Northeastern Naturalist 532 C. Gjerdrum, J. Loch, and D.A. Fifield 22001188 NORTHEASTERN NATURALIST 2V5(o4l). :2553,2 N–5o4. 44 The Recent Invasion of Cory’s Shearwaters into Atlantic Canada Carina Gjerdrum1,*, John Loch2, and David A. Fifield3 Abstract - Calonectris diomedea (Cory’s Shearwater) are observed in small numbers in Atlantic Canada every year. However, in 2016 and 2017, unusually high numbers were reported in both Nova Scotia and Newfoundland and Labrador, Canada. We used data from eBird and from standardized ship-based surveys to document the timing and magnitude of the increase. Results show that densities have been increasing in waters off Nova Scotia since 2007 and are now of similar magnitude to those observed on George’s Bank in the late 1980s, which suggests that birds are targeting cooler, more productive waters on the Scotian Shelf during migration. Continued monitoring using a variety of survey techniques will help identify important marine areas for this trans-equatorial migrant, and identify potential threats as birds move into areas not previously explo ited. Introduction Calonectris diomedea (Scopoli) (Cory’s Shearwater) breed in colonies located in the Mediterranean and Northeast Atlantic and are regular nonbreeding visitors to North American waters (Brown 1986, Nisbet et al. 2013). The Atlantic population, estimated at ~250,000 pairs (BirdLife International 2017), breeds primarily in burrows and rock crevices on islands of Portugal (Azores and Madeira) and Spain (Canary Islands) from late May (first eggs) through October (Granadeiro 1991) and feeds primarily on small fish (Xavier et al. 2011). Tracking devices show that individuals may forage up to 2000 km away from the colony during chick-rearing, targeting highly productive offshore areas characterized by high concentrations of chlorophyll-a and low sea-surface temperatures (SST), although much shorter foraging trips are more common (Paiva et al. 2010). At the end of the breeding season, which may last up to 5 months (Granadeiro 1991), the va st majority of the population migrates tens of thousands of kilometers to wintering areas in the South Atlantic, with a small proportion (4%) wintering in the North Atlantic (Dias et al. 2012). Wintering areas are dominated by coastal upwelling (i.e., the Benguela and Agulhas Currents off South Africa, the southern part of the Brazil Current, as well as the Canary Current in the North Atlantic; Gonzalez-Solis et al. 2007). The Cory’s Shearwater’s route and destination depend not only on their breeding location (Catry et al 2011), but also their age and experience (Missagia et al. 2015). 1Environment and Climate Change Canada, Canadian Wildlife Service, 45 Alderney Drive, Dartmouth, NS B2Y 2N6, Canada. 2Loch Consulting Services, Bedford, NS B4A4K4, Canada. 3Environment and Climate Change Canada, Wildlife Research Division, 6 Bruce Street, Mt. Pearl, NL A1N 4T3, Canada. *Corresponding author - carina.gjerdrum@canada.ca. Manuscript Editor: Gregory Robertson Northeastern Naturalist Vol. 25, No. 4 C. Gjerdrum, J. Loch, and D.A. Fifield 2018 533 Cory’s Shearwaters are observed in small numbers in Atlantic Canada every year from May through October and are assumed to be either sub-adult birds or failed breeders (Brown 1986). They are also common visitors to eastern US waters (Veit 1976); in the past, mean densities on Georges Bank exceeded 10 birds per km2 (Powers and Brown 1987), and numbers further south to Florida have reached into the thousands (Nisbet et al. 2013). In 2016 and 2017, unusually high numbers of Cory’s Shearwaters were reported by bird watchers in Nova Scotia (NS) and Newfoundland and Labrador (NL), Canada, primarily in the nearshore (D.A. Fifield, pers. observ.; Loch 2016). In this paper, we document the increase in abundance of Cory’s Shearwaters in Atlantic Canada in 2016 and 2017 using eBird, a global online database of bird checklists submitted by participants with skill levels varying from beginner to professional (Sullivan et al 2009), and compare the eBird results to data collected offshore of Atlantic Canada from standardized ship-based surveys conducted by the