Delaware Bay shorebird-horseshoe crab assessment report and peer review

              Delaware Bay Shorebird-Horseshoe Crab
Assessment Report and Peer Review
Prepared for the
Atlantic States Marine Fisheries Commission
by the
U.S. Fish and Wildlife Service
Shorebird Technical Committee
Peer Review Panel
June 2003
Information in the report was compiled by Brad A. Andres and available from U. S. Fish and
Wildlife Service, Division of Migratory Bird Management, 4401 N. Fairfax Dr., MBSP 4107,
Arlington, VA, 22203, USA or at http://migratorybirds.fws.gov/reports/reports.html
Report authors are listed in the Literature Cited. Some sections were drafted by Nellie Tsipoura
(Rutgers University), Joanna Burger (Rutgers University), Gregory Breese (U. S. Fish and
Wildlife Service), and Kimberly Cole (Delaware Coastal Management Programs). Shorebird
Technical Committee members provided review.
Suggested citation: U.S. Fish and Wildlife Service. 2003. Delaware Bay Shorebird-Horseshoe
Crab Assessment Report and Peer Review. U.S. Fish and Wildlife Service Migratory Bird
Publication R9-03/02. Arlington, VA. 99 p.
TABLE OF CONTENTS
A. Conclusions, Recommendations, and Peer Review..............................................1
1.0. PURPOSE AND APPROACH....................................................................................1
2.0. LONG-DISTANCE MIGRATION IN SHOREBIRDS ..............................................1
3.0. CONCLUSIONS..........................................................................................................2
3.1. Shorebird Use of Delaware Bay ......................................................................2
3.2. Shorebird Population Trends ...........................................................................2
3.3. Shorebird Population Threats ..........................................................................3
3.4. Shorebird Use of Horseshoe Crab Eggs...........................................................4
3.5. Availability of Horseshoe Crab Eggs ..............................................................4
3.6. Shorebird Weight Gain in Delaware Bay ........................................................5
3.7. Shorebird Survival ...........................................................................................6
4.0. RECOMMENDATIONS.............................................................................................7
4.1. Direct Management..........................................................................................8
4.1.1. Horseshoe crab egg abundance.........................................................8
4.1.2. Seasonal beach closures....................................................................8
4.1.3. Habitat protection and enhancement.................................................8
4.3. Needed Analyses............................................................................................10
4.2.1. Horseshoe crab egg abundance.........................................................9
4.2.2. Shorebird breeding-ground conditions .............................................9
4.2.3. Shorebird diet and energetics............................................................9
4.3. Improved Monitoring and Research ..............................................................10
4.3.1. Bay-wide horseshoe crab egg abundance .......................................10
4.3.2. Shorebird population surveys .........................................................10
4.3.2. Individually-marked shorebirds ......................................................10
4.3.4. Measurements of weight gain .........................................................10
4.3.5. Southern stop over quality ..............................................................10
5.0. SHOREBIRD TECHNICAL COMMITTEE MEMBERSHIP .................................11
6.0. PEER REVIEW PANEL PARTICIPANTS ..............................................................11
B. Shorebird-Horseshoe Crab Assessment .............................................................12
1.0. INTRODUCTION .................................................................................................................12
1.1. Shorebird Technical Committee ....................................................................12
1.2. Horseshoe Crabs, Shorebirds, and Delaware Bay .........................................12
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 ii
1.3. Economic Value of Crabs and Shorebirds .....................................................13
1.4. Evaluation Approach .....................................................................................13
2.0. DELAWARE BAY SHOREBIRDS..........................................................................14
2.1. Species Considered ........................................................................................14
2.2. Conservation Status and Protection ...............................................................14
2.3. Vulnerability of Long-distance Migrant Shorebirds and the Red Knot Focus15
2.4. Red Knot Distribution....................................................................................15
2.5. Red Knot Annual Cycle.................................................................................15
2.6. Distribution and Migration Routes of Other Species.....................................16
2.6.1. Ruddy turnstone ..............................................................................16
2.6.2. Sanderling .......................................................................................17
2.6.3. Semipalmated sandpiper .................................................................17
2.6.4. Dunlin .............................................................................................17
2.6.5. Short-billed dowitcher ....................................................................17
2.6.6. Long-billed dowitcher.....................................................................18
2.6.7. Least sandpiper ...............................................................................18
3.0. ABUNDANCE AND DISTRIBUTION OF HORSESHOE CRABS.......................18
4.0. POTENTIAL THREATS TO SHOREBIRDS ..........................................................20
4.1. Heavy Metal Concentrations in Shorebirds and Horseshoe Crabs ................20
4.2. Organic Compound Concentrations in Shorebirds and Horseshoe Crabs .....21
4.3. Disease in Shorebirds.....................................................................................22
4.4. Shoreline Changes in Delaware Bay..............................................................23
4.5. Sea Level Rise from Global Climate Change ................................................23
4.6. Arctic Breeding Ground Conditions ..............................................................23
4.7. South American Wintering Ground Conditions ............................................25
4.8. Human Disturbance to Shorebirds .................................................................25
4.9. Effect of Disturbance on Survival of Semipalmated Sandpipers...................26
4.10. Horseshoe Crab Bait Landings ....................................................................27
4.11. Changes in Horseshoe Crab Populations .....................................................27
5.0. ESTIMATES OF SHOREBIRD POPULATION SIZES AND TRENDS................29
5.1. Shorebird Population Sizes ............................................................................29
5.1.1. Coarse continental estimates...........................................................29
5.1.2. Re-sighting banded red knots in the 1980s .....................................29
5.1.3. Red knot band re-sighting in South America..................................30
5.2. Shorebird Population Trends .........................................................................30
5.2.1. Aerial surveys of red knots in South America ................................30
5.2.2. Spring aerial surveys in Delaware Bay...........................................31
5.2.3. International and Maritime Shorebird Surveys...............................32
5.2.4. Quebec migration checklists ...........................................................32
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 iii
6.0. IMPORTANCE OF DELAWARE BAY TO SHOREBIRD POPULATIONS ........33
7.0. HABITAT USE BY SHOREBIRDS AND HORSESHOE CRABS.........................34
7.1. Shorebird Use of Marine and Non-marine Habitats ......................................34
7.2. Red Knot Habitat Use and Movements in Delaware Bay..............................34
7.3. Shorebird Habitat Use on Cape May Peninsula, New Jersey ........................34
7.4. Shorebird Beach Use in Delaware .................................................................35
7.5. Influence of Beach Characteristics on Horseshoe Crab Reproductive Activity35
7.6. Beach Nourishment and Habitat Restoration for Crabs and Shorebirds .......36
7.7. Shorebird Habitat Use in Relation to Beach Characteristics and Abundance
of Horseshoe Crabs and Their Eggs...........................................................37
8.0. ABUNDANCE AND TRENDS OF HORSESHOE CRAB EGGS ..........................38
8.1. Bay-wide Egg Density in 1999......................................................................38
8.2. Egg Density on Delaware Beaches ................................................................38
8.3. Changes in Egg Density on New Jersey Beaches..........................................38
8.4. Egg Abundance Sampling Design Considerations ........................................39
9.0. SHOREBIRD DIET AND USE OF HORSESHOE CRAB EGGS...........................39
9.1. Shorebird Diet in Delaware Bay ....................................................................39
9.2. Stable Isotope Analysis Confirms Shorebird Dependance on Horseshoe Crab
Eggs in Delaware Bay................................................................................40
9.3. Functional Responses of Shorebirds Feeding on Horseshoe Crab Eggs .......40
9.4. Competition Between Shorebirds and Gulls for Horseshoe Crab Eggs ........41
9.5. Red Knots Use of Food Other than Horseshoe Crab Eggs ............................42
10.0 ENERGETIC REQUIREMENTS OF MIGRANT SHOREBIRDS.........................43
10.1. An Energetics Framework for Migrant Shorebirds .....................................43
10.2. Energetics of Sanderlings Migrating to Four Latitudes...............................44
10.3. Predicting Flight Ranges..............................................................................44
10.4. Fat-loading in islandica Red Knots .............................................................45
10.5. Effects of Weight on Metabolic Power Needed for Flight .........................45
10.6. Flight Energy Needs of rufa Red Knots Staging in Delaware Bay .............46
10.7. Assimilation Efficiency of Sanderlings Consuming Horseshoe Crab Eggs 46
10.8. Energy budget of Delaware Bay Shorebirds................................................47
10.9. Horseshoe Crab Egg Requirement of Delaware Bay Shorebirds ................47
11.0. SHOREBIRD WEIGHTS AND WEIGHT GAIN ..................................................48
11.1. General Capture Methods ............................................................................48
11.2. Organ Atrophy and Weight Change during Migration................................49
11.3. Red Knot Weights through the Annual Cycle .............................................50
11.4. Red Knot Weight Gains in Delaware Bay ...................................................51
11.4.1. Analytical approaches...................................................................51
11.4.2. Red knot arrival weights and weight gains ...................................51
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 iv
11.4.3. Red knot departure weights in Delaware Bay...............................53
11.5. Weights and Weight Gain in Ruddy Turnstones and Sanderlings...............53
11.6. Weights and Weight Gain in Semipalmated and Least Sandpipers.............54
12.0. RED KNOT SURVIVAL AND PRODUCTIVITY................................................55
12.1. Re-sighting Rates of Knots Banded in Florida and Argentina ....................55
12.2. Survival Rate................................................................................................55
12.3. Population Projections .................................................................................56
12.4. Juvenile Age Ratios ....................................................................................56
13.0. LITERATURE CITED ...........................................................................................57
14.0. TABLES ..................................................................................................................72
15.0. FIGURES.................................................................................................................91
C. Shorebird Technical Committee Terms of Reference – 2002 ..........................94
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 1
A. Conclusions, Recommendations, and Peer Review
1.0. PURPOSE AND APPROACH
The Atlantic States Marine Fisheries Commission asked the U. S. Fish and Wildlife Service to
form a Shorebird Technical Committee that would provide technical guidance, regarding effects
that horseshoe crab management actions could have on shorebird populations, to the Horseshoe
Crab Management Board. One of the immediate tasks of the Shorebird Technical Committee
was to produce a peer-reviewed report that synthesized unpublished and published information
on shorebird population trends, threats to shorebird populations, shorebird habitat use, shorebird
energetic requirements, and horseshoe crab egg abundance. Although several shorebird species
were considered in the report, attention primarily focused on the red knot (Calidris canutus rufa).
