Northeastern Naturalist 730 R.A. McKinney, K.B. Raposa, and C.L. Trocki 22001155 NORTHEASTERN NATURALIST 2V2(o4l). :2723,0 N–7o4. 54 Status and Distribution of Wintering Waterfowl in Narragansett Bay, Rhode Island, 2005–2014 Richard A. McKinney1,*, Kenneth B. Raposa2, and Carol L. Trocki3 Abstract - Surveys of wintering waterfowl can aid in both identifying estuarine habitats currently being used by species of conservation concern so that the sites can be targeted for protection and restoration, and in providing a baseline assessment from which the effects of future changes in wintering habitat can be assessed. In an effort to better understand the local distribution of wintering waterfowl during the period 2005–2014, we undertook a study of waterfowl abundance and distribution in Narragansett Bay, RI, a moderate-sized estuary located in the northeastern US within the Atlantic Flyway. Overall waterfowl abundance in the Bay ranged from 15,002 individuals in 2006 to 26,163 individuals in 2010 and averaged 20,062 ± 3393 individuals over the 10-y period. Species richness ranged from 1.80 to 10.8 per site; most of the sites with high species richness were located in the Upper Bay. Based on our counts from 67 ground locations, the Narragansett Bay waterfowl community was dominated by Aythya affinis (Lesser Scaup) and A. marila (Greater Scaup), Branta bernicla (Brant), and Branta canadensis (Canada Geese) over the survey period. Waterfowl-community composition indicated that the Upper Bay, an environment characterized by low wave-energy, shallow coves, sheltered embayments, and salt marshes, supported mostly dabbling ducks, geese, and swans. The Lower Bay, an environment characterized by higher wave-energy, rocky shorelines, and deeper open-water habitats, supported mostly sea ducks and other diving-duck species. Abundance over the survey period was relatively stable, and observed patterns of waterfowl distribution suggest that conservation actions to maintain shallow-water habitats, including efforts to protect and restore salt marsh habitat, will help to maintain resources needed by many of the waterfowl species wintering in the Bay. Introduction Estuaries in temperate climates contain a variety of habitats used by waterfowl (Anseriformes: ducks, geese, and swans), particularly in the non-breeding or wintering portion of their life cycle. Winter is a period when obtaining adequate food and cover is crucial to waterfowl survival and fitness, and the availability of suitable winter habitats throughout a species’ range may impact population dynamics (Weller 1988). Estuaries located on the east coast of the US, from Maine to the mid-Atlantic region, host substantial wintering populations of sea ducks, (members of the tribe Mergini), dabbling ducks, and diving ducks (Bellrose 1980; Gordon et al. 1989, 1998). Knowing the abundance and distribution of wintering waterfowl in specific habitats within these estuaries will help inform rangewide conservation and 1US Environmental Protection Agency, Office of Research and Development, National Health and Environmental Effects Research Laboratory, Atlantic Ecology Division, 27 Tarzwell Drive, Narragansett, RI 02882. 2Narragansett Bay National Estuarine Research Reserve, Prudence Island, RI 02872. 3Department of Natural Resources Science, University of Rhode Island, Kingston, RI 02881. *Corresponding author - mckinney.rick@epa.gov. Manuscript Editor: Noah Perlut Northeastern Naturalist Vol. 22, No. 4 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 731 management, particularly because many of these estuaries are heavily populated by humans, and waterfowl in these areas may be subject to both direct (e.g., hunting and recreation) and indirect (e.g., development and activity in surrounding areas) human disturbance. Coastal areas near urban centers often provide habitat for wintering waterfowl, but more study is needed because they may be subject to significant pressure from urbanization and human disturbance, which in turn may affect waterfowl distribution. Several studies have examined food habits of wintering waterfowl in New Hampshire estuaries (Stott and Olsen 1974), Jamaica Bay in New York (Burger et al. 1984), and coastal New Jersey (Tiedemann 1984), but they generally did not report overall waterfowl distribution or look at long-term patterns of utilization. However, Perry and Deller (1996) examined long-term distribution patterns of waterfowl in Chesapeake Bay and concluded that habitat degradation resulted in a decline in numbers of most duck species and a change in distribution of some species. The US Fish and Wildlife Service, Division of Migratory Bird Management, along with state wildlife agency biologists, conducts an annual mid-winter waterfowl survey of waterfowl along the Atlantic Flyway, which provides an aerial assessment of estuarine habitats. Several investigators have examined larger-scale (i.e., east-coast US) waterfowl distribution patterns using these data (e.g., Silverman et al. 2013, Zipkin et al. 2010); however, data from the mid-winter surveys may not be appropriate for examining local distribution and specific patterns of waterfowl habitat utilization (Eggeman and Johnson 1989). In