Canadian Wildlife Service (CWS) of Environment and Climate Change Canada. As Cory’s Shearwaters have been shown to aggregate at thermal fronts of the Gulf Stream during migration, where food availability is enhanced (Haney and McGillivary 1985), and to target productive feeding areas with low SST and high chlorophyll-a concentrations (Paiva et al 2010, 2013), we examined whether local-scale environmental conditions or larger-scale phenomena, as measured by the North Atlantic Oscillation (NAO) index, help explain the high numbers observed in 2016 and 2017. Field-site Description We complied eBird data from the Canadian provinces of Nova Scotia and Newfoundland and Labrador and their adjacent waters (Fig. 1). The study area for CWS ship-based surveys includes the western North Atlantic, primarily within Canada’s Exclusive Economic Zone (EEZ; 200 nm offshore). The area is dominated by shelf waters where the warm Gulf Stream waters traveling northward meet with the colder Labrador Current flowing south. The shelf break in this study is defined at the 1000-m isobath, where there exists a steep gradient between the shallower shelf habitat and the deep ocean floor (Fig. 1). The Scotian Shelf has an average depth of 90 m and extends up to 230 km offshore, while the Grand Bank waters are up to 150 m deep and extend almost 480 km offshore (Hutchings et al. 2012). Generally speaking, the shelf waters off NS and NL are highly productive year-round and support extensive fisheries (Hutchings et al. 2012). Methods We downloaded eBird data (http://ebird.org/) on 30 March 2018 for Cory’s Shearwater from both NS and NL. We compared abundance, which is described as the average number of Cory’s Shearwaters reported on all checklists submitted to eBird, throughout the annual cycle (January through December) from 2006 to 2017, the same period the CWS conducted ship-based seabird surveys for the Eastern Canada Seabirds at Sea monitoring program (ECSAS surveys; Gjerdrum et al. 2012). Northeastern Naturalist 534 C. Gjerdrum, J. Loch, and D.A. Fifield 2018 Vol. 25, No. 4 We conducted ECSAS surveys (2006–2017) from ships-of-opportunity including Canadian Coast Guard vessels, foreign oceanographic ships, industry-supply vessels, and ferries (Gjerdrum et al. 2012). Observers surveyed from the bridge of the ships, looking forward and scanning to a 90° angle from either the port or starboard side. We restricted our observations of birds to a transect band 300 m from the observer. We continuously recorded all birds on the water within the transect, and recorded flying birds using a series of snapshots, the frequency of which depended on the speed of the ship (Gjerdrum et al. 2012, Tasker et al. 1984). We conducted as many consecutive observation periods as possible during the daylight Figure 1. Survey area in the western North Atlantic showing the provinces of Nova Scotia (NS) and Newfoundland and Labrador (NL), and Canada’s Exclusive Economic Zone (EEZ) located 200 nm offshore. The shaded offshore area indicates NS waters used to quantify Cory’s Shearwater density across years, and the 1000-m isobath depicts the approximate location of the shelf break. Also shown are the locations for George’s Bank, the Scotian Shelf, Laurentian Channel, Grand Bank, and Flemish Cap, as well as the Western Scotian Shelf (WS), Eastern Scotian Shelf (ES), and Solas sites used to quantify sea-surface temperature (SST) and chlorophyll-a concentrations. Northeastern Naturalist Vol. 25, No. 4 C. Gjerdrum, J. Loch, and D.A. Fifield 2018 535 hours. We binned observations into 10-min observation periods in 2006, and 5-min periods from 2007 to 2017. We also examined ship-based survey data collected by CWS under the Programme intégré de recherches sur les oiseaux pélagiques (PIROP surveys; Brown et al. 1975) to examine the abundance and distribution of Cory’s Shearwaters in Atlantic Canada between 1965 and 1992. Unlike ECSAS survey methodology, PIROP surveys continuously counted all birds observed both in flight and on the water, and sightings were binned into 10-min observation periods. Prior to 1984 (78% of the PIROP survey effort), transect widths were