAL POSITIONING SYSTEM AND ACCELEROMETER SUMMARY STATISTICS .......................................................................................................................... 150 CHAPTER 5: POPULATION DYNAMICS OF GREAT LAKES MALLARDS .................... 153 ABSTRACT ............................................................................................................................ 153 INTRODUCTION .................................................................................................................. 154 STUDY AREA ....................................................................................................................... 158 METHODS ............................................................................................................................. 159 RESULTS ............................................................................................................................... 167 DISCUSSION ......................................................................................................................... 170 MANAGEMENT RECOMMENDATIONS .......................................................................... 177 TABLES ................................................................................................................................. 178 FIGURES ................................................................................................................................ 179 LITERATURE CITED ........................................................................................................... 190 CHAPTER 6: MANAGEMENT IMPLICATIONS ................................................................... 198 OVERVIEW OF STUDY RESULTS .................................................................................... 199 MANAGEMENT IMPLICATIONS ...................................................................................... 201 FUTURE RESEARCH ........................................................................................................... 207 LITERATURE CITED ........................................................................................................... 211 xii CHAPTER 1: INTRODUCTION Mallards (Anas platyrhynchos) are a generalist dabbling duck naturally distributed throughout the Holarctic (Johnsgard 1978). In North America, mallards primarily nested west of the Mississippi river prior to the early 1900s (Bent 1923). Mallard nesting density was greatest in the Prairie Pothole and Parkland regions and mallards mainly wintered in the Mississippi alluvial valley (Bellrose 1976). Widespread conversion of forest to agriculture in the eastern United States and Canada by the mid-1900s (Oswalt et al. 2019) increased landscape suitability for nesting mallards. Mallards expanded into eastern North America and likely outnumbered American black ducks (Anas rubripes) as the most abundant breeding duck in the Great Lakes region (Michigan, Minnesota, and Wisconsin) by the 1950s (Brewer et al. 1991, Chartier et al. 2013) and were the most-harvested duck in the Atlantic Flyway by 1969 (Huesmann 1974). In addition to breeding range expansion into eastern North America by wild mallards, domestic game-farm mallards, which had been domesticated from wild mallards in Eurasia (Lavretsky et al. 2020), were released in eastern North America beginning in the 1920s (Huesmann 1974, Hepp et al. 1988). Game-farm mallard releases since the 1920s were estimated between >200,000 (U. S. Fish and Wildlife Service 2013) and >500,000 (Huesmann 1974) annually. Wild mallard dispersal from the west and large-scale releases of domestic game-farm birds established mallards in eastern North America. MIDCONTINENT MALLARD MANAGEMENT In North America, mallards have been delineated into three populations for harvest management, which are defined by breeding geography and administrative flyway and include eastern, midcontinent, and western mallards (U. S. Fish and Wildlife Service 2023a). The midcontinent mallard population is the largest and comprises mallards nesting from the 1 Northwest Territories to the Great Lakes region (Figure 1.1). Midcontinent breeding mallard abundance estimates ranged from around 5.5 to 12 million since 1955 (U. S. Fish and Wildlife Service 2022). Beginning in 1995, annual duck hunting season frameworks for the Mississippi and Central flyways were established using Adaptive Harvest Management (AHM) of midcontinent mallards (U. S. Fish and Wildlife Service 2023a). In the AHM framework, regulatory alternatives are based on midcontinent mallard population size and pond (wetland) abundance estimated during the Waterfowl Breeding Population and Habitat Survey (WBPHS; U. S. Fish and Wildlife Service 2023b) with a goal of maximizing long-term sustainable harvest (U. S. Fish and Wildlife Service 2023a). Therefore, the midcontinent mallard population plays an important role in determining Mississippi and Central flyway duck hunting frameworks and considerable resources are devoted to population monitoring via aerial surveys, banding, and harvest surveys. GREAT LAKES MALLARDS The U. S. Fish and Wildlife Service (USFWS) added Great Lakes mallards (mallards surveyed during the breeding period in Michigan, Minnesota, and Wisconsin) to the midcontinent mallard population in 1997 under the assumption that Great Lakes mallard population dynamics were equivalent to those of mallards nesting in the U. S. and Canadian prairies (U. S. Fish and Wildlife Service 2023a). Great Lakes mallard abundance historically followed trends in the remainder of the midcontinent population. However, prairie-nesting mallard abundance increased following a decline in the early 2000s while Great Lakes mallard abundance continued to decline (Figure 1.2). Great Lakes region environmental conditions differ from those in prairie ecosystems and some evidence suggests drivers of Great Lakes mallard population dynamics differ than those for mallards nesting in the Prairie