this paper, we report results to date of an ongoing annual ground survey, initiated in 2005 by a group of biologists from local, state, and federal wildlife and environmental agencies, of waterfowl wintering in Narragansett Bay, RI, a moderate-sized urban east-coast estuary. The Narragansett Bay Winter Waterfowl Survey was completed in January each year from 2005–2014 by 8 teams composed of 2–4 observers who surveyed waterfowl at 67 locations throughout the Bay (NBWWS 2015). Each survey team was composed of scientists from participating organizations and led by a professional biologist, wildlife ecologist, or environmental scientist with experience identifying coastal bird species including waterfowl. The objectives of the survey were to (1) provide a continuing record of waterfowl abundance and distribution in Narragansett Bay and (2) supplement annual aerial surveys of Narragansett Bay carried out by staff of the Rhode Island Division of Fish Wildlife as part of the Mid-winter Waterfowl Survey. Here we examine survey data to document waterfowl distribution and discern specific patterns of habitat utilization to help provide insight into proximal factors influencing habitat use. We also summarize changes in waterfowl abundance for 20 species over the 10-year period, which is information that could help develop local conservation strategies for wintering waterfowl in the Bay. Field-site Description We selected 67 land-survey sites within 6 larger survey sections in Narragansett Bay (the Bay), a ~350-km2 urban estuary situated totally within Rhode Island. The Northeastern Naturalist 732 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 Vol. 22, No. 4 Bay extends ~40 km from its head at the mouth of the Blackstone River to the Atlantic Ocean south of Aquidneck Island, RI (Fig. 1). The head of the estuary is surrounded by the city of Providence (population ~178,000) and the bottom of the Bay is bounded by Sachuest Point on the east and Narragansett Point on the west. Figure 1. Narragansett Bay Winter Waterfowl Survey sections as color-coded areas with delineated site boundaries. The black circles represent the approximate observation locations for the surveys. Northeastern Naturalist Vol. 22, No. 4 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 733 Several other urban centers of lower population density line the shore of the estuary, with generally decreasing population density from head to mouth. We classified our sites by habitat type: coves, i.e., areas with narrow, restricted entrances including small, narrow, sheltered bays, inlets, creeks, or recesses in the coastline (0.3–50 ha area, n = 31); and open water sites, which included rocky headlands (n = 36). We also classified sites based on geographic location in the Bay. We considered sites within survey sections 2 and 3 (n = 24) to be Upper-Bay sites; they extended from the head of the Bay south to 18 km from the mouth of the Blackstone River (Fig. 1). The Upper Bay is an environment characterized by low wave-energy, shallow coves, sheltered embayments, and salt marshes. We designated as Mid Bay all sites within section 4 (18–26 km from the mouth of the Blackstone River; n = 8); they consisted primarily of open-water habitats.We classified sites within sections 1, 5, and 6 (n = 35) as Lower-Bay sites; these sites were located 26–40 km from the mouth of the Blackstone River. The Lower Bay is an environment characterized by higher wave-energy, rocky shorelines and deeper open-water habitats. We categorized a total of 36.8% of the sites in the Lower- and Mid-Bay sections as coves, whereas 65.2% of the sites in the Upper Bay were coves. All survey sites were public access points and we selected them to provide as complete coverage of the Bay as possible. Methods Observers used direct counts to record all waterfowl present at a site at the time of the surveys, all of which began at or after 0730 and ended by 1645. Researchers carried out the counts during a single day in early January each year from 2005 to 2014, scanning the entire area to be surveyed (i.e., cove or embayment) with binoculars and/or a spotting scope from a stationary point and counting all birds on the water surface plus those on the shore up to 50 m inland from the shoreline. Most sites required 10–20 min to survey. When the observers encountered large flocks of greater than 100 birds, they made estimates by counting in groups of 10 or 100; birds flying over a site were not counted. Observers also recorded weather conditions and evidence of human disturbance at each site, but we do not report those data here. We used survey data to generate abundance values for waterfowl. We determined abundance values as mean ± standard deviation and calculated relative standard deviation as (standard deviation/mean) x 100%. We figured species richness as the total number of species observed at a site. We aggregated closely related species for analysis, e.g., we report Aythya marila L. (Greater Scaup) and Aythya affinis Eyton (Lesser Scaup) abundances as scaup (Greater Scaup predominated in Narragansett Bay; P. Paton, University of Kingston, Kingston, RI, pers. comm.), and Melanitta americana L. (Black Scoter), Melanitta perspicillata