Atlantic Cooperative Wildlife Ecology Research Network at the University of New Brunswick

Introduction
Methods
Results / Discussion
    Tern Census
    Puffin Burrow Density Survey
    Common Eider Census
    Capture of Seabirds
    Productivity and Growth     
    Diet Study
    Leach's Storm Petrels
    Environmental Condidtions
    Tourist Questionnaire
    Additional Observations
Acknowldgements
Literature Cited

Introduction

There has been a long history of observations of seabirds on Machias Seal Island (MacKinnon and Smith 1985). Past studies conducted on the island have concentrated on Arctic Terns (Sterna paradisaea), including fledging success (Pettingill 1939), behaviour and ecology (Hawksley 1957), nesting ecology (Newell 1985), and nesting locations (Bunin and Boates 1994). In the 1980s and early 1990s the National Audubon Society, in conjunction with the Canadian Wildlife Service, employed a researcher to observe Arctic Tern provisioning on MSI (Amey 1998). The Atlantic Cooperative Wildlife Ecology Research Network (ACWERN) has been involved in seabird research on Machias Seal Island since 1995. Specific research projects on the island have focused on the use of seabirds as indicators of changes in the availability and commercial fisheries of herring (Amey 1998) and time budgeting behaviour of Arctic Terns (Paquet, in process). ACWERN’s research program is directed towards both management and monitoring and improving understanding of the links between the marine and terrestrial ecosystems on which they depend. The focus of this report is to summarise the general observations of reproductive success, growth of young, and foraging of the Common Terns (Sterna hirundo), Arctic Terns, Atlantic Puffins (Fratercula arctica) and Razorbills (Alca torda) as well as environmental conditions recorded between 1995 and 1999.

The numbers of Common and Arctic Terns nesting on the island are censused every two years by CWS. ACWERN personnel co-operate in carrying out this census. In 1998 the census protocol was modified to allow the results to be mapped by 30-m grid squares established in 1997. In the future this will allow changes in density and distribution to be related to habitat characteristics. Tern productivity and growth is monitored in ACWERN research plots, some of which are also used for diet studies.

Alcids are not censused accurately at present. Counts of all alcids visible from the top of the lighthouse are collected twice daily, but the relationship between these counts and breeding numbers is unknown. ACWERN plans to develop reliable methods to estimate breeding populations of both Puffins and Razorbills. Productivity and growth are monitored, and samples of banded breeding birds for the demographic project are being compiled by systematic banding of chicks and adults in designated plots.

The tern census, and diet and productivity monitoring, contribute to a regional cooperative program by the Gulf of Maine Seabird Working Group (GOMSWG) to monitor and restore seabird breeding populations throughout the Gulf of Maine and Bay of Fundy. A current project extending from the demographic project, to study the movement of Arctic Terns between islands, is being undertaken (Devlin, in progress). This project involves collaboration between GOMSWG, CWS and ACWERN, and fieldwork on MSI, Petit Manan Island ME, Matinicus Rock and Matinicus Seal Island ME, and The Brothers and Sable Island, NS.

Interactions with the marine ecosystem are addressed currently through systematic monitoring of diet and feeding rate in terns and alcids, using research plots established for this purpose. Intensive research projects are designed to develop and test hypotheses to explain the relationships suggested by the diet data (e.g. Amey 1998, Paquet in progress).

Studies of interactions with the terrestrial ecosystem are in preliminary stages now (1999). New colour aerial photography was flown in 1996, and these photographs were digitised and converted into georeferenced line maps using GPS ground control. Several resulting maps are included in this report (Figure 1, 2, and 10). A vegetation survey linked to this basemap was begun in 1998, and completed in 1999. Detailed studies of interactions between habitat and seabird biology are anticipated in the future.

