Population Studies of Seabirds on Alcatraz Island, 2004

Transcription

1 1 Population Studies of Seabirds on Alcatraz Island, 2004 Draft Report to the Golden Gate National Recreation Area (GGNRA) National Park Service (NPS) David Gardner Julie Thayer PRBO Conservation Science 4990 Shoreline Highway Stinson Beach, CA William W. Merkle Golden Gate National Recreation Area Fort Mason, Building 210 San Francisco, CA 94123

2 2 TABLE OF CONTENTS EXECUTIVE SUMMARY 4 INTRODUCTION 5 METHODS 6 RESULTS AND DISCUSSION 7 CONCLUSIONS AND RECOMMENDATIONS 11 ACKNOWLEDGEMENTS 12 LITERATURE CITED 12 FIGURES Figure 1. Figure 2. Alcatraz Island seabird breeding areas, survey observation points, and significant structures, Brandt's Cormorant (BRCO) and Western Gull (WEGU) breeding population size on Alcatraz Island, Figure 3. Dynamics of five Brandt's Cormorant sub-colonies on Alcatraz Island, Figure 4. Overall Brandt's (BRCO) and Pelagic (PECO) Cormorant productivity on Alcatraz Island, Figure 5. Pelagic Cormorant breeding population size on Alcatraz Island, Figure 6. Western Gull productivity at two sub-colonies on Alcatraz Island, Figure 7. Pigeon Guillemot population size on Alcatraz Island, Figure 8. Reactions of Brandt's Cormorants on Alcatraz Island to disturbance in TABLES Table 1. Brandt s Cormorant population size on Alcatraz Island, Table 2. Brandt's (BRCO), Pelagic Cormorant (PECO) and Western Gull (WEGU) reproductive phenology on Alcatraz Island, Table 3. Brandt's Cormorant productivity by sub-colony on Alcatraz Island, Table 4. Brandt's (BRCO), Pelagic Cormorant (PECO) and Western Gull (WEGU) 25

3 3 reproductive performance on Alcatraz Island, Table 5. Pigeon Guillemot reproductive phenology on Alcatraz Island, Table 6. Summary table of types of disturbances to Brandt's Cormorants on Alcatraz Island, APPENDICES A. Alcatraz Western Gull Census Area Descriptions. 28 B. Alcatraz Western Gull Census Map,

4 4 EXECUTIVE SUMMARY Alcatraz Island has become a breeding site for a number of seabird species in recent years. In 2004, we monitored the population size, breeding phenology, and productivity of Brandt s Cormorants, Pelagic Cormorants, Western Gulls, Pigeon Guillemots, and Black Oystercatchers on Alcatraz. Disturbance to wildlife is a concern on Alcatraz, given its status as a heavily-visited national park and its location in the center of the San Francisco Bay. While monitoring the cormorant sites and carrying out the twice-weekly census, we recorded and cataloged disturbances to seabird populations from March through mid- September, Brandt s Cormorant population size in 2004 numbered 752 breeding pairs, the largest yet recorded for this colony, representing a 29% increase from 584 breeding pairs seen in In 2003, Brandt s Cormorants did not breed in the Laundry Building sub-colony, however in 2004 this sub-colony was again used and expanded to include a large area above the cliffs. Cormorant numbers increased in other areas of the island as well. The Northern sub-colony expanded to 226 Brandt s Cormorant breeding pairs, stretching south from the Incinerator, along the cliff to Barker Beach. The first eggs of the season were laid between 11 and 17 March. Productivity of Brandt s Cormorants was 2.4 chicks fledged/pair in No double-brooding was attempted in The Pelagic Cormorant population has decreased over the years and appears to have stabilized around 12 to 14 breeding pairs since Productivity of Pelagic Cormorants in 2004 was the lowest ever documented at 0.7 chicks fledged/pair. Western Gull breeding numbers have remained fairly stable since 2003, increasing only slightly from 933 to 940 breeding pairs in Productivity of Western Gulls was 1.9 chicks fledged/pair, similar to recent years. We confirmed 33 Pigeon Guillemot nests sites in 2004, the highest number yet documented. One pair of Black Oystercatchers bred on the Seawall in 2004, fledging one chick. In 2004, 5 pairs of California Gull succeeded in nesting and hatching eggs. All five pairs were found close together on the walkway behind (south of) the rubble piles on the side of the Parade Ground. No successful fledging could be documented due to limited access for disturbance reasons; biologists were only able to survey a couple of times through the year. When last observed, many of the chicks were old enough to have begun growing both wing and mantle feathers. The presence of California Gulls has been noted at the start of the breeding season of 2002, but they disappeared prior to nest establishment. We documented only 43 disturbances to Brandt s Cormorants in 2004 during 340 hours of observation. This rate of disturbance is lower than that in the past five years of our observations. However, the proportion of major disturbances increased, resulting in cormorants flushing from breeding or roosting areas. Also a number of serious island based disturbances were observed in 2004; caused by both visitors, NPS and the U.S. Coast Guard. We recommend strict enforcement of the rule that no public visitation or construction activity be carried out within view of the Western cliffs after 15 February, especially if activities may influence the sensitive Pelagic Cormorant population. To minimize disturbances, we continue to encourage plans to post signs visible to both visitors on the island and to passing boaters that explain proximity restrictions with the threat of law enforcement. Alcatraz Island offers a unique opportunity for public to view seabird breeding activities up close, and this should be highlighted as a focal point of visitor education and outreach. Improved signage in sensitive areas and interpretive tours around the island will help educate visitors as well as protect breeding seabirds.

