STUDYING MIGRANT RAPTORS USING THE ATLANTIC FLYWAY. BLOCK ISLAND RAPTOR RESEARCH STATION, BLOCK ISLAND, RI: 2017 SEASON.

Transcription

1 STUDYING MIGRANT RAPTORS USING THE ATLANTIC FLYWAY. BLOCK ISLAND RAPTOR RESEARCH STATION, BLOCK ISLAND, RI: 2017 SEASON.

2 STUDYING MIGRANT RAPTORS USING THE ATLANTIC FLYWAY. BLOCK ISLAND RAPTOR RESEARCH STATION, BLOCK ISLAND, RI: 2017 SEASON. WILDLIFE SCIENCE CHANGING OUR WORLD SUBMITTED TO: The Bailey Wildlife Foundation Cambridge, MA and Scott Comings Associate State Director The Nature Conservancy P.O. Box High Street Block Island, RI SUBMITTED BY: Chris DeSorbo Director, Raptor Program Biodiversity Research Institute 276 Canco Road Portland, ME SUBMITTED ON: March 25, 2018 Biodiversity Research Institute Page ii

3 Biodiversity Research Institute (BRI) is a 501(c)(3) non-profit organization located in Portland, Maine. Founded in 1998, BRI's mission is to assess emerging threats to wildlife and ecosystems through collaborative research, and to use scientific findings to advance environmental awareness and inform decision makers. To obtain copies of this report contact: Biodiversity Research Institute 276 Canco Road Portland, ME USA (207) chris.desorbo@briloon.org FRONT PHOTO: Merlin with satellite transmitter. Photo credit: Lauren Gilpatrick/BRI SUGGESTED CITATION: DeSorbo, C. R., C. Persico and L. Gilpatrick Studying Migrant Raptors Using the Atlantic Flyway. Block Island Raptor Research Station, Block Island, RI: 2017 Season. BRI Report # submitted to The Nature Conservancy, Block Island, Rhode Island, and The Bailey Wildlife Foundation, Cambridge, Massachusetts. Biodiversity Research Institute, Portland, Maine. 35 pp. Biodiversity Research Institute Page iii

4 Table of Contents LIST OF FIGURES... V LIST OF TABLES... VI 1.0 EXECUTIVE SUMMARY INTRODUCTION STUDY AREA OBJECTIVES METHODS Research Station Establishment Raptor Banding and Sampling Instrumenting Raptors with Transmitters RESULTS AND DISCUSSION Intensity, Species Composition and Timing Migratory Connectivity: Migration Routes, Wintering Areas, Origins Evaluating Mercury Exposure in Migrant Raptors sampled on Block Island, RI Education, Research and Presentations LITERATURE CITED Biodiversity Research Institute Page iv

5 LIST OF FIGURES Figure 1. Removing a Merlin from a dho-gaza net Figure 2. Banding a Peregrine Falcon using a lock-on band (left), and banding a Merlin with a butt-end band (right) Figure 3. Measuring wing chord of a Peregrine Falcon (left) and examining wing molt on an adult female Merlin (right)... 4 Figure 4. Example of 12g solar PTT unit and 22g solar GPS PTT unit used to track migratory movements of Peregrine Falcons (left); a hooded Peregrine Falcon being fitted with a transmitter (right) Figure 5. Female hatching year Merlin fitted with a 5g satellite transmitter (left), male hatching year Peregrine Falcon (right) fitted with a 12g solar satellite transmitter at BIRRS Figure 6. Male hatching year Northern Harrier fitted with a 5g satellite transmitter in 2014 (left), female hatching year Northern Harrier fitted with a 5g satellite transmitter in 2017 (right) at BIRRS Figure 7. Location of automated telemetry arrays along the Atlantic coast relative to the Block Island Raptor Research Station (BIRRS) during 2014 and Figure 8. Female hatching year Merlin instrumented with backpack-style digitally encoded VHF radio tag Figure 9. Total number of diurnal raptors trapped by species and year at BIRRS, Figure 10. Total species composition of diurnal raptors captured at BIRRS, (n = 872) Figure 11. Average number of Peregrine Falcons captured by date, Figure 12. Average number of Merlins captured by date, Figure 13. Fall migration routes of Peregrine Falcons instrumented with satellite transmitters at BIRRS, Figure 14. Fall migratory movement of Peregrine Falcon HYM02 after departure from BIRRS, as indicated by satellite telemetry Figure 15. Consecutive annual fall migration routes of seven Peregrine Falcons instrumented with satellite transmitters at BIRRS Figure 16. Spring migration routes of eight female and three male Peregrine Falcons tracked using satellite telemetry Figure 17. Fall migration paths of ten hatching year female and one adult female Merlin tracked from BIRRS, using satellite telemetry Figure 18. Fall and spring migration paths of an adult female Merlin tracked from BIRRS, using satellite telemetry Figure 19. Southward Movements of eight (of 30) Merlins instrumented with digitally encoded VHF tags at BIRRS in 2015, as indicated by detections at automated telemetry arrays along the Atlantic coast Figure 20. Easterly Merlin movements detected by automated telemetry arrays during fall These six Merlins were not detected by arrays other than those displayed above Figure 21. Westerly movements of Merlins detected by automated telemetry arrays during fall These six Merlins were not detected at arrays other than those displayed above Figure 22. Fall movements of a hatching year male (2014) and a hatching year female (2017) Northern Harrier tracked from BIRRS, using satellite telemetry Biodiversity Research Institute Page v

6 Figure 23. Breast feather Mercury concentrations (µg/g) in eight species of hatching year migrant raptors sampled at BIRRS, Figure 24. Blood Mercury concentrations (µg/g) in six species of hatching year migrant raptors sampled at BIRRS, Figure 25. Sunset at BIRRS LIST OF TABLES Table 1. Annual species composition of diurnal raptors captured at BIRRS, Table 2. Wintering areas for 25 fall migrant Peregrine Falcons tracked from Block Island, RI (n = 23; ) and Monhegan Island (n = 2; 2010) using satellite telemetry Table 3. Last known locations of satellite transmitter instrumented Peregrine Falcons with presumed incomplete fall migration paths ( ) Biodiversity Research Institute Page vi