Due to the long life span of many seabirds, effective population monitoring must include examinations of the survival of adult birds as well as productivity. Thus, ACWERN has also begun a long-term demographic study of the terns and alcids nesting on the island. Tracking individuals is accomplished through marking individuals with bands and re-sighting them every year. Available programs such as SURGE and MARK require a sample of at least 200 adults marked and resighted every year. This is the target for both species of terns and alcids on MSI.

The research on MSI is subject to the following guidelines (CWS 1994):

• research must not interfere with normal activities on the island (including operation of the lighthouse), or visitors who may be present;
• research within the tern colony must be conducted only under good weather conditions, i.e. (once tern chicks have hatched) temperatures >11°C and dry vegetation;
• all research activity must relate directly to the needs for management or monitoring of seabirds, or provide a clear improvement to our present understanding.

Tern Census

Nests are counted by a field crew moving through an area with individuals walking about an arm’s length apart. Traditionally the census has been conducted in a clockwise direction and proceeds through regions of the island from the outermost shoreline area inward to the intermediate areas and the innermost vegetated areas. Each nest is marked with a wooden stirring stick. After the count is completed another pass of part of the island is conducted in which marked and unmarked nests are counted to determine a Lincoln Index (Newell and Mackinnon 1998). In 1998 the number of nests per 30-m grid square was also recorded. In the future this will allow changes in density and distribution to be related to habitat characteristics.

No census was done in 1999, however, species ratios were calculated. Two methods were used to determine the species ratio of Arctic and Common Terns nesting on Machias Seal Island in order to validate the sampling technique. The first method was one that the warden from the Canadian Wildlife Service (Jason Hudson) has used for the last 6 years. He used 14 different sites on the island (using blinds and other fixed points) and identified as many birds as possible landing on nests and loafing in a roost (it should be noted that only two roosting areas were observed and fewer than ~100 birds were identified while loafing). This method sampled the birds nesting in different areas of the island. The second method used this year was conducted by Laurel Bernard and Kate Devlin separate to that used by Jason Hudson. We also used fixed points on the island, such as blinds and the lighthouse tower, but we systematically identified birds nesting in all grid points, in all areas of the island. Only birds landing on nests were identified. The timing of the two methods overlapped (21-26 June), and we did not compare results until we had completed our surveys.

Puffin Burrow Density Survey

The number of puffins breeding on Machias Seal Island seems not to have been estimated other than by counts of individual birds; such estimates range in the 20th century from a few hundred to 3,500 (MacKinnon and Smith 1985, Table 4). Breeding pairs are difficult to count because most nests are inaccessible (beneath boulders) or are in soil adjacent to boulders and obscured by tall vegetation. Systematic methods used elsewhere (e.g. transects and quadrats; Walsh et al. 1995) need to be modified accordingly. An extra complication on MSI is introduced by the large colony (ca.3,000 pairs) of Arctic and Common Terns whose breeding would be compromised by intensive searching for puffin burrows between late May and early August (when puffin censuses are usually made).

In mid May (before the terns had arrived on the island) we ran transects along alternate 30 m grid lines across the island, following the approach of Rodway et al. (1996). Proceeding along each transect from one coast, we recorded the point at which we first found puffin burrows (including nest-sites under rocks). From that point until the last burrow along the transect, we counted all burrows within a 2m x 2m (i.e. 4 m²) quadrat every 5m along the transect. From these data we recorded: (i) the area of puffin habitat (i.e. containing nest burrows or cavities), and (ii) the nest density in nests/m².

Common Eider Census

During the island-wide tern census in 1998 the number of Common Eider (Somateria mollissima) nests (both hatched and not hatched) were counted by the field crew as they moved around the island. No census was done in 1999.

Capture of Seabirds

Each year as part of a long-term demographic study adult and young terns and alcids are trapped, banded, and measured.

Terns

Trapping of adult terns begins no sooner than 7 days after peak lay. A modified Weller treadle-trap is used (Weller 1957). The eggs are removed from the nest and temporarily replaced with wooden eggs of comparable size and colour. The real eggs are kept warm and protected in a carton and are not withheld from the nest for longer than 45 minutes.