5 5 INTRODUCTION A number of colonial waterbird species inhabit Alcatraz Island (37º 49'N, 122º 25'W) in San Francisco Bay. Alcatraz is a part of the Golden Gate National Recreation Area (GGNRA) and is managed by the National Park Service (NPS). Waterbird species of interest include Brandt's Cormorants (Phalacrocorax penicillatus), Pelagic Cormorants (P. pelagicus), Western Gulls (Larus occidentalis), Pigeon Guillemots (Cepphus columba), Black Oystercatchers (Haematopus bachmani), Black-crowned Night Herons (Nycticorax nycticorax), Snowy Egrets (Egretta thula), and Great Egrets (Casmerodius albus). The Brandt s Cormorant colony on Alcatraz is one of the few known estuarine breeding sites for this species. Pigeon Guillemots breed nowhere else in the San Francisco Bay, and the Western Gull, Black-crowned Night Heron, and Pelagic Cormorant colonies are among the largest in the Bay. This diversity of species, although protected by the Migratory Bird Treaty Act, National Park Service Management Policies, and NPS-77 Natural Resource Management Guidelines, exists in a delicate balance with the considerable human presence both on and around Alcatraz Island. Colonial waterbird populations on Alcatraz experience substantial disturbance from a number of different sources. Over 1 million visitors tour the island annually, and associated historic preservation and safety construction projects, public access to breeding areas, and special events may disrupt the breeding efforts of Alcatraz seabirds. Encroachment near the Alcatraz shoreline by large numbers of commercial and/or recreational boaters (e.g. tour boats, fishermen, kayakers), and uncontrolled aircraft overflights (e.g. air tour operators), may have similar effects. In addition, dredging and other projects which disturb and alter the subtidal environment are potentially disruptive to seabird populations, as these activities may remobilize contaminants, increase turbidity, and destroy essential foraging habitat. In 1993, GGNRA completed a management plan for Alcatraz Island, which included provisions for maintaining breeding populations of colonial waterbirds could cite the DCP This plan emphasized protection of the island's natural resources, while maintaining opportunities for visitor access, special events, and other island uses. The plan called for natural resource monitoring and the development of protocols to determine baseline information for key wildlife populations. Since 1996, PRBO Conservation Science (formerly Point Reyes Bird Observatory) in conjunction with GGNRA has conducted wildlife studies with the goals of (1) establishing the distribution, abundance, and reproductive performance of waterbird species, (2) assessing the extent and effects of various forms of disturbance, and (3) assisting management personnel in developing appropriate and effective policies to protect waterbird populations. This report details results of monitoring efforts during the 2004 breeding season. Our objectives in 2004 were to describe and measure the distribution, abundance, reproductive performance, and breeding phenology of Brandt's Cormorants, Western Gulls, Pelagic Cormorants, Pigeon Guillemots, and Black Oystercatchers, as well as to document the extent and effects of disturbance on these populations. Monitoring was focused, as in past years, upon on Brandt's Cormorants because this population is relatively new to Alcatraz and the San Francisco Bay (having established a breeding colony in 1991), and is suspected to be sensitive to human disturbance (Ainley and Lewis 1974, Boekelheide et. al.1990).

6 6 METHODS We conducted island- and boat-based surveys from March through August, Island surveys consisted of censuses and focal nest observations twice per week, on Saturdays and Wednesdays. In order to minimize disturbance to nesting birds, we made observations using binoculars (10 x 50) from concealed locations including a blind near Barker Beach, the Laundry Building, and the Model Industries Building (Figure 1). We also made remote observations of colonies using a spotting scope (Bushnell 20-60x). We monitored Western Gulls from observation points on the Cistern, Lighthouse Plaza and Model Industries Plaza (Figure 1). Boat surveys took place once every two weeks. We circled the island slowly at a distance of at least 100 meters from shore and made observations using binoculars and a digital camera with 300mm zoom lens. BRANDT S CORMORANT In each sub-colony visible from the island, we counted the total number of adult and immature Brandt s Cormorants twice weekly, between 08:00 and 11:00. These included the Southern (except for the Gap see Figure 1), Northern, Barker Beach and Laundry sub-colonies, as well as part of the Model Industries sub-colony. We observed the Gap, South Bricks, and parts of the Model Industries sub-colony not visible from the island every two weeks by boat. During each survey, we recorded the total number of adults, immature birds, nests with incubating birds, nests with chicks, and total number of chicks in each subcolony. We investigated Brandt s Cormorant reproductive phenology and performance in a sample of nests from the Southern, Barker Beach, Northern, Laundry and Model Industries sub-colonies. Groups of at least 20 visible nests were selected for each sub-colony, including both centrally-located and edge nests. Nests were numbered and mapped, and the contents of each nest recorded twice weekly (number of eggs and/or chicks), with no more than 4 days separating observations. We followed chicks until they wandered from nests and formed crèches (groups of chicks), at which point we considered them fledged, if at least 28 days old. We calculated clutch size, brood size, hatching success (percent of eggs hatched) and fledging success (percent of chicks surviving to fledge), and total productivity (chicks fledged per nest). PELAGIC CORMORANT Pelagic Cormorants were censused and monitored similarly and concurrently to Brandt s Cormorants (twice weekly between 08:00 and 11:00). Pelagic Cormorant nests are typically built on small, narrow cliff ledges; therefore, chicks have less opportunity to wander and crèche than do Brandt s Cormorant chicks. Consequently, we considered Pelagic Cormorant chicks fledged when they reached full feathering (all feathers grown in, capable of flight). For all other monitoring methods concerning reproductive performance and phenology, refer to Brandt s Cormorant methods above. WESTERN GULL We censused Western Gulls using standard protocols developed for Alcatraz in 1990 (Bell 1990). Two separate nest counts were obtained during the ten-day period following peak egg-laying (as determined from phenology in study plots). We divided the island into census areas and counted all nests in each area, recording nest contents whenever visible (Appendices A and B). We walked through accessible parts of the island and counted nests; boat surveys were used to survey inaccessible parts of the island. We assessed Western Gull reproductive performance and phenology by monitoring individual nest contents (number of eggs and/or chicks) weekly in study plots on the Cistern and the Parade Ground,