unlimited, but starting in 1984, sightings were restricted to a transect width of 300 m (B rown et al. 1975). Abundance (i.e., density) from ECSAS and PIROP surveys is reported in birds per linear km to allow for comparisons between the datasets. For each year, we computed density by dividing the total number of Cory’s Shearwater sightings by the total number of kilometers surveyed within NS waters (Fig. 1) between May and October (the months when Cory’s Shearwaters were found in these waters). We compared the proportion of sightings off the Scotian Shelf (beyond the 1000-m isobath) to the proportion on the shelf (less than 1000 m depth; Fig. 1) . To determine whether Cory’s Shearwater density varied as a result of regional variation in SST or chlorophyll-a concentration, we examined data from 3 North Atlantic sites (Fig. 1) monitored by the Atlantic Zone Monitoring Program (AZMP), Department of Fisheries and Oceans (DFO). All 3 sites are located in Nova Scotia waters: 2 on the Scotian Shelf (Western Scotian Shelf [WS] and Eastern Scotian Shelf [ES]), and one off the shelf (Solas) (Fig. 1). We derived SST values from information sensed by advanced very high-resolution radiometer (AVHRR) instruments on National Oceanic and Atmospheric Administration satellites (NOAA) and European Organisation of the Exploitation of Meteorological Satellites (EUMETSAT). We derived sea-surface chlorophyll-a concentrations from moderate resolution imaging spectroradiometer (MODIS) data collected on NOAA satellites. Data for both SST and chlorophyll-a concentration were provided by the Remote Sensing Unit at DFO’s Bedford Institute of Oceanography. To better quantify the magnitude of variation in chlorophyll-a concentrations, we used normalized anomalies (anomalies divided by the climatological standard deviation using the reference period 2003–2017). To determine whether Cory’s Shearwater density varied as a result of larger-scale oceanographic phenomena, we used the winter mean North Atlantic Oscillation (NAO) Index (https://climatedataguide.ucar.edu/climate-data/hurrell-north-atlantic- oscillation-nao-index-station-based; downloaded on 19 March 2018). The winter (December through March) NAO index is based on the difference in sea-surface pressure anomalies between station locations in Portugal and Iceland. Positive values tend to be associated with an increase in winter storms in the northern Atlantic Ocean and stronger than average westerly winds (Hurrell et al. 2003), which may affect spring migration strategies (i.e., Hüppop and Hüppop 2003) that could potentially influence densities of Cory’s Shearwaters observed in NS. Northeastern Naturalist 536 C. Gjerdrum, J. Loch, and D.A. Fifield 2018 Vol. 25, No. 4 We used multivariate adaptive regression splines (MARS) with package Earth (Milborrow 2018) in R 3.4.3 (R Core Team 2017) to explore the relationship between recent (ECSAS surveys 2006–2017) Cory’s Shearwater density and several environmental variables. MARS is a nonlinear regression modeling method (Friedman 1991, Friedman and Roosen 1995) that builds the model using a series of piecewise regressions (in this case, generalized linear models with gamma error structure and log-link function), all allowed their own slope. For this study, local-scale environmental variables included mean SST and mean chlorophyll-a concentration (normalized anomalies) from May to October across the 3 sites within NS (WS, ES, and Solas; Fig. 1) weighted by site area. We also included in the model the mean difference in SST between the off-shelf site (Solas) and on-shelf site (ES), the winter mean NAO index, as well as year to identify any points in time when abrupt changes in densities may have occurred. We used the same modeling approach (MARS) for the extended time series (PIROP plus ECSAS surveys) to determine whether recent increases in Cory’s Shearwater density were unprecedented. The only environmental variable included in this model was the winter NAO index because our time series of values for SST and chlorophyll-a concentration did not cover the period of PIROP surveys. Results Data from eBird between 2006 and 2017 showed that