L. (Surf Scoter), and Melanitta fusca L. (White-winged Scoter) abundances as scoters. We employed Student’s t-tests (PROC TTEST; SAS Institute, Carey, NC) to evaluate differences in waterfowl densities between estuaries and in species richness between habitat types. We used cluster analysis (CLUSTER, combined with a similarity-profile test, Northeastern Naturalist 734 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 Vol. 22, No. 4 SIMPROF) and similarity percentages (SIMPER) to compare winter waterfowlcommunities among the 6 survey sections and among the regions of the Bay. We employed CLUSTER to group survey sections based on the similarity of their waterfowl communities, SIMPROF to identify resultant section groups whose communities were statistically the same, and SIMPER to identify species that typified individual sections and regions as well as species that were most responsible for community dissimilarity among sections and between the 2 regions. Prior to each analysis, we square-root–transformed all data to dampen the weights of the most abundant species. We conducted each of these community analyses in PRIMER version 6.1.2 (Clarke and Gorley 2006). Results Waterfowl abundance Overall waterfowl abundance across the 67 survey sites in Narragansett Bay averaged 20,062 ± 3393 individuals from 2005 to 2014, with a relative standard deviation of 16.9%. Abundance ranged from 15,002 individuals in 2006 to 26,163 individuals in 2010 (Table 1 [on following page]). Scaup were the most frequently observed species in the Bay, and 5 of the top 10 most-abundant species were diving ducks (Table 2). Upper-Bay sections had the 2 highest mean overall abundances over the survey period, and accounted for over 60% of the waterfowl observed in the Bay (Table 3). The surveys found >60% of the scaup, Branta canadensis L. (Canada Geese), and Branta bernicula (L.) (Brant) in the Upper Bay, though most of the Somateria mollissima L. (Common Eider) were in Lower-Bay sections. Sea ducks averaged 5014 ± 658 individuals in the Bay, most of which were observed in the Lower Bay. Scaup had the highest proportional abundance throughout the survey period, with Brant and Canada Geese alternating between 2nd- and 3rd-most proportionately abundant. Scoters (r2 = 0.41, P = 0.05), Bucephala albeola L. (Bufflehead) (r2 = 0.41, P = 0.05), Lophodytes cucullatus L. (Hooded Merganser) (r2 = 0.40, P = 0.05), and Anas strepera L. (Gadwall) (r2 = 0.78, P < 0.001) abundances significantly Table 2. Percent distribution of the top 10 most-abundant species observed at the survey sites across the 6 survey sections. Survey Section Species 1 2 3 4 5 6 Scaup 0.3 25.3 61.7 8.8 0.1 3.8 Canada Goose 6.3 35.9 25.9 3.3 11.6 17.0 Brant 3.1 40.3 33.1 7.0 11.7 4.9 Common Eider 30.6 0.0 0.0 1.5 10.2 57.8 Common Goldeneye 9.6 14.1 22.4 11.7 22.2 20.0 American Black Duck 10.3 36.1 13.6 19.9 7.4 12.8 Bufflehead 21.4 17.6 15.8 4.6 26.4 14.2 Mallard 20.3 31.6 17.6 0.4 18.5 11.6 Red-breasted Merganser 20.2 26.8 19.3 6.3 5.1 22.3 Mute Swan 3.8 74.8 19.3 0.1 1.7 0.3 Northeastern Naturalist Vol. 22, No. 4 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 735 Table 1. Abundance by year and mean abundance (number of individuals ± standard deviation) of waterfowl species averaging at least 1.0 individual across the 67 survey sites in Narragansett Bay for the survey period 2005–2014. Species1 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Mean abundance Brant 1434 3808 2733 1853 2674 4004 2313 2020 2290 1321 2445 ± 899 Canada Goose 4008 1520 1179 2307 4882 2665 2565 2778 2545 3705 2815 ± 1118 Mute Swan 677 631 388 441 775 758 1128 198 526 499 602 ± 256 Wood Duck 0 0 0 0 11 0 0 0 0 0 1 ± 3 Gadwall 61 56 98 85 106 187 127 183 161 181 125 ± 51 American Wigeon 123 357 810 319 219 290 437 331 296 242 342 ± 185 American Black Duck 1474 1174 1199 1418 1358 1033 1113 1063 1097 738 1167 ± 215 Mallard 1478 368 616 935 1358 752 838 1050 766 1072 923 ± 333 Canvasback 1 0 0 2 19 12 0 0 0 0 3 ± 7 Scaup spp. 8257 2765 6712 7551 6253 11240 3106 5815 4554 9986 6624 ± 2759 Common Eider 2465 1911 987 726 1508 1247 1226 713 583 1090 1246 ± 585 Harlequin Duck 66 36 57 97 69 84 53 74 68 52 66 ± 17 Scoter spp. 135 151 301 318 262 546 558 1395 283 800 475 ± 383 Long-tailed Duck 1 0 0 5 3 0 1 0 0 2 1 ± 2 Bufflehead 470 625 738 1328 1530 1028 1608 1417 1143 1133 1102 ± 388 Common Goldeneye 849 834 1400 2142 1364 1263 1174 1000 1123 873 1202 ± 390 Barrow’s Goldeneye 0 0 1 0 0 0 0 2 6 2 1 ± 2 Hooded Merganser 70 45 187 132 171 337 296 240 177 267 192 ± 95 Common Merganser 0 0 0 0 23 57 206 63 0 1 35 ± 65 Red-breasted Merganser 427 721 1049 712 585 660 922 813 780 270 694 ± 227 Total waterfowl2 21,996 15,002 18,455 20,371 23,170 26,163 17,671 19,155 16,398 22,234 20,062 ± 3393 1Scientific names not provided in text: Cygnus olor (Gmelin) (Mute Swan), Anas sponsa (L.) (Wood Duck), Anas americana (Gmelin) (American Wigeon), Aythya valisineria (A. Wilson) (Canvasback), Histrionicus histrionicus (L.) (Harlequin Duck), Clangula hyemalis (L.) (Long-tailed Duck), Bucephala islandica (Gmelin) (Barrow’s Goldeneye), Mergus merganser L. (Common Merganser), Mergus serrator L. (Red-breasted Merganser). 