2013 NORTHEASTERN NATURALIST 20(2):219–228 Abundance and Distribution of Purple Sandpipers (Calidris maritima) Wintering in Maine Glen H. Mittelhauser1,*, Lindsay Tudor2, and Bruce Connery3 Abstract - We systematically surveyed the Maine coastline from Washington County to York County to provide baseline data concerning Calidris maritima (Purple Sandpiper) population status. Focusing on a particular region each winter, we conducted 66 winter surveys by boat along the entire coast of Maine between 2002 and 2007 plus three days surveying from the mainland between Kittery and Biddeford during the winter of 2005–2006. We tallied 13,318 Purple Sandpipers during these surveys. After accounting for birds present but not detected, we estimate that 14,000 to 17,000 Purple Sandpipers wintered annually in Maine between 2002 and 2007. Based on an assessment of historical records and data collected during this study, flocks of ≥250 Purple Sandpipers have been reported from 48 sites along the Maine coast. The area from Isle au Haut to Swans Island along the midcoast supports the highest concentrations of wintering Purple Sandpipers in Maine and the largest wintering concentration of Histrionicus histrionicus (Harlequin Ducks) in eastern North America, highlighting the potential importance of this geographic region. Introduction Calidris maritima (Brunnich) (Purple Sandpiper) breeds at northern latitudes from northern North America, through Greenland and Iceland, to northwestern Siberia and have the most northerly wintering distribution of any shorebird (Cramp and Simons 1983). In eastern North America, Purple Sandpipers winter along the coast from Newfoundland south to Virginia (Payne and Pierce 2002). Along the coast of Maine, Purple Sandpipers are a common winter resident (Palmer 1949) with a remarkably high site fidelity (Knight 1908, Mittelhauser et al. 2012, Palmer 1949). In contrast to most other shorebirds, they typically winter along wave-exposed, rocky shorelines instead of sheltered localities (Summers et al. 2002). Because Purple Sandpipers are typically recorded outside the spring and fall shorebird migration survey period in northeastern North America and in localities not monitored during Maritimes Shorebird and International Shorebird Surveys (Howe et al. 2000), estimates of population size and trends are largely unknown. A wintering population of 16,000 individuals has been estimated for this region based largely on an assessment of Christmas Bird Count data (Morrison et al. 2001, Payne and Pierce 2002). Population declines in this species have been reported in Scotland (Summers et al. 2001) and Belgium (Burton et al. 2008), where specific winter surveys have been conducted. Preliminary data suggest that numbers of Purple Sandpipers also may be declining at some sites in northeastern North America (Morrison et al. 2001). 1Maine Natural History Observatory, 317 Guzzle Road, Gouldsboro, ME 04607. 2Maine Department of Inland Fisheries and Wildlife, 650 State Street, Bangor, ME 04401. 3Acadia National Park, PO Box 177, Bar Harbor, ME 04609. *Corresponding author - purplesandpiper@gmail.com. 220 Northeastern Naturalist Vol. 20, No. 2 No coast-wide survey has ever been conducted in Maine to estimate the size of the wintering population. Purple Sandpipers are regularly recorded on Christmas Bird Counts in Maine; however, these counts typically include little of the primary wintering habitat (offshore islands) preferred by this species. The largest number reported for Maine during Christmas Bird Counts was 3417 birds during the winter of 1997–1998, but many offshore islands and ledges were not surveyed for this count. Our primary objective was to document the winter distribution and abundance of Purple Sandpipers along the coast of Maine. We also developed estimates of detectability and identified the more important coastal regions in Maine for wintering Purple Sandpipers. Field-Site Description The coast of Maine includes roughly 4500 islands and approximately 11,000 km of shoreline. We divided the coast into three regions. The eastern region from Eastport to Mount Desert Island (Fig. 1) includes the largest amount of intertidal marine habitats, with extensive mudflats and many bays, coves, and islands. However, portions of this eastern region (from Lubec to Cutler) are rock-bound, with relatively few islands and steep shorelines. The mid-coast region from Mount Desert Island to Pemaquid Point (Fig. 2) has Figure 1. Eastern region of Maine; maximum number and distribution of Purple Sandpipers reported between Eastport and Mount Desert Island since 1977. 