7 7 using binoculars and spotting scopes. Phenology was documented by determining mean dates of egglaying, hatching, and fledging at monitored nests. Fledging was assumed when chicks were fullyfeathered and therefore capable of flight. Reproductive performance was assessed by determining hatching success, fledging success, and total productivity. CALIFORNIA GULL California Gulls were monitored haphazardly due to the late discovery of their breeding attempts in California Gulls have begun to breed behind the rubble piles (Figure 1), and subsequently can only be monitored during the Western Gull censuses and by reports from the USGS Black-crowned Night Heron surveys. PIGEON GUILLEMOT We monitored Pigeon Guillemots during cormorant surveys and opportunistically, both from the island and by boat. We mapped and numbered nest crevices as they were identified throughout the season. Active nest sites were confirmed by observations of chicks or by parental delivery of fish to a crevice, indicating presence of a chick, or by presence of egg/chick remains found post-season. Probable nest sites were defined by regular attendance of adults. These methods have the potential for underestimation of breeding numbers, as we may not have documented nests which failed early in the breeding season. BLACK OYSTERCATCHER During each cormorant survey, we recorded the presence and behavior of adult Black Oystercatchers near the single known nest site on the Seawall, and we noted nest contents (eggs or chicks). We also recorded the locations and behaviors of any oystercatchers seen on or around other areas of the island. DISTURBANCE MONITORING During all monitoring activities, we documented disturbances to nesting seabirds. For each disturbance event, we described the event and its observed effects, including the approximate distance of the event from the colony and the number of birds affected. This report focused on details of events that caused noticeable disruption to cormorant breeding activities, as this species is considered sensitive to disturbance (Ainley and Lewis 1974, Boekelheide et. al. 1990). We classified disturbances as major, moderate, or minor. Major disturbance events caused cormorants to flush from breeding or roosting areas. Moderate disturbances caused agitation in cormorants such as fluffing, growling, threat gestures or standing up off nests. Disturbance was considered minor if cormorants only looked in the direction of the event. As in 2003, additional disturbance monitoring was carried out in 2004 on the North end of the island, to obtain baseline data for comparison with the potential future opening of the Model Industries Plaza and Laundry Building to limited public access during the seabird breeding season. RESULTS AND DISCUSSION BRANDT S CORMORANT Brandt s Cormorants continued their breeding population growth in 2004, increasing from 584 breeding attempts in 2003 to 752 in 2004, the largest number recorded for this colony to date (Figure 2). This

8 8 population growth equals approximately 29% growth since the previous year. Since 1997, the Brandt s Cormorant population has continued to grow at an impressive rate (β = 82.15, p < 0.005, R 2 = 0.85; n = 8). We observed 126 breeding attempts in the Southern sub-colony (including the Gap area), 226 in the Northern sub-colony, and 85 in the Laundry Building sub-colony (Table 1). We obtained maximum counts of 88 nests in the Model Industries sub-colony, 156 nests in the Barker Beach South sub-colony and 71 nests in the South Bricks sub-colony from a combination of island and boat surveys. The Northern sub-colony continued its expansion south of the Incinerator all the way to Barker Beach and included renewed breeding attempts north of the Incinerator. The increase in Brandt s Cormorant breeding population size may be due to a combination of factors. Nine years of high productivity (an average of 2.1 chicks fledged/pair), combined with the possibility of greater over-winter survivorship due to recent cold-water conditions (Boekelheide and Ainley 1989, Nur and Sydeman 1999) likely resulted in high recruitment each year. Southeast Farallon Island also experienced a population increase in 2004 (Warzybok et al. 2004). As in , Brandt s Cormorants constructed many nests in open view of visitor areas in This occurred above the Southern sub-colony near the Gap (visible from West Road), at the Incinerator (visible from West Road and Metal Detector), and most prominently both north and south of Barker Beach. Before 2001, only a small part of the Northern sub-colony was visible from areas regularly open to visitors. Reasons for colony expansion into these areas may include space limitations in established sub-colonies because of pop increase, coupled with possible preference of areas with less slope or other microhabitat characteristics for breeding. Brandt s Cormorant breeding activity began in February in 2004; display and nest building activity was already occurring when observations began on 14 February (Figure 3). There was a large abandonment of nests in the Northern sub-colony in early March for unknown reasons, but birds returned by 27 March. The first pairs laid eggs on or before 17 March in the Barker Beach South sub-colony, however the earliest mean lay date was in the Southern sub-colony on 10 April (s.d. = 24 days, n = 19; Table 2). Mean lay dates in the Northern, Model Industries and Barker Beach sub-colonies were earlier than last year, on 4 May (s.d. = 5 days, n = 19), 21 April (s.d. = 14 days, n = 26), and 16 April (s.d. = 15 days, n = 16) respectively. The Laundry Building sub-colony formed later in the season and subsequently had the latest mean lay date of 24 May (s.d. = 22 days, n = 6). Breeding phenology on Alcatraz has been consistently earlier than the nearby, large Farallon Island population since 1997 (PRBO unpublished data). The last dependent chick was observed on 16 September in the South Bricks sub-colony. Over the past ten years productivity has fluctuated around a mean of 2.1 chicks per pair (s.d. = 0.3, n = 10). Productivity rose once again to 2.4 in 2004 (s.d. = 1.2, n = 161; Figure 4). The Laundry sub-colony had the lowest productivity of all the sub-colonies in 2004, at 1.5 chicks per pair (s.d. = 1.3, n = 22; Table 3) while the Model Industries sub-colony had the highest productivity at 2.6 chicks per pair (s.d. = 1.1, n = 47; Table 3). For the past 3 years, Brandt s Cormorant productivity on Alcatraz has remained higher than that at other central California colonies which averaged < 1.8 chicks/pair (Thayer and Gardner 2005, Warzybok et al. 2004). This was the first year since 2000 in which no Brandt s Cormorants attempted a second clutch. Alcatraz has been, to the best of our knowledge, the only cormorant colony to attempt double-brooding apart from several records in Great Britain (Wanless and Harris 1997). Ainley et al. (1990) proposed that early breeding season conditions were the principal determinant as to whether second clutches appear in Cassin s Auklets (Ptychoramphus aleuticus), and Wanless and Harris (1997) came to a similar conclusion concerning double-brooding European Shags. Brandt s Cormorants began breeding earlier than ever