7 1.0 EXECUTIVE SUMMARY During fall seasons of , we operated the Block Island Raptor Research Station (BIRRS) on Block Island, Rhode Island to facilitate several ongoing research investigations on migrant raptors. The objectives of this study were to: (1) characterize the timing, intensity and species composition of the raptor migration at Block Island, (2) determine migration routes, stopover locations, overwintering locations and presumed origins of migrant raptors using Block Island during migration and (3) establish baseline mercury exposure levels in migrant raptors and make species comparisons. 1. We captured 872 individuals of eight different raptor species on Block Island throughout the fall seasons, Over three-quarters of raptors captured annually were falcons. In general, Merlins comprised roughly one-half (51%) of total captures, while Peregrine Falcons (peregrine hereafter) comprised slightly more than one-quarter (27%) of total captures annually. Sharpshinned Hawks, Cooper s Hawks, Northern Harriers, American Kestrels, and a single Red-tailed Hawk and a single Northern Goshawk comprised the remaining quarter (23%) of total captures. A notable increase in peregrine captures marked the 2014 season such that peregrines approached nearly half of all individuals captured during that year. Like most raptor migration research stations, the majority (97%) of raptors captured on Block Island were of the hatching year age class (Table 1, Figure 9 & Figure 10; pgs.11-12). 2. Captures on Block Island generally reflected known migration timing for Peregrine Falcon and Merlins, with the bulk of the peregrine captures occurring during the first 7-14 days of October. The Merlin migration on Block Island was more protracted compared to Peregrine Falcons, with the annual peak ranging from 17 September to 1 October (Figure 11 & Figure 12, pg.13). 3. We instrumented 33 peregrines with satellite transmitters on Block Island to learn about migration routes, stopover sites, wintering, and breeding areas. Combined with two peregrines tracked from Monhegan Island, Maine, these peregrines have generated the most extensive dataset available on peregrine migration in the Atlantic flyway (Figure 13, pg.15). 4. Peregrines wintered extensively in the Caribbean, as well as Central and South America. Preliminary data from our study suggests male peregrines may migrate further south compared to females (Table 2 & Figure 13, pgs. 15&16). 5. Twelve migration tracks generated by 11 satellite-tagged peregrines provided insights on spring migration routes and presumed/possible origins of fall migrants using the Atlantic flyway. Potential origins spanned from central Saskatchewan to Greenland. Satellite-tracked peregrines used little of the Atlantic Flyway during northward migration. Eight peregrines traveled through the central U.S. west of the Great Lakes, two traveled north from the mid-atlantic U.S., and one individual migrated through eastern Quebec and Nunavut to Greenland after overwintering on Block Island (Figure 16, p. 20). 6. This study is the first in which Merlins have been tracked using satellite telemetry. Eleven satellite-tagged Merlins (10 hatching year birds and 1 adult) provided information on fall migration routes along the Atlantic flyway. Merlins used both coastal and offshore habitats during migration. The adult Merlin instrumented overwintered in Puerto Rico and provided information on spring migration routes. We presume this Merlin originated from eastern Quebec or possibly Labrador (Figure 17 & Figure 18, pgs ). 7. This is the first study in which digitally encoded VHF radio transmitters in conjunction with automated telemetry arrays were used to study the migratory habits of Merlins. We fitted 80 Biodiversity Research Institute Page 1

8 Merlins with digitally encoded VHF radio tags during Of all tagged Merlins, 94% (75 of 80) were detected by 1 telemetry array, and 60% (48 of 80 tags) during fall migration. In addition to detections at Block Island, Merlins instrumented at BIRRS were detected at various arrays along the Atlantic coastline from Monomoy, MA to Fisherman Island, VA (the southernmost tower in fall 2015). In 2014, the majority of instrumented Merlins traveled southward; however, directions of 2015 movements were highly variable, with more localized detections to the east and west of Block Island (Figure 19, Figure 20 & Figure 21, pgs.25-27). 8. In a pilot effort, we fitted two hatching year Northern Harriers with satellite transmitters. Individuals in this small sample suggest Northern Harriers using Block Island in the fall maintain relatively local/regional movements versus long-distance movements observed in other migrant species. A hatching year male migrated north to Connecticut prior to spending nearly six weeks north of Buzzard s Bay in the Cape Cod region of MA, while a hatching year female migrated west to Long Island, NY. Limited data suggests hatching year Northern harriers may have very low survival rates in the fall (Figure 22, pg.28). 9. To date we have evaluated mercury (Hg) exposure in eight species of migrant raptors captured on Block Island. Overall patterns of Hg exposure among species were similar in blood and feather indices. Feather Hg concentrations were higher in Sharp-shinned Hawks and Peregrine Falcons compared to other species. Blood Hg concentrations were higher in Merlins compared to all other species. Adverse effect thresholds for Hg are not well-established for species sampled in this study; however, some individuals exceeded levels associated with reproductive effects in other species. Raptor sampling provided perspectives on Hg exposure in both natal areas and during migration (Figure 23 & Figure 24, pgs.30-31). 10. Our efforts on Block Island indicate that BIRRS is uniquely suited to address data gaps important to numerous conservation and management issues. The location of the station in the northern portion of the U.S. Atlantic Flyway (compared to numerous such stations in the mid-atlantic U.S. for example) is important because birds captured at northern latitudes provide information on habitat use along the largest portion of the Atlantic Flyway possible. Such information has become important in efforts to understand exposure risk of migrant raptors such as peregrines and others relative to proposed offshore wind energy areas along the Atlantic U.S. coast. Tracking data has additional relevance to conservation efforts in raptor breeding and non-breeding /wintering areas spanning from the arctic to South America. 11. The Block Island Raptor Research Station has provided exceptional opportunities for conducting environmental education and outreach programs. During the seasons, BRI and TNC partnered to conduct educational outreach programs focusing on the ecology and natural history of raptors, migration, contaminant exposure, and other environmental issues. Target audiences were of all ages including school groups, conservation professionals, ecotourism groups and government decision-makers. We will continue to report on project findings at professional scientific conferences, and to promote conservation through science and education. Biodiversity Research Institute Page 2

9 2.0 INTRODUCTION Migratory birds, particularly songbirds, shorebirds, and raptors, travel thousands of miles twice a year between breeding areas and wintering areas. Some birds travel en masse in spectacular migration events, while others travel inconspicuously on an individual basis. Many bird species use relatively well-documented continental-scale migration corridors, or flyways, to travel between breeding and wintering areas. The Atlantic Flyway along the eastern seaboard of the U.S. represents a major migratory throughway for tens of thousands of birds originating from arctic and temperate portions of eastern and central North America and elsewhere. Even species originating from Greenland, such as tundrius Peregrine Falcons, cross the Atlantic at high latitudes and use the Atlantic Flyway to reach wintering areas in the Caribbean or Central and South America. Information documenting the migratory habits of birds is important in developing effective conservation strategies. Conservation biologists recognize that many bird populations face threats in their breeding areas, wintering areas, and during migration, and that effective conservation measures can protect populations at each stage of their annual life cycle. In many species, flight routes used by many species remain poorly charted, links between populations in breeding areas and wintering areas remain poorly established, and important stopover habitats used to rest and refuel during migration are poorly documented. These notable data gaps stem from significant challenges in collecting information about birds that travel vast distances through remote areas annually. Notable developments in animal tracking technologies are rapidly advancing our understanding of the migratory ecology of birds. While traditional research approaches relied upon banding and opportunistic band encounters that occurred infrequently (i.e., 1-3% of birds banded were encountered), tracking technologies now offer the possibility of documenting daily even hourly movement patterns of individual birds continuously over time. Animal tracking data are increasingly being used by conservation biologists and regulators to fill-in substantial data gaps in our understanding of species needed to inform management and conservation decisions. Animal tracking data is routinely used to map migration routes, identify wintering and stopover habitats, and to inform wildlife risk assessments. The ability to use tracking data toward multiple and widely differing conservation and research applications quickly justifies the high cost of some types of transmitters. In 2009, BRI began searching for a suitable site from which migrant raptors using the Atlantic Flyway could be captured and studied to aid in migration studies, contaminant exposure assessments, and offshore wind energy wildlife risk evaluations. Relatively few sites exist in which raptors using coastal migration routes can be captured efficiently along the Atlantic coast; however, the majority were located in the Mid-Atlantic region (i.e., Cape May, NJ, Assateague Island, MD/VA, Kiptopeke, VA). Raptor research stations located in the northern portion of the Atlantic flyway are especially needed to learn about movements of raptors relative to offshore wind energy facilities because individuals need to be captured well before they encounter proposed facilities. Biodiversity Research Institute Page 1