Available information was greatest for the red knot and was less extensive for the ruddy
turnstone (Arenaria interpres morinella), sanderling (Calidris alba), semipalmated sandpiper
(Calidris pusilla), and least sandpiper (Calidris minutilla). Relatively little information existed
on the dunlin (Calidris alpina hudsonia) and short-billed dowitcher (Limnodromus griseus
griseus). Aside from the least sandpiper, which was chosen because of its contrasting use of
marsh habitats, all other species were selected because of their reliance on beach habitats and
their frequency of occurrence on Delaware Bay aerial surveys (1986–2002). After reviewing the
report, the Committee has generated this set of conclusions, management recommendations, and
information needs. The Committee used a concordance, or preponderance, of evidence approach
to evaluate the report’s contents. The report, conclusions, and recommendations were evaluated
by an independent Peer Review Panel, and their comments are included here as bolded text.
2.0. LONG-DISTANCE MIGRATION IN SHOREBIRDS
Many populations of shorebirds undertake a series of long-distance, non-stop flights to travel
between their wintering and breeding grounds. Because a shorebird often crosses vast stretches
of open water during migration, physiological and environmental conditions on departure can
directly, and immediately, affect its survival. The red knot is an extreme example of the long-hop
migration system and has one of the longest migrations of any bird. Besides adding 50% of
their body weight in fat reserves, red knots at Delaware Bay, and elsewhere, exhibit major
internal organ changes in response to the need to first accumulate fat and later to reduce flight
mass. The long-hop migration system of red knots, and other shorebird species, is highly
dependent on food availability at a limited number of stopover sites. Failure to gain sufficient
body mass at stopover sites, often during a short time span, can impair the health, productivity,
and survival of migrant shorebirds. Because arctic breeding grounds are generally food limited
in early summer when shorebirds first arrive, food-rich stopovers in the north-temperate region
are particularly important. At these sites, shorebirds are often under relatively strict time
constraints to add needed fat reserves.
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 2
3.0. CONCLUSIONS
3.1. Shorebird Use of Delaware Bay
Delaware Bay has been recognized by many scientists and organizations as one of the most
important and critical shorebird stopovers in the Western Hemisphere and, indeed, in the world.
Depending on the species, between 12 and 80% of the Atlantic flyway population of the six
beach-inhabiting shorebirds mentioned above (excluding least sandpiper) can be observed on
Delaware Bay’s beaches during northward migration. Far fewer numbers of shorebirds pass
through Delaware Bay during southward migration. For a given species, the proportion of the
population that uses Delaware Bay each spring may vary substantially among years. Compared
to 1986–1996, average shorebird use of Delaware Bay beaches, as measured by seasonal maxima
of aerial survey counts, has increased or remained stable during 1997–2002 for all six beach-inhabiting
species. During their northward migration in the Delaware Bay region, most
shorebird species use marine-influenced habitats — either salt marshes, tidal flats, or sand
beaches.
The Peer Review Panel generally agrees with these conclusions, except that a more
sophisticated analysis of the Delaware Bay shorebird use time-series data could have been
conducted. Data on shorebird-use days could be useful in constructing a total energy
budget for all northward-migrating shorebirds. The importance of accessible roosting sites
to migrant shorebirds is not mentioned.
3.2. Shorebird Population Trends
Based on a variety of sources, all available data indicate that the rufa red knot population has
decreased since the 1980s, but the magnitude of the decline is not precisely known. Besides the
red knot, the semipalmated sandpiper is the only other Delaware Bay shorebird species that has
relatively consistent patterns of population decreases among trend datasets. Because of unknown
turnover and detection rates, aerial survey data from Delaware Bay are not useful for estimating
population sizes of shorebirds in Delaware Bay.
The Peer Review Panel agrees that, although imperfect, patterns in the trend analyses
reasonably indicate a decrease, of some magnitude, in populations of rufa red knots and
semipalmated sandpipers. Most surveys of wintering and migrating red knots do not cover
the needed range of the population and complicate interpretation of changes in populations
at specific sites. Analytical methods used to summarize ISS data also lack rigor and may
only reveal general patterns of population change. Current and future surveys of
shorebird populations should undergo rigorous statistical review.
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 3
3.3. Shorebird Population Threats
The Shorebird Technical Committee evaluated information on the potential threats to shorebird
populations across their annual cycle. Testing for contaminants in shorebirds and crabs indicates
that metals and pesticides are not likely causing population reductions in shorebirds. Little
information exists on disease and parasite occurrence in red knots, particularly in Delaware Bay,
but there is no current evidence to suggest that these are major, potential problems. Although
environmental conditions vary considerably from year to year, arctic breeding habitats do not
appear to have changed in ways that would likely contribute to the observed reductions in red
knot survival and productivity. More information is needed to assess the effects that weather and
predation in the arctic have on rufa red knot population dynamics. Arctic environmental
conditions should also be evaluated for semipalmated sandpipers. Habitat conditions in
wintering areas have numerous potential threats, but these are not believed to have currently
affected key wintering sites. Food availability, however, has only been measured at a few South
American wintering or stopover sites. Beach nourishment is not having a negative effect on
shorebird use on Delaware beaches and is likely improving habitat quality; beach nourishment is
not widely practiced in New Jersey. Although no Bay-specific studies have been conducted,
repeated human disturbance likely reduces shorebird feeding efficiency in Delaware Bay.
Elsewhere, migrant shorebirds have been disturbed by dogs, self-propelled human recreation,
and vehicles. Human disturbance to semipalmated sandpipers feeding along the coast of
Massachusetts as they prepared for a long over-water flight, reduced their subsequent survival.
Gulls can potentially reduce food availability to shorebirds through direct and indirect
competition for crab eggs. Shorebirds, however, most often forage with other shorebirds, and
preliminary data and field observations suggest that the number of gulls using Delaware Bay
beaches has not substantially increased in recent years. Lastly, reduced numbers of horseshoe
crab eggs available for shorebird consumption, relative to the early 1990s, could reduce survival
and reproductive success in the six shorebird species that use Delaware Bay as the last stopover
prior to departing for their breeding grounds (see following sections).
The Peer Review Panel agrees that contaminants and parasites do not currently appear to
provide a major threat to shorebirds stopping at Delaware Bay. Further information is
needed to thoroughly evaluate whether or not changes in habitat quality on the breeding
and wintering grounds are contributing to declines in shorebird populations. However,
changes in breeding or wintering area conditions do not minimize the importance of
maintaining high quality north-temperate stopovers. Information presented in the report
is insufficient to determine if beach nourishment generally improves habitat quality for
spawning horseshoe crabs and foraging shorebirds. Although numerous studies have
demonstrated the immediate, disruptive effects of human disturbance to migrant
shorebirds, ultimate effects of disturbance on survival of shorebirds are not well-documented
and are usually inferred (including the Massachusetts semipalmated
sandpiper study referenced above). Increases in gull numbers do not superficially appear
to have direct or indirect influences on shorebird population changes, but more
quantitative information on effects of interference and exploitative competition between
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 4
gulls and shorebirds is needed. The life history of long-distance, long-hop shorebird
migrants indicates that the availability of abundant food resources at north-temperate
stopovers is critical for completing their annual cycle.
3.4. Shorebird Use of Horseshoe Crab Eggs
The importance of Delaware Bay as a spring shorebird stopover is likely due to the unique and
important food resource — horseshoe crab eggs. A variety of methods (stomach analyses,
captive feeding studies, and field observations) indicate that horseshoe crab eggs are a variable
component in the diet of numerous invertebrates and vertebrates (shorebirds, other birds, fish,
and turtles). Birds, and particularly shorebirds, are important predators of crab eggs. Stable
isotope analysis indicates that red knots are highly dependent on horseshoe crab eggs. Isotope
analysis of other shorebird species is currently underway. Red knots feed by pecking at surface
eggs and making shallow probes into beach sediments. Captive knots fed exclusively eggs
gained weight at rates that were similar to those observed in wild birds. Egg consumption was
estimated at 18,000 eggs per day and rates of knot weight gain ranged from 2.6 to 8.0 grams per
day while they were in Delaware Bay. Daily weight gains of rufa red knots in Delaware Bay are
the highest reported for any stopover site or knot population. At other stopovers throughout the
world, knots generally feed on molluscs or bivalves. Although Bay beaches were reported to
have low invertebrate prey densities, detailed evidence does not exist to thoroughly evaluate
whether or not alternative shorebird foods exist in high enough abundances to meet the energetic
needs of red knots and other migrant shorebirds while in the Delaware Bay region.
The Peer Review Panel believes that the importance of Delaware Bay to shorebirds is due
to a number of factors such as an abundant primary food resource (crab eggs), the
availability of secondary food resources, and availability of safe roost sites. Stable isotope
analysis indicates that red knots feed almost exclusively on horseshoe crabs while at
Delaware Bay. Although this result does not necessarily indicate a “dependency” on this
food, crabs should be assumed to be critically important unless a viable alternative prey
base is shown to exist. A comprehensive review of migrant shorebird foraging behavior
and diet is needed to thoroughly evaluate the importance of Delaware Bay, and its food
resources, to shorebirds; caloric value of alternative foods should be determined. No
information was presented on the specific egg or larval life stage was being consumed by
shorebirds. Foraging behavior of knots, in particular, at sites other than Delaware Bay
could provide insights into the importance of the Bay’s horseshoe crabs to shorebirds. The
habitat section of the report should have included more information, if available, on the
correlation between beach use by shorebirds and the distribution of horseshoe crab
spawning females and eggs.
3.5. Availability of Horseshoe Crab Eggs
Although a sampling plan has been devised, no Bay-wide, systematic survey of egg availability
has yet been conducted. Geographically limited surveys conducted in May, variably over the last
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 5
four years, do not provide conclusive evidence of a trend in the abundance of surface eggs
available to shorebirds. Likewise, there are not ample data to assess whether or not surface
horseshoe crab eggs occur in abundances that will support Delaware Bay populations of migrant
shorebirds. Although counts of spawning crabs have not changed between 1999 and 2002, trawl
survey indices of all age-classes of crabs are now lower than they were in the early 1990s.
Further analysis of egg data collected on New Jersey beaches and additional information on the
temporal and spatial distribution of surface and sub-surface eggs is needed to thoroughly
evaluate if there has been a significant trend in horseshoe crab egg abundance. Further
refinement of the total shorebird energy budget is needed to determine how many eggs are
required across the entire spring season.