9 9 recorded on Alcatraz in 2004 however, suggesting an alternative factor may influence double-brooding attempts. PELAGIC CORMORANT The Pelagic Cormorant population has decreased, though not significantly (β = -0.76, p > 0.1; n = 8) over the years and appears to have stabilized around 12 to 14 breeding pairs since Thirteen Pelagic Cormorant breeding attempts were recorded in 2004 (Figure 5). Nests were located on cliffs below the Model Industries Building, as well as a lone site at Barker Beach (Figure 1). Pelagic Cormorants started building nests in mid February in Egg laying was asynchronous, starting on 21 April and continuing through 16 June (Table 2). Mean lay and fledge dates were 16 May (s.d. = 22, n = 7) and 10 August (s.d. = 19, n = 3), respectively. Unlike Brandt s Cormorants, Pelagic Cormorant productivity was lowest ever recorded on Alcatraz, at 0.7 chicks/pair (s.d. = 1.0, n = 12; Figure 4). This is of particular concern, especially since productivity at other central California colonies in 2004 was higher with > 0.8 chicks/pair on Ano Nuevo Island (Thayer and Gardner 2005), and > 2.3 chicks/pair on the Farallones (Warzybok et al. 2004). Several factors may have negatively affected breeding Pelagic Cormorants in recent years. First of all, the warm-water conditions of 2003 and 2004, may have adversely affected reproductive success through a mechanism of reduced prey availability. Secondly, the continued growth of the Brandt s Cormorant population has resulted in inter-specific competition with Pelagic Cormorants for nest sites. Thirdly, a fence was erected in 2003 just 30 feet from the edge of the cliff where Pelagic Cormorants breed, which may have disturbed early breeding activity. Additionally, visitor access in 2004 was allowed down to that fence throughout the breeding season, including the very sensitive nest-initiation stage. Finally, other staff activity in the Model Industries Building and surrounding plaza during the breeding season has increased in recent years potentially adversely influencing Pelagic Cormorant nesting (see Disturbance Monitoring below). WESTERN GULL For the first time in five years, the Western Gull breeding population growth appears to have stabilized. Approximately 933 nesting pairs were counted in 2003, whereas in 2004 there were 940 breeding pairs (Figure 2). Fluctuations were apparent at various areas on the island, and a small general increase was seen around Alcatraz. The small growth the in the population was additionally supplemented by the reduction of nest removal as part of a control effort by the NPS. The first Western Gull eggs in 2004 were observed on 14 April, and the mean lay date was 7 May (s.d. = 5.3, n = 79; Table 2), extremely consistent with the past several years, The average hatch date was 3 June (s.d. = 5.8, n = 82). Western Gull hatching success was 0.9 (s.d. = 0.3, n = 92) and fledging success was 0.9 (s.d. = 0.3, n = 84; Table 4). Productivity was 1.9 chicks per pair in 2004 (s.d. = 1.1, n = 103; Figure 6), similar to recent years. CALIFORNIA GULL In 2004, 5 pairs of California Gull succeeded in nesting and hatching eggs. All five pairs were found close together on the walkway behind (south of) the rubble piles on the side of the Parade Ground. No successful fledging could be documented due to limited access for disturbance reasons; biologists were only able to survey a couple of times through the year. When last observed, many of the chicks were old enough to have begun growing both wing and mantle feathers. The presence of California Gulls has been noted at the start of the breeding season of 2002, but they disappeared prior to nest establishment.

10 10 PIGEON GUILLEMOT In 2004, we first detected Pigeon Guillemots rafting in the water off Alcatraz on 17 March, and attending nest sites on the island on 17 April (Table 5). A high count of 52 adults was recorded both on 16 April and on 10 July, the highest number of Pigeon Guillemots observed since our surveys began in We confirmed 33 Pigeon Guillemot nest sites in 2004, and located an additional 7 probable nest sites, the highest numbers yet recorded (Figure 7). BLACK OYSTERCATCHER A single pair of Black Oystercatchers bred on the Seawall in Adults were seen at the Seawall and around the island regularly from 27 March to 3 July. Eggs were laid between 8 and 12 May in a nest concealed within vegetation on the Seawall. The two chicks were observed on 26 May, and a third chick was seen on 2 June. The only chick left by 23 June was fully feathered by 3 July and after fledging was seen occasionally foraging in the inter-tidal below the Seawall with both parents. The chick was last seen in early August. Black Oystercatchers on Alcatraz have fledged only four chicks since 1997 (one each in 1997, 2001, 2003 and 2004). DISTURBANCE MONITORING The rate of disturbance to seabirds on Alcatraz decreased in 2004 to 0.13 disturbances per hour, compared to a mean of 0.74 disturbances per hour for the past five years (Table 6). The percentage of major disturbances however has increased, accounting for 49% of the total observed, while moderate disturbances represented 7% and minor disturbances 44% of the total (Figure 8). Nevertheless it is possible that regularly disturbed colonies are likely to persist only if the advantages for seabirds to stay at the colony outweigh the disadvantages. For instance, prey availability near Alcatraz Island may be high or predictable in some years, prompting cormorants to breed there, but environmental conditions negatively affecting feeding conditions may intensify the effects of disturbance felt by the birds, resulting in colony abandonment as at North sub-colony during the 1998 El Niño event. Extra surveys on the North end of the island resulted in additional hours of disturbance data collection. We observed 22 disturbances at the North end, over 50% of total disturbances observed on the whole island, with a slightly smaller proportion of major disturbances, and a larger proportion of minor disturbances than on the rest of the island. Of these disturbances, only 45% were marine-based, a large reduction from 77% observed in The remaining disturbances observed from the North end of the island were predominantly caused by air traffic. As in past years, disturbances that resulted in the most severe reactions from birds included the presence of kayaks and other watercraft moving slowly, quietly and/or erratically while close to the shoreline of the island. Seven of eight observed kayak/canoe/row-boat related disturbances accounted for a third of the incidences in which we observed birds flushing from the island in An example of such disturbances was seen on 14 February. A large kayak race took place, with more than 43 out-rigger kayaks passing within 10m of all the west side of Alcatraz, flushing 3 breeding Pelagic Cormorants, 20 breeding and >150 roosting Brandt s Cormorants. In contrast to 2003, large party-boat disturbances were less commonly observed disturbing breeding seabirds in 2004, comprising only 5% of total disturbances compared with 21% from last year (Table 6). Even so, we observed large numbers of party-boats within 10m of the island, suggesting potential habituation by some cormorants to frequent motorized boat traffic.