10 Over the last decade, the need for a raptor research station north of the Mid-Atlantic U.S. states has become evident as offshore wind energy facilities are being built or proposed in state (< 3 nautical mi from shore) and federal (>3 nautical mi from shore) waters throughout the Atlantic Flyway (BOEM 2015), and the risks these facilities pose to migrating birds remain unknown. While raptors are a primary bird group associated with collision risks at terrestrial-based wind energy projects, they are generally overlooked when considering potential impacts of wind energy projects in offshore settings. Peregrine Falcons, Merlins, and several other species capable of enduring lengthy over-water flight are commonly encountered offshore. With an increasing number of offshore wind energy facilities being proposed along the Atlantic flyway, further evaluations of the degree to which migrating raptors might encounter or collide with offshore wind energy facilities are warranted. With support from a grant from the U.S. Department of Energy (DOE), BRI and numerous collaborators initiated a study in 2012 to estimate wildlife densities, distribution and movement patterns along the Mid-Atlantic U.S. coast, particularly in offshore waters containing Delaware, Maryland and Virginia wind energy areas. In 2012, BRI established the first raptor research station on Block Island, RI, the Block Island Raptor Research Station (BIRRS), as part of the DOE study. Block Island was strategically attractive for this study because it was situated north of many proposed offshore wind energy areas along the Atlantic Flyway; however, none had previously attempted to catch migrant raptors there. BRI s efforts on Block Island revealed that BIRRS was not only an optimal location in which migrating Peregrine Falcons could be studied, but that it also offered unique opportunities to learn about other raptors such as Merlins, Northern Harriers, and others. With the conclusion of support from the DOE grant during seasons, BRI has continued to use BIRRS to band, track, and sample raptors to learn about their ecology and to inform conservation and management decisions. In 2014, BRI partnered with The Nature Conservancy (TNC) and researchers at the University of Rhode Island (URI) to strategize how to continue operation of BIRRS. Support from The Bailey Wildlife Foundation, TNC, the Overlook Foundation, the Ocean View Foundation, the Bluestone Foundation, and others has enabled BRI to continue operating BIRRS during the fall seasons. To date we have deployed transmitters on 33 Peregrine Falcons. Summaries of peregrine movement data for seasons indicated that peregrines commonly used the offshore environment, and that use (vertical and horizontal) of the Mid-Atlantic Study Area containing Delaware, Maryland and Virginia WEAs was highly variable. Findings of the peregrine study component and other related efforts were released in October 2015 and can be found on BRI s webpage at: (DeSorbo et al. 2015). Starting in 2014, we initiated further research efforts on BIRRS, including: (1) evaluating mercury exposure in migrant raptors, (2) expanding satellite tracking efforts to include Merlins and Northern Harriers, (3) evaluating migratory and stopover habits of Merlins using automated telemetry (often referred to as NanoTags ) to learn about migratory and stopover habits of Merlins. Additionally, we substantially increased educational outreach efforts. In this report, we summarize cumulative findings from efforts at BIRRS and continue to reflect upon its value in conservation, outreach and raptor research. Biodiversity Research Institute Page 2

11 3.0 STUDY AREA Block Island is located 13 miles (21 km) south of Point Judith, Rhode Island and 14 miles (23 km) due east from Montauk, NY. The island is approximately 9.7 square miles (52.2 km²). Just over 46 percent of the island is protected from development. The northern three-quarters of coastline consist of sand dunes and beaches broken up by low clay-based bluffs. Coastlines in the southern portion of the island are characterized by high clay-based bluffs and extensive sand beaches. The Block Island Raptor Research Station is situated on private conservation land in the southwest corner of Block Island. The banding station was surrounded from the south and west by pasture land sectioned by hedgerows, and to the north and east small open spaces are interspersed with groups of conifers. 4.0 OBJECTIVES The objectives of this study are to: 1. Characterize the timing, intensity, and species composition of the raptor migration at Block Island. 2. Determine raptor migration routes, overwintering locations, migratory stopover areas, and presumed origins of migrant raptors captured on Block Island during migration. 3. Establish baseline Hg exposure levels in Block Island s migrant raptors and make species comparisons. 5.0 METHODS 5.1 Research Station Establishment In fall 2011, BRI explored Block Island to find a location suitable for establishing a raptor research station. Many factors were considered, including property ownership, public use, habitat type, topography, and location relative to the presumed flightpath of fall migrant raptors. With these considerations in mind, we selected a privately owned property in the southwestern portion of Block Island. The station was generally modeled after similar well-established stations elsewhere (i.e, Cape May, NJ, Monhegan Island, BRI; DeSorbo et al. 2012) which rely upon using lures to attract raptors to the station and then capturing them using a combination of mist nets, bow nets, and dho-gaza nets (Figure 1). Figure 1. Removing a Merlin from a dho-gaza net. Biodiversity Research Institute Page 3

12 5.2 Raptor Banding and Sampling Upon capture, we removed raptors from traps, banded them using U.S. Geological Survey (USGS) leg bands, and collected standard morphometric data (natural wing cord, tarsus width, body mass, tail, and culmen) following standard protocols (Pyle 2008) (Figure 1, Figure 2, Figure 3). Each bird was given a rudimentary health evaluation which included classifying both the crop and body index condition (BIC) into four qualitative classes (0-3). We plucked 3-4 feathers from the rump and breast of all individuals. Feathers are used for analyses of mercury (Hg), and will be further considered for analyses of other metals, stable isotopes, genetics, and/or archival. We collected blood samples from a portion of captured individuals following standard protocols (Fair et al. 2010). Many birds are also photo-documented. Figure 2. Banding a Peregrine Falcon using a lock-on band (left), and banding a Merlin with a butt-end band (right). Figure 3. Measuring wing chord of a Peregrine Falcon (left) and examining wing molt on an adult female Merlin (right). Biodiversity Research Institute Page 4

13 5.3 Instrumenting Raptors with Transmitters Satellite Transmitters Satellite telemetry has revolutionized our understanding of animal movements by providing a means of tracking movements of animals traveling vast distances and in remote regions that were otherwise impossible or impractical (Seegar et al. 1996). Transmitters attached to individuals can fix both GPS and Doppler or Argos locations (with varying degrees of accuracy), which can be received by globally orbiting satellites and relayed to users. GPS locations are generally associated with higher levels of accuracy compared to Argos locations (Douglas et al. 2012). We attached transmitters (Platform Transmitter Terminals, PTTs) to a subset of captured raptors, prioritizing birds that were visibly healthy and heavier, such that the transmitter package remained 3.9% of bird body mass (peregrines) or 3.0% (Merlins). Transmitters were instrumented to individuals with backpack-style harnesses made of 0.25 in (6.35 mm; Peregrine Falcon) or in (4.76 mm; Merlin) Teflon ribbon (Bally Ribbon Mills, Bally, PA) sewn with Spiderwire Dyneema thread and having a fixed point of the harness centered over the mid-keel of the bird (Kenward 2001, Steenhof et al. 2006, Walls and Kenward 2007, Fair et al. 2010) (Figure 4). All satellite transmitters fixed Doppler (or Argos ) locations; larger transmitters on female peregrines also fixed GPS locations. The location error for GPS locations typically ranges from 5 15m. Argos locations are classified into 7 different location classes by CLS America according to their associated error (Douglas et al. 2012, CLS 2016). Implausible locations were removed from our location dataset using the Douglas Argos Filter (DAF) (Douglas et al. 2012). Figure 4. Example of 12g solar PTT unit and 22g solar GPS PTT unit used to track migratory movements of Peregrine Falcons (left); a hooded Peregrine Falcon being fitted with a transmitter (right). We fitted individuals of the following three species with PTTs as follows during at BIRRS: Peregrine Falcons: We fitted 33 satellite transmitters to peregrines captured on Block Island between the years (Figure 5). Included in summaries here also are two hatching year (HY) females fitted with transmitters on Monhegan Island in 2010 (DeSorbo et al. 2012). Thus, data from 35 instrumented peregrines were available for analyses. Of these individuals, Biodiversity Research Institute Page 5