The chicks are captured and banded by hand. The island is searched slowly, and all tern chicks old enough to identify are banded, and the grid square in which each was banded is recorded. The numbers of dead chicks and chicks in each grid square too small to identify (and therefore to band) are also recorded. The criteria for identifying species are based on MSI growth curves measured by Krista Amey: if wing-length > 50mm and tarsus > 17mm, Common Tern; if tarsus < 17mm, Arctic Tern.

Alcids

Trapping of adult alcids has been conducted early in the season before the terns begin to nest (i.e. late April and May). Two drop-in boxes with swivelling lids are placed in loafing areas in the nesting colony. When the birds use them to take off from or as perches they are caught in the box. Other adults are captured in burrows when the nests are checked for chicks. The adults are often present when the chicks are young. The alcid chicks are banded 3-7 days before expected fledging (35 days for Puffins and 15 days for Razorbills). Most Puffin chicks (~250 per year) are banded when they come to the lighthouse on the night they fledge.

Productivity and Growth

Terns

Four study plots are monitored each year. Each plot is 10 m x 10 m and contains between 10 and 25 nests. A minimum of 30 nests each of Common and Arctic Terns are included in the growth and survival monitoring. The plots are fenced with a combination of chicken wire and landscape fabric (Nisbet and Drury 1972, Walsh et al. 1995). The wire fencing is set up before the terns arrive to nest on the island. There are also permanent blinds set up at each of the plots (Table 1 summarises the use of different blinds, Figure 1 identifies locations of blinds on the island). As the terns begin to establish their nests, the clutch size and species associated with each nest are recorded. The nest initiation date is the date the first egg was laid, and peak hatch is the mean date the first chick in each nest hatched. The hatch date of the chicks is predicted and the plots are visited only to monitor the establishment of new nests. Hatching success is reported as the average number of chicks that hatch per nest. Productivity is reported as the average number of fledglings produced per nest initiated.

Once the chicks begin to hatch, the plots are visited every other day and the chicks are banded, weighed and measured until they reach fledging. Measurements of chicks include weight, head-bill length, tarsus length and wing chord. During the 1995-1997 field seasons the plots were visited every other day to measure and weigh chicks. In 1998 and 1999, plots were visited less frequently and measurements were taken only twice during the linear phase of wing and weight growth (ages 5 to 15 days). There is some discrepancy as to the age at which tern chicks can be considered "fledged". Drury (1990) and Walsh et al. (1995) use 15 days of age where as the CWS wardens have used 20 days. Both are recorded to make the data comparable. If a chick is alive on day 14 or 19 it is considered to be alive the next day and therefore fledged. If it was found dead after day 20, it is subtracted from those that made it to day 20. If the chick was alive at day 15 but then not resighted, it is not included in the numbers that were alive at day 20; thus productivity represents the minimum number of chicks fledged per nest. The day a chick hatches is considered to be Day 0.

Alcids

Seventy-five Puffin burrows have been marked and their locations mapped. Thirty-seven of these are assigned to measuring growth rates and the remaining 38 for measuring productivity. Similarly 73 Razorbill nests have been marked and mapped, 15 are assigned to measuring growth rates and the remaining 58 for productivity. The puffin nests are marked with numbered red or white painted stakes and the razorbill nests are marked with numbers painted in black on rock faces. See Figure 1 for general locations of study plots.

The spread of egg laying among alcids is greater (30 days or more) than among terns. To cover this range the marked nest sites are checked three times at the beginning of the breeding season. On the first check, carried out at the first opportunity (by mid-May), all marked sites are visited. On the second check, 1 week later, only those burrows, which were empty on the first check or had eggs that were cold, are visited. On the third visit, 1 week after the second check, only those burrows, which were empty on the second check, are visited. Any sites that are still empty at the third visit are deemed inactive and are not visited again that season.