The Peer Review Panel believes that knowledge about the spatial and temporal patterns of
horseshoe crab egg densities is critical to understanding how crab management affects
migrant shorebird populations. Specifically, a clearer understanding of how eggs become
available to shorebirds is needed. Energetic considerations indicate that horseshoe crab
eggs are only profitable to shorebirds if they occur in high surface densities. The
excavation and transport of eggs to the beach surface might only occur when spawning
females occur in very high densities, and there may be a threshold female crab density at
which sufficient numbers of eggs become available on the surface. Little appears to be
known about the depletion of surface eggs attributable to shorebird, and other bird,
predation. Depletion of surface eggs would be consistent with the hypothesis that crab eggs
are a limiting resource for shorebirds. The Panel agrees that information from trawl
surveys, given gear limitations for adequately sampling large numbers of crabs, indicates
that horseshoe crabs in Delaware Bay are currently at lower levels than they were in the
early 1990s. Uncertainty in recent estimates of sizes of horseshoe crab age classes
precludes reasonable comparison of recruitment rates and harvest levels. The report
would have benefitted from thorough analyses of datasets already collected on changes in
egg densities on New Jersey beaches. An unified bioenergetics model for Delaware Bay
shorebirds will be needed to integrate the information about available food with the
requirements of shorebirds.
3.6. Shorebird Weight Gain in Delaware Bay
There is agreement that a smaller percentage of rufa red knots are making threshold departure
weights by the end of May in recent years. These results are not dependent on inclusion of 1997,
a year when shorebird-banding did not begin until 22 May. The different analytical approaches
used to determine weight gains of Delaware Bay red knots (average weights of time-dependent
catches, cohort analysis, and individual recaptures) have generated two hypotheses regarding
decreases in rates of weight gain between 1997 and 2002 — either a greater proportion of red
knots are arriving later in Delaware Bay in recent years, or red knots are increasingly unable to
find sufficient food. In the first analytical approach, rates of weight gain in knots decreased
through time, but in the latter two approaches they did not. Evidence suggests that rates of
weight gain by semipalmated sandpipers have decreased in recent years, while rates of weight
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 6
gain in least sandpipers, a more marsh-foraging species, remained stable. Patterns of decreasing
(average) rates of weight gain were less consistent for ruddy turnstones and were not apparent in
sanderlings. Ruddy turnstones can excavate eggs to feed on, and sanderlings are thought to
commute regularly between Atlantic Ocean and Delaware Bay feeding sites. No hypotheses, as
an alternative to decreased horseshoe crab egg availability, have been formulated to explain
changes found in weight gains of semipalmated sandpipers. Semipalmated sandpipers do not
winter in the same regions of South America as red knots. More information on the condition of
South American stopovers and observations of individually marked birds are needed to fully
discriminate between these two alternatives. Late arrival of knots could be caused by changes in
spring weather patterns or by their inability to build fat stores at South American stopovers. Red
knots can physiologically compensate for late arrival by increasing their rates of fat deposition
while in Delaware Bay.
The Peer Review Panel believes that the two hypotheses forwarded to explain changes in
weight gain in Delaware Bay red knots are not mutually exclusive, but instead represent
two factors which operate in tandem to affect departure weights from Delaware Bay. Both
factors operate within the same year, although their relative importance may vary among
years. The existing data, however, are not adequate to evaluate their relative importance
for any year of record. But in any case, Delaware Bay must provide the food resources
shorebirds need to adequately gain fat mass to make the flight to the arctic. That a lesser
proportion of red knots are making minimal departure weights suggests that food
resources in Delaware Bay may not be adequate. Similar feeding rates observed among
species of different size supports the finding that the larger red knots should be most
sensitive to decreases in food availability. The shorebird banding program in Delaware
Bay would greatly benefit by a more cooperative approach to design and analysis.
Procedures used in both analyses of weight gain were not documented adequately enough
in supporting reports to allow independent evaluation. Patterns of weight gain were more
clearly presented for semipalmated and least sandpipers. Unfortunately, attempts to
estimate growth rate based on independent samples of body mass are inherently flawed, as
assumptions must be made to accommodate the uncertainty in arrival times of birds.
These assumptions lead to the possibility of conflicting results and additional controversy.
Adjusting field methods to emphasize the collection of multiple measurements on
individual birds would increase the sample of individually-marked birds and would
ultimately strengthen conclusions about annual changes in rates of weight gain.
3.7. Shorebird Survival
Shorebird return rates (on southward migration) relative to stopover departure weights indicate
that the inability to gain sufficient weight at stopover sites can reduce survivorship for red knots
(Calidris canutus) and semipalmated sandpipers (Calidris pusilla), which supports the link
between stopover conditions and population trends. Recent estimates of adult survival and
productivity of rufa red knots are substantially lower than estimates for knot populations
wintering in Europe and Australia; these knot populations also breed in arctic regions and
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 7
undertake long-distance, long-hop migrations. Sustained low levels of vital rates could cause a
drastic decline in the knot population. Evidence generated through population modeling,
however, was insufficient to evaluate the probabilities of extinction under the current range of
demographic values.
The Peer Review Panel supports the conclusion that low-weight red knots had a lower
return rate, but found the estimates of adult survival to be highly variable among periods.
Further details of the analytical procedures used for estimating survival rates are needed to
thoroughly evaluate these results for application to management decisions. To fully
evaluate the biological significance of survival rates and juvenile ratios, better information
on non-breeding distribution and movements of juveniles is needed. Because estimates
among years were from different sites, the variability of these estimates among sites should
be evaluated. Overall, the Panel believes that design and analysis of future mark-resight/
recapture studies could be improved to remove ambiguities in interpretation of results and
to take better
advantage of the large number of banded birds. Use of field-readable, individually-numbered
color flags should be thoroughly evaluated.
4.0. RECOMMENDATIONS
Horseshoe crab management actions already taken (for example, bait bags, harvest reductions,
alternative bait development, designation of the Carl N. Shuster, Jr. Horseshoe Crab Reserve)
have likely improved conservation of crabs and shorebirds. Despite these actions, and the
stability of spawning horseshoe crab numbers over the last four years, the population of red
knots, and perhaps other species, has declined. As a general management action, the U. S.
Shorebird Conservation Plan suggests that any declining shorebird population should be
stabilized and then restored to population levels of the late 1970s and early 1980s. Accordingly,
shorebirds in Delaware Bay should be managed to maintain current population sizes, and
decreasing populations should be stabilized and then increased.
Based on the shorebird and crab information currently available, the Shorebird Technical
Committee therefore recommends that the Horseshoe Crab Management Board pursue a
management strategy that is more risk-averse to shorebirds. Using an adaptive approach,
continued or improved monitoring programs for shorebirds, horseshoe crabs, and horseshoe crab
eggs are needed to evaluate results of management actions and to provide guidance for future
selection of management alternatives. The Shorebird Technical Committee supports the
cooperative effort of the Horseshoe Crab Technical Committee and the Horseshoe Crab Stock
Assessment Committee to develop and implement various crab surveys. Specific
recommendations of the Shorebird Technical Committee follow, which were generally supported
by all Committee members. Peer Review Panel comments are also included, as bolded text,
below.
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 8
4.1. Direct Management
4.1.1. Horseshoe crab egg abundance
Until further information is available on whether or not current egg abundances are sufficient for
shorebirds to reach threshold departure weights, the Committee recommends further reductions
in bait landings for New Jersey, Delaware, and Maryland. Although the Committee realizes
there currently are no biological reference points on which to base reduction amounts, total
reductions in the range of 50 to 75% below the Reference Period Landings might be considered.
Committee members could not reach consensus on the amount of reduction, if any, that would be
considered risk-averse. Because crabs caught in Federal waters from New York and to Virginia
ca be landed in any of the mid-Atlantic states, in New York and Virginia might also be
considered. Mandatory use of bait bags and development of alternative baits could contribute to
reduced bait use of horseshoe crabs.
The Peer Review Panel supports a reduction in harvest but suggests that this action be
viewed as an interim solution until integrated and comprehensive models are constructed
to set reasonable biological objectives for shorebirds. Although the Panel is unsure about
the amount of the reduction that is immediately needed, the numerous indications of
shorebird population declines suggests that harvest rates should be at or below the current
levels. Based on very limited data, a 75% reduction would ensure recruitment of female
crabs into the breeding population at the low bound of the population estimate of
primiparus female crab; a 66% reduction would allow no population growth at this level.
Development of conservation methods to use bait crabs most efficiently is worthwhile.
Landings in states other than New Jersey, Delaware, and Maryland should be carefully
tracked.
4.1.2. Seasonal beach closures
To increase abundance and availability of horseshoe crab eggs for feeding shorebirds, restrict
hand harvest of horseshoe crabs, vehicles, humans, and dogs on State- and Federally-owned
beaches important to shorebirds from 1 May to 7 June, the period of highest shorebird use, along
the Delaware Bay shoreline of Delaware and New Jersey. Evaluate the effectiveness of
restrictions.
The Peer Review Panel believes that this is a reasonable short-term action to increase the
number of horseshoe crab eggs available to migrant shorebirds. Evaluation of these
restrictive measures should be undertaken.
4.1.3. Habitat protection and enhancement
Encourage Delaware and New Jersey to continue environmentally responsible beach
nourishment and other enhancement projects that increase high quality habitat for spawning
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 9
crabs and feeding shorebirds. Consider long-term protection measures, including easements and
acquisition, for beaches that are important for crab spawning and shorebird foraging. Evaluate
the effectiveness of beach enhancement activities.
The Peer Review Panel believes further evaluation of the effects of beach nourishment on
horseshoe crab spawning and invertebrate infauna are warranted before broad-scale
activities are undertaken. If results of these evaluations, preferably using a before-and-after
experimental design, are favorable, specific prescriptions of “environmentally
responsible” practices should be developed. Evaluations and prescriptions should be
sensitive to the geographic scale of application. Long-term protection of beaches would
likely be a beneficial conservation measure.
4.2. Needed Analyses
4.2.1. Horseshoe crab egg abundance
Complete analyses of horseshoe crab egg abundance data that have already been collected on
New Jersey beaches to further evaluate evidence of a change in egg abundance.
4.2.2. Shorebird breeding-ground conditions
Compile information on annual weather conditions and predation pressure on breeding grounds
to assess short- and long-term effects on red knot survival and reproduction and on semipalmated
sandpiper population change. Report information on density, hatching success, and habitat use
on breeding grounds.
4.2.3. Shorebird diet and energetics
Complete stable isotope analysis for remaining Delaware Bay shorebird species to quantify their
dependence on horseshoe crab eggs. Develop the best possible estimate of the total energy
needed and horseshoe crab eggs required by all migrant Delaware Bay shorebirds. Complete
analysis of information on alternative foods available to Delaware Bay shorebirds to determine if
other energy sources exist that could supplement horseshoe crab eggs. Report on role of
nocturnal foraging.
The Peer Review Panel encourages efforts to expedite the reporting and analysis of all
previously-collected data pertinent to topics addressed in the report. The Panel also
encourages the involvement of biometricians in these analyses.