14 two were adult females, seventeen were hatching year males, and sixteen were hatching year females. Merlins: We fitted eleven satellite transmitters to Merlins on Block Island in 2014 and 2017 (Figure 5). All transmitters were 5g units, fit to one adult female and ten hatching year females. Northern Harrier: In a pilot effort, we fitted one 6g satellite transmitter to a hatching year male Northern Harrier in In 2017 we fitted one 5g satellite transmitter to a hatching year male and one 5g satellite transmitter to a hatching year female (Figure 6). Figure 5. Female hatching year Merlin fitted with a 5g satellite transmitter (left), male hatching year Peregrine Falcon (right) fitted with a 12g solar satellite transmitter at BIRRS. Figure 6. Male hatching year Northern Harrier fitted with a 5g satellite transmitter in 2014 (left), female hatching year Northern Harrier fitted with a 5g satellite transmitter in 2017 (right) at BIRRS. Biodiversity Research Institute Page 6

15 5.3.2 Digitally Encoded VHF Radio Tags ( NanoTags ) In 2014, BRI began collaborating with researchers at the University of Rhode Island to study the migratory habits of Merlins using the Atlantic Flyway using automated VHF telemetry. During , we instrumented 80 Merlins (20 females, 19 males in 2014; 17 females, 13 males in 2015; 5 females, 6 males in 2016) with digitally encoded VHF radio tags to gain perspectives on the Merlin migration and stopover areas along the Atlantic coast. Commonly referred to as NanoTags, digitally encoded transmitters emit an individual-specific VHF radio burst that can be detected at automated tower arrays erected by a network of research partners. Due in large part to its open source and collaborative nature, and low cost compared to some wildlife tracking options, the MOTUS wildlife tracking system network is growing rapidly in North America, particularly along the Atlantic coast including the Eastern Seaboard (Figure 7), Canadian Maritimes, and the Great Lakes ( This tracking network enables researchers to economically gather information on wildlife movement patterns and stopover timeframes on both a large and/or small geographic scale, providing that individuals fly within the detection range of arrays. Figure 7. Location of automated telemetry arrays along the Atlantic coast relative to the Block Island Raptor Research Station (BIRRS) during 2014 and * Automated Telemetry Arrays are not permanent structures, some are moved to suit the associated project needs and some are taken down in the fall to be reinstalled in the spring or summer of the following year due to permitting restrictions. The network of arrays is generally growing each year. Arrays shown on this map were active during the timeframe of the Merlin migration (mid- September through the end of November). As indicated (above) three 2014 arrays were removed or moved; overall 9 additional arrays were installed in Detection range up to 15 km, but varies widely according to many factors including receiver antenna type and size, transmitter size, topography, weather conditions, etc. Biodiversity Research Institute Page 7

16 A subset of captured Merlins were instrumented with NanoTags (Lotek Wireless, Newmarket, Ontario, Canada). Individuals that were visibly healthy and heavier were selected for instrumentation, such that the female and male backpacks remained 3.0% of bird body mass and tail mount units remained 1% of bird body mass. Females were fitted with backpacks (NTQB-6-2; 2.6 g prior to customization and fitting) to achieve a greater unit lifespan in hopes of registering at automated telemetry arrays during first spring and/or second fall migrations. Tail mount tags (NTQB-3-2; 0.67g prior to customization and fitting) were attached to a central retrix (tail feather) of both males and females using a light cured resin (Clear Cure Goo, Southlake TX). Merlins were harnessed with backpacks using 0.18 in. (4.76 mm) Teflon ribbon (Bally Ribbon Mills, Bally, PA) sewn with Spiderwire Dyneema thread. Harnesses were centered over the mid-keel of individuals similar to designs described elsewhere (Kenward 2001, Steenhof et al. 2006, Walls and Kenward 2007, Fair et al. 2010). After manufacturer customization and harnessing, male backpacks weighed approximately 2.5 g, and female backpacks weighed approximately 4.5 g. Tail mount units fitted to both males and females weighed approximately 1.6 g at deployment. Transmitter lifespan is estimated by the manufacturer to be approximately 84 days and 459 days for smaller and larger transmitters, respectively. Figure 8. Female hatching year Merlin instrumented with backpack-style digitally encoded VHF radio tag. Biodiversity Research Institute Page 8

17 Number of individuals captured 6.0 RESULTS AND DISCUSSION 6.1 Intensity, Species Composition and Timing We captured a total of 872 individual diurnal raptors representing eight species during fall capture efforts in the years (Table 1). Lower trapping numbers in 2013 and 2015 were considered to be largely related to multiple days of easterly or northeasterly winds typically considered to lessen the number of migrant birds reaching Block Island. Southerly winds also contribute to fewer raptors reaching the island; it is also presumed that a southerly wind will extend stopover times for birds currently on the island leading to a greater chance of capturing individuals. Overall, capture totals were relatively similar within species annually (Figure 9). Table 1. Annual species composition of diurnal raptors captured at BIRRS, Species Code Total AMKE 5 (4%) - 4 (2%) 3 (2%) 3 (2%) 2 (2%) 17 (2%) COHA 8 (6%) 7 (6%) 8 (4%) 11 (9%) 10 (7%) 13 (10%) 57 (7%) MERL 74 (52%) 59 (49%) 87 (41%) 67 (52%) 87 (61%) 65 (50%) 439 (50%) NOGO 1 (1%) (0%) NOHA 10 (7%) 16 (13%) 9 (4%) 8 (6%) 4 (3%) 2 (2%) 49 (6%) PEFA 35 (25%) 25 (21%) 91 (43%) 20 (16%) 28 (20%) 45 (35%) 244 (28%) RTHA 1 (1%) (0%) SSHA 7 (5%) 14 (12%) 11 (5%) 19 (15%) 10 (7%) 3 (2%) 64 (7%) Total 141 (100%) 121 (100%) 210 (100%) 128 (100%) 142 (100%) 130 (100%) 872 (100%) NOGO RTHA AMKE COHA SSHA NOHA PEFA MERL Species code Figure 9. Total number of diurnal raptors trapped by species and year at BIRRS, * Species codes follow American Ornithologist Union convention: Northern Goshawk (NOGO), Red-tailed Hawk (RTHA), American Kestrel (AMKE), Cooper s Hawk (COHA), Sharp-shinned Hawk (SSHA), Northern Harrier (NOHA), Peregrine Falcon (PEFA), and Merlin (MERL). Biodiversity Research Institute Page 9