Each marked burrow that has an egg is revisited at the end of a full incubation period (40 days for Puffins and 35 days for Razorbills; Harris and Birkhead 1985, Harris and Wanless 1989). If the egg has hatched, the culmen, tarsus and wing chord of the chick are measured, the chick is weighed, and the age is estimated based on past measurements of wing chord of known aged chicks (see Figure 7 and 8). If the egg has not hatched, it is recorded if the egg is warm or an adult is present, and the burrow is re-visited 1 week later.

The Puffin productivity nests are visited only once more, 35 days after a successful hatch check. At this time the chick is banded and measured. This check defines the success of the nest. Starting in 1997, subsets of the Puffin and Razorbill nests followed for productivity were selected for the monitoring of chick growth. Puffin growth rate nests are visited twice only during the period of linear growth (10-30 days). Razorbill productivity nests are visited no later than 15 days after a successful hatch check. Razorbill growth nests are visited twice during the period of linear growth (5-15 days).

Diet Study

Terns

There are four observation areas per tern species designated for the diet study (Table 1, Figure 1). Each of these sites has a permanent elevated blind that looks down into the plot. Each observation stint is conducted for three hours. During this time 5-10 nests within 10 m of the blind are observed. Any food items brought in by adult terns and fed to chicks are recorded. The amount, size and type of prey item are recorded as well as the time and the nest to which it is delivered. Only those prey items that a chick actually swallows are reported here. The size of the prey item is recorded in relationship to the size of the tern’s bill-length (e.g. 1 bill-length, 2 bill-lengths, etc.).

Alcids

There are 3 main sites per alcid species designated for the diet study (Figure 1, Table 1). At each site a 10 x 10m area is observed. Since it is difficult to identify individual burrows in the rocks, any feedings within the area are recorded. The time of delivery, the number of food items, prey species and size are recorded. Prey size is estimated by comparing to the length of the Puffin or Razorbill bill length (culmen) (e.g. 1 bill-length, 2 bill-lengths, etc.). If possible a photograph of the bird carrying prey items is taken to verify prey identification.

Leach’s Storm Petrels

A study of the distribution and contents of Leach’s Storm Petrel (Oceanodroma leucorhoa) nests was made by Dorothy McFarlane in 1998 (Aug. 10-11) and in 1999 (Aug. 24-26). The island was searched systematically and all the burrows encountered were investigated, where possible. Any adults or chicks encountered were banded. The locations of the nests were recorded on a map.

Environmental Conditions

Weather variables that are recorded twice daily (at 0900 and 2100) include: general conditions, wind speed and wind direction, visibility, cloud cover, and current temperature. General conditions are descriptions such as sunny, foggy, warm, windy, that characterise the day. Temperature (ºC) is recorded from a digital thermometer, which is located in the front porch of the CWS house, with its probe positioned in the shade under the front steps. Wind direction and speed (knots) are obtained from an anemometer, the dials for which are located in the lighthouse keepers’ house. Visibility is estimated using landmarks of known distance (nearby Gull Rock, North Rock, Grand Manan, Mainland, Gannet Rock).

Weather variables that are recorded once daily (at 2100) include: minimum and maximum temperature, rainfall and sea-surface temperature. The minimum and maximum temperatures are recorded from a digital thermometer (°C). Amount of rainfall (mm) is measured using the standard rain gauge located on the lawn between the two houses. When conditions are not too rough, sea-surface temperature (°C) is measured at 12:00 from the end of the tramway, using a standard sea-surface temperature bucket.

Tourist Questionnaire

In 1999, tourists that visited the island were encouraged to fill out a visitor questionnaire. The questionnaire asked some basic information including: where they were from, education level, occupation, age group, tour boat arrived on, the reason for the visit, and trip planning information. They were also asked their opinion on the research and banding taking place on Machias Seal Island.

Additional Observations

In addition to the seabird research on the island the ACWERN crew also records any species of bird that is seen on or near the island during the May-August field season. An effort is also made to observe if any of the species are also nesting on the island.