Pothole Region (Munro 2 and Kimball 1982, Coluccy et al. 2008). The Great Lakes influence regional temperature and precipitation (Scott and Huff 1996), and wetland hydrology is generally less seasonally dynamic than in the prairies due to a temperate climate and moderating lake effects (Euliss et al. 2004, Simpson et al. 2005). Additionally, harvest rates were generally greater for Mississippi than for Central Flyway mallards, suggesting different anthropogenic and ecological influences on Great Lakes mallard vital rates (Coluccy et al. 2008). Despite extensive annual monitoring and research, wildlife managers are unsure what factors have primarily contributed to declining Great Lakes mallard abundance. Declining mallard abundance is a concern for wildlife management agencies because mallards and other waterfowl have ecological (Ackerman 2002, Kleyheeg et al. 2019), social (NAWMP 2018), and economic (Carver 2013, 2015, Vrtiska et al. 2013) value. In the Mississippi Flyway and Great Lakes states, mallards are the most-harvested duck species (Raftovich et al. 2020). Locally produced mallards are particularly important for Great Lakes duck hunters with an estimated 58-83% of their mallard harvest derived from mallards hatched within the region (Arnold and De Sobrino 2010). Waterfowl hunters contribute substantial support for conservation (Vrtiska et al. 2013, Carver 2015) through duck stamp purchases and Pittman-Robertson excise taxes which are vital in funding habitat management and research to conserve wetland wildlife. Mallards are also an ecologically important species whose abundance is related to wetland quantity and quality (Soulliere et al. 2017, U. S. Fish and Wildlife Service 2022). As a generalist species, mallards use a variety of wetland types during breeding and the nonbreeding periods (Soulliere et al. 2017) and may serve as an indicator of wetland abundance and function relevant to other wetland wildlife. Therefore, recovering mallards is a priority for waterfowl managers in the Great Lakes region (Soulliere et al. 2017). 3 RESEARCH QUESTIONS, GOAL, AND OBJECTIVES Although midcontinent mallards are one of the most highly monitored wildlife populations in the world, Great Lakes mallards are less studied and reasons for population decline are unknown. Previous research (e.g., Simpson et al. 2007, Singer 2014, Singer et al. 2016, Boyer et al. 2018, Palumbo and Shirkey 2022) focused on demographic rates and their relationships to ecological factors and harvest somewhat independently. However, Coluccy et al. (2008) concluded that Great Lakes mallard population growth rate should be most influenced by female nonbreeding survival, ducking survival, and nest survival. Mallard hen and brood survival were inversely related to forest cover in the Great Lakes (Simpson et al. 2007, Boyer et al. 2018). While forest cover remained relatively constant in the upper Midwest over the last three decades (Oswalt et al. 2019), the Upper Mississippi/Great Lakes (UMGL) Joint Venture (JV) region lost an estimated 1.4 million ha of grassland/herbaceous and hay/pasture cover types between 2001 and 2016 (Yang et al. 2018, Soulliere et al. 2020). Upland nesting cover is an important factor in mallard nest success (Stephens et al. 2005, Bortolotti et al. 2022) and productivity (Specht and Arnold 2018). However, long term mallard productivity estimates appeared to be relatively stable in Michigan, Wisconsin, and Minnesota from 1961–2011 (Singer et al. 2016). Long-term adult female mallard survival estimates in Minnesota and Wisconsin were also stable (D. Fowler, Wisconsin Department of Natural Resources, unpublished) and harvest did not appear linked to reduced Great Lakes mallard abundance (Singer 2014). Further, recent work suggested female midcontinent mallards should have the capacity to at least partially compensate for harvest mortality through density dependence in reproduction and mortality (Riecke et al. 2022). Without a clear link between survival, productivity, and population decline, researchers and managers identified several questions that warranted further examination. 4 Mallard distribution may have changed at one or more spatial scales. Anecdotal observations by wildlife managers and researchers suggested the number of mallards utilizing urban areas increased in the early 2000s and some banding operations began to target these birds (D. Avers, Michigan Department of Natural Resources and B. O’Neal, Franklin College, personal communications). Aerial breeding waterfowl surveys are likely ineffective in detecting mallards using large urban areas, potentially resulting in abundance estimates that are biased low. Additionally, mallards using urban developed land cover during part of the annual cycle may have different demographic rates, possibly resulting in unmeasured individual heterogeneity in population parameters such as survival and productivity. In addition to changing use of land cover types, mallard breeding distribution could be shifting at a regional scale in response to habitat or climatic changes. Breeding period fidelity and hen mallard dispersal were relatively unstudied for Great Lakes mallards and these parameters are difficult to estimate using banding data when live recapture rates are low (Dooley et al. 2019). Capacity to estimate dispersal probability of hen mallards hatched in the Great Lakes to other breeding areas such as the Hudson Bay lowlands (Brook et al. 2021) or prairies was thus limited by available data. Assessing fine-scale and regional movements and fidelity are hence important to obtain a more complete picture of population processes. Moreover, recent evidence demonstrated introgression of domestic game-farm mallard genes into wild populations in Europe and eastern North America (Söderquist et al. 2017, Lavretsky et al. 2020). Mallards in the U. S. portion of the Atlantic Flyway were a hybrid swarm consisting of ~90% wild x game-farm mallard hybrids (Lavretsky et al. 