Cory’s Shearwaters were reported more frequently in NS (a total of 13,164 Cory’s Shearwater sightings) compared to NL waters (95 sightings; Table 1). In NS, they were absent from eBird Table 1. Sightings of Cory’s Shearwaters observed in Nova Scotia (NS) and Newfoundland and Labrador (NL) from eBird checklists submitted between 2006 and 2017. Abundance (abund.) is the average number of Cory’s Shearwaters reported per checklist submitted in a given year . Nova Scotia Newfoundland and Labrador # checklists # checklists reporting reporting Total Cory’s Total # Total Cory’s Total # Year individuals Shearwater checklists Abund. individuals Shearwater checklists Abund. 2006 0 0 1680 0.000 0 0 241 0.000 2007 0 0 1923 0.000 0 0 258 0.000 2008 0 0 1424 0.000 0 0 417 0.000 2009 3 1 1415 0.002 0 0 585 0.000 2010 12 3 1897 0.006 0 0 1101 0.000 2011 20 2 3235 0.006 0 0 1649 0.000 2012 31 10 4945 0.006 0 0 1626 0.000 2013 24 7 6044 0.004 0 0 2627 0.000 2014 178 15 11,288 0.016 1 1 6500 0.000 2015 841 46 15,546 0.054 9 1 7458 0.001 2016 5185 127 18,866 0.275 65 6 9382 0.007 2017 6870 189 24,686 0.278 20 11 10,953 0.002 Total 13,164 400 92,949 95 19 42,797 Northeastern Naturalist Vol. 25, No. 4 C. Gjerdrum, J. Loch, and D.A. Fifield 2018 537 records in 2006 through 2008, uncommon from 2009 through 2013, and relatively frequent from 2014 to 2017, particularly in the latter 2 years (Table 1). Individuals have been documented in eBird as early as the second week of June and as late as the first week of December (Fig. 2A). Prior to 2016, peak abundance occurred in August (2010 and 2012) and September (2011, 2013–2015). In contrast, abundance in both 2016 and 2017 peaked in October (83.6% of the sightings), and the total number of reported sightings was 6–8 times greater in these years compared to 2015 (Table 1, Fig. 2A). Checklists submitted for NL during the same time period (2006–2017) indicated that Cory’s Shearwaters have only been reported in that region since 2014, but compared to NS have remained relatively uncommon (Table 1). According to the Figure 2. Cory’s Shearwater abundance from (A) eBird checklist submissions for Nova Scotia, Canada, over 48 weeks (January through December) in 2006–2015 (solid line) compared to 2016 (dotted line) and 2017 (dashed line), and (B) monthly at-sea survey densities from PIROP 1965–1992 (dashed-dot line), ECSAS 2006-2015 (solid line), ECSAS 2016 (dotted line), and ECSAS 2017 (dashed line) in waters of f Nova Scotia. Northeastern Naturalist 538 C. Gjerdrum, J. Loch, and D.A. Fifield 2018 Vol. 25, No. 4 NL records, Cory’s Shearwater abundance spiked in NL in 2016, with 65 submitted records compared to just 9 in 2015 and 20 in 2017 (Table 1). ECSAS ship-based surveys (2006–2017) documented a total of 2724 Cory’s Shearwaters from May through October (Fig. 2B), primarily (68.5% of sightings) within NS waters (Fig. 3A, B). Prior to 2016 in NS, between 1 and 136 individuals were counted every year, peaking in August (2010, 2013, 2014), September (2007) or October (2006, 2008, 2009, 2011, 2012, 2015). Between 2006 and 2015, the Figure 3. At-sea survey data showing density of Cory’s Shearwaters (birds km-1) from May through October from (A) ECSAS surveys between 2006 and 2015, (B) ECSAS surveys in 2016 and 2017, and (C) PIROP surveys between 1965 and 1992. The light gray shaded area indicates waters off Nova Scotia, Canada, used to quantify Cory’s Shearwater abundance across years, and the 1000-m isobath depicts the approximate location of the shelf break. We also show (D) the sea-surface temperature (SST) anomaly in October 2016 (reference period October 2006–2015) with reference to the Western Scotian Shelf (WS), Eastern Scotian Shelf (ES), and Solas sites to highlight the relative difference in SST between the off-shelf and on-shelf areas. Northeastern Naturalist Vol. 25, No. 4 C. Gjerdrum, J. Loch, and D.A. Fifield 2018 539 highest density in a given year varied from 0.003 birds km-1 in August 2010 (1 sighting over 388 km) to 0.15 birds km-1 in September 2011 (63 sightings over 431 km), and just over half (58.1%) of the sightings occurred off the shelf, beyond the 1000-m isobath (Fig. 3A). In contrast, ECSAS surveys in 2016 documented a total of 994 individuals in NS waters, and 343 in 2017; the highest density was recorded in October 2016 (0.97 birds km-1; 360 sightings over 368 km), a density that was more than 6 times greater than the highest density recorded in any of the previous 10 years. The vast majority of sightings in 2016 and 2017 (81.8%) were recorded on the Scotian Shelf (Fig. 3B), despite an almost identical distribution of effort between the on-shelf and off-shelf habitats between the 2 time periods (2006–2015: 69.6% of km surveyed on-shelf; 2016–2017: 71.7% of km surveyed on-shelf). Of the 69 Cory’s Shearwaters observed on ECSAS surveys within NL waters (within the EEZ), all but one individual were sighted near the shelf break, in the Laurentian Channel (Fig. 3A, B). PIROP at-sea surveys conducted between 1965 and 1992 documented 6222 Cory’s Shearwaters, primarily outside Canadian waters (92.7% of sightings), on George’s Bank and in the Gulf of Maine (Fig. 3C). In NS waters, where 419 (6.7%) of the sightings occurred, Cory’s Shearwaters were observed from June through October. The maximum number of individuals observed in NS in any 1 year was 93 in 1988; of these 90 were observed in August (0.19 birds km-1). Most (80.2%) of the Cory’s Shearwaters observed in NS were on the shelf, although PIROP surveys were rarely conducted beyond the shelf break (only 8.0% of km surveyed were beyond the shelf break). A total of 46 Cory’s Shearwaters were recorded in NL waters during PIROP surveys (22 of which were reported in August 1980), amounting to 9.9% of all the Cory’s Shearwater sightings in Canadian waters during that time period. These sightings were all on the Grand Bank, NL. The only predictor variable selected by the model as an important determinant of Cory’s Shearwater density (2006–2017) was the difference in SST between the off-shelf (Solas) and on-shelf (ES) sites (R2 = 0.71; Fig. 3D). The model identified an inflection point at 7.3 °C; in other words, there was no detectable relationship between the difference in SST (between the off-shelf and on-shelf sites) and Cory’s Shearwater density below 7.3 °C, but the model indicated a positive relationship above that temperature value (Fig. 4A). The model selected year (and not winter NAO index) as an important determinant of Cory’s Shearwater density when PIROP and ECSAS density estimates were combined (R2 = 0.40). Inflection points included the years 1988 and 2007, indicating that Cory’s Shearwater density has been increasing in NS waters since 2007, reaching a similar magnitude to that observed in 1988 (Fig. 4B) . Discussion Our results using both eBird records and ship-based CWS survey data show that Cory’s Shearwater abundance in Atlantic Canada has increased over time. While we recognize that the use of the eBird database to document trends can be problematic, for example, due to variation in the distribution of effort, or experience Northeastern Naturalist 540 C. Gjerdrum, J. Loch, and D.A. Fifield 2018 Vol. 25, No. 4 of the individuals submitting the checklists, the ECSAS surveys corroborated both the intensity of the 2016/2017 Cory Shearwater irruption documented by eBird, as well as its timing. In NS, in particular, where the bulk of the sightings occurred, average monthly density recorded during ship-based surveys in 2016 was at least 6 times higher than the highest density recorded in any previous year surveyed. A much higher proportion of those birds were observed on the shelf, as opposed to the Figure 4. Multivariate adaptive regression splines (MARS) results showing (A) the effect of the difference in SST between the off-shelf and on-shelf sites on annual density of Cory’s Shearwater (birds km-1) from May through October 2006–2017, and (B) the effect of year on annual density of Cory’s Shearwater (birds km-1) from May through October combining PIROP (1965–1992) and ECSAS (2006–2017) survey data. Northeastern Naturalist Vol. 25, No. 4 C. Gjerdrum, J. Loch, and D.A. Fifield 2018 541 deeper off-shelf waters, suggesting that Cory’s Shearwaters moved closer to shore in 2016 and 2017 compared to previous