Tern Census

In 1998 the biannual tern census was conducted from 13-14 June. During the survey 2991 nests were counted. The Lincoln index indicated an error of 9.4%, which would bring the number of nests to 3272 (Newell and Mackinnon 1998). This number shows an increase over the 1994 and 1996 counts (Table 2). The increase may reflect a true increase in nesting population or may reflect the movement of terns from other islands. Early in the 1998 season a Peregrine Falcon on Petit Manan Island (the nearest tern colony in the Gulf of Maine) may have forced some terns to leave and nest elsewhere.

The ACWERN study examining the movement of Arctic Terns between colonies will help determine how much exchange there is between different islands. This will aid in the interpretation of yearly fluctuations in nesting numbers not only on MSI but also on other colonies.

In 1998, based on Jason Hudson’s observations of terns in 14 areas of the island, the species ratio of terns nesting on the island was determined to be 70% Arctic Terns and 30% Common Terns. Prior to 1996 fewer sites were observed in determining the species ratio, and Diamond (1999) determined that the actual proportion of Common Terns nesting on MSI was underestimated by about 50%. The actual proportion of Common Terns nesting on MSI has probably been consistently between 26% and 30% since 1996 (Diamond 1999).

Comparing the results of the two methods of species ratio counts done in 1999, reveals that the sampling method used by J. Hudson obtained results almost exactly the same as the method that counted the number of each species of tern nesting in each grid square (see Table 3). Thus, using a sampling of points around the island is an effective way of determining species ratio.

Puffin Burrow Density Survey

The data are shown in Table 4. Overall the burrow density averaged 0.5 burrows per square metre; the total area of puffin-burrow habitat was estimated at about 19,800 m² which gives an overall estimate of 9,900 burrows on the island (see Figure 2).

We could not estimate occupancy rate directly, as Rodway et al. (1996) did, because of the tern colony which covers the entire nesting area of puffins. However an indication is given by the proportion of the burrows marked in 1998 in study plots, which were occupied in 1999, i.e. 63/82 = 77%. Applying this to the total above suggests a population of 7,600 occupied burrows. We believe this is a considerable overestimate and will calibrate it based on the mark-recapture study we also carried out of puffin fledglings banded in burrows and recaptured at the light. This analysis is still in progress but suggests a figure closer to 2,000 pairs. The discrepancy probably arises from two causes: (i) our burrow count was carried out (necessarily, because of the terns) so early in the season that we include burrows that may not have been occupied for several years, and (ii) significant variation in density in different parts of the island.

Common Eider Census

In 1998, a total of 121 Common Eider nests were counted during the tern census. Eider nests were not counted in 1999. According to Newell and Mackinnon (1998) the population of Common Eiders nesting on MSI has essentially tripled since 1996. There is the opportunity to acquire some baseline information on the nesting eiders.

Capture of Seabirds

Terns

For the long-term demographic study of the terns and alcids nesting on MSI at least 200 adult birds need to be captured, banded and resighted. The total numbers of all seabirds that have been captured and banded are summarised in Table 5, and the band numbers used are given in Appendix II.

The number of adult Common Terns that have been captured and banded is still under 200 after the 1999 season although there were more captured this year than any other year. A primary focus in the 1999 season was to capture as many adult Arctic Terns as possible for the metapopulation/movement study, thus 253 were captured in this year alone. Of the 253 breeding Arctic Terns captured on nests, 28 were already banded; of these, 8 had been banded on other islands (Petit Manan, Matinicus Rock and Matinicus Seal). Field-readable bands were placed on all Arctic Terns banded in 1999 as this will aid the re-sighting and identification of marked birds. In the future more emphasis on Common Terns will be needed to reach the projected goal of 200 marked breeding adults in the population.

There were 562 tern chicks banded in 1999, 61% of which were Arctic Terns, and 39% were Common Terns. The ratio of Common Terns is somewhat higher than that found in the species ratio count. This could be due to the fact that the Arctic Terns nested a week earlier than the Common Terns so more of the Arctic chicks were flying at the time of the mass chick banding (July 12-15) and thus were missed, and to the substantially greater productivity of Common Terns in 1999.