2013 G.H. Mittelhauser, L. Tudor, and B. Connery 221 a highly irregular coastline. Penobscot Bay dominates this region and contains a wide variety of marine and estuarine habitats and numerous islands. In the southern region from Pemaquid Point to Kittery (Fig. 3), Casco Bay is a predominant feature with large shallow bays with intertidal flats and the occurrence of over 400 nearshore islands that provide additional intertidal areas. South of Casco Bay, topographic relief is slight, with few islands but many sandy beaches and salt marshes. Offshore areas along the entire coast typically remain ice-free in winter, although during extremely cold weather in mid-winter, all but the most offshore and wave-exposed shorelines freeze over and portions of those intertidal areas are occasionally covered with a coating of ice. Methods We counted Purple Sandpipers along all wave-exposed rocky shorelines on the coast, and also included protected shorelines not considered typical wintering habitat for this species. We excluded ice-encased bays and shorelines and nonrocky shorelines with primarily muddy or sandy substrates. We concentrated our efforts on a particular coastal region each winter, completing the entire coast in five winter seasons. Figure 2. Mid-coast region of Maine; maximum number and distribution of Purple Sandpipers reported between Mount Desert Island and Muscongus Bay since 1977. 222 Northeastern Naturalist Vol. 20, No. 2 We focused our survey efforts each winter within two temporal survey windows: 7 December to 31 January, and 15 March to 30 April. We excluded the mid-winter timeframe (1 February to 14 March) because high winds, rough seas, and low atmospheric temperatures typical of this time period can alter bird movements (Mittelhauser et al. 2012). We chose the starting and ending dates for the survey windows based on initial survey data and assessment of early- and lateseason bird observations (Mittelhauser et al. 2006). We conducted surveys during suitable weather from boats ranging in size from 5-m skiffs for nearshore surveys to 12-m fishing boats for more offshore locations. We surveyed the area from Cape Elizabeth to Kittery within the southern region, an area with few offshore islands and ledges, from the shoreline using a spotting scope during the winter of 2005–2006. We did not survey the Isle of Shoals area because of our difficulty in accessing this offshore site, but instead used Christmas Bird Count data for this location during our survey window for this portion of the coast. During surveys by boat, we followed the mainland shore and circled islands and ledges as close as was safely possible and spotted, tallied, and recorded birds as observed. During surveys, we typically used two observers. Our field protocol was to count and recount a flock of birds until a “good” count was made. We conducted surveys during all daylight hours when weather conditions were Figure 3. Southern region of Maine; maximum number and distribution of Purple Sandpipers reported between Pemaquid Point and Kittery since 1977. 