18 Overall, annual species composition measures were similar among years was noteworthy for an unusual increase in the total number and relative proportion of peregrines captured compared to other years. Mean species composition summaries generally show 50% of annual captures are Merlins, 25% are peregrines, and the remaining 25% are comprised of Northern Harriers, Sharp-shinned Hawks, Cooper s Hawks, American Kestrels, Red-tailed Hawk and Northern Goshawk (Figure 10) NOGO RTHA AMKE, 2% COHA, 7% SSHA, 7% NOHA, 6% MERL, 50% PEFA, 28% Figure 10. Total species composition of diurnal raptors captured at BIRRS, (n = 872). Age Class: The vast majority (97.5%) of all raptors captured on Block Island were young of the year (hatching year age class). Adults were captured in three of the eight species at BIRRS; six peregrines (2.6% of all peregrines trapped), six Merlins (2.7%), and two Northern Harriers (5.7%). Timing: Distinct timing patterns were evident for Peregrines and Merlins. For the other species captured, the timing was highly variable, and small sample sizes preclude powerful conclusions about migration timing on Block Island or the region. The peak days of peregrine capture during the seasons (daily averages) occurred between 4-13 October (Figure 11). This ten-day range accounted for 56% of all peregrines captured. Seventy-five percent of all peregrine captures occurred from 27 September to 13 October. The timing of the Merlin flight as indicated by capture data, was notably more consistent and seasonally protracted compared to peregrines. Peak Merlin captures occurred between 18 September and 2 October (Figure 12). This 15-day time span accounted for 61% of all Merlins trapped during the season. 75% of all Merlins captured occurred between 18 September and 6 October (19 days). Biodiversity Research Institute Page 10

19 Figure 11. Average number of Peregrine Falcons captured by date, Figure 12. Average number of Merlins captured by date, Biodiversity Research Institute Page 11

20 In general, migration count data, rather than daily capture data such as used here, is more accurate in characterizing the timing and intensity of raptors passing through an area during migration. Standardized migration counts have not been possible during this study due to prioritization of staffing toward other study objectives. Here, we use capture totals as a surrogate for some information that might be better understood through standardized counts. The Hawk Migration Association of North America (HMANA) provides regularly updated annual hawk count information for over 275 North American hawk watch sites ( Site-specific information on this website can be of value to the general public and researchers in placing local migration patterns in a broader regional context. Overall, migration timing patterns for raptors captured on Block Island were consistent with general knowledge of migration timing for each species; however, comparable and localized migration count data from other offshore sites are rare (DeSorbo et al. 2012). Limited annual capture totals for some species, such as the Accipiter hawks (Sharp-shinned Hawks, Cooper s Hawks) and American Kestrels on Block Island preclude use of this data in characterizing migration timing. 6.2 Migratory Connectivity: Migration Routes, Wintering Areas, Origins Peregrine Falcons Fall Migration Routes and Wintering Areas The datasets of the 33 peregrines fitted with satellite transmitters on Block Island, combined with the two individuals fitted with transmitters on Monhegan Island, comprise the most substantial information to date on peregrine movement patterns along the Atlantic Flyway. This dataset is especially unique in that hatching year peregrines have rarely been studied using satellite telemetry. Overall, migration routes used by fall migrants captured on Block Island tended to follow a coastal route along the Atlantic coastline (Figure 13). Overland migration routes of instrumented peregrines generally reflected second fall migration fight paths of individuals originating in areas north of Hudson Bay or west of the Great Lakes. Preliminary analyses of these migration routes and the extent to which they can inform use of offshore wind energy areas along the Atlantic coast can be found in DeSorbo et al. (2015). Fall migrant peregrines overwintered in locations across a wide geographic range, spanning throughout the Bahamas, Central America, and South America (Figure 13, Table 2). Of the 33 birds instrumented on Block Island and the two from Monhegan Island, 25 individuals are considered to have reached wintering areas, while transmitters on ten individuals ceased transmitting during migration (Table 2, Table 3). Preliminary patterns in our sample suggest that males captured at Block Island may use more southerly wintering areas than females. For example, while the majority of females (71%; 10 of 14 individuals) overwintered in the Bahamas and Cuba, eight of the eleven (72%) instrumented males traveled to Central and South America (Figure 13, Table 2). At present, a male traveling to Argentina has traveled further south than any other individual in this study. Of the ten instrumented peregrines that stopped transmitting prior to reaching wintering areas, all were within the latitudinal span of the continental U.S. between Rhode Island and Florida (Table 3). Some of the individuals in this cohort of instrumented peregrines demonstrate the notable Biodiversity Research Institute Page 12

21 challenges facing birds during migration; including starvation, hurricanes, and predation (Table 2). Of these individuals, 60% were males, and 40% were females. One noteworthy male traveled on an impressive and unique transoceanic journey, 1500 km to the east into the Atlantic, following release (Figure 14). Travel speeds by this individual as indicated by transmitter location estimates suggest this individual was resting on offshore vessels before transmissions ceased. Figure 13. Fall migration routes of Peregrine Falcons instrumented with satellite transmitters at BIRRS, Biodiversity Research Institute Page 13

22 Table 2. Wintering areas for 25 fall migrant Peregrine Falcons tracked from Block Island, RI (n = 23; ) and Monhegan Island (n = 2; 2010) using satellite telemetry. Number of Individuals Wintering Area a Male Female Both % (Both) Comment Bahamas / Turks and Caicos Is b Cuba / Jamaica c D.R. / P.R. / Haiti / Brit. Virg. Is d Central America e South America - N. of Equator f South America - S. of Equator g Florida h Rhode Island i Total Comment: a. Wintering area described as the furthest south location where the bird spent multiple days in one area b. HYF01 (Turks and Caicos), HYF02 (Crooked Is., Bahamas), HYF11 (Spanish Wells, Bahamas), HYM08 (Moss Town, Bahamas); c. ADF01 (Jamaica), HYF04 (Cuba), HYF07 (Cuba), HYF09 (went to the Dominican Republic, then back to Cuba), HYF14 (Cuba); d. HYF06 (Dominican Republic), HYF08 (Dominican Republic), HYM07 (Spanish Town, British Virgin Islands), HYM10 (Haiti); e. HYF03 (Honduras), HYM03 (Nicaragua), HYM04 (Isla de San Andres), HYM17 (Panama); f. HYF05 (Columbia), HYM01 (Columbia), HYM06 (Venezuela), HYM14 (Guyana); g. HYM12 (Argentina), HYM13 (Bolivia); h. HYF15 (South East FL); i. ADF02 (Wintered on Block Isl.) Figure 14. Fall migratory movement of Peregrine Falcon HYM02 after departure from BIRRS, as indicated by satellite telemetry. * Speeds presented in figure measured between first and last location of daily clusters of locations. Biodiversity Research Institute Page 14

23 Table 3. Last known locations of satellite transmitter instrumented Peregrine Falcons with presumed incomplete fall migration paths ( ). Number of Individuals Last Transmition Area a Male Female Both % (Both) Comment Rhode Island b New York c Virginia d North Carolina e South Carolina f Georgia g Florida h Total Comments: a. Areas in which transmitters instrumented to peregrines last transmitted during migration between Block Island and Florida. b. HYF16 (South Kingstown); c. HYM11 (Eastern Long Island); d. HYM05 (Presumed mortality, transmitter recovered under eagle nest); e. HYM02 (Ship Rider, offshore from NC), HYM16 (Offshore from Wilmington); f. HYF13 (Parris Island), HYM09 (Charleston); g. (HYM15 (Blackbeard Island); h. HYF10 (Cocoa), HYF12 (found emaciated in Titusville); both stopped transmitting in Cape Canaveral. Seven instrumented Peregrine Falcons in our study were tracked during two consecutive fall seasons, enabling comparisons of annual migration routes (Figure 15). Of those seven individuals, four migrated from locations generally west of the Great Lakes during their second fall migration; therefore these individuals did not provide insights on inter-annual migration routes for much or all of Atlantic coast. Of the three individuals that did follow a significant portion of the Atlantic coast during multiple migrations, two originated from northern Quebec, while the third originated from eastern (mainland) Nunavut, Canada and migrated to the Atlantic coast from the southern tip of Hudson Bay. Overall, Atlantic migration routes used by these three individuals were similar between consecutive years. Two of the three individuals (HYF02 and HYM14) appeared to use more inshore migration routes during the second fall compared to that used in the first. Migration routes were notably similar between consecutive fall seasons for the third peregrine (HYF15) Peregrine Falcons Fidelity to Wintering Areas Of the Peregrine Falcons tracked using satellite telemetry during this study, six (4 females and 2 males) enabled evaluations of fidelity to wintering areas. Of these six individuals, five returned to the same specific wintering areas used during consecutive winter seasons in the following areas: Aklins Island, Bahamas; Turks and Caicos; the Dominican Republic; western Haiti, and Miami, FL. The peregrine that did not show winter site fidelity, HYM12, traveled further south during its first winter than any other bird tracked during this study (to Argentina, see Figure 13). Biodiversity Research Institute Page 15