years, where they were more likely to be spotted by bird watchers and citizen scientists using eBird checklists to document their sightings. Cory’s Shearwaters observed in Atlantic Canada are assumed to be sub-adults (Brown 1986, Nisbet et al. 2013), but it is possible that some of those observed in May were pre-laying females. Females tracked from Berlenga Island (Portugal) were shown to travel as far west as the Grand Bank and NL Shelf during the prelaying period (April–May) to meet the nutritional requirements to form an egg (Paiva et al. 2013). Cory’s Shearwaters have also been tracked from the Azores during chick-rearing, and long trips to replenish reserves depleted during chick feeding have extended as far as the Flemish Cap (Magalhães et al. 2008). In 2016, ocean productivity around the Azores was low, which may have forced birds to explore more productive areas further offshore (V.H. Paiva, IMAR/CMA, Coimbra, Portugal, pers. comm.), although there is no direct evidence that breeding birds from the Azores have traveled as far west as the Scotian Shelf, where most of our sightings were reported. Sightings in Atlantic Canada after the breeding season (October–November) may also include breeders; inexperienced birds (i.e., young breeders) off the Portuguese coast have been shown to visit sites in the western North Atlantic before migrating to the Brazilian Current for winter, and are assumed to be exploring new routes and stopover sites as they learn to optimize their foraging strategies (Haug et al. 2015, Missagia et al. 2015). Results from our analyses suggest that Cory’s Shearwater density in NS waters has been increasing since 2007, but the high densities observed in 2016 and 2017 do not appear to be unprecedented. Cory’s Shearwater densities were similarly high in 1988, and the highest density on record was in August of that year. However, all of the birds observed in 1988 in NS were seen on George’s Bank, on the far western edge of NS waters, and not on the Scotian Shelf, where the bulk of the recent sightings were concentrated. We have not observed a similar increase in abundance for other shearwater species, such as the Ardenna gravis (O’Reilly) (Great Shearwater) or A. griseus (Gmelin) (Sooty Shearwater), which are common in NS waters between April and November, nor have we seen an influx of the much less common Puffinus lherminieri Lesson (Audubon’s Shearwater) (CWS, Dartmouth, NS, Canada, unpubl. data), which, like the Cory’s Shearwater also breeds in the northeast Atlantic. Although population-trend information for the Cory’s Shearwater is incomplete, there is no evidence that they are increasing (BirdLife International 2017). As such, we hypothesize that the recent increase in abundance reported in this study is the result of a redistribution of birds onto the Scotian Shelf. PIROP surveys suggest that the Gulf of Maine and George’s Bank were historically important habitat for Cory’s Shearwaters, much more so than the Scotian Shelf (Fig. 3C). The recent increase in density within NS waters may therefore be a result of a redistribution from the Gulf of Maine, where SST has increased faster than in any other ocean on the planet (Pershing et al. 2015). However, Cory’s Shearwater abundance has also been unusually high in recent years in waters off the Northeastern Naturalist 542 C. Gjerdrum, J. Loch, and D.A. Fifield 2018 Vol. 25, No. 4 northeastern US, partly related to an increase in Ammodytes (sand lance) abundance and is also statistically linked to climate, as indexed by the NAO (unpubl. data; R. Veit, R.R. Veit, City University of New York, Staten Island, NY, pers. comm.). Our results show that more Cory’s Shearwaters migrate through NS waters in years when SST off the shelf is high relative to SST on the shelf, and suggest that the birds may be targeting more productive feeding areas in the nearshore. Such flexibility has been shown by Cory’s Shearwaters on both their breeding and wintering grounds, demonstrating their capacity to adapt to local changes in their environment (Dias et al. 2011, Paiva et al. 2013), and supporting their candidacy as an indicator species of underlying