2Additional species observed: Somateria spectabilis (L.) (King Eider); Anas acuta L. (Northern Pintail); Aythya americana (Eyton) (Redhead); Anas crecca L. (Green-winged Teal); Anas penelope L. (Eurasian Wigeon); Aytha collaris (Donovan) (Ring-necked Duck); Oxyura jamaicensis (Gmelin) (Ruddy Duck) Northeastern Naturalist 736 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 Vol. 22, No. 4 increased with survey year (Fig. 2). Scoter abundance was comprised of 73.3% Black Scoter, 20.3% Surf Scoter, and 6.3% White-winged Scoter. Abundances of Common Eider (r2 = 0.46, P = 0.03), and Anas rubripes Brewster (American Black Duck) (r2 = 0.60, P = 0.01) significantly decreased with survey year (Fig. 2). The mean density of waterfowl in Narragansett Bay for the period from 2009 to 2013 was 54.7 ± 10.8 individuals km2. Waterfowl species richness Species richness ranged from 1.8 at the Prudence Island lighthouse to 10.8 at Sachuest Point (Table 4). Mean per-section species richness ranged from 3.84 ± 0.49 for section 4 (Mid Bay) to 6.10 ± 0.44 for section 2 in the Upper Bay. Sections 2 and 3 (6.05 ± 0.50 species) in the Upper Bay had significantly higher mean species richness than section 4 (t-test: t = 2.12, df = 16, P = 0.002; and t = 2.12, df = 16, P = 0.004, respectively), and richness in section 2 was significantly higher than section 1 (4.10 ± 0.55 species; t-test: t = 2.08, df = 21, P = 0.005). Seven of the 10 sites with the highest mean species richness were located in the Upper Bay, while 9 of the 10 sites with the lowest mean species richness were located in either in the Lower Bay or the Mid Bay. We classified 6 of the 10 sites with the highest Table 3. Mean abundance (number of individuals ± standard deviation) of waterfowl observed across the 6 survey sections in Narragansett Bay for the survey period 2005–2014. 1 = Narragansett, 2 = West Bay, 3 = East Bay, 4 = Prudence, 5 = Sakonnet, and 6 = Aquidneck. Sections 2 and 3 were in the Upper Bay, 4 was in the Mid Bay, and 1, 5, and 6 were in the Lower Bay. Survey Section Species 1 2 3 4 5 6 Common Eider 381 ± 427 0 ± 0 0 ± 0 18 ± 33 127 ± 243 720 ± 363 Harlequin Duck 0.3 ± 1 0 ± 0 0 ± 0 0 ± 0 4 ± 4 61 ± 17 Long-tailed Duck 0.5 ± 2 0 ± 0 0 ± 0 0 ± 0 0 ± 0 0.7 ± 1 Scoters 96 ± 205 1 ± 4 11 ± 34 3 ± 4 11 ± 11 354 ± 283 Common Goldeneye 115 ± 50 169 ± 151 269 ± 143 141 ± 58 267 ± 109 241 ± 96 Barrow’s Goldeneye 0.2 ± 0.4 0.2 ± 0.6 0.1 ± 0.3 0.2 ± 0.4 0 ± 0 1 ± 1 Bufflehead 236 ± 132 195 ± 95 174 ± 112 50 ± 36 291 ± 133 157 ± 73 Hooded Merganser 28 ± 27 77 ± 56 73 ± 56 12 ± 13 0.8 ± 2 2 ± 3 Red-breasted 5 ± 12 20 ± 53 11 ± 19 0 ± 0 0.1 ± 0.3 0 ± 0 Merganser Common Merganser 5 ± 62 20 ± 136 11 ± 66 0 ± 19 0.1 ± 24 0 ± 81 Scaup 157 ± 37 1833 ± 2397 3485 ± 3088 628 ± 1732 43 ± 19 370 ± 204 Canvasback 0.2 ± 1 0.2 ± 1 1 ± 3 0.2 ± 2 0 ± 0 0 ± 0 American Black 120 ± 114 421 ± 181 158 ± 81 233 ± 84 86 ± 52 150 ± 112 Duck Mallard 188 ± 154 292 ± 158 162 ± 68 3 ± 3 171 ± 90 107 ± 80 American Wigeon 0.4 ± 1 245 ± 144 73 ± 68 1 ± 3 6 ± 9 17 ± 22 Gadwall 5 ± 12 66 ± 40 15 ± 27 1 ± 3 14 ± 23 24 ± 21 Wood Duck 0.2 ± 2 0 ± 0 0 ± 0 0.2 ± 2 0 ± 0 1 ± 1 Mute Swan 23 ± 29 450 ± 221 117 ± 63 0.4 ± 1 10 ± 11 2 ± 3 Canada Goose 176 ± 138 1012 ± 415 730 ± 294 94 ± 121 325 ± 285 479 ± 313 Brant 75 ± 75 985 ± 465 809 ± 634 172 ± 129 286 ± 210 119 ± 118 Total 1611 ± 560 5786 ± 2513 6099 ± 3176 1359 ± 1745 1642 ± 476 2806 ± 613 Northeastern Naturalist Vol. 22, No. 4 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 737 species richness as coves, and 8 of the 10 sites with the lowest species richness as open-water sites. Mean species richness was greater in coves (5.75 ± 1.38 species) than at open-water sites (4.92 ± 2.02 species; t-test: t = 1.67, df = 62, P = 0.03). Waterfowl distribution The waterfowl-community composition of sections in the Upper Bay (sections 2 and 3) was similar, and different than those in the Mid (section 4) and Lower (sections 1, 5, and 6) Bay (Fig. 3), respectively. Canada Geese were ubiquitous, but numbers varied throughout the Bay; however, this species was common to waterfowl communities at all survey sections. Upper-Bay waterfowl communities Figure 2. Annual changes in abundance for waterfowl species that significantly increased or decreased in Narragansett Bay during the survey period from 2005 to 2014: (A) Gadwall, (B) scoter, (C) Bufflehead, (D) Hooded Merganser, (E) American Black Duck, (F) Common Eider). Note that the scales for plots (A) and (D) dif fer from (B), (C), (E), and (F). Northeastern Naturalist 738 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 Vol. 22, No. 4 Table 4. Mean waterfowl-species richness at 67 survey sites across the 6 survey sections in Narragansett Bay for the survey period 2005–2014. Geographic location within the Bay: U = Upper-Bay sites located in survey sections 2 and 3, M = Mid-Bay sites located in section 4, and L = Lower-Bay (L) sites located in sections 1, 5, and 6. C = coves or areas off the larger bay with narrow, restricted entrances including small, narrow, sheltered bays, inlets, creeks, or recesses in the coastline; O = open water sites. [Table continues on following page]. Site Section Location Habitat Species richness Sachuest NWR 6 L O 10.20 Bullock’s Cove 3 U C 8.60 Apponaug