24 Figure 15. Consecutive annual fall migration routes of seven Peregrine Falcons instrumented with satellite transmitters at BIRRS. Biodiversity Research Institute Page 16

25 6.2.3 Peregrine Falcons spring migration routes and insights on natal origins Of the 35 peregrines tracked during this study, nearly one-third (n = 11) have provided information on spring migration patterns. These individuals offer insights on the potential natal origins of individuals (assuming second-year birds return to their natal areas), as well as information linking use of breeding areas, wintering areas, and migration stopover sites by individuals. The strongest evidence clearly linking wintering and breeding populations derives from the two adult female peregrines captured on Block Island: ADF01 overwintered in Jamaica and migrated to Saskatchewan, Canada, while ADF02 overwintered in the vicinity of Block Island and migrated to eastern Greenland (Figure 16). Overall, these two regions roughly span the longitudinal range of areas visited by spring migrating peregrines in this study. Numerous peregrines captured during their first year of life traveled to regions spanning from areas as far west as Saskatchewan to regions throughout Nunavut, Canada (including Baffin Island), and as far east as arctic regions of Quebec, Newfoundland and Labrador (Figure 16). Of the eleven peregrines tracked northward in the spring, only three (ADF01, wintered on Block Island; HYF15, wintered in FL, and HYF02, wintered in the Bahamas) used travel routes along the Atlantic coast during their northward migrations (Figure 16). The remaining 8 peregrines yielding information on spring migration routes traveled through the central U.S. west of the Great Lakes. One Peregrine, HYF07, was tracked northward during two consecutive spring seasons. This individual migrated through the central U.S. and spent the summer in southern Saskatchewan, but during its second spring migration migrated through Ontario and Hudson Bay (where it appeared to be riding ice flows) to northern Quebec (Figure 16). Biodiversity Research Institute Page 17

26 Figure 16. Spring migration routes of eight female and three male Peregrine Falcons tracked using satellite telemetry. Biodiversity Research Institute Page 18

27 6.2.4 Merlins Migration Routes, Wintering Areas, and Presumed Origins (satellite telemetry) Between the 2014 and 2017 fall seasons, we fitted twelve female Merlins (1 adult, 11 hatching year individuals) with 5g satellite transmitters. To our knowledge, these individuals represent the only individuals in this species to be tracked globally using satellite telemetry. Two transmitters fitted to hatching year Merlins did not provide useful data due to either early mortality or unit failure. Of the ten remaining individuals, 60% (1 adult, 5 hatching year individuals) provided insights on wintering areas spanning from Rhode Island to the Caribbean (Figure 17). Of those six individuals, half migrated to the Greater Caribbean Region. One Merlin (the adult) was tracked to Puerto Rico, while two hatching year Merlins (HYF07 and HYF09) migrated along the Atlantic coast and overwintered in Cuba. One (HYF08) migrated along the Atlantic coast to coastal North Carolina where it wintered on coastal islands and agricultural land on the southern end of Pamlico Sound; one (HYF10) overwintered near New London, Connecticut; and one (HYF11) overwintered locally on Conanicut Island in Jamestown, Rhode Island (Figure 17). Four hatching year Merlins did not complete migrations prior to the cessation of transmissions, likely due to either mortality or transmitter failure. Of these four, two last transmitted in southern Florida; one last transmitted near Virginia Beach, Virginia; and one last transmitted over the Atlantic Ocean off Long Island, New York. This pattern, also observed in peregrines and numerous other migrant birds, likely reflects the challenges migrants face during migration and the relatively high mortality rate during the first migration event. Of seven Merlins tracked down the Atlantic seaboard (six hatching year birds and one adult), two showed clear indications of offshore habitat use. This finding has relevance to ongoing efforts to understand movement patterns of birds in relation to offshore wind energy facilities being proposed along the Atlantic coastline. Migration counts on Monhegan Island, ten miles east off the Maine coast, documented that Merlins were by far the most abundant diurnal raptor species observed offshore in that region (DeSorbo et al. 2012), a finding that is consistent with annual species composition measures reported in this study. One Merlin (adult female, ADF01) instrumented with a satellite transmitter in this study provided insights on spring travel routes from wintering areas, as well as insights on likely origins. From Block Island, ADF01 migrated along the Atlantic coastline to Florida, where it then crossed over to Cuba, to the Dominican Republic and then to Puerto Rico, where it overwintered. Satellite communications ceased for much of the winter until the spring, at which point it retraced much of its southward migration path north to Petit-Mecatina, Quebec (Figure 18). This individual transmitted from this general location for approximately 20 days until transmissions ceased on 2 June Based on observed movement patterns, we speculate that this individual likely originated from eastern Quebec or Newfoundland/Labrador, possibly from the general region in which it last transmitted. This individual is likely the first Merlin for which a complete or nearcomplete migration track has been documented (Figure 18). Biodiversity Research Institute Page 19

28 Figure 17. Fall migration paths of ten hatching year female and one adult female Merlin tracked from BIRRS, using satellite telemetry. Biodiversity Research Institute Page 20

29 Figure 18. Fall and spring migration paths of an adult female Merlin tracked from BIRRS, using satellite telemetry. Biodiversity Research Institute Page 21

30 6.2.5 Merlins Migration Patterns along the Atlantic Seaboard Using Automated Radio- Telemetry Receivers A total of 80 digitally encoded VHF radio transmitters ( NanoTags ) were deployed on Merlins captured at BIRRS between the years (39 in 2014; 30 in 2015, 11 in 2016). These tags provided first-time perspectives about the migratory ecology of this poorly studied species. Overall, 94% of instrumented Merlins were detected by 1 automated telemetry array (92% in 2014 [n = 36]; 97% in 2015 [n = 29]; 91% in 2016 [n = 10]), and 60% (48 of 80) were detected by 3 telemetry arrays. No females were detected during their first spring or second fall migration (only the ~ 4 g tags put on females had a predicted lifespan capable of spring/second fall detections; see methods). Our study characterizes previously undocumented movements of Merlins during migration. Of 69 individuals tagged in analyzed, 39% were detected by a telemetry arrays along the Eastern Seaboard, confirming southerly movements for this portion of individuals (Figure 19). The southernmost detections were in New Jersey (n = 6) and Virginia (n = 20), reflecting limitations of the tower network. Interestingly, 14% of the remaining individuals were documented to travel in an easterly direction from BIRRS (n = 9) (Figure 20), while another 16% were documented to travel in a westerly direction relative to their Block Island deployment site (n = 10) (Figure 21). None of these individuals were detected at any other telemetry arrays south of New York. Less than 1% (n = 4) of tags were not detected at any array; the remaining 30% (n = 20) of the tags were detected only at Block Island and/or Montauk, Long Island. The overall detection pattern of Merlins tagged at BIRRS suggests that a portion of Merlins either use inland migration routes, or that they overwinter inland along the Atlantic coast in areas where automated telemetry arrays were not present. Information on bird sightings in ebird supports this explanation; a filtered search on Merlin sightings during the last four years during winter months shows that Merlins commonly overwinter in southern New England and New York with coastal concentrations. Biodiversity Research Institute Page 22