11 11 Low flying aircraft, especially helicopters, also made up a large percentage of disturbances (33%; Table 6). This proportion of total disturbances was similar to 2003, however the frequency of disturbance was lower throughout the year. There were noticeably fewer flights from aircraft that were contacted by the NPS as repeat offenders last year. For example, based on FAA identification numbers, a specific air tour operator that was the cause of 11% of all disturbances recorded in 2003, was not noted to cause any disturbances in Over-flight disturbances were generally not as severe as water-based disturbances, and did not usually cause birds to flush. Aircraft can be seen and heard often long before they come close to Alcatraz, and it appeared that birds were not often surprised by these events. The increased signage of closed areas that was erected in 2002 did not appear as successful in 2004 as it was in We were twice notified of severe infractions into enclosed areas that occurred while we were not on the island. On 18 May, C. Hellwig/NPS observed 50 Brandt s Cormorants flush and over 25 Western Gulls alarm calling when three teenagers went into an off-limits area west of the Laundry Building. In a separate incident on the same day, B. Batom/NPS noted 2 adults entering a closed area near the Northern sub-colony, at the Incinerator. He stated that they disturbed many birds. Two other island-based disturbances were recorded near the Brandt s and Pelagic Cormorant breeding sub-colonies. A Coast Guard contractor conducting bi-annual maintenance resulted in the flushing of roosting Brandt s Cormorants near the North Foghorn. The second disturbance occurred by the NW entrance to the second floor of the Laundry Building, as a NPS maintenance crew of two worked on building new stairs up to the entrance. This disturbance caused Brandt s Cormorants to look up every now and then, and occasionally stand up off the nest. No Pelagic Cormorant could be seen by the observer, so disturbance on that species is unknown. This disturbance raised communication issues, because no NPS natural resources staff were informed of the maintenance, and therefore were not able to suggest correct procedures and monitoring to reduce potential disturbance. To provide a visual barrier between breeding birds and newly-opened public access at the Model Industries Plaza, a double layer shade cloth was installed on the fence between the Model Industries Building and the Laundry Building. This visual barrier was an improvement over the single layer of shade cloth utilized in 2003, however, by the end of the season, the material was very tattered and flapping in the wind. Constant effort was required to keep the shade cloth tied down sufficiently to prevent disturbance to breeding birds. An improvement should be considered, because this visual barrier is extremely important due to the initiation of the new Upper-Laundry Building Brandt s Cormorant subcolony. Over 70 pairs of Brandt s Cormorants bred along the top of the cliff edge in 2004, less than 50 feet from the double-shade cloth and fence. CONCLUSIONS AND RECOMMENDATIONS Only 13 Pelagic Cormorant breeding attempts were recorded in 2004, and productivity was lowest ever recorded at 0.7 chicks/pair. The Pelagic Cormorant is a very sensitive species and may be influenced by multiple factors, including human disturbance, environmental stochasticity, and interspecific competition with Brandt s Cormorants for limited breeding habitat. To protect a shrinking Pelagic Cormorant population from disturbance, construction activities around the cliffs are minimized as much as possible before breeding season, and ceased completely after 15 February, as mandated in the Alcatraz Final Environmental Impact Statement of 2001 (NPS/GGNRA), and perhaps that rule should apply also to visitor activities. Brandt s Cormorants continued their breeding population growth in 2004, increasing to 752 breeding pairs, the largest number recorded on Alcatraz since monitoring began in The Pigeon Guillemot

12 12 breeding population also increased suggesting potential higher food availability in a changing bay environment (more saline) may be advantageous to these pelagic seabird species. Sustained high productivity in recent years as evidenced in Brandt s Cormorants, combined with the possibility of greater over-winter survivorship in recent productive cold-water conditions likely resulted in high recruitment to the Alcatraz breeding populations. Disturbance to seabirds decreased in 2004 compared to the past five years, yet major disturbances accounted for a larger proportion of total disturbances. Additional North end monitoring resulted in extra hours of disturbance data. Kayaks, row boats and canoes were the main cause of many of the major disturbances that occurred during the early part of the breeding season. However, aircraft flyovers were less frequent, especially repeat offenders from 2003 that were contacted in person by NPS/GGNRA staff. To lessen the potential of colony-threatening disturbances, we recommend improved signs with the threat of law enforcement ticketing in sensitive closed areas. We also encourage existing plans to install signs visible to watercraft around the island warning to maintain a 100m buffer around the island during the breeding season of March through August. Improved and standardized communication with repeat and potential offenders and such as outrigger canoe clubs, kayak clubs, yacht harbors and local tour companies may also reduce severe disturbances. Island based disturbance to cormorants by U.S. Coast Guard personnel and contractors occurred twice during the breeding season. We urge that access to the foghorns for necessary bi-annual service be scheduled before and after the breeding season to minimize disturbance to breeding birds. There is a need to set up consistent scheduling and follow-up communication between NPS/GGNRA natural resources staff and the U.S. Coast Guard and contractors. We recommend strict enforcement of the rule that no public visitation or construction activity be carried out within view of the Western cliffs after 15 February, especially if activities may influence pre-breeding or early-nesting activities of Pelagic Cormorants. Visual barriers between humans and breeding birds, especially in the Model Industries Plaza area, must be more rigid and durable to make them effective. Additionally, improved signage in sensitive areas and enhanced interpretive tours around the island will help protect breeding seabirds as well as educate visitors about wildlife on this unique island. ACKNOWLEDGEMENTS This study was funded by Golden Gate National Parks Conservancy in cooperation with PRBO Conservation Science. In addition, we received generous financial support from The Osher Foundation in 1997 through We would also like to thank NPS Rangers and other employees, as well as Blue and Gold Fleet personnel, for their interest, encouragement, and logistical support. Special thanks to Chris Hellwig for many hours of service. Thanks to Meredith Elliot for help with boat support. This is PRBO contribution number LITERATURE CITED Ainley, D. G. and T. J. Lewis The history of the Farallon Island marine bird populations, Condor 76: Ainley, D. J., R. J. Boekelheide, S. H. Morrell and C. S. Strong Cassin s Auklet. Pp in Seabirds of the Farallon Islands. (Ainley, D.J. and R.J. Boekelheide, eds.). Stanford University

13 13 Press, Palo Alto, CA. Bell, D Alcatraz Island Western Gull nest survey, Unpublished report to the National Park Service, Golden Gate National Recreation Area. Museum of Vertebrate Zoology, University of California, Berkeley, CA. Bell, D Alcatraz Island Western Gull nest survey, Unpublished report to the National Park Service, Golden Gate National Recreation Area. Museum of Vertebrate Zoology, University of California, Berkeley, CA. Boekelheide, R. J. and D. G. Ainley, Age, resource availability, and breeding effort in Brandt s Cormorant. The Auk 106: Boekelheide, R. J., D. G. Ainley, S.H. Morrel, and T.J. Lewis Brandt's Cormorant. pp in Seabirds of the Farallon Islands: Ecology, Dynamics, and Structure of An Upwelling System Community. (Ainley, D.J. and R.J. Boekelheide, eds.). Stanford University Press, Palo Alto. Brown, M. E Population monitoring for Western Gulls (Larus occidentalis) on Alcatraz Island, California. Unpublished report to the National Park Service, Golden Gate National Recreation Area. Biology Department, University of Dallas, TX. Carter, H.R., G.J. McChesney, J.E. Takekawa, L.K. Ochikubo, D.L. Whitworth, T.W. Keeney, W.R. McIver, and C.S. Strong Population monitoring of seabirds in California: aerial photographic surveys of breeding colonies of Common Murres, Brandt s Cormorants, and Double-crested Cormorants. Unpublished Final Report, U.S. Geological Survey, Biological Resources Division, California Science Center, Dixon, CA. 3. Nur, N. and W. J. Sydeman Survival, breeding probability, and reproductive success in relation to population dynamics of Brandt s Cormorants Phalacrocorax penicillatus. Bird Study 46: Thayer, J.A., and D.G. Gardner Long-term studies of seabirds on Año Nuevo Island and Mainland, Unpublished report to State of California Department of Parks and Recreation, Bay Area District, Año Nuevo State Reserve. Point Reyes Bird Observatory, Stinson Beach, CA. Warzybok, P.M., R.W. Bradley, and W.J. Sydeman Population size and reproductive performance of seabirds on Southeast Farallon Island, Unpublished final report to the USFWS Farallon National Wildlife Refuge, CA. Wanless, S. and M. P. Harris Successful double-brooding in European Shags. Colonial Waterbirds 20(2):