ecological processes at this sc ale. Recent developments in tracking technologies, such as reductions in instrument size and weight, prolonged battery life, and improved attachment methods, have dramatically increased our understanding of the spatial ecology of Cory’s Shearwater throughout the annual cycle (e.g., Catry et al. 2011, Dias et al. 2011, Gonzàlez-Solís et al. 2007, Missagia et al. 2015). Ship-based surveys however, remain important for identifying important marine areas, especially for non-breeders and populations not targeted by tracking studies. The ship-based survey results presented in our study reveal the use of the Scotian Shelf for Cory’s Shearwater migration, which is an area where instruments have not yet tracked breeding birds. These results also validate citizen scientist reports such as those submitted by users of eBird who are often the first to report unusual sightings and document significant changes in distribution and abundance. Continued monitoring using a variety of survey techniques will be essential not only to identify important marine areas for this and other trans-equatorial migrants, but also to identify vulnerability to human activities (e.g., Lieske et al. 2014), such as those posed by fishing, oil and gas production, shipping, and wind farms, especially as birds move into areas not previously exploited. The redistribution of Cory’s Shearwaters into Canadian waters, if it persists, suggests Canada’s responsibility for the conservation of this species will increase. Acknowledgments We thank all of the observers who conducted the surveys at sea as well as science staff, and ships’ officers and personnel for support at sea, especially the Canadian Coast Guard and the Department of Fisheries and Oceans. In particular, we thank Carla Caverhill (DFO) for helping obtain and interpret the remote-sensing data and images. We acknowledge the financial support (2006–2009) of the Environmental Studies Research Funds (ESRF, http:// www.esrfunds.org). Input from 2 anonymous reviewers greatly improved the manusc ript. Literature Cited BirdLife International. 2017. Species factsheet: Calonectris borealis. Available online at http://www.birdlife.org. Accessed 14 November 2017. Brown, R.G.B. 1986. Revised Atlas of Eastern Canadian Seabirds. Canadian Wildlife Service, Ottawa, ON, Canada. 111 pp. Brown, R.G B., D.N. Nettleship, P. Germain, C.E. Tull, and T. Davis. 1975. Atlas of Eastern Canadian Seabirds. Canadian Wildlife Service, Ottawa, ON, Canada. 220 pp. Northeastern Naturalist Vol. 25, No. 4 C. Gjerdrum, J. Loch, and D.A. Fifield 2018 543 Catry, P., M.P. Dias, R.A. Phillips, and J.P. Granadeiro. 2011. Different means to the same end: Long-distance migrant seabirds from 2 colonies differ in behaviour, despite common wintering grounds. PLoS ONE 6:e26079. Dias, M.P., J.P. Granadeiro, R.A. Phillips, H. Alonso, and P. Catry. 2011. Breaking the routine: Individual Cory’s Shearwaters shift winter destinations between hemispheres and across ocean basins. Proceedings of the Royal Society B 278:178 6–1793. Dias, M.P., J.P. Granadeiro, and P. Catry. 2012. Do seabirds differ from other migrants in their travel arrangements? On route strategies of Cory’s Shearwater during its trans-equatorial journey. PLoS ONE 7:e49376 Friedman, J.H. 1991. Multivariate adaptive regression splines. The Annals of Statistics. 19:1–47. DOI:10.1214/aos/1176347963. Friedman, J.H., and C.B. Roosen. 1995. An introduction to multivariate adaptive regression splines. Statistical Methods in Medical Research 4:197–217. Gjerdrum, C., D.A. Fifield, and S.I. Wilhelm. 2012. Eastern Canada Seabirds at Sea (ECSAS) standardized protocol for pelagic seabird surveys from moving and stationary platforms. Canadian Wildlife Service Technical Report Series No. 515. Atlantic Region, Ottawa, ON, Canada. 37 pp. Gonzàlez-Solís, J., J.P. Croxall, D. Oro, and X. Ruiz. 2007. Trans-equatorial migration and mixing in the wintering area of a pelagic seabird. Frontiers in Ecology and the Environment 5:297–301. Granadeiro, J.P. 1991. The breeding biology of Cory’s Shearwater, Calonectris diomedea, on Berlenga Island, Portugal. Seabird 13:30–39. Haney, J.C., and P.A. McGillivary. 