Cove 2 U C 8.30 Bristol Harbor 3 U C 8.10 Lower Providence River 2 U O 7.70 Gaspee Point 2 U O 7.60 Sabin Point 3 U O 7.30 Narragansett Central 1 L C 7.30 Sakonnet Point 5 L O 7.20 Warwick Cove 2 U C 7.20 Wickford Harbor 1 L C 7.20 Providence River NW 6 L O 7.00 Brenton Point 6 L O 7.00 Colt State Park 3 U O 6.70 Jamestown West 6 L O 6.50 Greenwich Cove 2 U C 6.50 Kikemuit River 3 U C 6.40 Brush Neck Cove 2 U C 6.40 Sandy Point 6 L O 6.30 Watchemoket Cove 3 U C 6.20 Beavertail 6 L O 6.10 Makeral/Sheffield Coves 6 L C 6.10 Bissel Cove 1 L C 6.10 Newport East 6 L O 5.80 Jamestown East 6 L O 5.80 Fort Adams 6 L O 5.70 Portsmouth Cove 5 L C 5.70 Sapowet 5 L C 5.70 Warren River 3 U C 5.70 Fogland Point 5 L O 5.60 Upper Barrington River 3 U C 5.60 Potowomut 2 U C 5.60 Passeonkquis 2 U C 5.60 McCorrie Point 6 L O 5.50 Hull Cove 6 L C 5.40 Greenwich Bay North 2 U O 5.40 Pawtuxet Cove 2 U C 5.40 Nannaquacket 5 L C 5.30 T-dock 4 M O 5.30 Coggeshal Cove 4 M C 5.30 Sakonnet River/Island Park 5 L C 5.20 Mary Donovan 5 L C 5.20 Plum Point 1 L O 5.20 Tiverton Harbor 5 L C 5.10 Potter Cove 4 M C 5.10 Northeastern Naturalist Vol. 22, No. 4 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 739 were also characterized by Brant, scaup, and American Black Duck, while Lower- Bay communities were characterized by Bufflehead, Bucephala clangula L. (Common Goldeneye) and Common Eider (Table 3). Differences in abundance of Brant, scaup, and Canada Geese were primarily responsible for observed differences between Upper- and Lower-Bay waterfowl communities, as well as those between the Mid-Bay section and the Upper Bay (Table 3). Differences in abundance of Common Eider, scaup, and Anas platyrhynchos L. (Mallard) were primarily responsible for observed differences between Lower- and Mid-Bay waterfowl communities. Table 4, continued. Site Section Location Habitat Species richness Bullock’s Point 3 U C 4.80 Upper Providence River 2 U O 4.70 Prudence West 4 M O 4.50 Bonnet Point 1 L O 4.40 Jamestown North 6 L O 4.10 J.L. Lewis Park 3 U C 3.90 Newport Harbor 6 L C 3.80 Narragansett South 1 L O 3.60 Ferry Landing 4 M O 3.50 Sakonnet River NW 5 L O 3.30 Barrington Beach 3 U O 3.20 Narragansett North 1 L O 3.00 Point Judith 1 L O 2.90 Mount View 2 U O 2.80 Allen Harbor 1 L C 2.80 Nag Pond 4 M C 2.60 Providence Point 4 M O 2.60 Quonset Point 1 L O 2.30 Davisville 1 L O 2.30 Arnold Point 6 L O 2.10 Casey Point 1 L O 2.10 Prudence Lighthouse 4 M O 1.80 Figure 3. Results of cluster analysis of winter waterfowl communities in the 6 survey sections in Narragansett Bay during the period from 2005 to 2014. Northeastern Naturalist 740 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 Vol. 22, No. 4 Discussion Ten years of abundance data depicted a winter waterfowl-community in Narragansett Bay dominated by scaup; 13 additional species had mean abundances greater than 100 individuals in each of the 10 years. Diving ducks (sea ducks and scaup), dabbling ducks (Anas spp. and Aix spp.), geese, and swans were similarly abundant in the Bay, which may be a reflection of the equal availability of their preferred habitats (McKinney 2004). Studies of food habits and habitat use by wintering waterfowl in northeastern US estuaries (e.g., Huang 2010, Stott and Olsen 1974) have demonstrated a close association of species with preferred foraging habitats. Sea ducks appeared to exhibit foraging-habitat preferences in Narragansett Bay because we most-often observed them in the Lower Bay in association with rocky coastlines and extensive open water, and dabbling ducks occurred mostly in the Upper Bay in shallow, salt marsh- dominated coves and sheltered embayments. Waterfowl were more abundant in the Upper Bay, which may have have been a result of higher wintering-population sizes of species that utilize these habitats, from the absence of hunting activity in the Upper Bay, or the shelter from wind and waves afforded by the shallow-water embayments. Rhode Island state hunting regulations prohibit discharge of a firearm within 152.4 m (500 ft) of an occupied dwelling; dense residential and commercial development in close proximity to the shoreline renders most of the Upper Bay coastal areas off-limits to hunting. Waterfowl abundance is known to decrease in areas subject to hunting disturbance (Evans and Day 2002, Madsen 1998, Owen 1993). Hunting disturbance was previously found to be the most important factor determining waterfowl distribution along a human-disturbance gradient in Narragansett Bay (McKinney et al. 2006). The mean wintering population of scoters, a group of species whose abundance increased during the survey period in Narragansett Bay, represented 1.4% of the estimated mean wintering population of 33,510 ± 17,956 individuals reported within the Atlantic Flyway (USFWS 2014b). We saw an increase in the abundance of scoters wintering in Narragansett Bay, and an increase in abundance over the same time period was also observed in Boston Harbor, a smaller urban estuary located 90 km to the northeast that supported a similar assemblage of waterfowl species as we observed in Narragansett Bay (TASL 2014). However, there were no clear trends in scoter abundance over the survey period for