14 14 Observation Points Pigeon Guillemot breeding sites Common Raven nest Model Industries sub-colony Laundry Building sub-colony Brandt s Cormorant breeding areas Brandt s Cormorant roosting areas Pelagic Cormorant breeding areas Black Oystercatcher breeding area California Gull breeding area Northern sub-colony Barker Beach sub-colony Southern sub-colony including the Gap South Bricks sub-colony Figure 1. Alcatraz Island seabird breeding areas, survey observation points, and significant structures, 2004.

15 BRCO Number of breeding pairs WEGU Figure 2. Brandt's Cormorant (BRCO) and Western Gull (WEGU) breeding population size on Alcatraz Island, Data represent the minimum number of breeding pairs on Alcatraz. Breaks in data indicate a change in observers and/or census methodology. BRCO data in 1991 from R. Farwell/GGNRA, pers. obs.; in 1993 from D. Hatch/GGNRA (unpubl. data); in 1994 from R Hothem/USGS and W. Reyes/GGNRA, per. obs.. WEGU data in 1990 from Bell (1990); in 1991 from Bell (1991); in 1995 from D. Hatch and Fish/GGNRA (unpubl. data); in from Brown (1997); in 1998 from M. Brown/Univ. of Dallas (unpubl. data).

16 Southern sub-colony (S) Barker Beach sub-colony (BB) Northern sub-colony (N) Laundry Building sub-colony (LB) Model Industries sub-colony (MI) /8 2/18 2/28 3/9 3/19 3/29 4/8 4/18 4/28 5/8 5/18 5/28 6/7 6/17 6/27 7/7 7/17 7/27 8/6 8/16 8/26 9/5 9/15 Number of active nests 1 st pre-breeding activity at S, BB & MI: 14 Feb 1 st pre-breeding activity at N: 28 Feb 1 st eggs laid at MI: Mar 1 st pre-breeding activity at LB: Mar Last chicks at S: 25 Aug Last chicks at BB, N, LB & MI: 1 Sept Last chick at South Bricks: 16 Sept 1 st eggs laid at BB: Mar 1 st eggs laid at S: Mar 1 st eggs laid at LB: Apr 1 st eggs laid at N: Apr Figure 3. Dynamics of five Brandt's Cormorant sub-colonies on Alcatraz Island, Number of active Brandt's Cormorant nests (well-built nests with eggs/chicks or birds in incubation posture) are shown for each sub-colony.

17 17 3 BRCO (6) PECO (100) (187) (138) (6) (98) (10) (161) Number of chicks fledged. 2 1 (181) (231) (92) (16) (10) (208) (139) (12) Figure 4. Overall Brandt's (BRCO) and Pelagic (PECO) Cormorant productivity on Alcatraz Island, Sample size is in parentheses productivity was calculated from total chick counts productivity was calculated from number of chicks fledged per pair followed.

18 Number of breeding pairs Figure 5. Pelagic Cormorant breeding population size on Alcatraz Island, Data in 1996 from M. Parker/USFWS aerial surveys.

19 19 3 Number of chicks fledged/pair 2 1 (63) (54) (52) (76) (43) (49) (52) (47) (55) (47) (43) (60) Parade Ground Cistern Figure 6. Western Gull productivity at two sub-colonies on Alcatraz Island, Sample size in parentheses.

20 20 60 Maximum Adult Raft Count Total Confirmed Breeding Sites Total Probable Breeding Sites 46 Total number Figure 7. Pigeon Guillemot population size on Alcatraz Island,

21 21 Percent Occurrence (%) (45) (61) (95) (42) (22) (118) (20) (32) (155) (34) (11) (22) (34) (25) (90) Major Moderate Minor (21) (19) (3) Year Figure 8. Reactions of Brandt's Cormorants on Alcatraz Island to disturbance in Major disturbance events caused cormorants to flush from breeding or roosting areas. Moderate disturbances caused agitation in birds such as fluffing, growling, threat gestures or standing up. Disturbance was considered minor if birds only looked in the direction of the event.

22 Table 1. Brandt s Cormorant population size on Alcatraz Island, YEAR TOTAL Southern Sub-Colony Northern Sub-Colony POPULATION SIZE (breeding attempts) Laundry Sub-Colony Model Indstr. Sub-Colony South Bricks Sub-Colony 4 Gap Area 4 Barker Beach Sub-Colony METHOD no data no data no data no data no data no data no data incidental observation SOURCE R. Farwell/GGNRA, pers. obs no data no data no data no data no data no data no data no data no data no data no data 0 0 ground survey no data no data no data no data no data no data incidental observation 1995 ~ ~ 100 ~ ground & boat photographic surveys D. Hatch/GGNRA, unpubl. data R. Hothem & W. Reyes, pers. obs. D. Hatch/GGNRA, W. Sydeman/PRBO, unpubl. data no data no data 0 ground survey PRBO data ground & boat surveys PRBO data ground & boat surveys PRBO data ground & boat surveys PRBO data ground & boat surveys PRBO data ground & boat surveys PRBO data ground & boat surveys PRBO data ground & boat surveys PRBO data ground & boat surveys PRBO data 1 Observed during 1993 Western Gull/ general seabird ground survey. 2 Observed during 1994 Black-crowned Night Heron survey; apparently only one of two Brandt's Cormorant sub-colonies was noted. 3 Carter et al reported 218 Brandt's Cormorant nests on Alcatraz in 1995, based on aerial photographic surveys. 4 Visible only during boat surveys.