Alcids

Since 1995, 325 adult Puffins have been banded (Table 5). Due to the trapping method used from 1995 to 1997 more non-breeding birds than breeding birds may have been captured in those years. The data collected when banding (e.g. presence/absence of a brood patch) need to be closely examined to determine the extent of non-breeding birds captured. In 1998 a new study plot (Northwest Notch) was established for the capturing of adult puffins in nests checked also for productivity. The majority of the adult puffins banded in 1999 were breeding birds captured in burrows.

ACWERN plans on developing a more effective capture method for Razorbills in the future because the current methods of capturing Razorbills (including the use of the drop boxes and capturing by hand) has resulted in the capture of very few adult Razorbills (Table 5). In addition to the capturing of more adult birds more effort will be made to re-sight marked individuals.

Productivity and Growth

Terns

The timing of peak laying for Common and Arctic Terns on MSI is generally 6-15 June (Table 6), although the Arctic Terns’ peak was May 31 this year. Peak laying of Arctic Terns tends to fall a few days to a week before the peak laying time for Common Terns. Peak hatching of terns is generally between the last few days of June and the first week of July.

The terns on MSI on average fledge between 0.6-1.0 chicks per nest for both species (Table 8). The productivity in 1999 was average for Arctic Terns but higher than average for Common Terns with a five-year high of 1.2 fledglings per nest. The 1996 season stands out as a year with very poor fledging success. This appears to have been related to weather (Amey 1998). Many other tern colonies to the south in the Gulf of Maine also reported very poor fledging success in 1996 (Hokama et al. 1996). The growth rates of Common and Arctic Terns during the 1996 season, as well as the 1998 season, also appear to be lower than for other years (Table 9, Figure 3, 4, 5 and 6). We are currently building a data base of measurements from known-aged chicks to improve our ability to discriminate between Common and Arctic Tern chicks when encountered outside of the established study plots.

Alcids

Timing of egg laying for Puffins on MSI generally falls at the end April to mid-May (Table 7). The first puffin chicks are usually found during the first two weeks of June. The laying time for Razorbills appears to be spread over a longer time period and generally falls between the second week of May and the first week of June. In the nests that are observed for productivity there is generally around 80% hatching success, and fledging success for puffins has ranged from 60-80% (Table 8). Fledging success for Razorbills has been more variable, ranging from 40-70% (Table 8). The Razorbill chicks are on the island for only approximately 2 weeks, and since we limit the number of times that we visit the burrows, we may underestimate the number of chicks that actually fledge (especially in 1998). In 1997 we began compiling measurements of known-aged chicks so that we will be better able to predict when a given chick will fledge. Future effort is needed to measure more known-aged chicks. The graphs in Figures 7 and 8 illustrate the growth measurements of known-age Puffin and Razorbill chicks for 1997 and 1999. The measurement of wing chord appears to give the clearest indication of age of the chicks. Linear regression equations for age based on wing chord for the linear period of growth are given.

Diet Study

Over 580 person-hours have been spent observing tern feedings and 387 hours have been spent observing alcid feedings since 1995 (Tables 10). Between 1995 and 1997 the major prey item fed to both Common and Arctic Tern chicks was herring (Table 11, Figure 9). In 1998 and 1999, although herring was still the major prey item fed to tern chicks, there was also an increase in the number of hake or rockling that was brought in. The Common Terns also brought in a much higher percentage of pollack in 1999 than in previous years; it was the second most common food item after herring. This may account for the high breeding success in Common Terns this year, as the pollack were larger than the herring (Table 12). Arctic Terns brought in very few pollack and were not as successful as the Common Terns. The pollack appeared in the diet later in the season, and the Common Terns nested a week later than the Arctic Terns.