2020). Hybridization with game-farm mallards may lead to maladaptive traits or behaviors, resulting in lower survival or fecundity for admixed individuals. Using early banding data, Lincoln (1934) suggested that 5 hand-reared domestic mallards released into the wild had lower survival and shorter dispersal distances than did wild mallards. More recent survival estimates have consistently been lower for hand-reared and domestic than for wild mallards (Brakhage 1953, Schladweiler and Tester 1972, Smith 1999, Osborne et al. 2010, Söderquist et al. 2013). Game-farm mallards had a longer breeding period and longer incubation time than wild mallards (Prince et al. 1970, Cheng et al. 1980), traits which may be detrimental in natural settings where nesting hens are exposed to predators. There is concern that releases of domestic mallards are contributing to declines in wild mallard populations (Söderquist et al. 2014, 2017, Lavretsky et al. 2020). Understanding behavioral and demographic consequences of hybridization between wild and domestic game- farm mallards is important given releases of game-farm mallards in eastern North America (Lavretsky et al. 2023). To address these questions and identify factors limiting Great Lakes mallards, a comprehensive view of mallard movements and population dynamics was needed. The research goal was to estimate hen mallard survival, productivity, resource selection, and fidelity to the Great Lakes region in relation to banding location, genotype, molt and natal location, and age to identify factors limiting abundance and develop management recommendations to recover Great Lakes mallards. Project objectives were to 1) identify the influence of resource selection, genotype, and age on hen mallard breeding and nonbreeding survival using GPS-GSM transmitter data and known-fate models; 2) assess the effects of nest site land cover, predicted suitability, genotype, age, and nest initiation date on nest success and productivity using GPS-GSM transmitter and banding data; 3) quantify differences in nonbreeding season selection of land cover and wetland types between hen mallards marked in urban and rural areas, wild and admixed genotypes, and after hatch year and hatch year birds using location data from GPS-GSM transmitters and remote-sensed spatial data; and 4) 6 determine the relative importance of urban and rural marking location, genotype, molt migration incidence, nest success, and age on probability of hen mallard fidelity to the Great Lakes region using GPS-GSM transmitter data and estimates of molt and natal origin derived from secondary feather stable isotope values. DISSERTATION CONTENT This dissertation is comprised of this introductory chapter, four research chapters intended for publication in peer reviewed journals, and a conclusion and management implications chapter. The research chapters include co-author contributions and therefore use plural pronouns; however, I take sole responsibility for their content in this dissertation. Chapter 2 examines movements and resource selection of mallards in relation to individual genotypes. Chapter 3 contains survival estimates for GPS-marked female mallards. Chapter 4 focuses on breeding period ecology including hen mallard incubation initiation, nest survival, and fidelity. Chapter 5 integrates banding, band recovery, and aerial survey data into an integrated population model for Great Lakes mallards. Chapter 6 summarizes the primary findings and management implications of the research. 7 FIGURES Figure 1.1. North American mallard breeding population survey areas. Data from U. S. Fish and Wildlife Service. 8 Figure 1.2. Midcontinent and Great Lakes (Michigan, Minnesota, and Wisconsin) breeding mallard abundance estimates (millions) and 95% confidence intervals (shaded region) from the waterfowl breeding population and habitat survey and Great Lakes state surveys, 1991–2023. Data are from the U. S. Fish and Wildlife Service, Michigan Department of Natural Resources, Minnesota Department of Natural Resources, and Wisconsin Department of Natura Resources. 9 LITERATURE CITED Ackerman, J. T. 2002. Of mice and mallards: Positive indirect effects of coexisting prey on waterfowl nest success. Oikos 99:469–480. Arnold, T. W., and C. De Sobrino. 2010. An updated derivation of mallard harvest for midwestern waterfowl hunters. Midwest Fish and Wildlife Conference. Indianapolis, Indiana, USA. Bellrose, F. C. 1976. Ducks, geese, and swans of North America. Second Ed. Stackpole Books, Harrisburg, PA. Bent, A. C. 1923. Life histories of North American wild fowl. Bulletin 126. Smithsonian Institution, Washington, D.C. Bortolotti, L. E., R. B. Emery, L. M. Armstrong, and D. W. Howerter. 2022. Landscape composition, climate variability, and their interaction drive waterfowl nest survival in the Canadian Prairies. Ecosphere 13:1–14. Boyer, R. A., J. M. Coluccy, R. A. Montgomery, K. M. Redilla, and S. R. Winterstein. 2018. The effect of habitat on the breeding season survival of Mallards (Anas platyrhynchos) in the Great Lakes region. Canadian Journal of Zoology 96:700–706. Brakhage, G. K. 1953. Migration and mortality of ducks hand-reared and wild-trapped at Delta, Manitoba. Journal of Wildlife Management 17:465–477. Brewer, R., G. A. McPeek, and R. J. Adams. 1991. The Atlas of Breeding Birds of Michigan. Michigan State University Press, East Lansing. Brook, R. W., L. A. Pollock, K. F. Abraham, and G. S. Brown. 2021. Bird trends from long-term observation data at sites in the hudson bay lowlands. Avian Conservation and Ecology 16. Carver, E. 2013. Birding in the United States: a demographic and economic analysis: Addendum to the 2011 National Survey of Fishing, Hunting, and Wildlife-Associated Recreation. Arlington, VA. Carver, E. 2015. Economic impact of waterfowl hunting in the United States: Addendum to the 2011 National Survey of Fishing, Hunting, and Wildlife-Associated Recreation. Falls Church, VA. . Chartier, A. T., J. J. Baldy, and J. Brenneman. 