2013 G.H. Mittelhauser, L. Tudor, and B. Connery 223 Table 1. Regional summaries of counts of Purple Sandpipers along the coast of Maine based on surveys, 2002–2007; 95% CI based on detectability study, which suggested that our surveys missed 6.5 to 29.4 percent of the birds present. Location Winter surveyed Actual count 95% CI Seal Island to Monhegan Island (Fig. 2) 2002–2003 1073 1143–1388 Petit Manan Point to Vinalhaven (Fig. 2) 2003–2004 5243 5584–6784 Eastport to Schoodic Point (Fig. 1) 2004–2005 2754 2933–3564 Searsport to Cape Small (Fig. 2) 2005–2006 1730 1842–2239 Cape Elizabeth to Kittery (Fig. 3) 2005–2006 388 413–502 Pemaquid Point to Cape Elizabeth (Fig. 3) 2006–2007 2130 2268–2756 Overall 13,318 14,183–17,233 favorable and were not limited to a particular tide cycle. We used a GIS coverage of the Maine coast to estimate the length of shorelines surveyed and calculate the number of birds per km of shoreline. To estimate numbers of individuals present but not detected during our surveys, we used a double-observer technique (Thompson 2002) during two surveys conducted in Frenchman Bay in 2009. In the survey skiff, we had a crew of four observers: a two-person team of primary observers and a two-person team of secondary observers. Primary observers were positioned in the bow of the boat and secondary observers were positioned near the back of the boat to minimize the chance that primary observers would key in on birds detected by the secondary observers. During the survey, the primary observers pointed to and announced all birds detected. The secondary observers recorded the birds detected by the primary observers and separately recorded any additional birds they detected. The primary observers had no previous knowledge of Purple Sandpiper numbers or distribution along the survey route, but had one or two years of experience with this species along other portions of the coast. The secondary observers each had five years of experience documenting Purple Sandpiper numbers and distribution and radio-tracking individuals along the Frenchman Bay survey route (Mittelhauser et al. 2012). For difficult counts where all birds were not visible at the same time, we flushed the birds to get a better count of the number of birds present. We compiled a database of historical Purple Sandpiper records and distribution for the coast of Maine since the 1970s, based on a review of national and local bird newsletters summarizing Maine bird sightings, unpublished survey data (Maine Department of Inland Fisheries and Wildlife, Bangor, ME, unpubl. data), and anecdotal records from various winter bird surveys along the Maine coast since 1988. Results We conducted 66 winter surveys by boat (13 during December, 15 during January, 12 during March, and 26 during April) along the coast of Maine focusing on a particular region each winter. Pooling data across years from non-overlapping surveys, we detected and recorded 13,318 Purple Sandpipers wintering on the coast of Maine (Table 1): 2754 birds along the eastern region (Fig. 1), 6316 birds along the mid-coast region (Fig. 2), and 4248 birds along the southern region 224 Northeastern Naturalist Vol. 20, No. 2 Table 2. Maine locations where ≥250 Purple Sandpipers have been reported at least once since 1977. Maine Island Max. Data summary Area/location Registry # count Date n Mean ± SE Eastern region (see Fig. 1) Batson Ledges 79-613 250 1 Feb 2005 3 92 ± 79 Black Rock 79-622 300 8 Mar 1989 6 71 ± 47 West Carrying Place Cove 79-775 300 10 Apr 1989 1 n/a Bald Rock 59-192 300 27 Nov 1989 2 213 ± 63 Old Soaker 59-304 300 27 Nov 1989 8 70 ± 32 Flat Island 79-621 320 1 Feb 2005 4 193 ± 73 Seaduck Rock 79-607 350 7 Dec 1990 2 175 ± 175 Big Nash Island 79-626 350 19 Apr 2005 3 230 ± 63 Green Island 79-929 400 12 Apr 2004 6 158 ± 80 Petit Manan Island 79-933 400 2 Dec 1993 20 62 ± 21 Little Cranberry I: Bar Point 59-313 400 8 Mar 1989 2 200 ± 200 Egg Rock 59-301 425 18 Jan 2004 8 128 ± 58 Shag Ledge near 79-763 450 4 Jan 2005 2 225 ± 225 Egg Rock 79-935 650 8 Jan 1990 9 131 ± 74 Mid-Coast region (see Fig. 2) Robinson Rock 63-341 250 17 Jan 1986 3 107 ± 74 Green Island Seal Ledges near 63-655 250 22 Apr 2002 2 125 ± 125 Ship and Barges Ledge near 59-341 300 27 Nov 1989 3 210 ± 38 Spirit Ledge 59-998 300 24 Feb 1991 14 108 ± 27 Green Ledge (Stinson Neck) 59-949 300 17 Jan 1986 5 124 ± 58 Shingle Island 59-959 300 17 Jan 1986 3 100 ± 100 Marsh Cove Ledges 63-220 300 27 Feb 1986 18 81 ± 17 Green Ledge 63-266 300 27 Feb 1986 12 38 ± 25 The Cow Pen 63-283 300 27 Feb 1986 23 93 ± 19 Smuttynose Island 59-931 300 20 Dec 1985 17 99 ± 19 Isle au Haut: Merchant Pt 63-230C 300 23 Nov 2001 56 47 ± 7 Isle au Haut: Trial Point 63-227 300 8 Apr 2004 21 139 ± 21 Metinic Green Island 63-585 300 13 Dec 