It should be noted that the food items reported are those that were actually swallowed by the chicks. In 1999, in the case of the Arctic Terns, the hard-to-swallow butterfish composed only 5% of the diet, but they made up 10% of the items presented to the chicks.

Herring composed the majority of the Puffins’ and Razorbills’ diet in 1999 (Table 11, Figure 9). The Puffins appear to have shifted away from herring and more towards hake and sandlance in 1997, and almost entirely away from herring to hake in 1998. There may have been some underestimating of the amount of sandlance brought in by Puffins in 1998, however, examination of the photographs taken of puffins carrying fish confirms an abundance of hake/rockling in the prey items.

In most years the terns brought in prey items between 3.5 and 7 cm in length (Table 12). Young tern chicks tend to have difficulty swallowing any items that are much larger than this, although as the young approach fledging age, larger prey items can be swallowed. The size of prey items brought in by alcids is also linked to what the chicks are capable of swallowing. Puffins tend to bring in prey items 5 to 10 cm in length (Table 13) and Razorbills bring in prey items 3.5 to 10 cm in length (Table 14). In 1999, Razorbills brought in a much higher percentage of extra large items (>7 cm) than in previous years.

Leach’s Storm Petrels

Seventy-five petrels were banded in 1998 (25 of which were double banded) and 122 were banded in 1999, the majority of which were chicks (Table 15). Occupied petrel burrows were found around the perimeter of the eastern, northern and western parts of the island, where most of the burrows were associated with large boulders, as well as just inland of puffin burrows (Figure 10). Petrels used areas with tall weeds, as long as there were underground rocks to protect the burrow. Several burrows were found closely associated with, or within, puffin burrows. A pile of decaying boards was also used. It is estimated that there are at least 150 breeding pairs of petrels on the island.

Environmental Conditions

The 1999 field season was the driest season in the last 5 years, with just over 100 mm of rain between mid-May and the end of July (Table 16). There was relatively little fog in 1999, despite one stretch of 11 days of fog in late July. Although day-to-day variations in air temperature and sea-surface temperature may affect the behaviour of the seabirds nesting on MSI on a daily basis, there has been little difference in monthly trends between years (Table 16, Figure 11).

Tourist Questionnaire

There were 181 surveys completed over June and July, 1999, of which 88% of respondents were from the United States and the rest from Canada. The majority of respondents had gone to university (81%) and the largest occupation group (27%) indicated that they were retired. Forty-six percent of respondents were older than 55.

Over half (53%) indicated that they were regular members of a conservation or bird organization. Figure 12 shows the opinions about banding and research on the island. Generally, most people are in favour of research and banding, although banding was not as highly supported as research.

Additional Observations

Appendix I lists the bird species observed on or near MSI in the 1999 field season. In 1999 we observed courtship behaviour of the Common (Thin-Billed) Murre, however no nests were discovered. In 1999 no tree swallows nested on the island, although individuals were observed on the island in May. Two Laughing Gulls were consistently observed on the island from late May to mid-July, with as many a six at one time. The pair were observed landing in the Calamagrostis sp. (blue-joint grass) patch at the north end of the island but a search revealed no nest. Many of the warblers noted on the list passed through during migration in May. Several raptors did visit the island, mainly in May, and we witnessed a peregrine carrying off a puffin.

Two patches of Beach Head Iris (Iris setosa) were discovered on the island by Kate Devlin and the identification was confirmed by Hal Hinds who described it as an unusual find. The patches are located in the central portion of the island, one of which is beside the Windpole plot.

Without the help and encouragement of many people and institutions the study of seabirds on Machias Seal Island would not be possible. We would like to sincerely thank the Canadian Wildlife Service (Atlantic Region), the Lighthouse-Keepers past and present, Jason Hudson and other past CWS island wardens, Captain Andy Patterson, Captain Peter Wilcox, Captains John and Barna Norton, Reg Newell, Dr. Steve Kress (National Audubon Society), island research crews especially Krista Amey, Julie Paquet, Sue Tattersall, Nikki Benjamin, Dorothy McFarlane and others, as well as all the other people who have been involved in the past and present with the work on MSI.
 