31 Figure 19. Southward Movements of eight (of 30) Merlins instrumented with digitally encoded VHF tags at BIRRS in 2015, as indicated by detections at automated telemetry arrays along the Atlantic coast. Biodiversity Research Institute Page 23

32 Merlins instrumented with NanoTags in this study provided rare perspectives on Merlin travel rates and information on the direction of departure of Merlins from Block Island. Average travel times between Block Island and detections at telemetry arrays in New Jersey and Virginia were 5.6 and 7.6 days or 64.2 km/day and 76.3 km/day, respectively. Detection patterns suggested that the majority of Merlins leave Block Island to the south or southwest (Figure 19); however, north and northeast departures were also evident. Gaps in automated telemetry array coverage in the northwestern quadrant of the island may have resulted in northwestern departures that were undetected. Further analyses of stopover patterns of migrant Merlins on Block Island and elsewhere are still being analyzed. Overall, broad-scale migration patterns of Merlins are poorly studied to date. Merlins commonly follow the Atlantic coastline, but lacking detections of individuals at some key telemetry arrays is suggestive of variable use of coastal and offshore habitats during migration. These findings are consistent with those indicated by satellite telemetry, which verify use of offshore habitats by migrant Merlins. The relatively large sample size of Merlins studied with NanoTags in this study allowed characterization of different migratory strategies that are otherwise poorly documented by other approaches such as banding, and costly to document using satellite telemetry. Figure 20. Easterly Merlin movements detected by automated telemetry arrays during fall These six Merlins were not detected by arrays other than those displayed above. Biodiversity Research Institute Page 24

33 Figure 21. Westerly movements of Merlins detected by automated telemetry arrays during fall These six Merlins were not detected at arrays other than those displayed above Northern Harriers perspectives on migration We fitted one hatching year male Northern Harrier with a 6g satellite transmitter in Unlike almost all of the other raptors fitted with satellite transmitters in this study, this bird did not appear to migrate southward in the fall. Northern Harrier HYM01 departed the island from the southwestern corner, traveled down a portion of Long Island, New York, until it headed northward through central Connecticut, where it spent roughly 5 days before continuing on to Carver, Massachusetts just north of Buzzard s Bay (Figure 22). Northern Harrier HYM01 transmitted from the general vicinity of Carver, Massachusetts until its last transmission on 28 November, As is often the case in animal tracking studies, causes for ceased transmissions remain unclear. In 2017 we fitted one hatching year female Northern Harrier (HYF01) with a satellite transmitter. This bird appeared to leave the north end of the island and traveled to Connecticut, but it then flew south again, crossing Long Island Sound and then traveling southwest along the coast towards New York City, before it turned back to Short Beach Island along the south coast of Long Island where it stayed until the last transmission on 19 November 2017 (Figure 22). The limited sample described above, plus one instrumented individual that died on Block Island shortly after instrumentation, suggest that Northern Harrier movements are relatively regional/local in nature. Limited data may also suggest that hatching year Northern Harriers have a higher mortality rate compared to other hatching year raptors studied using satellite telemetry in this study. Biodiversity Research Institute Page 25

34 Figure 22. Fall movements of a hatching year male (2014) and a hatching year female (2017) Northern Harrier tracked from BIRRS, using satellite telemetry. Biodiversity Research Institute Page 26

35 6.3 Evaluating Mercury Exposure in Migrant Raptors sampled on Block Island, RI Background: Mercury Exposure in Wildlife Mercury (Hg) pollution is a well-established environmental problem at a global scale. Mercury occurs naturally in our environment, but it is also produced through a wide variety of industrial activities. It can be deposited into ecosystems directly, or more commonly, through atmospheric deposition. Once deposited, Hg can be persistent in ecosystems and it readily accumulates in organisms. Bird tissues are commonly sampled and analyzed to gather insights on spatial and temporal patterns of Hg exposure, and to evaluate potential for adverse impacts. Blood reflects recent dietary exposure to Hg, and feathers can reflect a combination of recent dietary exposure and cumulated body burdens (Evers et al. 2005). Elevated Hg exposure is associated with a wide variety of adverse effects on animal behavior, reproduction, and physiology. Traditionally, toxicologists primarily considered Hg a risk for aquatic-based wildlife; however, recent studies have suggested that Hg can also pose health risks to birds with non-piscivorous diets, including those with diets comprised of terrestrial sources of protein such as passerines (Rimmer et al. 2005, Jackson et al. 2011). Recent studies have also further demonstrated that species vary widely in their sensitivity to Hg (Heinz et al. 2009), and therefore exposure to Hg at levels traditionally considered low may be of concern for some species Mercury Exposure in Migrant Raptors at Block Island Hg in breast feathers: Mercury concentrations in feathers represent Hg exposure during the time of feather growth. In hatching year birds, this exposure reflects dietary Hg exposure during the period of nestling development. We evaluated mercury (Hg) exposure in breast feathers in eight species of diurnal migrant raptors sampled at Block Island, RI during fall migration (Figure 23). Like most raptor migration stations, the majority of individuals captured for sampling were of the hatching year age class. Qualitatively, mean breast feather Hg concentrations among species exhibited some expected patterns based on known dietary habits, such as lowest exposure in Redtailed Hawks, and generally higher Hg exposure in the falcons (Merlins, American Kestrels and Peregrine Falcons) compared to many remaining species sampled. Feather Hg concentrations in hatching year raptors differed significantly across the six raptor species groups with a sample size of 15 (p< , χ5 = 70.2). Nonparametric statistical comparisons between species pairs revealed significant differences in feather Hg exposure between numerous species. Notably, Sharp-shinned Hawks exhibited higher mean breast feather Hg concentrations compared to Merlins, Cooper s Hawks, and Northern Harriers, but feather Hg concentrations did not differ compared to peregrines or American Kestrels. Variability was notably higher in American Kestrels compared to other species; however, our sample size of this species remains limited. Our sample sizes in adult age classes generally preclude statistical comparisons between age classes; however, preliminary feather Hg comparisons between age classes suggests that after hatching year (AHY) Merlins have higher Hg exposure levels and higher variability compared to hatching year individuals. Higher variability in breast feather Hg exposure concentrations in after hatching year Merlins likely arises from the difficulty in distinguishing hatching year breast feathers from those grown during the first feather molt (after hatching year feathers). Biodiversity Research Institute Page 27