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 10
4.3. Improved Monitoring and Research
4.3.1. Bay-wide horseshoe crab egg abundance
Support implementation of the Bay-wide egg survey to determine abundance of, and ultimately
trend in, horseshoe crab eggs on Delaware Bay beaches. Information is needed on egg
deposition and movements to understand what makes eggs available to shorebirds on Delaware
Bay beaches.
4.3.2. Shorebird population surveys
Continue, and expand, the aerial survey of South American wintering grounds of red knots to
identify additional concentration areas and track population changes. Include areas with winter
aggregations of semipalmated sandpipers. Develop and evaluate other counting and
demographic methods to track populations of shorebirds.
4.3.3. Individually-marked shorebirds
Increase marking and scan-sampling of red knots on wintering grounds and in Delaware Bay to
track changes in population size, annual survival, and reproductive success. Expand efforts to
include semipalmated sandpipers. Use individually color-flagged and radio-tagged shorebirds to
determine movements into and within Delaware Bay to evaluate the late-arrival hypothesis.
4.3.4. Measurements of weight gain
Continue to monitor shorebird weights in Delaware Bay, while minimizing disturbance to
foraging shorebirds. Agree on standard data collection techniques, for both sides of Delaware
Bay, and record wing length and time after capture that weighing takes place. Develop a
common, Bay-wide database and agree on analytical approaches.
4.3.5. Southern stop over quality
Assess habitat quality of stopovers south of Delaware Bay to determine if South American sites
are providing enough food resources for migrant red knots and other shorebird species to gain
the weight needed to undertake trans-ocean flights.
The Peer Review Panel believes that virtually all management, research, and monitoring
programs would benefit from being placed within a more holistic and comprehensive
framework in which models are used to provide coherent structure for both combining
existing information and predicting consequences of management activities. Currently,
many of the research and monitoring efforts are fragmented and isolated, and it is unclear
whether appropriate information is presently collected to best aid management decisions.
The Panel encourages the Shorebird Technical Committee to work with all partners and
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 11
stakeholders to develop a more comprehensive and integrated research and monitoring
program. Theoretical models should be developed for core components of this program
that would include: 1) integrated shorebird energetics and horseshoe crab egg availability,
2) shorebird demographics, and 3) monitoring design and analysis. Even in the absence of
detailed quantitative information, explicit, well-developed models can illustrate the most
likely explanatory hypotheses, identify speculative and real data linkages, highlight key
gaps in current knowledge, and clarify specific goals and objectives. For many of the
research and monitoring components, more emphasis should be placed on the use of
information collected on individually-marked shorebirds, including radio-tagged birds. A
premium should be placed on the development of robust survey and experimental designs.
5.0. SHOREBIRD TECHNICAL COMMITTEE MEMBERSHIP
Karen Bennett Shorebird biologist, Delaware Division of Fish and Wildlife
Gregory Breese Shorebird biologist, U. S. Fish and Wildlife Service
Joanna Burger Shorebird biologist, Rutgers University
David Carter Coastal zone manager, Delaware Coastal Management Program
Robert Gorrell Fisheries biologist, National Marine Fisheries Service
Brian Harrington Shorebird biologist, Manomet Center for Conservation Sciences
Marshall Howe Shorebird biologist, U. S. Geological Survey
Stewart Michels Fisheries biologist, Horseshoe Crab Technical Committee
Mike Millard Fisheries biologist, U. S. Fish and Wildlife Service
David Mizrahi Shorebird biologist, New Jersey Audubon Society
Lawrence Niles Shorebird biologist, New Jersey Division of Fish and Wildlife
Nellie Tsipoura Shorebird biologist, National Resource Defense Council (formerly)
Brad Andres Coordinator, Shorebird biologist, U. S. Fish and Wildlife Service
6.0. PEER REVIEW PANEL PARTICIPANTS
Dr. H. Jane Brockmann University of Florida, Department of Zoology
Dr. Chris S. Elphick University of Connecticut, Department of Ecology and
Evolutionary Biology
Dr. James D. Fraser Virginia Polytechnic Institute & State University, Department of
Fisheries and Wildlife Sciences
Dr. Patrick G. R. Jodice South Carolina Cooperative Fish and Wildlife Research Unit,
Clemson University
Dr. Erica Nol Trent University, Biology Department
Dr. Adrian H. Farmer U. S. Geological Survey, Fort Collins Science Center
Dr. James D. Nichols U. S. Geological Survey, Patuxent Wildlife Research Center
Dr. John R. Sauer U. S. Geological Survey, Patuxent Wildlife Research Center
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 12
B. Shorebird-Horseshoe Crab Assessment
1.0. INTRODUCTION
1.1. Shorebird Technical Committee
The Atlantic States Marine Fisheries Commission asked the U. S. Fish and Wildlife Service to
form a Shorebird Technical Committee that would provide technical guidance, regarding effects
that horseshoe crab management actions could have on shorebird populations, to the Horseshoe
Crab Management Board. Members and Terms of Reference of this committee are provided
along with this report. The immediate task of the committee is to produce a peer-reviewed report
that reviews and synthesizes unpublished and published information on shorebird populations,
shorebird habitat use, shorebird energetic requirements, threats to shorebird populations, and
horseshoe crab egg abundance. From this report, the committee will generate a set of
conclusions, management recommendations, and research needs. The report and
recommendations will also undergo an independent peer review.
1.2. Horseshoe Crabs, Shorebirds, and Delaware Bay
Reported commercial landings of horseshoe crabs (Limulus polyphemus) on the Atlantic coast of
the U. S. increased dramatically, relative to the previous 4 decades, in the mid 1990s (Figure 4 in
Walls et al. 2002). Horseshoe crabs are most abundant between Virginia and New Jersey
(Shuster 1982), and Delaware Bay supports the largest concentration of spawning individuals
(Shuster and Botton 1985, Botton and Ropes 1987). Delaware Bay also supports large
aggregations of shorebirds (>500,000 individuals) during spring migration and is one of the most
numerically important spring stopover sites in North America (Clark et al. 1993). Timing of
shorebird arrival coincides with the availability of an abundant food source — the eggs released
by spawning horseshoe crabs — that is used to build fat reserves for non-stop flights to breeding
grounds in the Canadian arctic (Myers 1986). Hence, concern has been raised about the negative
effect that crab harvest might have on shorebirds during spring migration (see Berkson and
Shuster 1999). Although several actions have recently been taken to conserve horseshoe crab
populations (restrictions on harvest, delineation of a no-fishing reserve, use of bait bags, and
development of alternative baits), the current status of horseshoe crabs, shorebirds, and their
relationship remains unclear (see Walls et al. 2002).
1.3. Economic Value of Crabs and Shorebirds
Horseshoe crabs are commercially harvested for use in the biomedical industry (where crabs are
bled and usually returned to the ocean) and as bait in the American eel (Anguilla rostrada) and
“conch” (really a whelk, Busycon spp.) pot fisheries (Atlantic States Marine Fisheries
Commission 1998a*). Eels are then used for either finfish bait or human consumption. An
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 13
economic analysis indicates that the annual social welfare benefit (the benefit to consumers
because they are able to purchase goods and services below their willingness to pay) of the
fishery along the entire Atlantic coast is about $150 million for the biomedical industry and $21
million for the commercial eel and whelk fisheries (Manion et al. 2000*; 1999 dollars).
Regional economic outputs (New Jersey, Delaware, Maryland) are valued (1999 dollars) at $2.2
– 2.8 million for the eel/whelk fisheries, $26.7 – 34.9 million biomedical industry, and $6.8 –
$10.3 million for recreational birding (Manion et al. 2000*). Another study estimated that 6,000
– 10,000 recreational birders visited New Jersey’s Delaware Bay beaches in the spring and
contributed a gross economic value (total gross output + consumers’ surplus) of 11.8 – 15.9
million to local communities (Eubanks et al. 2000*). Overall, the biomedical use of horseshoe
crabs is the most economically valuable across the entire Atlantic coast, and the regional value of
crabs to recreational birding is at least, if not greater, than the commercial value.
1.4. Evaluation Approach
Under the precautionary principle (Buhl-Mortensen and Welin 1998), Smith et al. (2002c)
suggest that it would be risk prone to assume species’ risk is low unless a statistical power
analysis had shown that a study design was powerful enough to detect biologically meaningful
change. Peterman and M’Gonigle (1992) outline 3 outcomes when statistical power is
incorporated into decision-making: 1) a biologically meaningful and statistically significant
decline results in harvest restrictions, 2) no evident decline and high power results in no harvest
restrictions, and 3) a biologically meaningful, statistically non-significant decline and low power
increases species’ risk. In the latter case, high uncertainty should trigger harvest restrictions as a
risk-averse strategy. Power analyses generally address singular datasets. To judge an overall
effect when multiple studies or datasets test a singular null hypothesis, a concordance of
evidence approach is a reasonable way to evaluate overall effects (Andres 1999). Thus, a
preponderance of evidence in one direction or the other should result in clear management action
(including no action). Therefore, the committee will use the concordance, or preponderance, of
evidence approach described above to evaluate the report’s contents. Because many regression
analyses are sensitive to the time period selected, and results varied widely depending on starting
year, analyses of some population data were compared among 2 groups — before 1997 and after,
and including, 1997. More intensive shorebird and horseshoe crab studies were generally
initiated during, or after, 1997. An “*” after the year of a citation indicates that the material is an
unpublished report, a submitted manuscript, or an abstract.
2.0. DELAWARE BAY SHOREBIRDS
2.1. Species Considered
Although several shorebird species will be considered in this report, attention will primarily
focus on the red knot (Calidris canutus). Available information is greatest for the red knot and is
less extensive for the ruddy turnstone (Arenaria interpres), sanderling (Calidris alba),
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 14
semipalmated sandpiper (Calidris pusilla), and least sandpiper (Calidris minutilla). Relatively
little specific information exists on the dunlin (Calidris alpina), short-billed dowitcher
(Limnodromus griseus), or long-billed dowitcher (Limnodromus scolopaceus). Information
presented is specific to taxa that use Delaware Bay. Aside from the least sandpiper, species were
selected because of their reliance on beach habitats and their frequency of observation on
Delaware Bay aerial surveys from 1986 to 2002 (see Clark et al. 1993; Table 5.4): semipalmated
sandpiper (40%), ruddy turnstone (29%), red knot (17%), sanderling (6%), dunlin (6%), and
long-/short-billed dowitcher (2%). The least sandpiper was chosen because of its contrasting us
of marsh habitats, rather than beaches, which indicates less of a dietary reliance on horseshoe
crab eggs. Long-billed dowitchers are only rarely observed on Delaware Bay beaches.