2013. Michigan Breeding Bird Atlas II. A. T. Chartier, J. J. Baldy, and J. Brenneman, editors. River Run Press, Kalamazoo, MI. Cheng, K. M., R. N. Shoffner, R. E. Phillips, and F. B. Lee. 1980. Reproductive performance in wild and game farm mallards. Poultry Science 59:1970–1976. Poultry Science 10 Association Inc. . Coluccy, J. M., T. Yerkes, R. Simpson, J. W. Simpson, L. Armstrong, and J. Davis. 2008. Population dynamics of breeding mallards in the Great Lakes states. Journal of Wildlife Management 72:1181–1187. Dooley, J. L., M. L. Szymanski, R. J. Murano, M. P. Vrtiska, T. F. Bidrowski, J. L. Richardson, and G. C. White. 2019. Age class dynamics of Canada geese in the central flyway. Journal of Wildlife Management 83:938–953. Euliss, N. H., J. W. Labaugh, L. H. Fredrickson, D. M. Mushet, M. K. Laubhan, G. A. Swanson, T. C. Winter, D. O. Rosenberry, and R. D. Nelson. 2004. The wetland continuum: A conceptual framework for interpreting biological studies. Wetlands 24:448–458. Hepp, G. R., J. M. Novak, K. T. Scribner, and P. W. Stangel. 1988. Genetic distance and hybridization of black ducks and mallards: A morph of a different color? The Auk 105:804–807. Huesmann, H. W. 1974. Mallard-black duck relationships in the northeast. Wildlife Society Bulletin 2:171–177. Johnsgard, P. A. 1978. Ducks, geese, and swans of the world. University of Nebraska Press, Lincoln, NE. Kleyheeg, E., W. Fiedler, K. Safi, J. Waldenström, M. Wikelski, and M. L. van Toor. 2019. A comprehensive model for the quantitative estimation of seed dispersal by migratory mallards. Frontiers in Ecology and Evolution 7:1–14. Lavretsky, P., N. R. McInerney, J. E. Mohl, J. I. Brown, H. F. James, K. G. McCracken, and R. C. Fleischer. 2020. Assessing changes in genomic divergence following a century of human-mediated secondary contact among wild and captive-bred ducks. Molecular Ecology 29:578–595. Lavretsky, P., J. E. Mohl, P. Söderquist, R. H. S. Kraus, M. L. Schummer, and J. I. Brown. 2023. The meaning of wild: Genetic and adaptive consequences from large-scale releases of domestic mallards. Communications biology 6:819. Springer US. Lincoln, F. C. 1934. Restocking of marshes with hand-reared mallards not proved practicable. Yearbook of Agriculture. Munro, R. E., and C. F. Kimball. 1982. Population Ecology of the Mallard: VII. Distribution and Derivation of the Harvest. Resource Publication 147. Laurel, Maryland, USA. . NAWMP. 2018. 2018 North American Waterfowl Management Plan (NAWMP) update : connecting people, waterfowl, and wetlands. 11 Osborne, C. E., B. L. Swift, and G. A. Baldassarre. 2010. Fate of captive-reared and released mallards on eastern Long Island, New York. Human-Wildlife Interactions 4:266–274. Oswalt, S. N., W. B. Smith, P. D. Miles, and S. A. Pugh. 2019. Forest resources of the United States. USDA Forest Service General Technical Report WO-97. Washington, D.C. USA. . Palumbo, M. D., and B. T. Shirkey. 2022. Harvest mortality of mallards (Anas platyrhynchos) banded at Lake St. Clair, CAN and western Lake Erie, USA. Waterbirds 45:51–61. Prince, H., P. B. Sieoel, A. N. I. George, and W. Cornwell. 1970. Inheritance of egg production and juvenile growth in mallards. The Auk 87:342–352. Raftovich, R. V., K. K. Fleming, S. C. Chandler, and C. M. Cain. 2020. Migratory bird hunting activity and harvest during the 2018 and 2019 and 2019-2020 hunting seasons. Laurel, Maryland, USA. Riecke, T. V., B. S. Sedinger, T. W. Arnold, D. Gibson, D. N. Koons, M. G. Lohman, M. Schaub, P. J. Williams, and J. S. Sedinger. 2022. A hierarchical model for jointly assessing ecological and anthropogenic impacts on animal demography. Journal of Animal Ecology 91:1612–1626. Schladweiler, J. L., and J. R. Tester. 1972. Survival and behavior of hand-reared mallards released in the wild. Journal of Wildlife Management 36:1118–1127. Scott, R. W., and F. A. Huff. 1996. Impacts of the Great Lakes on regional climate conditions. Journal of Great Lakes Research 22:845–863. Elsevier. . Simpson, J. W., T. J. Yerkes, B. D. Smith, and T. D. Nudds. 2005. Mallard duckling survival in the Great Lakes region. Condor 107:898–909. Simpson, J. W., T. Yerkes, T. D. Nudds, and B. D. Smith. 2007. Effects of habitat on mallard duckling survival in the Great Lakes region. Journal of Wildlife Management 71:1885– 1891. Singer, H. V. 2014. Factors affecting productivity and harvest rates of Great Lakes mallards. Michigan State University, East Lansing, Michigan. Singer, H. V., D. R. Luukkonen, L. M. Armstrong, and S. R. Winterstein. 2016. Influence of weather, wetland availability, and mallard abundance on productivity of Great Lakes mallards (Anas platyrhynchos). Wetlands 36:969–978. Wetlands. . Smith, D. B. 1999. Survival, behavior, and movements of captive-reared mallards released in 12 Dorchester County, Maryland. Louisiana State University. Söderquist, P., J. Elmberg, G. Gunnarsson, C. G. Thulin, J. Champagnon, M. Guillemain, J. Kreisinger, H. H. T. Prins, R. P. M. A. Crooijmans, and R. H. S. Kraus. 2017. Admixture between released and wild game birds: a changing genetic landscape in European mallards. European Journal of Wildlife Research 63. European Journal of Wildlife Research. Söderquist, P., G. Gunnarsson, and J. Elmberg. 2013. Longevity and migration distance differ between wild and hand-reared mallards in Northern Europe. European Journal of Wildlife Research 59:159–166. Söderquist, P., J. Norrström, J. Elmberg, M. Guillemain, and G. Gunnarsson. 2014. Wild mallards have more “goose-like” bills than their ancestors: A case of anthropogenic influence? PLoS ONE 9:1–14. Soulliere, G. J., M. A. Al-Saffar, J. M. Coluccy, R. J. Gates, H. M. Hagy, J. W. Simpson, J. N. Straub, R. L. Pierce, M. W. Eichholz, and D. R. Luukkonen. 2017. Upper Mississippi river and Great Lakes region joint venture waterfowl habitat conservation strategy. Bloomington, MN. Soulliere, G. J., M. A. Al-Saffar, K. R. VanBeek, C. M. Tonra, M. D. Nelson, D. N. Ewert, T. Will, W. E. Thogmartin, K. E. O’Brien, S. W. Kendrick, A. M. Gillet, J. R. Herkert, E. E. Gnass Giese, M. P. Ward, and S. Graff. 