2002 5 94 ± 58 Southern Mark Island 63-260 300 10 Dec 2003 3 183 ± 93 Shabby Island 59-996 350 5 May 2003 17 167 ± 28 Johns Island Dry Ledge 59-484 350 29 Nov 1998 5 151 ± 60 Dogfish Island 63-471 360 23 Dec 2003 12 92 ± 58 Green Ledge 63-493 400 7 Apr 2005 7 128 ± 62 Outer Scrag Ledge 63-205 400 28 Mar 2001 10 132 ± 38 Scraggy Island 59-836 400 7 Apr 2004 15 206 ± 42 White Horse 63-293 400 7 Feb 2005 6 200 ± 74 Hog Island Ledge 65-176 500 18 Mar 1983 2 250 ± 250 Heron Island 59-480 600 24 Feb 2004 20 136 ± 37 Mahoney Island 59-933 600 31 Mar 2004 15 173 ± 40 Scraggy Ledge 63-209 600 22 Dec 1985 3 208 ± 196 Burnt Ledge near 63-271 650 20 Dec 1985 5 209 ± 113 Black Ledge 59-482 700 10 Dec 2003 10 92 ± 68 Wheat Island 63-269 800 17 Jan 1986 3 267 ± 267 Brimstone Island 59-479 900 25 Nov 1998 13 99 ± 69 Southern region (see Fig. 3) White Bull 55-628 250 7 Dec 1979 3 108 ± 74 Biddeford Pool n/a 300 12 Dec 2000 90 61 ± 6 Pemaquid Point n/a 300 2 Feb 2000 15 89 ± 24 Richmond Island 55-579 600 7 Mar 2005 3 234 ± 152 Halfway Rock 55-502 1200 11 Apr 2007 3 640 ± 342 2013 G.H. Mittelhauser, L. Tudor, and B. Connery 225 (Fig. 3). The average flock observed contained 50 birds (n = 319, SE = 6.1); the largest flock observed contained 1200 birds and was observed on 11 April 2007 on Halfway Rock in Casco Bay (Fig 3). Our counts underestimated the actual number of birds because all sandpipers present during our surveys were not detected. Using a double-observer approach, we estimated that 56% of flocks were correctly detected and counted, 13% of flocks were not detected by the primary observers (all missed flocks had fewer than 10 birds), and 31% of flocks had 17% more birds in them than were recorded by the primary observers. Overall, we estimate that 17.9% (95% CI: 6.5–29.4) of the birds present during our surveys were not detected or counted. Expanding the 95% confidence interval for birds present but not detected during our surveys, we estimate that 14,000 to 17,000 Purple Sandpipers were present in Maine during the winters of 2002–2007 (Table 1). We calculated 2.3–3.8 Purple Sandpipers/km of surveyed shoreline throughout the entire coast of Maine calculated across all years of this study. Of the areas surveyed, the area from Vinalhaven to Petit Manan Point in the mid-coast region had the highest densities of birds. We identified 48 islands along the Maine coast that supported flocks of ≥250 of Purple Sandpipers: 14 locations in the eastern region, 29 locations in the midcoast region, and 5 locations in the southern region (Table 2). Discussion To estimate the number of Purple Sandpipers wintering along the coast of Maine, we conducted systematic surveys between 2002 and 2007, focusing our efforts on certain regions per winter season using a within-winter survey window spanning nearly five months. We typically observed birds among the more waveexposed islands and ledges, although we found some birds wintering in protected bays on muddy habitats, similar to the findings of Cayford and Waters (1996) during winter surveys in Great Britain. It is important to assess the degree of site fidelity and local movements of Purple Sandpipers, both within-winter and return rates during subsequent winters, to assess the validity of our regional population surveys and coastwide population estimate. Mittelhauser et al. (2012) found that the within-winter fidelity of Purple Sandpipers to particular sites in Maine was extremely high. The three most-visited sites of an individual over the course of the winter, which accounted for approximately 80% of all observations, averaged less than 2 km apart. Maximum within-winter movements of an individual of 25 km has been reported, although within-winter movements greater than 10 km does not appear to be a common occurrence, at least on the coast of Maine (Mittelhauser et al. 2012). Relocation of Purple Sandpipers from one location to another during subsequent winters is reported to be no greater than within-season fidelity (Atkinson et al. 1978, 1981; Mittelhauser et al. 2012). Of 285 banded birds in the UK, 78% moved 0–5 km, 15% moved 6–15 km, 6% moved 16–25 km, and 1% (2 