1985. Aggregations of Cory’s Shearwaters (Calonectris diomedea) at Gulf Stream fronts. Wilson Bulletin 97:191–200. Haug, F.D., V.H. Paiva, A.C. Werner, and J.A. Ramos. 2015. Foraging by experienced and inexperienced Cory’s Shearwater along a 3-year period of ameliorating foraging conditions. Marine Biology 162:649–660. Hüppop, O., and K. Hüppop. 2003. North Atlantic Oscillation and timing of spring migration in birds. Proceedings of the Royal Society of London B. 270:233–240. DOI:10.1098/rspb.2002.2236. Hurrell, J.W., Y. Kushnir, G. Ottersen, and M. Visbeck 2003. An overview of the North Atlantic Oscillation. Pp. 1–35, In J.W. Hurrell, Y. Kushnir, G. Ottersen and M. Visbeck (Eds.). The North Atlantic Oscillation: Climatic Significance and Environmental Impact, American Geophysical Union, Monograph 134, Washington, DC. 279 pp. DOI:10.1029/134GM01. Hutchings, J.A., I.M. Cote, J.J. Dodson, I.A. Fleming, S. Jennings, N.J. Mantua, R.M. Peterman, B.E. Riddell, and A.J. Weaver. 2012. Climate change, fisheries, and aquaculture: Trends and consequences for Canadian marine biodiversity. Environmental Reviews 20:220–311. Lieske, D.J., D.A. Fifield, and C. Gjerdrum. 2014. Maps, models, and marine vulnerability: Assessing the community distribution of seabirds at-sea. Biological Conservation 172:15–28. Loch, J. 2016. Autumn bird reports: Tubenoses through cormorants. Nova Scotia Birds 59(1):16.. Magalhães, M.C., R.S. Santos, and K.C. Hamer. 2008. Dual-foraging of Cory’s Shearwaters in the Azores: Feeding locations, behaviour at sea, and implications for food provisioning of chicks. Marine Ecology Progress Series 359:283–293. Milborrow, S. 2018. earth: Multivariate adaptive regression splines. R package version 4.6.2. Available online at https://CRAN.R-project.org/package=earth. Northeastern Naturalist 544 C. Gjerdrum, J. Loch, and D.A. Fifield 2018 Vol. 25, No. 4 Missagia, R.V., J.A. Ramos, M. Louzao, K. Delord, H. Weimerskirch, and V.H. Paiva. 2015. Year-round distribution suggests spatial segregation of Cory’s Shearwaters based on individual experience. Marine Biology 162:2279–2289. Nisbet, I.C.T., R.R. Veit, S.A. Auer, and T.P. White. 2013. Marine Birds of the Eastern United States and the Bay of Fundy. Nuttall Ornithological Club, Cambridge, MA. 198 pp. Paiva, V.H., P. Geraldes, I. Ramírez, A. Meirinho, S. Garthe, and J.A. Ramos. 2010. Oceanographic characteristics of areas used by Cory’s Shearwaters during short and long foraging trips in the North Atlantic. Marine Biology 157:1385–1399. Paiva, V.H., P. Geraldes, I. Ramirez, A.C. Werner, S. Garthe, and J.A. Ramos. 2013. Overcoming difficult times: The behavioural resilience of a marine predator when facing environmental stochasticity. Marine Ecology Progress Series 486:277–288. Pershing, A.J., M.A. Alexander, C.M. Hernandez, L.A. Kerr, A. Le Bris, K.E. Mills, J.A. Nye, N.R., Record, H.A. Scannell, J.D. Scott, G.D. Sherwood, and A.C. Thomas. 2015. Slow adaptation in the face of rapid warming leads to collapse of the Gulf of Maine cod fishery. Science 350:809–812. DOI:10.1126/science.aac9819. Powers, K.D., and R.G.B. Brown. 1987. Seabirds. Pp. 372–384, In R.H. Backus and D.W. Bourne (Eds.). Georges Bank. MIT Press, Cambridge, MA. 593 pp. R Core Team. 2017. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available online at https:// www.R-project.org/. Sullivan, B.L., C.L. Wood, M.J. Iliff, R.E. Bonney, D. Fink, and S. Kelling. 2009. eBird: A citizen-based bird observation network in the biological sciences. Biological Conservation 142:2282–2292. Tasker, M.L., P. Hope Jones, T. Dixon, and B.F. Blake. 1984. Counting seabirds at sea from ships: A review of methods and a suggestion for a standardized approach. Auk 101:567–577. Veit, R.R. 1976. Recent changes in the range and distribution of Cory’s Shearwaters in North American waters. Bird Observer of Eastern Massachusetts 4:72–73 . Xavier, J.C., M.C. Magalhães, A.S. Mendonҫa, M. Antunes, N. Carvalho, M. Machete, R.S. Santos, V.H. Paiva, and K.C. Hamer. 2011. Changes in diet of Cory’s Shearwaters Calonectris diomedea breeding in the Azores. Marine Ornithology 39:129–134.