2020. Upper Mississippi / Great Lakes Joint Venture Landbird Habitat Conservation Strategy – 2020 Revision. Bloomington, MN. Specht, H. M., and T. W. Arnold. 2018. Banding age ratios reveal prairie waterfowl fecundity is affected by climate, density dependence and predator–prey dynamics. Journal of Applied Ecology 55:2854–2864. Stephens, S. E., J. J. Rotella, M. S. Lindberg, M. L. Taper, and J. K. Ringelman. 2005. Duck nest survival in the Missouri Coteau of North Dakota: Landscape effects at multiple spatial scales. Ecological Applications 15:2137–2149. U. S. Fish and Wildlife Service. 2013. Review of captive-reared mallard regulations on shooting preserves. Washington, D. C. U. S. Fish and Wildlife Service. 2022. Waterfowl population status, 2022. Washington, D.C. USA. U. S. Fish and Wildlife Service. 2023a. Adaptive Harvest Management: 2024 Hunting Season. US Dept. Interior, Washington, DC 54pp. …. Washington, D.C. USA. . U. S. Fish and Wildlife Service. 2023b. Waterfowl Population Status, 2023. Washington, D.C. 13 USA. Vrtiska, M. P., J. H. Gammonley, L. W. Naylor, and A. H. Raedeke. 2013. Economic and conservation ramifications from the decline of waterfowl hunters. Wildlife Society Bulletin 37:380–388. Yang, L., S. Jin, P. Danielson, C. Homer, L. Gass, S. M. Bender, A. Case, C. Costello, J. Dewitz, J. Fry, M. Funk, B. Granneman, G. C. Liknes, M. Rigge, and G. Xian. 2018. A new generation of the United States National Land Cover Database: Requirements, research priorities, design, and implementation strategies. ISPRS Journal of Photogrammetry and Remote Sensing 146:108–123. Elsevier. . 14 CHAPTER 2: GENETICS AFFECT GREAT LAKES MALLARD MOVEMENT AND RESOURCE SELECTION ABSTRACT Abundance data collected during spring aerial surveys suggested breeding mallard (Anas platyrhynchos) populations in the Great Lakes region (Michigan, Minnesota, and Wisconsin, USA) declined by >30% between 2000–2022. Understanding Great Lakes mallard movement ecology in the context of changing environmental and anthropogenic factors could reduce uncertainty in the drivers of declining mallard abundance. Recent detection of widespread introgression of domestic game-farm mallard genes into mallard populations in eastern North America raises concern about the role of hybridization in population declines. Our objective was to examine local and migratory movement and resource selection during the non-breeding period for female Great Lakes mallards in relation to genotype and urban development using GPS-GSM transmitters and remote-sensed spatial data. We captured, obtained a genetic sample, and marked 592 female mallards with GPS-GSM transmitters in Michigan, Wisconsin, Ohio, Indiana, and Illinois, USA from 2021–2023. We used ddRAD sequencing to estimate genotypes of GPS- marked mallards. We used linear mixed models and general linear models to examine factors affecting daily movement distance and autumn migration probability, respectively. We implemented a step-selection function and conditional logistic regression to estimate relative selection strength of land cover types relevant to mallard resource use. Genetic analyses indicated 44% of GPS-marked birds were wild mallards and 56% were wild x domestic game-farm mallard hybrids. Hybrid mallards had shorter daily movement distances, were less likely to migrate, and had higher selection for urban developed land cover than did wild mallards. Sedentary behavior and selection of urban areas raises concerns regarding the ecological fitness of hybrid mallards 15 and their impact on the regional mallard population. INTRODUCTION At the North American scale, mallards have been delineated into 3 populations for harvest management, defined by breeding geography and administrative flyway, including eastern, midcontinent, and western mallards (U. S. Fish and Wildlife Service 2023a). The midcontinent population is the largest and comprises mallards breeding from the Northwest Territories in Canada to the Great Lakes region. The Waterfowl Breeding Population and Habitat Survey (WBPHS) and Great Lakes state (Michigan, Minnesota, and Wisconsin) breeding-period waterfowl surveys are conducted annually to estimate spring waterfowl abundance by species (U. S. Fish and Wildlife Service 2023b). Trends in Great Lakes mallard abundance were historically similar to trends in the midcontinent population. However, prairie-nesting mallard abundance increased following a decline in the early 2000s while Great Lakes mallard abundance continued to decline (Figure 2.1). Breeding population size of Great Lakes mallards peaked near 1.2 million in the year 2000 based on state aerial surveys and declined to around 0.5 million in 2023 (U. S. Fish and Wildlife Service 2023a). Regionally produced mallards are particularly important to hunters in the Great Lakes region as an estimated 58-83% of their mallard harvest is derived from mallards produced within the region (Arnold and De Sobrino 2010). Declining Great Lakes mallard abundance is thus a concern for wildlife management agencies involved in hunting recreation as well as those implementing the North American Waterfowl Management Plan (Soulliere et al. 2017). Recent research revealed mallards in the U. S. Atlantic Flyway are a hybrid swarm comprised of approximately 90% wild x domestic game-farm mallard hybrids (Lavretsky et al. 2020, 2023). Eastern mallard breeding abundance declined from about 1.4 million in 1998 to 1.2 16 million in 2023 (U. S. Fish and Wildlife Service 2023b), with the decline driven by mallards occurring in the U. S. portion of the flyway (U. S. Fish and Wildlife Service 2017), and hybridization with game-farm mallards is hypothesized to be a contributing factor (Lavretsky et al. 2020, Roberts et al. 2023). Game-farm mallard releases since the 1920s were estimated between >200,000 (U. S. Fish and Wildlife Service 2013) and >500,000 (Huesmann 1974) annually, although