birds) moved 26–40 km away from their original banding location during 226 Northeastern Naturalist Vol. 20, No. 2 subsequent winters (Atkinson et al. 1978). Similarly, Rehfisch et al. (2003) reported that of 254 birds marked in one winter in the UK, recaptures during the subsequent winter averaged 0.4 km from the original capture location. Because of this extreme fidelity in wintering Purple Sandpipers both within and between winters, count data compiled over multiple winters are unlikely to be substantially biased (Atkinson et al. 1978, 1981). Our efforts to focus our surveys to a particular section of coast each winter helped to minimize any bias in our population estimates because of bird movements. Annual fluctuations in population levels also can confound the interpretation of regional counts compiled over multiple years to generate a population estimate. Summers et al. (1975) reported that counts of Purple Sandpipers in coastal regions can vary by up to 25% between years, particularly in relation to the number of first-year birds present in the population. During our limited fall banding efforts on the coast of Maine from 2001 to 2007, the percent of juveniles captured were 7% (n = 42) during the winter of 2002–2003, 8% (n = 12) during 2004–2005, 16% (n = 25) during 2005–2006, and 33% (n = 16) during 2006–2007. Although our sample sizes are small during these capture efforts, the results suggest our counts during the winters of 2005–2006 and 2006–2007 may have been inflated because of the larger numbers of juvenile birds in the population. During these two winters, we surveyed the region south of Penobscot Bay along the southern coast and found relatively low densities of Purple Sandpipers. We suggest that during future surveys, the percent of juvenile birds in the population also be monitored annually. Purple Sandpipers can be difficult to detect and count when surveyed from a boat, especially during less than ideal wind, wave, and lighting conditions. Using a double-observer approach (Thompson 2002), we estimated that we did not detect 13% of Purple Sandpiper flocks during our surveys yet all undetected flocks had fewer than 10 birds. This finding corresponded with our field observations of this species that flocks with fewer than five birds can be difficult to point out to all members of our survey or banding crews. Flocks with more than 10 birds were typically readily visible to all members of our crew. Estimates of Purple Sandpipers per km of shoreline varied considerably along the Maine coast depending on the region considered in the estimate. For the entire coast of Maine, we estimated 2.3–2.8 birds/km of shoreline, which is considerably less than the 4.3 birds/km of shoreline reported for Troms County, Norway (Summers et al. 1990). Maximum bird densities on the Maine coast, however, were 4.5–5.5 birds/km of shoreline along the mid-coast region from Vinalhaven to Petit Manan Point. Morrison et al. (2001) estimated 16,000 birds wintering in all of eastern North America, based on an assessment of Christmas Bird Count data. Comparison of our counts with Maine Christmas Bird Count data clearly shows that the Christmas Bird Count is not a good indicator of wintering Purple Sandpiper abundance or distribution, at least in Maine. Maine has a high responsibility for this species because it supports a reasonably large proportion of the population in eastern North America; thus, a monitoring plan is needed for evaluating threats to habitat 2013 G.H. Mittelhauser, L. Tudor, and B. Connery 227 such as winter habitat loss from climate change and anthropogenic causes. This project provided baseline data on wintering Purple Sandpiper distribution and abundance in Maine, a necessary first step when developing an efficient and effective monitoring plan. The knowledge gained from this study will contribute to adaptive management strategies and conservation decisions for this wintering shorebird species in Maine. Acknowledgments We thank Brad Allen, Brian Benedict, Sarah Bockian, Yaniv Brandvain, D. Cadbury, Jason Czapiga, John Drury, Sam Edmonds, Chris Fichtel, Megan Gahl, Dave Hiltz, Donna Kausen, Jonathan