Literature Cited

Amey, K. D. D. 1998. Seabirds as indicators of changes in availability and commercial weir landings of herring. Unpublished M.Sc. Thesis, University of New Brunswick, New Brunswick.

Bunin, J. S. and Boates. 1994. Effects of nesting location on breeding success of Arctic Terns on Machias Seal Island. Canadian Journal of Zoology 72:1841-1847.

Canadian Wildlife Service. 1994. Draft curator manual and management strategy for Machias Seal Island Migratory Bird Sanctuary. CWS. December 1994. 40 pp.

Diamond, A.W. 1999. Implications of the tern census on Machias Seal Island. Bird Trends 7:34-35.

Drury, W.H. 1990. The Gulf of Maine Tern Management Plan. Unpublished report to the Gulf of Maine Tern Working Group.

Harris, M. P. and T. R. Birkhead. 1985. Breeding ecology of the Atlantic Alcidae. Pp 156-204 in: The Atlantic Alcidae: The evolution, distribution and biology of the auks inhabiting the Atlantic Ocean and adjacent water areas. Eds. D.N. Nettleship and T.R. Birkhead. Academic Press Inc., London.

Harris, M. P. and S. Wanless 1989. The breeding biology of Razorbills, Alca torda, on the Isle of May. Bird Study 36:105-114.

Hawksley, O. 1957. Ecology of a breeding population of Arctic Terns. Journal of Field Ornithology 28:57-92.

Hokama, K., D. Ramil, and R. Borzik. 1996. Minutes of the Gulf of Maine Seabird Working Group, Bremen, Maine, 12 August 1996.

MacKinnon, C. M. and A. D. Smith. 1985. A summary of historical information on the Seabirds of Machias Seal Island. Canadian Wildlife Service, Sackville, New Brunswick

Newell, R. B. 1985. Nesting ecology of Arctic Terns Sterna paradisaea Pontoppidan in relation to habitat on Machias Seal Island. Unpublished M.Sc. Thesis, Acadia University, Wolfville, Nova Scotia.

Newell, R. B. 1994. 1994 Machias Seal Island Tern Census. Canadian Wildlife Service Unpublished Report.

Newell, R. B. 1996. 1996 Machias Seal Island Tern Census. Canadian Wildlife Service Unpublished Report.

Newell, R. and D. Mackinnon. 1998. Machias Seal Island Tern Census 1998. Unpublished report to Canadian Wildlife Service. July 1998.

Nisbet, I.C.T. and W.H. Drury. 1972. Measuring breeding success in Common and Roseate Terns. Bird Banding 43:97-106.

Paquet, J. and A.W. Diamond. 1998a. Seabird demographic and diet research on Machias Seal Island. Environmental Impact Assessment. Unpublished report to Canadian Wildlife Service. 18pp.

Paquet, J. and A.W. Diamond. 1998b. Behavioural responses of Arctic Terns (Sterna paradisaea) to foraging conditions. Environmental Impact Assessment. Unpublished report to Canadian Wildlife Service. 17pp.

Pettingill, O. S. 1939. History of one hundred nests of Arctic Terns. Auk 56:420-428.

Rodway, M.S., H.M. Regehr, and J.W. Chardine. 1996. Population census of breeding Atlantic Puffins at Great Island, Newfoundland in 1993-1994. Technical Report Series No. 263. Environmental Conservation Branch, Canadian Wildlife Service, Atlantic Region.

Walsh, P.M., D.J. Halley, M.P. Harris, A. del Nevo, I.M.W. Sim, and M.L. Tasker. 1995. Seabird monitoring handbook for Britain and Ireland. JNCC / RSPB / Seabird Group, Peterborough.

Weller, M. W. 1957. An automatic nest-trap for waterfowl. Journal of Wildlife Management. 21:456-458.

December 1999