23 23 Table 2. Brandt's (BRCO), Pelagic Cormorant (PECO) and Western Gull (WEGU) reproductive phenology on Alcatraz Island, Only nests for which accurate data could be obtained were used in calculations. Actual ranges may be wider due to re-lays and hard-to-see nests. Data refers to first attempts only, and to first eggs and first chicks of a clutch/brood. For BRCO, fledging was assumed when chicks were large enough to wander from their nests, since after that time it was difficult to assign chicks to particular nest sites. For PECO and WEGU, fledging was assumed once chicks were fully feathered. EGG-LAYING DATE mean ± s.d. (n) (range) HATCHING DATE mean ± s.d. (n) (range) FLEDGING DATE mean ± s.d. (n) (range) Southern Sub-Colony 10 Apr ± 24 (19) 6 May ± 15 (15) 27 Jun ± 13 (26) (20 Mar - 26 Jun) (21 Apr - 19 Jun) (9 Jun - 24 Jul) Barker Beach Sub-Colony 15 Apr ± 15 (16) (17 Mar - 1 May) 20 May ± 15 (7) (17 Apr - 2 Jun) 3 Jul ± 13 (17) (9 Jun - 14 Jul) Brandt's Cormorant Northern Sub-Colony Laundry Building Sub-Colony 4 May ± 5 (19) (21 Apr - 13 May) 24 May ± 22 (6) (14 Apr - 16 Jun) 1 Jun ± 4 (8) (26 May - 5 Jun) 23 Jun ± 13 (6) (12 Jun - 17 Jul) 12 Jul ± 7 (30) (26 Jun - 28 Jul) 29 Jul ± 11 (15) (7 Jul - 21 Aug) Model Industries Sub-Colony 21 Apr ± 14 (26) (27 Mar - 19 May) 1 Jun ± 12 (19) (1 May - 16 Jun) 12 Jul ± 13 (40) (12 Jun - 11 Aug) ALL SUB-COLONIES 22 Apr ± 20 (86) (17 Mar - 26 Jun) 26 May ± 19 (55) (17 Apr - 17 Jul) 10 Jul ± 15 (128) (9 Jun - 21 Aug) Pelagic Cormorant 16 May ± 22 (7) (21 Apr - 16 Jun) 19 Jun ± 32 (3) (19 May - 21 Jul) 10 Aug ± 19 (3) (28 Jul - 1 Sep) Western Gull Cistern Parade Ground ALL SUB-COLONIES 7 May ± 6 (56) (28 Apr - 26 May) 7 May ± 5 (23) (28 Apr - 19 May) 7 May ± 5 (79) (28 Apr - 26 May) 3 Jun ± 6 (50) (19 May - 23 Jun) 4 Jun ± 6 (32) (26 May - 16 Jun) 3 Jun ± 6 (82) (19 May - 23 Jun) 30 Jul ± 6 (47) (14 Jul - 18 Aug) 1 Aug ± 7 (36) (14 Jul - 25 Aug) 31 Jul ± 6 (83) (14 Jul - 25 Aug)

24 Table 3. Brandt's Cormorant productivity by sub-colony on Alcatraz Island, PRODUCTIVITY YEAR Southern Sub-Colony Northern Sub-Colony Laundry Sub-Colony Model Industries Sub-Colony Barker Beach Sub-Colony TOTAL METHOD 1995 (chicks/site) 2.6 (262/100) 1.1 (89/81) (0) (0) (0) 1.9 (351/181) colony-wide, island-based and aerial photographic surveys 1996 (chicks/site) 1.7 (215/126) 2.2 (230/105) (0) (0) (0) 1.9 (445/231) colony-wide, island-based surveys 1997 mean ± s.d. (n) 2.4 ± 1.2 (76) 2.0 ± 0.8 (24) no data no data (0) 2.3 ± 1.1 (100) focal-site analysis 1998 mean ± s.d. (n) 1.6 ± 1.0 (83) (0) 1.7 ± 0.5 (6) 2.0 ± 0.0 (3) (0) 1.6 ± 1.0 (92) focal-site analysis 1999 mean ± s.d. (n) 2.6 ± 1.0 (93) 2.1 ± 1.0 (53) 2.1 ± 1.0 (17) 2.0 ± 0.8 (24) (0) 2.3 ± 1.0 (187) focal-site analysis 2000 mean ± s.d. (n) 2.5 ± 1.0 (81) 2.2 ± 1.4 (19) 2.5 ± 1.3 (17) 2.4 ± 0.9 (21) (0) 2.4 ± 1.1 (138) focal-site analysis 2001 mean ± s.d. (n) 2.2 ± 1.2 (102) 1.7 ± 1.3 (80) 2.5 ± 1.3 (13) 2.2 ± 1.3 (13) (0) 2.0 ± 1.3 (208) focal-site analysis 2002 mean ± s.d. (n) 2.7 ± 0.9 (43) no data 2.8 ± 0.8 (23) 2.0 ± 1.0 (23) no data 2.5 ± 1.0 (98) focal-site analysis 2003 mean ± s.d. (n) 1.9 ± 1.1 (54) 1.2 ± 1.2 (21) (0) 1.9 ± 1.2 (44) 2.3 ± 0.7 (20) 1.9 ± 1.1 (139) focal-site analysis 2004 mean ± s.d. (n) 2.5 ± 1.2 (37) 2.5 ± 1.1 (35) 1.5 ± 1.3 (22) 2.6 ± 1.1 (47) 2.4 ± 1.1 (20) 2.4 ± 1.2 (161) focal-site analysis