2.2. Conservation Status and Protection
The U. S. Shorebird Conservation Plan describes 6 factors of vulnerability (population trend,
relative abundance, breeding threats, non-breeding threats, breeding distribution, and non-breeding
distribution) that were used to determine the conservation concern of North American-breeding
shorebird populations (Brown et al. 2001*). Combinations of these factors were used
to designate the conservation concern of shorebird populations as: highly imperiled, high
concern, moderate concern, low concern, or not at risk. This type of assessment was used by the
U. S. Fish and Wildlife Service (2002*) to develop a Congressionally-mandated list of Birds of
Conservation Concern. Of the 8 species mentioned in Section 2.1, the red knot, ruddy turnstone,
and sanderling are listed as species of high conservation concern in the U. S. Shorebird
Conservation Plan (Brown et al. 2001*), and the red knot and short-billed dowitcher (primarily
due to central and western populations) are listed as Birds of Conservation Concern by the U. S.
Fish and Wildlife Service (2002*).
All migrant species are protected in the U. S. under the statutes of the Migratory Bird Treaty Act,
as amended, and are recognized in international agreements such as the Western Hemisphere
Convention and the Convention on Arctic Flora and Fauna. Because of its value to birds,
Delaware Bay has received international recognition as a Western Hemisphere Shorebird
Reserve Network site of hemispheric importance (>500,000 shorebirds annually), a Wetland of
International Importance under the Ramsar Convention (>1% of a flyway waterbird population),
and an Important Bird Area of global significance (because of large aggregations).
2.3. Vulnerability of Long-distance Migrant Shorebirds and the Red Knot Focus
Piersma and Baker (2000) outlined several critical life history traits of migrant shorebirds that
include: low productivity, long lifespan, trophic specialization, gregariousness,
immunospecialization, sometimes strong sexual selection, long flights, metabolic adaptations for
flight endurance, a precise annual cycle clock, orientation mechanisms, geographic bottlenecks
(reliance on a small number of wintering and stopover sites), and reduced genetic variability.
The red knot epitomizes these critical life history traits, and their trophic specialization on
marine environments makes them vulnerable to perturbations to these habitats, particularly at
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 15
geographic bottlenecks. Piersma and Baker (2000) suggested that populations of long-distance,
long-hop migrant shorebirds, such as the red knot, are mainly constrained by access to high
quality non-breeding habitats, a concept previously championed by Myers (1983).
Hitchcock and Gratto-Trevor (1997) modeled a local decline of semipalmated sandpipers and
found that out of 5 variables (fecundity, adult survivorship, juvenile survivorship, delayed
recruitment, and immigration), adult survivorship had the most significant influence on the
population decline. Reductions in adult survival, through over-hunting and possibly stopover
habitat change, are suggested to have caused the drastic decreases, and possible extinction, of
Eskimo and slender-billed curlews (Gill et al. 1998, Gretton 1991). Piersma and Baker (2000)
suggest that the probability of death by exhaustion or infection increases exponentially and
reproduction decreases logarithmically as energy stores at stopover departure time and body
mass on breeding ground arrival decrease. Because changes in population size are so sensitive to
levels and variation in adult survival, conservation of high quality stopover and wintering sites is
critical. Historical population bottlenecks may have caused the low genetic variability currently
observed in some shorebird populations (Piersma and Baker 2000).
2.4. Red Knot Distribution
The red knot breeds in arctic regions of Siberia, Alaska, Canada, and Greenland and is the largest
arctic-nesting sandpiper (i.e. in the genus Calidris) in North America. Three populations of red
knots are found in North America: the subspecies C. c. islandica breeds in the northeastern high
Canadian arctic and Greenland, migrates through Iceland, and winters in western Europe; C. c.
roselaari likely breeds in Alaska and migrates along the Pacific coast and likely through interior
North America; and C. c. rufa breeds in the central Canadian arctic and migrates primarily along
the eastern coast of North America (Piersma and Davidson 1992). Most rufa individuals winter
along the coasts of South America, and the largest number of individuals are found along the
Chilean and Argentine shorelines of Tierra del Fuego (Morrison and Ross 1989a). Breeding
origins of knots wintering in the southern U. S. and migrating through the interior of the
continent are not completely known (Harrington 2001).
2.5. Red Knot Annual Cycle
Southward migration of adult red knots begins in mid-July when between 5,000–15,000 birds
have been observed in James Bay, Canada (Morrison and Harrington 1992). Adult knots arrive
on the Atlantic coast of North America from mid-July to early August. Juveniles depart later
than adults and migrate through eastern North America from late August to mid-September.
Concentrations of fall migrants are more disperse than during spring migration (see section 3.3).
September aggregations of 1,800–12,000 knots have recently been reported along the coast of
Georgia (Harrington and Winn 2001*). Knots banded in Georgia generally winter in Florida
(likely C. c. roselarii), where the mean wintering population is about 6,300 " 3, 400 (SD)
individuals (Harrington et al. 1988). Individuals wintering in southwest Florida have high site
fidelity (Below 2001*). Rufa knots depart the northeastern U. S. by late August and early
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 16
September to undertake a trans-Atlantic Ocean flight to arrive on the north coast of the
Suriname, French Guiana, and Brazil. From there, they overfly central Brazil, stop briefly in the
Pantanal (on the Rio Negro’s salt lakes late September to early October), reach maximum
abundance in Lagoa do Peixe in October, and arrive in Tierra del Fuego by early November.
Northward migration in Argentina begins in mid-February and persists through early April.
From mid-February to mid-March, 5,000–7,000 knots were present daily in Bahía de San
Antonio Oeste, Argentina (González et al. 2001). Main passage through Lagoa do Peixe, Brazil,
(used by about 7,000 knots) occurs from mid-April through the first week of May (Nascimento
2001*). Birds depart the Maranhão coast of northeastern Brazil, where >10,000 knots have been
observed (Nascimento 2001*), during early to mid-May. April aggregations of $6,000 knots
have been noted in South Carolina (Harrington and Winn 2001*) and peak counts of 7,710–
8,955 knots have been recorded on the outer coast of Virginia (Truitt et al. 2001*), where birds
banded in Argentina (27 knots), Delaware Bay (27) and Brazil (4) were observed. Large
numbers of birds (maximum counts range from 19,445 to 95,490 knots) arrive in Delaware Bay
during the second week of May and usually depart by the end of May or early June. Passage
flights of knots have been observed in James Bay, Canada, (but not landing) in late May and
early June (Morrison and Harrington 1992). Knots arrive on their Southampton Island breeding
grounds during the first 10 days of June (P. Smith, Canadian Wildlife Service, personal
communication). Incubation is 21–22 days, and both parents incubate the 4-egg clutch (see
Harrington 2001). Fledging period is estimated to be about 18 days (see Harrington 2001).
Females may depart the breeding grounds before males (see Harrington 2001).
2.6. Distribution and Migration Routes of Other Species
2.6.1. Ruddy turnstone
A Holarctic species, 3 populations of ruddy turnstones breed in North America: A. i. intepres
breeds in western and northern Alaska and winters on Pacific islands and the Pacific coast of
North America, a disjunct population A. i. intepres breeds in the Canadian high arctic and
winters in Europe, and A. i. morinella breeds in the central and low Canadian arctic, into
northeastern Alaska, and migrates primarily along the eastern coast of North America, including
through Delaware Bay (Nettleship 2000). Highly coastal in its habitats, morinella winters in the
southern U. S., throughout the Caribbean, and along the northern and eastern coasts of South
America south (a few) to Tierra del Fuego (Morrison and Ross 1989a). Turnstones wintering on
the western coasts of Central and South America may be either morinella or interpres (Nettleship
2000). The greatest winter aggregations of morinella occur in northern South America
(Morrison and Ross 1989a).
2.6.2. Sanderling
Breeding distribution of the sanderling is similar to that of the red knot, but no subspecies have
been described (MacWhirter et al. 2002). The wintering distribution is much broader than the
knot —sanderlings are found along the shorelines of every continent except Antarctica
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 17
(MacWhirter et al. 2002). Sanderlings nesting in the northeastern Canadian High arctic are
thought to winter in Europe, and other birds breeding in the eastern arctic likely use eastern
Atlantic and interior flyways (MacWhirter et al. 2002). The population passing through
Delaware Bay probably winters in the southeastern U. S., Caribbean, and South America
(Morrison et al. 2001). The greatest aggregations of wintering birds are found along the Pacific
coast, rather than the Atlantic coast, of South America (Morrison and Ross 1989a).
2.6.3. Semipalmated sandpiper
The semipalmated sandpiper breeds throughout the well-vegetated tundra of arctic and sub-arctic
regions of North America. Although populations have not differentiated to the point of
subspecies recognition, a decreasing cline in body size occurs from east to west (Gratto-Trevor
1992). Semipalmated sandpipers that use Delaware Bay are thought to nest in the eastern
Canadian arctic and use the Atlantic flyway to travel to wintering grounds along the Caribbean
and Atlantic coasts of South America (Harrington and Morrison 1979). Winter aggregations are
greatest along the northern coast of South America (Morrison and Ross 1989a).
2.6.4. Dunlin
The breeding distribution of the dunlin is one of the most cosmopolitan of all small sandpipers.
Populations in North America have differentiated into 3 subspecies: C. a. arcticola breeds in
northern Alaska and northwest Canada and winters in southeastern Asia, C. a. pacifica breeds in
western Alaska and winters primarily along the west coast of North America, and C. a. hudsonia,
which passes though Delaware Bay, breeds in the eastern and central Canadian arctic and winters
on the Atlantic and Gulf of Mexico coasts (Warnock and Gill 1996). Few dunlins of any
subspecies winter south of Mexico (Warnock and Gill 1996). More dunlins may be found in
marshes than on beaches of Delaware Bay (Burger et al. 1997).
2.6.5. Short-billed dowitcher
The short-billed dowitcher is restricted to North America, where 3 recognizable subspecies
occur: L. g. griseus breeds in eastern Canada and winters in Central and South America, L. g.
hendersoni breeds in Central Canada west of Hudson Bay and winters in on the Atlantic and
Gulf of Mexico coasts, and L. g. caurinus breeds in southern Alaska and winters along the
Pacific coast from California to South America (Jehl et al. 2001). Short-billed dowitchers in
Delaware Bay are likely L. g. griseus. More short-billed dowitchers might use Delaware Bay
marshes than beaches (Burger et al. 1997).
2.6.6. Long-billed dowitcher
The long-billed dowitcher is monotypic throughout its range in northeastern Russia, Alaska, and
northwestern Canada (Takekawa and Warnock 2000). Its breeding range is more northern than
the congeneric short-billed and Asiatic dowitchers (L. semipalmatus). Long-billed dowitchers
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 18
winter on the Pacific coast from southern British Columbia to El Salvador and eastward to North
Carolina (Takekawa and Warnock 2000). Most spring dowitchers in Delaware Bay are short-billeds.