precise estimates are lacking. Historically, game-farm mallards were released primarily in the Atlantic Flyway (Huesmann 1974, Hepp et al. 1988) although 17.5% of captive- reared mallard releases from licensed shooting preserves occurred in the Mississippi Flyway in 2001 (U. S. Fish and Wildlife Service 2013). The proportion of wild x game-farm mallard hybrids detected at a continental scale generally decreases westward, with relatively few hybrids detected west of the Mississippi River (Lavretsky et al. 2023). However, banding data revealed that since 1990, 23% of mallards banded in the U. S. portion of the Atlantic Flyway were recovered in the Mississippi Flyway, revealing a pathway for domestic mallard genes to move westward (Lavretsky and Sedinger 2023). While only 4% of mallards sampled in the southern Mississippi Flyway were hybrids (Davis et al. 2022), 65% of hunter-harvested mallards sampled in northwest Ohio were hybrids (Schummer et al. 2023). Large scale releases of game-farm mallards and declining mallard populations in eastern North America warrant consideration of the effects of hybridization on individual behaviors and population ecology. Release of artificially selected animals can cause changes in morphology, behavior, and demography that negatively impact wild populations (Champagnon et al. 2012). However, quantifying the population impact of hybridization between captive-bred and wild individuals can be difficult as the additive effects of introgression on fitness may be slow to develop (Tufto 2017). Large-scale, long-term releases of domestic mallards are predicted to negatively impact wild 17 populations if gene flow transfers inheritable traits possessed by domestic ducks that are maladaptive in free-ranging populations (Söderquist et al. 2017, Lavretsky et al. 2020). Gene flow could have maladaptive consequences if domestic traits confer morphological or behavioral characteristics which reduce survival or reproductive capacity for admixed individuals. Using early banding data, Lincoln (1934) posited that hand-reared domestic mallards released into the wild had lower dispersal distances than those of wild mallards. More recent studies found that domestic mallards were typically harvested within three km of release locations (Osborne et al. 2010) and had significantly shorter migration distances (Söderquist et al. 2013) and smaller home range areas (Smith 1999, Bengtsson et al. 2014) than observed in wild mallards (Yetter et al. 2018). Whereas mallards can be considered facultative migrants, migratory decisions have been linked to weather cues that affect energy expenditure and food availability (Schummer et al. 2010, Weller et al. 2022), which have likely evolved to enable selection of areas that enhance survival. If hybridization with game-farm mallards diminishes migration and or the ability to select vital resources, survival could be reduced. Thus, quantifying the effects of hybridization on local movement, migration, and resource selection can aid understanding of factors affecting Great Lakes mallard population dynamics. Our objective was to examine local and migratory movement and non-breeding period resource selection of hen mallards in relation to their genotype and association to urban development using GPS-GSM transmitters and remotely sensed spatial data. We predicted that hybrid (wild x domestic game-farm) mallards would be more sedentary and less likely to migrate than wild mallards, but that selection of land cover types would be similar for wild and hybrid mallards. Evaluating whether movement and resource selection differs between wild and hybrid mallards would aid understanding of potential consequences of hybridization. Quantifying 18 differences in individual behaviors is important in developing models that examine factors affecting demographic rates and ultimately drivers of population dynamics. In addition to informing Great Lakes mallard population modelling, movement and resource selection information is useful in prioritizing habitat conservation and managing harvest at local and flyway scales. STUDY AREA We captured and marked female mallards with GPS-GSM transmitters in Bird Conservation Regions (BCRs; Bird Studies Canada and NABCI 2014) 12 (Boreal Hardwood Transition), 22 (Eastern Tallgrass Prairie), and 23 (Prairie Hardwood Transition) in the Great Lakes states of Michigan, Wisconsin, Ohio, Indiana, and Illinois, USA (hereafter, Great Lakes region; Figure 2.2). Bird Conservation Regions are landscape planning units comprised of similar ecosystems and bird communities and are relevant for landscape conservation planning in the Upper Mississippi/Great Lakes (UMGL) Joint Venture (JV) region (Soulliere and Al-Saffar 2021). The northern UMGL JV is dominated by undeveloped natural communities with extensive upland and wetland forest and lakes (BCR 12) that transition to a mixed landscape of forests, lakes, and more herbaceous wetlands interspersed with agriculture and urban development (BCR 23). The south half of the JV region (BCR 22) is dominated by row crop agriculture and urban landscapes, with smaller proportion coverage in forest and grassland communities; about 90% of historic wetlands have been drained and most remaining wetlands in BCR 22 are associated with riverine systems (Soulliere et al. 2017). Regional temperatures and precipitation across much of the study area are influenced by the Great Lakes and generally consist of cold, snowy winters and hot, humid summers (Scott and Huff 1996). Moderating lake effects result in wetland conditions that are generally more stable and less seasonally dynamic than in the midcontinent prairies and parklands 19 (Euliss et al. 2004, Simpson et al. 2005). Mallards were the most abundant nesting duck in the Great Lakes region (U. S. Fish and Wildlife Service 2023b) and predicted nesting densities were greatest in BCR 23 (Soulliere et al. 2017). We monitored movements of GPS-marked mallards within North America, which