25 25 Table 4. Brandt's (BRCO), Pelagic Cormorant (PECO) and Western Gull (WEGU) reproductive performance on Alcatraz Island, Actual ranges may be wider due to re-lays and hard-tosee nests. Data refer to first attempts only. CLUTCH SIZE mean ± s.d. (n) BROOD SIZE mean ± s.d. (n) HATCHING SUCCESS mean ± s.d. (n) FLEDGING SUCCESS mean ± s.d. (n) Southern Sub-Colony 3.8 ± 0.6 (16) 2.5 ± 1.5 (17) 0.8 ± 0.3 (12) 0.7 ± 0.3 (16) Western Gull Brandt's Cormorant Barker Beach Sub-Colony 3.4 ± 0.7 (10) 2.7 ± 1.9 (7) 0.8 ± 0.4 (6) 0.7 ± 0.4 (10) Northern Sub-Colony 3.8 ± 0.4 (10) 3.8 ± 0.4 (5) 1.0 ± 0.0 (3) 0.7 ± 0.3 (8) Laundry Building Sub-Colony 3.7 ± 0.5 (9) 1.0 ± 1.6 (10) 1.0 ± 0.0 (2) 0.7 ± 0.3 (8) Model Industries Sub-Colony 3.5 ± 0.8 (29) 2.2 ± 1.6 (23) 0.8 ± 0.4 (19) 0.7 ± 0.3 (28) ALL SUB-COLONIES 3.6 ± 0.7 (74) 2.3 ± 1.7 (62) 0.8 ± 0.3 (42) 0.7 ± 0.3 (70) Pelagic Cormorant 4.0 ± 0.0 (3) ~ 2.1 (7) ~ 0.3 (3) ~ 0.6 (6) Cistern 2.7 ± 0.7 (67) 2.3 ± 1.0 (55) 0.9 ± 0.3 (54) 0.8 ± 0.3 (45) Parade Ground 2.9 ± 0.2 (39) 2.5 ± 0.8 (41) 0.9 ± 0.2 (38) 0.9 ± 0.3 (39) ALL SITES 2.8 ± 0.6 (106) 2.4 ± 0.9 (96) 0.9 ± 0.3 (92) 0.9 ± 0.3 (84)

26 26 Table 5. Pigeon Guillemot reproductive phenology on Alcatraz Island, Adults are censused from April through June, and activity on the water and at nesting areas on the south cliffs is noted until no remaining guillemots are sighted. Crevices cannot be regularly monitored, as many are in cormorant colonies or out of reach. Presence of chicks is confirmed by delivery of fish to the nest site by the parent or by incidental sightings of chicks. First adults seen rafting on water First adults seen on cliffs/ at nest sites First fish delivery seen Last fish delivery seen 17 Mar 17 Apr 11 Jun 9 Jul

27 27 Table 6. Summary table of types of disturbances to Brandt's Cormorants on Alcatraz Island, Number of disturbances observed Type of Disturbance (205.5 hours) (226.5 hours) (313.4 hours) (216.4 hours) (251.3 hours) (95.5 hours) (224 hours) (339.7 hours) Marine traffic (51%) (30%) (49%) (48%) (38%) (33%) (62%) (40%) External: Air traffic (27%) (19%) (29%) (30%) (49%) (34%) (32%) (33%) Other (3%) (4%) (5%) (12%) (4%) (0%) (1%) (9%) Human interference (7%) (5%) (1%) (0%) (0%) (19%) (2%) (12%) Island-Based: Interspecies event (4%) (24%) (13%) (6%) (5%) (10%) (3%) (2%) Rockslide or natural event (0%) (1%) (0%) (0%) (0%) (0%) (0%) (0%) Unknown Cause: (8%) (16%) (2%) (4%) (2%) (3%) (1%) (5%) Total: Frequency of disturbances The total hours observed was reduced in 2002 due to observer inconsistency. 2 In 2003 and 2004, the total includes extra disturbance monitoring on the North End of the island. 3 This frequency represents the minimum number of disturbances per hour. Observers could not see the whole island at once, therefore the actual disturbance rate to cormorants is likely to be higher.

28 28 APPENDIX A: ALCATRAZ WESTERN GULL CENSUS AREA DESCRIPTION Monitoring Western Gulls on Alcatraz presents problems of disturbance to gulls, Black-crowned Night Herons, and nesting seabirds, particularly cormorants. It is safest to census early in the morning or in late afternoon to avoid heat-stressing of gull eggs and chicks. Be cognizant of heat stress symptoms (continuous panting, drooping appearance, and torpor) and vacate areas accordingly. Young chicks less than two weeks old area especially at risk. Limit your time in each area. Be careful not to step on eggs or young chicks. Chicks hide in well-protected areas, and are also well camouflaged and thus hard to see on the ground even when in the open. When working, move slowly and deliberately through each area. This helps maximize awareness and minimize risk of panic runs by older chicks. If a larger chick bolts beyond your immediate area, crouch quietly until it settles. Two counts should be obtained one week apart, during the ten-day period following peak egg-laying, when chicks begin to hatch. Peak egg-laying occurs approximately two and one-half weeks after egglaying begins. Counts should be done at the same time of day, either early morning or early evening when most gulls are in the colony. Census Area Descriptions AT: Agave Trail The agave area along the southern end of Alcatraz as well as the S, SW, and W shoreline extending from the large warning sign at eucalyptus trees south past the south Fog Horn, up to the Tunnel Bush and ending at the Gap between BRCO nesting areas 1,2 and 3,4 (where the trail is falling away and eroding). BB: Barker Beach A large area extending from the Cell House west to the water line. Includes walkways, some gardens, agave, and steep cliffs. Boundaries include the southeast rec yard wall and fence extending towards beach, and on other side, and imaginary line drawn from the corner of the cell house down to the walkway next to the blind. Behind Rubble The areas on the backside of the rubble piles NOT VISIBLE FROM LIGHTHOUSE PLAZA. Includes nests on the Parade Ground cement not visible from lighthouse plaza. BL: Below Laundry The thin flat cement walk and grass between the Laundry Building and the cliffs leading down to the Seawall. A chain-link fence used to be in place along the western perimeter of this area. C: Cistern Encompasses the flat table-like area supporting the cistern and water tower just north of the Recreation Yard. Extends from the rec yard wall to the steep cliffs as its western and northern boundary. Includes the grassy slope and cement walkway along its NE boundary. This area consists mostly of grass, rubble, some shrubs and the cement cistern. CH: Cell House Roof only. D: Dock The loading dock to the tourist entryway in front of building 64. E: Eucalyptus The eucalyptus grove and trail at the Southeast end of Alcatraz. FH: Foghorn The small rocky slope area and shoreline between Model Industries and the water as the northern point of Alcatraz. Includes the foghorn. BOAT SURVEY if BCNH disturbance is too great. G: Gardens The historic gardens west of the rec yard. Mostly steep terrain with mixed shrub, agave, and cement walks. Upper boundary is walkway between metal detector and BB. Lower boundary is old walkway at the base of slope. I: Incinerator The flat grassy area with some ruins around the incinerator above the west shore. BOAT SURVEY