either the 3 states immediately to the south of Narragansett Bay (Connecticut, New York, New Jersey), or the 2 states to the north of Boston Harbor (New Hampshire and Maine) (USFWS 2014b). Increases in scoter abundance in Narragansett Bay and Boston Harbor may reflect changes in local habitat quality or environmental conditions, or perhaps a shift in distribution within the flyway; however, further study is needed to determine which of these factors may be driving patterns of scoter distribution. Wintering populations of the 3 other species whose abundance in Narragansett Bay increased during the survey period (Gadwall, Bufflehead, and Hooded Merganser) represented less than 1% of the Flyway population; no discernable trends in their population sizes in the Flyway were noted over the period corresponding to our survey (USFWS 2014b). Northeastern Naturalist Vol. 22, No. 4 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 741 Common Eider showed a corresponding decline in average wintering abundance in the Atlantic Flyway similar to what we observed in Narragansett Bay over the same time period (USFWS 2014b). The decline we observed may reflect these regional trends, which in turn, may to some extent be influenced by ongoing impacts of avian cholera, which continues to affect Common Eider on their breeding grounds (Environment Canada 2014). There were also reports of an avian influenza affecting Common Eider wintering in southern New England during the latter part of the survey period (Fraser 2012). This disease may have contributed to declines, although it’s not clear to what extent it is impacting the wintering population. Declines in American Black Duck could also be attributed to changes in regional population size, shifts in distribution of wintering birds across available locations, or changes in habitat quality or local environmental conditions. However, without documentation of changes in these factors over the survey period, it is impossible to attribute observed trends to any particular factor. The differences we observed in species richness in the Upper, Mid, and Lower Bay may have been due to the distribution of habitat types at specific sites, for example, Upper-Bay sites were predominantly classified as coves, while Mid- and Lower-Bay sites were predominantly open water. Cove sites in the Bay had significantly higher species richness, and this may, in part, result from the variety of different habitat types—including sheltered embayments with a variety of vegetation types including salt marshes, open shoreline, and shallow open water—present in these relatively low-wave–energy environments. Several studies have reported that habitat heterogeneity is an important determinant of wintering waterfowl species richness (Erwin 1996, Ma et al. 2007, Perry et al. 2007). Lower-Bay sites were predominantly hardened shorelines with fewer shallow-water habitats and more open-water habitats, which tended to be deeper and therefore may have been less favorable for foraging, particularly for dabbling ducks. Differences in the overall amount of shoreline may also influence waterfowl species richness. Suter (1994) found that shoreline length was a primary determinant of wintering waterfowl species richness in lakes in Switzerland, along with availability of distinct habitat types characterized by water depth and prey availability. Our results suggest the Narragansett Bay waterfowl community was characterized by scaup (33.0%), Canada Goose (14.0%), and Brant (12.2%). Scaup are listed as species of high conservation concern within the New England/mid-Atlantic coast Bird Conservation Region (BCR 30) (Steinkamp 2005), and scaup in Narragansett Bay represent 3.2% of those reported in the Atlantic Flyway during the MWS (USFWS 2014b). Their abundance in the Bay is currently lower than in the mid-1900s when more than 20,000 birds were reported (D. Ferrin, Berkshire Community College, Pittsfield, MA, unpubl. manuscript); this trend mirrors declines in the North American breeding population documented during the latter half of the 20th century (Afton and Anderson 2001). Contaminant effects, lower female survival, and reduced recruitment linked to variability in food resources have been proposed as possible factors contributing to the decline of the species (Cohen 1998); potential exposure to contaminants may be a concern in urban estuaries such Northeastern Naturalist 742 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 Vol. 22, No. 4 as Narragansett Bay (Austin et al. 2000). Brant in Narragansett Bay are listed as a species of highest conservation concern in BCR 30, and based on MWS data, the Narragansett Bay wintering population represents 2.2% of the total breeding population (USFWS 2014a). Habitats used by these species, including shallow coves, salt marshes, open shoreline, shallow open water, and small, sheltered embayments, should be considered for protection and restoration. Within Narragansett Bay, the Upper Bay supported mainly dabbling ducks, geese, and swans. In contrast, the Lower Bay mainly supported sea ducks that are better adapted to these high-wave–energy habitats. In our surveys, scaup were unique in that they tended to congregate in