2.6.7. Least sandpiper
The least sandpiper has the broadest and most southern distribution of any Calidris sandpiper
breeding in North America; their range stretches across the northern boreal and sub-arctic region
from Newfoundland to western Alaska (Cooper 1994). Populations have not differentiated to the
point of subspecies recognition, but birds using the Atlantic flyway, including Delaware Bay,
likely breed in eastern Canada (Morrison et al. 2001) and winter in the southeastern U. S.,
Caribbean, and northern South America. Winter aggregations are greatest along the northern
coast of South America (Morrison and Ross 1989a). Least sandpipers tend to use marshes,
rather than shorelines, of Delaware Bay during spring migration and are not recorded in large
numbers on aerial beach surveys (see Clark et al. 1993).
3.0. ABUNDANCE AND DISTRIBUTION OF HORSESHOE CRABS
The horseshoe crab ranges from the Yucatan Peninsula to Maine and is most abundant between
Virginia and New Jersey (Shuster 1982). The Delaware Bay hosts the largest concentration of
spawning horseshoe crabs worldwide (Shuster and Botton 1985). Within Delaware Bay,
spawning horseshoe crabs have been reported from Woodland Beach to Cape Henelopen in
Delaware and from Sea Breeze to Cape May in New Jersey (Smith et al. 2002b,c). Some
spawning may occur farther up the estuary but is probably restricted by salinity and the
increasing presence of salt marsh and peat banks (Shuster and Botton 1985). Botton et al. (1988)
observed fewer spawning crabs in proximity of peat beds. Density of spawning crabs on beaches
varies annually (Smith et al. 2002b), although beaches within the lower to middle portion of
Delaware Bay tend to support the highest spawning concentrations.
The high concentration of breeding crabs may be attributable to the abundance of sheltered,
coarse-grained, well-drained sandy beaches that are conducive to spawning and egg incubation.
In addition, large intertidal flats adjoining, or in close proximity, to these beaches likely provide
important nursery habitat. High, wide, low-tide terraces also dissipate wave energy and create
narrow, steep beaches. Low wave energy associated with tidal creeks may explain why high
concentrations of horseshoe crab spawning have been observed in sandy areas within tidal
creeks. Botton et al. (1988) estimated that only 10% of the New Jersey shoreline in Delaware
Bay provided optimal horseshoe crab spawning habitat. However, horseshoe crabs are
opportunistic and use other habitats that are less conducive to egg survival. Shuster (1982)
suggested that beach temperature, moisture level, and oxygen concentration affected horseshoe
crab egg viability. Eggs remain in the sand for 2–4 weeks before hatch. Crabs have been known
to spawn subtidally, but the extent to which this occurs is unknown (Atlantic States Marine
Fisheries Commission 1998a*). Female crabs burrow into sediments to lay their eggs. Kraeuter
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 19
and Fegley (1994) found that mean depth of sediment mixing (11 cm) corresponded closely to
the mean carapace height of female crabs.
Mature horseshoe crabs move inshore from deeper portions of the bay and coastal waters in late
spring to spawn (Atlantic States Marine Fisheries Commission 1998a*). Spawning in Delaware
Bay may occur as early as April and last into July (Shuster and Botton 1985), with peak
spawning activity typically occurring around the new and full moons in May or June. Spawning
is usually higher on the highest of the 2 daily tides, which typically occur at night in Delaware
Bay. Male horseshoe crabs often precede females to a beach and await the arrival of females
(Shuster 1996). Maximum concentrations of spawning crabs may differ temporally between the
New Jersey and Delaware sides of the Bay. For example, in 1999 maximum horseshoe crab
spawning occurred in mid-May in New Jersey, but peaked in early June in Delaware (Smith et al.
2002c). Spawner abundance (adult females) during 1999–2000 was higher in Delaware than in
New Jersey, but was higher in New Jersey in 2002 (Smith and Bennett 2003*). Previously,
authors have reported higher spawning concentrations in New Jersey (Shuster and Botton 1985).
Smith et al. (2002c) found that lunar phase (new/full) and wave height had the most significant
effects on spawning activity, but effective modeling of spawning activity included a combination
of time, place, weather, and tide height. In terms of an optimal design to survey spawning crabs,
an increase in the number of sampled beaches had the greatest effect on reducing the CV
(coefficient of variation) of the estimate of spawning females. Thus, spawning varied spatially
and temporally and was moderated by wave height
Two years of Peterson disc tagging in Delaware Bay showed that horseshoe crabs spawn
multiple times over a season, with males spawning more frequently than females, and that crabs
appear to exhibit limited beach fidelity from year to year (Eyler and Millard 2002*). A
combined acoustic and radio-tag study conducted by Brousseau et al. (2002*) also showed strong
within-season fidelity to spawning beaches; 91% of the 23 crabs successfully tracked returned to
spawn on beaches where they were initially tagged in the same year. Although sample sizes
were low and observation duration was relatively short, the study also found that tagged female
crabs remained between 50 and 250 m offshore from their known spawning beaches.
Besides providing food to shorebirds, horseshoe crab eggs and larvae are seasonal foods for fish
[particularly striped bass (Morone saxatilis) and white perch (Morone americana)], crabs, and
gastropods (Shuster 1982). Contributions of horseshoe crab eggs and larvae to the diet of these
species is generally unknown (Atlantic States Marine Fisheries Commission 1998a*). Buckel
and McKown (2002) found horseshoe crab eggs and juveniles in 42% of stomachs, which
comprised 44% of identifiable prey items, of age 1 striped bass collected in beach seines in Long
Island and Staten Island. Lutcavage and Musick (1985) determined that the most common prey
of loggerhead turtles (Caretta caretta) in Chesapeake Bay were adult and sub-adult horseshoe
crabs, which can represent $42% of the diet (Lutcavage 1981). Botton (1993) observed gulls
feeding on live adult horseshoe crabs that were stranded on exposed beaches. Gulls attacked the
exposed book-gills of overturned crabs. Through transect surveys, mortality was estimated at
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 20
7,760 crabs/km, and gull predation was suggested to be the most important source of mortality to
crabs when they were exposed on spawning beaches.
4.0. POTENTIAL THREATS TO SHOREBIRDS
4.1. Heavy Metal Concentrations in Shorebirds and Horseshoe Crabs
Data from the 1990s indicated that the levels of metals in body feathers of 3 species of
shorebirds from Delaware Bay were generally not high enough to directly affect birds
themselves (Burger et al. 1993). However, mercury levels were relatively high (red knot = 1.1
ppm, sanderling = 2.8 ppm) and suggested a need for further monitoring. Burger et al. (2002b)
examined the levels of arsenic, cadmium, chromium, lead, manganese, mercury and selenium in
the eggs, leg muscle, and carapace musculature (hereafter called apodeme, the fleshy part of the
carapace) in female horseshoe crabs from 4 beaches in New Jersey and 4 beaches in Delaware to
determine whether there were location differences in metal levels, and whether these levels were
high enough to cause effects in birds that eat them. If the crabs were obtaining heavy metals in
the period immediately before egg laying, and sequestering them in their eggs, then the eggs
from female crabs that nest farther north in the bay, where industrialization is greater, should
have higher levels. Eggs were examined because they could be compared to levels reported
earlier from the same study area (Burger 1997), and they are the major food resource for
shorebirds migrating through the bay. Overall, there were some differences in metal levels of the
crabs collected in New Jersey and Delaware, but the differences were generally not great and
there was no consistent pattern in the bay. Previous work demonstrated horseshoe crab egg
sensitivity to heavy metal toxicity (Botton et al. 1998, Botton 2000, Itow et al. 1998a, 1998b).
Manganese concentrations in Delaware crabs (but not the eggs) were >2x than those from New
Jersey. There were some location differences for all 3 tissues (except eggs in Delaware) for both
New Jersey and Delaware. Although the differences were significant, they were generally not
great; there were no order of magnitude differences among collection sites. Contaminant levels
were generally low. The levels of contaminants found in horseshoe crabs were well below those
known to cause adverse effects in the crabs themselves or in organisms that consume them or
their eggs. Contaminant levels have generally declined in the eggs of horseshoe crabs from
1993–2000 in Delaware Bay, suggesting that contaminants are not likely to be a problem for
secondary consumers. While it is important to examine the levels of metals in horseshoe crabs
from Delaware Bay, it is equally important to understand contaminant patterns along the east
coast of North America. This study is reported below.
Burger et al. (2002a) examined the levels of metals (arsenic, cadmium, chromium, lead,
manganese, mercury, and selenium) in the eggs, leg muscle, and apodeme of 100 horseshoe
crabs collected at 9 sites from Maine to Florida. Crabs were collected from the spawning
beaches during 2000. Only large females (n = 5–16 per location) were collected to control for
possible sexual differences and to increase the likelihood of obtaining egg samples. Arsenic
levels were the highest, followed by manganese and selenium, and levels for the other metals
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 21
averaged below 100 ppb for most tissues. Arsenic and mercury levels were highest in the leg
muscle, cadmium, lead, manganese, and selenium levels were highest in eggs, and chromium
levels were highest in the apodeme. There were significant geographical differences for all
metals in all 3 tissues. No one geographical site had the highest levels of >2metals. Arsenic,
with the highest levels overall, was highest in Florida in all 3 tissues. Manganese levels were
highest in Massachusetts for eggs and apodeme, but not leg, which was highest in Port Jefferson,
New York. Selenium was highest in apodeme from Florida, and in eggs and leg muscle from
Prime Hook, Delaware. The patterns among locations and tissues were not as clear for the other
metals because the levels generally averaged below 100 ppb. The levels of contaminants found
in horseshoe crabs, with the possible exception of arsenic in Florida, and mercury from Barnegat
Bay and Prime Hook, were below those known to cause adverse effects in the crabs themselves,
or in organisms that consume them or their eggs, even in relatively large quantities. These
results indicate that site-specific data are essential for managers to evaluate the potential threat
from contaminants to both the horseshoe crabs and to their consumers.