occurred in an area bounded by approximately 51.5⁰N and 32⁰N, and 75.2⁰W, and 99.8⁰W. METHODS Mallard Capture and Data Collection We captured ducks using baited traps, rocket- and spring-propelled nets, handheld nets, and via night-lighting from 4 March – 4 October, 2021–2023. We aged ducks as AHY (after hatch year; adult), HY (hatch year; juvenile capable of flight) or L (local; juvenile incapable of flight) and sexed birds via plumage characteristics (Carney 1992). We banded mallards with a size 7 United States Geological Survey (USGS) aluminum leg band. We attached 20 g Ornitela GPS-GSM (global system for mobile communications) transmitters (OrniTrack-E20 4GCT C48; Ornitela, Vilnius, Lithuania) dorsally to female mallards via 4.5 mm-wide elastic straps. Transmitters were attached with two separate elastic loops in 2021 and via an X-shaped design consisting of a single piece of elastic in 2022 and 2023. We weighed each hen mallard to the nearest 10 g and attached transmitters only to individuals >700 g so transmitters comprised <3% of body mass (mean = 1.9%). Transmitters were distributed in approximate proportion to estimated breeding period mallard abundance by BCR (Soulliere et al. 2017). We classified transmitter deployment locations as urban if the proportion of developed land cover (low, medium, or high intensity developed, or developed open space as classified by the 2021 National Land Cover Database [Dewitz 2023]) within a 7 km (mean of maximum daily net displacement) radius of the site was >0.5, or rural if ≤0.5. We measured total head, bill (culmen), total tarsus, and wing chord (Dzubin and Cooch 20 1992) of GPS-marked females. We drew approximately 0.1 ml of blood from the tarsal vein and clipped approximately 3 mm of the first secondary wing feather of one wing for genetic and isotope analyses, respectively. Ducks were released immediately after processing at capture locations. Approval to capture, band, and attach transmitters was provided by Michigan State University institutional animal care and use committee (IACUC) permit PROTO202100046 and USGS Bird Banding Laboratory permit 03110. Transmitters were programmed to record a location every 30 minutes, 2 hours, or 4 hours when battery charge was >50%, <50% and >25% , or <25%, respectively, and uploaded data once every 24 hours when connected to cell networks. Transmitter data were forwarded to Movebank (Wikelski and Kays 2019) for storage. We performed data preparation and analyses in Program R (R Core Team 2023) and used the move package (Kranstauber et al. 2018) to retrieve GPS data from Movebank. We defined the non- breeding period as 16 August to 29 February based on Coluccy et al. (2008) and the earliest date of nest incubation observed in this study (13 March) and included data collected during this timeframe from 2021 to 2024 in movement and resource selection analyses. We censored GPS locations where a satellite fix was not obtained or only one satellite was successfully contacted, and/or locations with a horizontal dilution of precision (HDOP) <5 (D’Eon and Delparte 2005). We monitored individuals from marking to reported harvest by a waterfowl hunter, mortality or transmitter loss indicated by transmitter temperature and accelerometer data and verified by field observation when possible, or transmitter failure (transmitter stopped sending data with no indication of mortality). We excluded individuals from analyses that provided GPS data for <7 days after release. DNA Extraction, Sequencing, and Genetic Ancestry Analyses We extracted genomic DNA from blood samples using a DNeasy Blood & Tissue kit following 21 the manufacturer’s protocols (Qiagen, Valencia, CA, USA). DNA quality was visually assessed on a 1% agarose gel to ensure high molecular weight bands. For mitochondrial DNA (mtDNA), we polymerase chain reaction (PCR) amplified and Sanger sequenced the control region across samples using primers L78 and H774 (Sorenson and Fleischer 1996, Johnson and Sorenson 1999), and following protocols outlined in (Lavretsky et al. 2014b). Final products were sequenced on an ABI 3730 (Applied Biosystems, Life Technologies, Carlsbad, California, USA) machine at the University of Texas at El Paso Border Biomedical Research Centers (BBRC) Genomic Analysis Core Facility. Sequences were then aligned and edited using SEQUENCHER v. 4.8 (Gene Codes Corporation, Ann Arbor, MI, USA). We note that mtDNA control region sequences for reference wild and domestic mallards from previous studies were included in the analyses (Lavretsky et al. 2014a, b, 2019a, Lavretsky 2020a). Reference samples included known wild, domestic game-farm, and domestic Khaki Campbell mallards. We constructed a median- joining haplotype network to visualize mtDNA structure as calculated in the program POPART (Leigh and Bryant 2015). Mallards are characterized by the old world (OW) A and new world (NW) B mitochondrial (mtDNA) haplogroups, which distinguish individuals of Eurasian or North American descent, respectively (Ankney et al. 1986, Avise et al. 1990, Lavretsky et al. 2014a). Importantly, being of Eurasian descent, all domestically-derived mallards carry OW A haplotypes, and thus, are a distinguishing marker when assessing whether game-farm mallard introgression occurred within a wild mallard lineage in North America (Lavretsky 2020b, Lavretsky et al. 2020). For nuclear DNA, we followed ddRAD-seq (double-digest restriction site associated DNA sequencing) library protocols outlined in DaCosta and Sorenson (2014) and Lavretsky et al. (2015). In short, genomic DNA was enzymatically fragmented using SbfI and EcoRI restriction 22 enzymes, and Illumina TruSeq compatible barcodes ligated for future de-multiplexing. The barcode-ligated fragments were then size selected using optimized double-sided bead selection protocols (Hernández et al. 2021). Libraries were then quantified with a Qubit 3 Flourometer (Invitrogen, Carlsbad, CA, USA) and pooled in equimolar amounts and sent to