large flocks in open-water habitat in the Upper Bay. This result is somewhat contradictory to reports of food habits which suggest that wintering scaup eat primarily aquatic vegetation (Jones and Drobney 1986, Thompson et al. 1988), and although they dive for their food, vegetation would presumably be sparse in open-water habitats. However, studies of wintering scaup in Connecticut estuaries suggested the possibility of a diet shift and documented that bivalves and snails accounted for significant portions of their diet (Cohen 1998, Cronan 1957, Eccleston 1999). These findings are consistent with our observations of scaup using open-water habitats in the Bay, although we did not include any behavioral observations in this study that could confirm that scaup were feeding in the areas where we saw them. The mean density of waterfowl in Narragansett Bay during the latter half of the survey period (2009–2013) was lower than that reported for Boston Harbor (92.9 ± 9.7 individuals km-2; t = 2.62, df = 9, P = 0.002; TASL 2014). However, it is important to note that our abundance data, as well as those from surveys in Boston Harbor, did not include estimates of detection probability and hence can only be considered an index of abundance for each location. Quantitative comparison between estuaries is not possible, but other studies have suggested latitude, or proximity to northern waterfowl-breeding grounds, and orientation within the Flyway may influence the long-term average number of wintering waterfowl in a given water body (e.g., Zipkin et al. 2010). A multitude of local factors may influence wintering-waterfowl abundance and habitat utilization in estuaries throughout the range of a species including: (1) breeding success (changes in annual recruitment levels); (2) mortality rates (changes in annual survival, hunting mortality); and (3) climate change (range expansion and contraction). Unfortunately we cannot evaluate any of these largerscale drivers of population abundance with the abundance data we collected. Our survey results suggest that conservation actions to maintain shallow-water habitats in the Bay, including efforts to protect and restore salt marsh habitat, will help to provide resources needed by wintering-waterfowl species. Similarly, efforts to protect Lower-Bay shorelines from extensive development will help minimize disturbance effects on sensitive species that utilize these areas. Bay-wide winterwaterfowl monitoring should continue to evaluate overall waterfowl abundance and species trends as species and habitats continue to adapt to changes in climate and land use. Specific factors such as food resources and specific behavior of wintering species should also be investigated. Northeastern Naturalist Vol. 22, No. 4 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 743 Acknowledgments The concept for the Narragansett Bay Winter Waterfowl Survey originated with Scott McWilliams and Peter Paton; they were instrumental in its development and implementation. Many thanks also to those who coordinated surveys of the various sections over the years, including W. Berry, M. Chintala, T. Gleason, H. Hopkins, R. Kenney, W. Munns, B. Sherman, and K. Vigness-Raposa. We are grateful to the many people who participated in the survey during the 10 years since its inception who are too numerous to mention. We thank A. Oczkowski, R. Pruell, and C. Glinka for their reviews and comments on earlier versions of the manuscript. Mention of trade names or commercial products does not constitute endorsement or recommendation. Although the research described in this article has been funded wholly by the US Environmental Protection Agency, it has not been subjected to Agency-level review; therefore, it does not necessarily reflect the views of the Agency. This is ORD Tracking Number ORD-0011584 of the Atlantic Ecology Division, National Health and Environmental Effects Research Laboratory, Office of Research and Development, US Environmental Protection Agency. Literature Cited Afton, A.D., and M.G. Anderson. 2001. Declining scaup populations: A retrospective analysis of long-term population and harvest –survey data. Journal of Wildlife Management 65:781–796. Austin, J.E., A.D. Afton, M.G. Anderson, R.G. Clark, C.M. Custer, J.S. Lawrence, J.B. Pollard, and J.K. Ringelman. 2000. Declining scaup populations: Issues, hypotheses, and research needs. Wildlife Society Bulletin 28:254–263. Bellrose, F. 1980. Ducks, Geese, and Swans of North America, 3rd Edition. Stackpole Books, Harrisburg, PA. 540 pp. Burger, J., J.R. Trout, W. Wander, and G.S. Ritter. 1984. Jamaica Bay studies VII: Factors affecting the distribution and abundance of ducks in a New York estuary. Estuarine, Coastal and Shelf Science 19:673–689. Clarke, K.R., and R.N. Gorley. 2006. PRIMER v6: User Manual/Tutorial. PRIMER-E, Plymouth, UK. 190 pp. Cohen, J. 1998. Greater Scaup as bioindicators of contaminants in Long Island Sound. M.Sc. Thesis. University of Connecticut, Storrs, CT. 117 pp. Cronan, J.M. 1957. Food and feeding habits of the scaups in Connecticut waters. Auk 74:459–468. Eccleston, K.A. 1999. Food sources as a factor in the decline of Greater Scaup and Lesser Scaup ducks. M.Sc. Thesis. University of Connecticut, Storrs, CT. 89 pp. Eggeman, D.R., and F.A. Johnson. 