4.2. Organic Compound Concentrations in Shorebirds and Horseshoe Crabs
Maghini (1996*) collected sand, horseshoe crab eggs, and ruddy turnstones, at 2 locations, Port
Mahon and South Bower Beach, along the Delaware shoreline. Sites were selected to sample
resident and migrant horseshoe crab populations, which could be exposed to different
contaminant sources. Sediment and egg samples at each site were collected 1–4 June 1992 along
10 (non-randomly selected) transects located perpendicular to the shoreline. Sand within 25 cm
of the surface was collected at 10 stations along each transect. Horseshoe crab eggs were also
collected along the 10 transects. Twenty-two turnstones were shot at Port Mahon, and none were
collected from South Bower Beach. Chemical analyses were conducted by the Geochemical
Environmental Research Group at Texas A&M University. Quality assurance measures were
conducted by the laboratory and considered satisfactory. Many samples had concentrations of
organic compounds that were below the limits of detectability. Maghini (1996*) found that
concentrations of DDE and PCBs in turnstones were at background concentrations, but 2
carcasses had concentrations of DDT that suggested recent exposure. Although concentrations
of lead, mercury, and cadmium were detectable in sand and tissue samples, most were within
background concentrations. Arsenic and selenium concentrations were elevated in turnstone
tissues, but were similar to other species that fed on marine invertebrates and fish. Similar
concentrations in horseshoe crab eggs suggest that they were the likely route of exposure.
Conclusions were that concentrations of trace metals and organochlorines presented low
toxicological risk. However, wider geographic and taxonomic sampling, component analysis of
arsenic in eggs, and measurement of selenium concentrations in livers of turnstones were
suggested. Little is known about chemical concentrations in shorebird wintering areas.
4.3. Disease in Shorebirds
Piersma (1997) suggested that shorebirds may make a trade-off between investments in
immunofunctioning and growth (chicks) or sustained exercise. Some shorebird species appear to
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 22
be restricted to parasite-poor habitats (seashores, the arctic). Red knot chicks raised in the high
arctic had daily energy expenditures that were 1.5x higher than temperate shorebirds of the same
mass, yet grew at a faster rate. For migrant birds, optimal areas are separated seasonally by long-distances.
If long-distance shorebirds are adapted to use parasite-poor habitats, they may be
particularly susceptible to parasites and pathogens. Captive red knots only remained healthy
after sea water was flushed through their holding cages, which suggested that they may be
particularly susceptible to common avian pathogens. Figuerola (1999) found that haematoza
infection rates in waterbirds, when controlling for phylogeny and population size, were greater in
freshwater species than in those inhabiting saline habitats. Low reproductive success could be a
cost associated with breeding in the climatically-marginal, but parasite-low, arctic. Increased
adult survival afforded by inhabiting areas of low parasite loads may offset these costs.
The Southeastern Cooperative Wildlife Disease Study (2002*) sampled 905 shorebirds from
Delaware beaches in 2000 and 501 shorebirds (and 75 fecal samples) in 2001 for occurrence of
influenza viruses. Virus was isolated from 5 species. The ruddy turnstone (n = 368) had the
highest incidence rate (>13%), and lesser incidence rates (<5%) were found in red knots (n =
620), dunlins (n = 164), semipalmated sandpipers (n = 107), and short-billed dowitchers (n = 68).
Fecal samples collected off the ground in areas of turnstone activity revealed isolation of 5
viruses. Preliminary results from 2002 were similar. One interesting note is that a turnstone in
Delaware Bay that did not have the virus on 21 May tested positive when it was recaptured on 28
May.
In 1997, dead and dying red knots (46), white-rumped sandpipers (11), and sanderlings (3) were
discovered in the area of Lagoa do Peixe, Brazil (Baker et al. 1998). All of the 35 collected
knots were infected by hookworms (Acanthocephala spp.). About 150 knots found sick or dead
in western Florida had their digestive tract infected by an unidentified sporozoan-type protozoan
parasite (Woodward et al. 1977). Although no dramatic die-offs have been observed over the
last 2 decades, information on parasite loads of Delaware Bay’s shorebirds is lacking and should
be evaluated. Following Piersma’s (1997) hypothesis, Delaware Bay beaches could provide
important, low-parasite environments needed by foraging red knots.
4.4. Shoreline Changes in Delaware Bay
Shoreline habitat change can reduce horseshoe crab spawning habitat and consequently shorebird
feeding habitat. Residential development along Delaware Bay’s beachfront can have negative,
direct and indirect, effects on foraging and roosting shorebirds. Storm damage and longshore
transport of sand can greatly affect beach characteristics. Bulkheads may block access to
intertidal spawning beaches, and seawalls and groins can intensify local shoreline erosion and
prevent natural beach migration (Atlantic States Marine Fisheries Commission 1998a*). Over
the last 100 years, beaches in New Jersey have eroded at a rate of 0.3–3.7 m/year and in
Delaware at a rate of 0.3–7.9 m/year (mean = 0.9–1.5 m/year, U. S. Army Corps of Engineers
1991*, 1997*), and are presently at 2–6 m/year (Galofre 2002*). Increased turbidity, siltation,
and peat exposure caused by erosion creates anaerobic conditions in horseshoe crab nests and
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 23
reduces egg survivorship (Botton et al. 1988). Few crabs tend to spawn on beaches with much
peat. Natural and human creation of inlets (e.g., Thompson’s and Moore’s Beaches) may have
channeled crabs into marshes where they were harvested or failed to successfully reproduce (see
<www.delawarebay.com>). Sand nourishment on beaches can increase habitat for spawning
horseshoe crabs if sediment types match natural beaches favorable to breeding horseshoe crabs
(see section 7.6). Monitoring and management of beach conditions will likely be needed to
sustain habitats for spawning crabs and foraging shorebirds.
4.5. Sea Level Rise from Global Climate Change
Galbraith et al. (2002) used U. S. Environmental Protection Agency data on historical sea level
rise to predict sea level change at sites important to shorebirds. Assuming global temperature
changes of 2oC (50% chance) or 4.7oC (5%), resultant sea level rise would be 0.34 m (50%
chance) or 0.77 m (5%). Local rates of historical sea level change were used with the Sea Level
Affecting Marshes Model (SLAMM 4) to predict local effects of sea level rise by 2100. Based
on historical rates, sea level in Delaware Bay would rise 0.3 m by 2100, with a 50% chance of
rising 0.6 m. With these rates of sea level rise, tidal flats in Delaware Bay would decrease by
23% under a historical rise and a predicted 50% chance of a 57% loss. A corresponding increase
in salt marsh (.10%) would occur. These estimates do not account for any mitigation measures
undertaken (e.g., seawalls). If losses of this magnitude occurred, Delaware Bay might not be
able to support historical levels of shorebird use. Increased “storminess” associated with global
climate change could further alter Delaware Bay’s shoreline habitats.
4.6. Arctic Breeding Ground Conditions
Reproduction in arctic-breeding birds is known to be highly variable. Inter-annual variability in
the reproductive success of shorebirds is usually attributed to weather or predation. Variability
in predation on shorebird nests has been suggested as an indirect consequence of the cyclical
abundance of lemmings. When lemmings are abundant, predators primarily rely on them as
food; when lemmings are scarce, predators switch to other sources like birds. Blomqvist et al.
(2002) used a 50-year series of fall banding data of red knots (C. c. canutus) migrating through
the Baltic Sea in southern Sweden (Ottenby), and other information in the literature, to test the
“bird-lemming hypothesis”. They predicted that: 1) juvenile red knot numbers would correlate
with lemming fluctuations, 2) adult red knot numbers would not correlate with lemming
numbers, and 3) post-breeding migration of adults would be earlier in years of high predation
pressure. As an alternative hypothesis, they examined the correlation of climatic oscillations and
breeding success.
At Ottenby, Blomqvist et al. (2002) found no significant (P > 0.05) long-term trend in the
number of adult or juvenile knots and no significant correlation between the annual numbers of
adults and juveniles. Predation index from the Taimyr region of Russia was significantly and
negatively associated with median knot passage date at Ottenby; proportional den use by foxes
correlated with lemming abundance in arctic Russia. Predation index was significantly and
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 24
negatively correlated with the number of juvenile knots captured at Ottenby, but not with the
number of captured adults. Numbers of juveniles captured at banding stations in South Africa
and Germany were also negatively associated with the predation index. Fourier analysis of the
time series of juvenile captures revealed a periodicity of 3 years, which matched the median date
of adult passage and lemming abundance in the Russian arctic. May weather did not correlate
with any shorebird population variables. Blomqvist et al. (2002) found that patterns in Swedish
knots were similar for curlew sandpiper (Calidris ferruginea) and likely extend to numerous
other arctic-breeding species (see Underhill et al. 1993). Productivity of red knots and other
shorebirds on eastern Southampton Island appears to be similarly correlated to abundance to of
lemmings (P. Smith, Canadian Wildlife Service, unpublished data).
Zöckler and Lysenko (2000) used a climate change model (HadCM2GSal), with a 1% increase
of CO2/year, and Dynamic General Vegetation Models to examine effects of climate change on
Holarctic waterbird populations. Of all biomes, tundra areas are expected to suffer the greatest
climate-related habitat change. Major habitat changes for Calidris sandpipers, particularly in the
low Canadian arctic, are predicted. Southampton Island is predicted to undergo major tundra
loss, while part of northeastern Canada and Greenland are predicted to cool. Habitat changes
have not yet occurred, but temperature changes are underway. Temperatures have risen by 1.3oC
over the last 30 year at Resolute, Canada (Falkingham et al. 2001*). Mean July temperature in
breeding areas was positively, but not significantly, correlated (r = 0.3) with the percentage of
juvenile islandica red knots observed in the subsequent season on wintering grounds. Boyd
(1992), however, suggested that a relationship existed between mean June temperature in
northeastern Canadian arctic and the total number of knots observed in Great Britain the
subsequent winter. More recently, Boyd and Piersma (2001) found that cold arctic summers
affected both productivity and adult survival of knots wintering in Britain.
Little information exists on the biology or productivity of rufa red knots on their breeding
grounds. Knots (20 of 165) radio-tagged in Delaware Bay were relocated on breeding grounds
on Southampton and Prince William Islands, Canada (Niles et al. 2001*). Knots tended to use
low elevation, barren tundra located within 50 km of the coast. Eleven nests in sparsely
vegetated tundra (e.g., eskers, frost boils), often associated with Dryas, were found in 2000.
Topographical placement of nests may depend on the amount of snow cover when birds arrive,
but nests are usually located #180 m of isolated wetlands. Nest density on Southampton Island
ranged from 0.85 nests/km2 in 2000 to 0.58 nests/km2 in 2002 (Niles et al. 2003*). No dramatic
weather events occurred on Southampton Island during the breeding seasons of 1999–2002 (L.
Niles, personal communication).