Novogenetics LTD (Sacramento, California, USA) for 150 base-pair, single-end chemistry sequencing on an Illumina HiSeq X. Raw Illumina reads were de-multiplexed using the ddRADparser.py script of the BU ddRAD-seq pipeline (DaCosta and Sorenson 2014) based on perfect barcode/index matches. Comparable sequences from previously published wild and domestic mallards were included and served as respective references (Lavretsky et al. 2014a, b, 2019b, 2020). For each sample, we first trimmed or discarded sequences of poor quality using TRIMMOMATIC (Bolger et al. 2014), and then remaining quality reads aligned to a chromosomal-level reference wild mallard genome (Lavretsky et al. in press) using the BURROWS WHEELER ALIGNER v. 07.15 (Li and Durbin 2011). Samples were then sorted and indexed in SAMTOOLS v. 1.7 (Bolger et al. 2014) and combined using the mpileup function with the following parameters “-c –A -Q 30 -q 30.” All steps through mpileup were automated using a custom Python script (Lavretsky et al. 2020). Next, we used VCFTOOLS v. 0.1.15 (Danecek et al. 2011) to filter variant call format (VCF) files for any base-pair missing >10% of samples that also included a minimum base-pair depth of 5X (i.e., 10X per genotype) and quality per base PHRED scores of ≥30. All nuclear population structure was based on independent bi-allelic ddRAD-seq autosomal single nucleotide polymorphisms (SNPs), and without using a priori assignment of individuals to populations or species. The final dataset was obtained by using VCFTOOLS (Danecek et al. 2011) to first extract bi-allelic SNPs, and then PLINK v.1.9 (Purcell et al. 2007) to filter for singletons (minimum allele frequency: 0.0014), any SNP missing ≥10% of data across 23 samples, as well as any SNPs found to be in linkage disequilibrium (LD). We randomly excluded all but one SNP for any positions found to be in significant LD (r2 > 0.5). Population structure was first visualized using a principal components analysis (PCA) as implemented in PLINK v.1.9 (Purcell et al. 2007). Next, the program ADMIXTURE 1.3 (Alexander et al. 2013) was used to attain per sample maximum likelihood estimates of population assignments for each individual, with datasets formatted for the ADMIXTURE analyses using PLINK v.1.9 (Purcell et al. 2007), and following steps outlined in Alexander and Lange (2011). ADMIXTURE analyses were run for population models of K of 2 and 3 with a 10-fold cross validation, incorporating a quasi- Newton algorithm to accelerate convergence (Zhou et al. 2011). Each analysis used a block relaxation algorithm for point estimation and terminated once the change in the log-likelihood of the point estimations increased by <0.0001. Finally, standard deviations around each point estimate were calculated based on 1,000 bootstrap replicates. Ancestry assignments and their standard deviation were used to recategorize samples as feral game-farm, wild mallard, and to filial classes of hybrids (Schummer et al. 2023) under expected genotypes in generational backcrosses and uncertainty on assignment probabilities. Thus, we classified individuals with ≥ 0.92 wild assignment probability as wild and all others as hybrids. Assignment probabilities are also interpretable as an estimate of the proportion of an individual’s genes that are of wild ancestry, and thus we also considered proportion wild genome as a continuous covariate in movement analyses. Movement We modeled the mean sum of step lengths (i.e., movement distance) in each 24-hour period during non-migratory movements in relation to the estimated proportion of wild mallard genes within individuals using linear mixed models in the lme4 package (Bates et al. 2015). First, we 24 inspected the sampling rate of GPS fixes and excluded 6 individuals with a median fix rate >30 minutes (highest programmed location frequency) because estimation of step lengths assumes the length of time between pairs of consecutive locations is equal. We resampled the remaining data to a location every 30 minutes to ensure time between each location was consistent and only used successive locations to calculate step lengths. We calculated step lengths as the linear distance between consecutive GPS locations using the package amt (Signer et al. 2019). Step lengths provide an estimate of movement distance over a given time period and don’t require the assumption that individuals are range resident (plots of variance in position over time reach an asymptote, indicating the amount of space used eventually becomes constant), an assumption needed to estimate home range area (Noonan et al. 2019). We excluded step lengths collected during autumn or spring migration that were >30 km when bird trajectories were directed in a northward or southward direction. Thus, the daily sum of step lengths provided an index of total non-migratory distance moved in each 24-hour period and facilitated comparison of local daily movement distances. Inspection of the distribution of step length sums and QQ plots suggested a log transformation of daily movement distance was appropriate because residuals were not normally distributed. We modeled log-transformed daily step length sum (daily movement distance) as a function of proportion wild genotype with a random intercept of mallard ID to account for random individual variation. We included linear and quadratic effects of proportion wild genome as covariates on daily movement distance, resulting in three candidate models. We ranked models using Akaike’s Information Criterion (Burnham and Anderson 2002) adjusted for small sample size (AIC(cid:2913)) using the package MuMIn (Barton 2019). We calculated predicted daily movement distances from the top-ranked model treating proportion wild genome as a continuous variable. To predict how backcrossing of hybrid with wild mallards influences daily movement 25 distance, we also fit a model treating genotype as a discrete predictor where the generational categories of feral, first generation (F1),