1989. Variation in effort and methodology for the midwinter waterfowl-inventory in the Atlantic Flyway. Wildlife Society Bulletin 17:227–233. Environment Canada. 2014. Disease in Common Eider: Avian cholera. Available online at https://www.ec.gc.ca/mbc-com/default.asp?lang=En&n=D65C11B0-1. Accessed 27 March 2015. Erwin, R.M. 1996. Dependence of waterbirds and shorebirds on shallow-water habitats in the mid-Atlantic coastal region: An ecological profile and management recommendations. Estuaries 19:213–219. Evans, D.M., and K.R. Day. 2002. Hunting disturbance on a large shallow lake: The effectiveness of waterfowl refuges. Ibis 144:2–8. Northeastern Naturalist 744 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 Vol. 22, No. 4 Fraser, D. 2012. Virus blamed for Eider Duck deaths. Cape Cod Times, 10 March 2012. Available online at http://www.capecodonline.com/apps/pbcs.dll/article?AID=/20120310/ NEWS/203100335. Accessed September 2015. Gordon, D.H., B.T. Gray, R.D. Perry, M.B. Prevost, T.H. Strange, and R.K. Williams. 1989. South Atlantic coastal wetlands. Pp. 57–92, In L.M. Smith, R.L. Pederson, and R.M. Kaminski (Eds.). Habitat Management for Migrating and Wintering Waterfowl in North America. Texas Tech University Press, Lubbock, TX. 574 pp. Gordon, D.H., B.T. Gray, and R.M. Kaminski. 1998. Dabbling duck-habitat associations during winter in coastal South Carolina. Journal of Wildlife Management 62:569–580. Huang, M. 2010. Wintering habitat use, survival, and time and energy budgets of Black Ducks (Anas rubripes) in Connecticut. Final report to Black Duck Joint Venture. Available online at http://blackduck.cmi.vt.edu/research.php?rtype=projects&PNum=One. Accessed September 2015. Jones, J.J., and R.D. Drobney. 1986. Winter feeding ecology of scaup and Common Goldeneye in Michigan. Journal of Wildlife Management 50:446–452. Ma, Z., X. Gan, C. Choi, K. Jing, S. Tang, B. Li, and J. Chen. 2007. Wintering bird communities in a newly formed wetland in the Yangtze River estuary. Ecological Research 22:115–124. Madsen, J. 1998. Experimental refuges for migratory waterfowl in Danish wetlands. II. Tests of hunting-disturbance effects. Journal of Applied Ecology 35:386–397. McKinney, R.A. 2004. Habitat relationships of waterfowl wintering in Narragansett Bay. Rhode Island Natural History Society Bulletin 11:3–6. McKinney, R.A., S.R. McWilliams, and M.A. Charpentier. 2006. Waterfowl-habitat associations during winter in an urban eastern North Atlantic estuary. Biological Conservation 132:239–249. Narragansett Bay Winter Waterfowl Survey (NBWWS). 2014. Narragansett Bay Research Reserve. Available online at http://www.nbnerr.org/waterfowl.htm. Accessed August 2015. Owen, M. 1993. The UK shooting-disturbance project. Wader Study Group Bulletin 68:35–46. Perry, M.C., and A.S. Deller. 1996. Review of factors affecting the distribution and abundance of waterfowl in shallow-water habitats of Chesapeake Bay. Estuaries 19:272–278. Perry, M.C., A.M. Wells-Berlin, D.M. Kidwell, and P.C. Osenton. 2007. Temporal changes of populations and trophic relationships of wintering diving ducks in Chesapeake Bay. Waterbirds 30:4–16. Silverman, E.D., D.T. Saalfeld, J.B. Leirness, and M.D. Konef f. 2013. Wintering sea-duck distribution along the Atlantic coast of the United States. Journal of Fish and Wildlife Management 4:178–198. Steinkamp, M. 2005. New England/mid-Atlantic coast bird conservation region (BCR 30) implementation plan. Atlantic Coast Joint Venture, US Fish and Wildlife Service, Laurel, MD. 251 pp. Stott, R.S., and D.P. Olson. 1973. Food-habitat relationship of sea ducks on the New Hampshire coastline. Ecology 36:468–477. Suter, W. 1994. Overwintering waterfowl on Swiss lakes: How are abundance and species richness influenced by trophic status and lake morphology? Hydrobiologia 279–280:1–14. Take a Second Look (TASL). 2014. TASL online. Available online at http://032acf2.netsolhost. com/tasl.htm. Accessed 27 March 2015. Northeastern Naturalist Vol. 22, No. 4 R.A. McKinney, K.B. Raposa, and C.L. Trocki 2015 745 Thompson, B.C., J.E. Tabor, and C.L. Turner. 1988. Diurnal-behavior patterns of waterfowl wintering on the Columbia River, Oregon and Washington. Pp. 153–167, In M.W. Weller (Ed.). Waterfowl in Winter. University of Minnesota Press, Minneapolis, MN. 624 pp. Tiedemann, J.A. 1984. Habitat preference of Canvasback Ducks, Aythya valisineria, observed in New Jersey, USA. Bulletin of the New Jersey Academy of Sciences 29:25–28. US Fish and Wildlife Service (USFWS). 2014a. Mid-winter Waterfowl Survey. Division of Migratory Bird Management. Available online at https://migbirdapps.fws.gov/mbdc/ databases/mwi/aboutmwi_allflyways.htm. Accessed August 2015. USFWS. 2014b. Waterfowl population status, 2014. US Department of the Interior, Washington, DC. Weller, MW. 1988. Waterfowl in Winter. University of Minnesota Press, Minneapolis, MN. 624 pp. Zipkin, E.F., B. Gardner, A.T. Gilbert, A.F. O’Connell, J.A. Royle, and E.D. Silverman. 2010. Distribution patterns of wintering sea ducks in relation to the North Atlantic Oscillation and local environmental characteristics. Oecologia 163:893–902.