Keller, Darren Kelly, Michael Langlois, Aaron Lewis, Marilee Lovitt, Beethany Murray, J. Naumann, Kipp Quinby, Greg Runge, Dave Schick, Linda Welch, Chris West, and MaryEllen Wickett for countless hours devoted to locating and recording Purple Sandpiper observations often during harsh winter weather conditions. Portions of this study were financially supported through grants from the National Park Service, Maine Department of Inland Fisheries and Wildlife, and the Maine Outdoor Heritage Fund. Acadia National Park and Petit Manan National Wildlife Refuge generously provided logistical support throughout the study. We thank Tom Hodgman for reviewing an earlier version of this manuscript and Jean-Pierre Savard for helping to improve the final manuscript. Literature Cited Atkinson, N.K., M. Davies, and A.J. Prater. 1978. The winter distribution of Purple Sandpipers in Britain. Bird Study 25:223–228. Atkinson, N.K., R.W. Summers, M. Nicoll, and J.J.D. Greenwood. 1981. Population, movements, and biometrics of the Purple Sandpiper, Calidris maritima, in eastern Scotland. Ornis Scandinavica 12:18–27. Burton, N.H.K., J. Blew, K. Colhoun, J. Cortes, B. Deceuninck, K. Devos, F. Hortas, L. Mendes, L. Nilsson, D. Radović, M.M. Rehfisch, M. van Roomen, C. Soldatini, O. Thorup, and D.A. Stround. 2008. Population status of waders wintering on Europe’s non-estuarine coasts. Pp. 95–101, In N.H.K. Burton, M.M. Rehfisch, D.A. Stroud, and C.J. Spray (Eds.). The European Non-estuarine Coastal Waterbird Survey. International Wader Studies 18. International Wader Study Group, Thetford, UK. Cayford, J.T., and R.J. Waters. 1996. Population estimates for waders Charadrii wintering in Great Britain, 1987/88–1991/92. Biological Conservation 77(1996):7–17. Cramp, S., and K.E.L. Simmons (Eds.). 1983. The Birds of the Western Palearctic. Volume 3. Oxford University Press, Oxford, UK. Howe, M., J. Bart, S.C. Brown, C.S. Elphick, R.E. Gill, B.A. Harrington, C. Hickey, R.I.G. Morrision, S. Skagen, and N. Warnock. 2000. A comprehensive monitoring program for North American shorebirds. Manomet Center for Conservation Sciences, Manomet, MA. Knight, O.W. 1908. The Birds of Maine. Charles H. Glass Company, Bangor, ME. 693 pp. Mittelhauser, G.H., L. Tudor, and B. Connery. 2006. Distribution and ecology of Purple Sandpipers wintering in the Acadia National Park region, Maine 2001–2004. NPS Technical Report NPS/NER/NRTR–2006/048. Boston, MA. Mittelhauser, G.H., L. Tudor, and B. Connery. 2012. Within-year movements and site fidelity of Purple Sandpipers during the nonbreeding season. Journal of Field Ornithology 83:32–40. 228 Northeastern Naturalist Vol. 20, No. 2 Morrison, R.I.G., R.E. Gill, Jr., B.A. Harrington, S. Skagen, G.W. Page, C.L. Gratto- Trevor, and S.M. Haig. 2001. Estimates of shorebird populations in North America. Canadian Wildife Service Occasional Paper, Number 104. Ottawa, ON, Canada. Palmer, R.S. 1949. Maine Birds. Bulletin of the Museum of Comparative Zoology, Cambridge, MA. 656 pp. Payne, L.X., and E.P. Pierce. 2002. Purple Sandpiper (Calidris maritima). No. 706, In A. Poole and F. Gill (Eds.). The Birds of North America. The Birds of North America, Inc., Philadelphia, PA. Rehfisch, M.M., H. Insley, and B. Swann. 2003. Fidelity of overwintering shorebirds to roosts on the Moray Basin, Scotland: Implications for predicting impacts of habitat loss. Ardea 91:53–70. Summers, R.W., N.K. Atkinson, and M. Nicoll. 1975. Wintering wader populations on the rocky shores of eastern Scotland. Scottish Birds 8:299–308. Summers, R.W., K-B Strann, R. Rae, and J. Heggas. 1990. Wintering Purple Sandpipers, Calidris maritime, in Troms County, northern Norway. Ornis Scandinavica 21:248–254. Summers, R.W., M. Nicoll, and W. Peach. 2001. Numbers, migration phenology and survival of Purple Sandpipers, Calidris maritima, at Gourdon, Eastern Scotland. Bird Study 48:139–46. Summers, R.W., L.G. Underhill, and A. Simpson. 2002. Habitat preferences of waders (Charadrii) on the coast of Orkney Islands. Bird Study 49:60–66. Thompson, W.L. 2002. Towards reliable bird surveys: Accounting for individuals present but not detected. Auk 119:18–25.