36 B B B B A A Figure 23. Breast feather Mercury concentrations (µg/g) in eight species of hatching year migrant raptors sampled at BIRRS, * Mean ± 95% confidence interval displayed. Pairs of species that share the same letter were not significantly different using a Wilcoxon Multiple Comparison Procedure (only species with 9 samples, in the hatching year age class were included in this test). Species codes follow American Ornithologist Union convention: Red-tailed Hawk (RTHA), Northern Goshawk (NOGO), Cooper s Hawk (COHA), Northern Harrier (NOHA), American Kestrel (AMKE), Merlin (MERL), Peregrine Falcon (PEFA) and Sharp-shinned Hawk (SSHA). Numbers in parentheses indicate sample size within species/age class groups. Hg in blood: Blood Hg concentrations provide perspectives on recent dietary exposure to Hg during migration. Overall, blood Hg exposure patterns in raptor species groups were similar to those in feathers (Figure 24). Mean blood Hg concentrations differed across species groups (p< , χ5 = 134.7). Mean blood Hg concentrations in Cooper s Hawks and Northern Harriers were similar and significantly lower in comparison to means in Sharp-shinned Hawks, Peregrine Falcons and Merlins. Notably, mean blood Hg concentrations were statistically higher in Merlins than all other species. Unlike feather Hg comparisons, in which Sharp-shinned Hawks had the highest feather Hg concentrations of all species sampled, Merlins contained the highest blood Hg concentrations of all species in our analyses. These findings suggest that while Sharp-shinned Hawks may be exposed to higher Hg concentrations during nestling development compared to other species, Merlins are exposed to the highest concentrations of Hg during migration of the species studied. Biodiversity Research Institute Page 28

37 D B C C A A A B Figure 24. Blood Mercury concentrations (µg/g) in six species of hatching year migrant raptors sampled at BIRRS, * Mean ± 95% confidence interval displayed. Pairs of species categories that share the same letter were not significantly different using a Wilcoxon Multiple Comparison Procedure (only species with 9 samples, in the hatching year age class were included in this test). Numbers in parentheses indicate sample size within species/age class groups. See Figure 23 for species code explanations Interpretations and Conclusions - Mercury Efforts during this study have established baseline blood and feather Hg exposure concentrations in eight migrant raptor species using the Atlantic Flyway. Overall, mean Hg exposure patterns among species exhibited some expected patterns based on known dietary habits and trophic level, such as lowest exposure in Red-tailed Hawks, and generally higher exposure in the falcons (Merlins, American Kestrels and Peregrine Falcons) compared to the remaining species sampled. The exposure to all species of Hg included in this study, despite their different habitat associations, exemplifies the pervasiveness of Hg in our environment across multiple prey guilds and habitats. Findings that Sharp-shinned Hawks exhibited the highest Hg concentrations among all species sampled, and that American Kestrels might exhibit similar Hg concentrations is interesting given general conservation concerns of population declines for both of these species (Viverette et al. 1996, Farmer et al. 2008, Smallwood et al. 2009). Blood and feather Hg concentrations associated with adverse effects have not been established for the raptors sampled in our study. Overall, adverse effect thresholds for Hg in feathers and blood remain poorly established for most raptors at environmentally realistic levels. Feather Hg concentrations have been reported to be notably higher in many Bald Eagle (Haliaeetus leucocephalus) and Osprey (Pandion haliaetus) populations in North America compared to those reported in this study (Bowerman et al. 1994, Anderson et al. 2008, Desorbo et al. 2009, Biodiversity Research Institute Page 29

38 Scheuhammer et al. 2012); however, links between tissue exposure levels and reproductive impairment in wild populations has been poorly demonstrated. Rutkiewicz et al. (2011) found clear evidence of neurological impairment in Bald Eagles in the Midwestern U.S. associated with Hg exposure. Adverse effects of Hg on Common Loon (Gavia immer) reproduction and physiology have been demonstrated in New England and Canada (Burgess and Hobson 2006, Evers et al. 2008). Birds vary in their sensitivity to mercury (Heinz et al. 2009). Therefore, exposure levels and established adverse effect thresholds in Bald Eagles, Ospreys, or Common Loons may not be appropriate for comparisons to species in this study. Recent studies have documented that forest songbirds and invertebrates are effective in accumulating Hg, and Hg exposure has been liked to adverse effects in some passerines (Jackson et al. 2011). For example, Jackson et al. (2011) found that blood and feather Hg concentrations of 0.7 µg/g and 2.4 µg/g respectively were associated with a 10% reduction in nest success in Carolina Wrens (Thryothorus ludovicaianus). In our study, individual Merlins, Sharp-shinned Hawks, Peregrine Falcons, and American Kestrels exceeded one or both of these blood and feather Hg concentrations (Figure 23, Figure 24). Further study is needed to determine whether Hg levels associated with adverse effects in songbirds are relevant to raptors. Raptors with an evolutionary association with aquatic systems, such as Peregrine Falcons and Merlins, may have a lower sensitivity to Hg effects compared to species associated with uplands and terrestrial ecosystems, such as Sharp-shinned Hawks. Findings in this study indicating adult Merlins have higher Hg exposure compared to younger age classes is suggestive that Hg exposure in many raptors sampled in this study will increase over time for individuals that survive into their next feather molt. Our findings indicate that individuals are being exposed to Hg in both breeding areas and during migration. Further efforts to improve our understanding of the exposure and potential impacts of Hg species sampled in this study are warranted. 6.4 Education, Research and Presentations BRI and TNC have partnered to host formalized environmental education and outreach programs on Block Island to promote habitat and wildlife conservation and informed decision-making through field research. Over the past five years, we have hosted or provided programs for dozens of school groups, eco-tourism groups, conservation professionals, fellow researchers, and lawmakers (including Rhode Island s former Governor, Lincoln Chaffee). BRI contributes annually to the James Stover Series Nature Walk and other events sponsored by TNC. In 2015, BRI completed the Mid-Atlantic Baseline Study, an evaluation of wildlife distribution and movement patterns relative to Mid-Atlantic offshore wind energy areas. Information on the overall study can be found at: Summaries and report chapters can be found at: Our report chapter evaluating peregrine movements relative to offshore wind energy areas can also be found at: along with a science communication about the Block Island Raptor Research Station, this report, downloadable maps, and other information. Biodiversity Research Institute Page 30

39 An update on our 2015 summary findings was presented at the 50 th annual conference of the Raptor Research Foundation (RRF) in Cape May, NJ, Oct 16 20, The abstract for this presentation can be found in the program materials listed on the RRF website (p.32): Figure 25. Sunset at BIRRS. Acknowledgements Original funding support for this project in was provided by the Department of Energy, Award no. DE-EE Continuing field efforts beyond 2013 would not have been possible without key support commitments by The Nature Conservancy, The Bailey Wildlife Foundation, and the Ocean View Foundation. Additional support from the Bluestone Foundation, the Overlook Foundation and individual donors were crucial in supporting operational costs and additional transmitter purchases. We thank Adam Gravel of Stantec Consulting Services Inc., for donating a satellite transmitter and associated data transmission costs to the project. Sincere thanks to Keith and Kay Lewis for their support, for being the common bond to many other likeminded individuals on Block Island, and for the use of their property to support various project needs. Thanks to the following individuals for their various important contributions to field efforts: Rick Gray, Al Hinde, Deneb Sandack, Jeff Johnson, Ken G. Wright, Fred Tilly, LeRoy Fink, Dustin Riordan, Jon Atwood, Bruce Duarte, Nigel Grindley, Cathy Joyce, Ken G. Wright. We also thank Adam Smith, Clara Cooper-Mullins and Scott McWilliams, University of Rhode Island. Special thanks to Mike Yates and Bill Seegar of Earthspan; David Douglas, USGS Alaska Science Center; and Keith LeSage of GeoTrak Inc. Karine Aigner, Karine Aigner Photography, and TNC kindly shared images used in this report. Satellite transmitters deployed on Monhegan Island, Maine, were included in peregrine migration maps in this study; that project was supported by the Maine Outdoor Heritage Fund and the Davis Conservation Foundation. Biodiversity Research Institute Page 31