4.7. South American Wintering Ground Conditions
In general, much of the Patagonia and Tierra del Fuego coast line is in good ecological condition
(see descriptions at <http://www.ramsar.org> and <http://www.whsrn.org>). However, oil
exploration and its associated infrastructure pose risks for migrant shorebirds that depend on
intertidal feeding areas. Some wells have been placed in intertidal areas and development of oil
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 25
industry infrastructure has lead to water and wind erosion of beach environments. Spills from oil
storage and transfer facilities and oil tankers’ illegal ballast discharges are probably the greatest
threat to migrant and wintering shorebirds. Although the region is still sparsely populated, much
of the human population in concentrated in coastal areas, and pollution from untreated sewage is
increasing and may have a future, negative effect on shorebirds. Season tourism brings needed
cash to the region, but recreational beach activities (walking, shellfish collecting, vehicles, dogs)
can disturb feeding and roosting shorebirds. Negative human disturbance effects are often
greatest near cities. Installation of an ash plant (for the production of glass) could negatively
affect shorebirds that use Bahía San Antonio Oeste. The plant could release $250,000 tons of
calcium chloride into the bay annually, that could destroy the clams, mussels, oysters and other
food sources upon which migrating shorebirds depend. Lagoa do Peixe is a large, shallow
coastal lagoon in southern Brazil that has a highly variable, natural hydrology. Depending on
rainfall and winds, the lagoon can dry up completely during the austral summer. Thus, shorebird
use can be highly variable among years. Further north along the Maranhão coast of Brazil,
shrimp farming could alter coastal systems in a way that is detrimental to migrant shorebirds.
Despite potential threats, the southern wintering grounds of red knots do not appear to have
changed dramatically in the last decade.
4.8. Human Disturbance to Shorebirds
Nesting and migrant shorebirds are susceptible to disturbance caused by human activities.
Human disturbance can force shorebirds to: 1) shift to feeding areas with fewer numbers of
humans (Burger and Gochfeld 1991), 2) entirely abandon an area (Pfister et al. 1992, Smit and
Visser 1993), or 3) increase vigilance, movement, or escape flights (flushing). Disturbance can
therefore reduce feeding time and increase energy requirements at a time when migrant birds
need fuel for migration (Hockin et al. 1992, Davidson and Rothwell 1993, Lafferty 2001).
Distance to birds was the best measure of disruption to foraging sanderlings on California
beaches (Thomas et al. 2003). Free-ranging dogs also disrupted foraging behavior and birds
were completely excluded from beaches with intense vehicular use. The chronic disturbance of
shorebirds can disrupt their behavior and cause them to use the energy they are trying to store for
migration in an escape flight, thus affecting their energy balance and potentially their survival
(Helmers 1992, Hockin et al. 1992, Davidson and Rothwell 1993, Harrington and Drilling 1996,
Brown et al. 2001*, Gill et al. 2001, Lafferty 2001, West et al. 2002).
Disturbance, frequently measured by flushing rate, has a greater effect on migratory bird species
than on resident species (Burger and Gochfeld 1991). Anecdotal observations of shorebird
researchers in Delaware (Carter et al. 2002*) and numerous published studies have noted
negative human disturbance effects on shorebirds caused by: 1) walking and jogging (Burger
1981), 2) windsurfing and hunting (Madsen 1998), 3) dog-walking, bird-watching, and shell-fishing
(Goss-Custard and Verboven 1993), 4) automobiles, boats and all-terrain vehicles
(Rodgers and Smith 1997), 5) personal watercraft and outboard-powered boats (Rodgers and
Schwikert 2002), and 6) aircraft (Koolhaas et al. 1993). Flushing distances have been shown to
vary between types of disturbance, individual birds, and species. Researchers associated with
national and regional shorebird conservation plans identified the high priority need to gain more
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 26
information on how human disturbance affects shorebirds (Clark and Niles 2000*, Oring et al.
2000*).
Although no specific studies have been conducted to quantify disturbance effects to shorebirds
in Delaware Bay, repeated disturbance along its beaches likely reduces shorebird feeding
efficiency thereby increasing energy expenditure and reducing energy intake. Efforts to reduce
and minimize human disturbance from recreational and commercial activities, and from research
studies, in Delaware Bay are ongoing. New Jersey has implemented regulations which reduce
the potential for disturbance associated with the horseshoe crab fishery by curtailing the hand
harvest from its beaches. Bird observation platforms in New Jersey and Delaware have been
built to allow for viewing of shorebirds with minimal disturbance. Actions have also been
adopted to minimize any potential disturbance impacts of research associated with the catching
and observing of shorebirds.
4.9. Effect of Disturbance on Survival of Semipalmated Sandpipers
Pfister et al. (1998) present one of the few studies of a migratory species that demonstrates a
relationship between body mass and annual return rate to a site of a migratory species.
Semipalmated sandpipers were captured, color-marked, and measured during fall migration at
Plymouth Beach. Massachusetts, in 1985 and 1986. Beaches was surveyed extensively during
those 2 years for banded birds to determine the minimum length of stay for individuals. From
Plymouth beach, semipalmated sandpipers are thought to make over-water crossings of >3,000
km. Using body fat estimates, length of stay, and a linear regression model derived from
sandpiper banding and recapture data at this site during the 15 years from 1971 to 1984, the
authors calculated percent body fat of 255 individual sandpipers departing from the site. During
1986 and 1987 surveys were conducted to determine how many birds banded the previous year
returned to the site. A logistic regression model was used to relate return to the staging site (1 =
return, 0 = no return) to the estimated fat levels at departure. Because of possible biases in the
methods used to estimate fat at departure, an alternate method was also used to test the
hypothesis that return rates are associated with fat levels. In this method, the authors used the
difference between of actual length of stay and the time in days that would be needed to attain
40% body fat (based on linear regression of fat deposition rates from previously collected data)
as an index of the likelihood that birds would attain the desirable departure weight before
migration. Birds were separated into 3 risk groups based on how many days short they were of
attaining that level of 40% body fat. In both the estimated fat levels at departure and the risk of
not attaining favorable fat levels at departure models, regression analysis revealed that fat level
at departure had a significant association with return rate. The authors suggest that the
association between fat levels and annual return rate is due to differences in return rates caused
by fat depletion during the non-stop flight over water. The results support the idea that
disturbance reducing the feeding efficiency of shorebirds at staging areas can reduce the ability
of these migrants to attain high fat levels for their migratory flights and therefore may lead to
their mortality.
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 27
4.10. Horseshoe Crab Bait Landings
The Atlantic States Marine Fisheries Commission adopted a Fishery Management Plan for
Horseshoe Crab in 1998. It limited landings in New Jersey, Delaware, and Maryland (in
recognition of that these states had already acted to reduce harvest levels) to existing harvest
levels, encouraged other states to reduce harvest, and recommended development of a coast-wide
cap on commercial bait landings in 2000. Adopted in 2000, Addendum 1 established landings
for the 1995–1997 reference period and state-specific 25% reductions in 2000 landings from the
reference period (Atlantic States Marine Fisheries Commission 2000*). It was recognized that
some states had already reduced harvest >25% below the reference period, and these states were
encouraged to maintain their current reductions (about 211,000 crabs in Maryland and 297,680
crabs in New Jersey). States that harvested <1% of the coast-wide landings were exempted from
the 25% reduction (reviewed annually). In addition, Addendum 1 asked the National Marine
Fisheries Service to establish a horseshoe crab sanctuary at the mouth of Delaware Bay. In 2001,
the sanctuary was established and now protects 3,885 km2 of crab habitat from harvest. Also in
2001, Addendum II of the fishery management plan was adopted to establish procedures for
inter-state transfer of harvest quotas (Atlantic States Marine Fisheries Commission 2001*).
After adoption of Addendum I in 2000, coast-wide reductions in crab bait landings ranged from
37 to 58%, and bait landings were reduced 34–75% in New Jersey, Delaware, and Maryland
(Table 4.1). Some unknown portion of crabs that breed in Delaware Bay are likely landed in
Maryland. Because horseshoe crabs have a delayed sexual maturity of about 9 years, changes in
population size that resulted from increased harvest in the mid-1990s and subsequent restrictions
not yet been realized.
4.11. Changes in Horseshoe Crab Populations
The Atlantic States Marine Fisheries Commission Horseshoe Crab Stock Assessment Sub-committee
(Millard et al. 2000*) recommended that 3 surveys, as interim measures until a stock
assessment is completed (Atlantic States Marine Fisheries Commission 1998b*,c*), be evaluated
to determine short-term trends of horseshoe crab populations: 1) re-designed Delaware Bay
spawning survey, 2) Delaware trawl survey, and 3) National Marine Fisheries Service fall trawl
survey.
The Delaware Bay Horseshoe Crab Spawning Survey was substantially modified in 1999 to
provide a statistically reliable survey of spawning crabs. In 2002, volunteers conducted 243 tide-based
surveys on 23 beaches of New Jersey (10 beaches) and Delaware (13 beaches). An index
of spawning activity is calculated as the number of spawning females within 1 m of high tide on
beach index sites. Smith et al. (2002c) recommended that females be used to assess spawning
activity because: 1) female abundance is the most direct measure of reproductive potential, 2)
distribution of females is less variable than males, and 3) counting females alone is more cost-effective.
In 2002, spawning, which peaked in late May, tended to be somewhat higher in New
Jersey than in Delaware (Smith and Bennett 2003*). Since 1999, spawning activity has
remained unchanged in New Jersey (slope = 0.06, SE = 0.04, P = 0.29) and in Delaware (slope =
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 28
-0.08, SE = 0.03, P = 0.16). Substantial shifts in spawning concentrations were noted from
previous years. In 2002, for example, there were large increase in spawning activity on New
Jersey beaches in the upper bay. Increases on some New Jersey beaches may or may not
compensate for declines elsewhere. However, bay-wide spawning activity has been stable over
the past 4 years, indicating some degree of compensation. Smith et al. (2002c) found that the
number of sampled beaches and temporal stratification were the most important determinants of
achieving the power needed to detect changes in spawning activity.
The Delaware 30-foot trawl survey has been conducted consistently between March and
December since 1990; horseshoe crab information is restricted to the April–July period. The
State of Delaware has also conducted a 16-foot trawl survey, for the last 11 years, that targets
juvenile (<160 mm wide) and young-of-the-year crabs. The National Marine Fisheries Service
(NMFS; 2002*) has conducted a fall trawl survey along the Atlantic coast since 1977.
Horseshoe crab information was restricted to the region between New York and Cape Hatteras,
and only stations #27 m deep were used to calculate crab abundances. Gear for the NMFS
survey changed dramatically over the course of a few years in the mid-1980s and invalidated
analysis of the complete time series. Geometric means of annual all crab catches in the 30-foot
trawl have decreased since 1990 (linear regression, R2 = 0.661, P = 0.0007; S. Michels,
unpublished data). Although counts in most recent years appear to be stable, the lowest recorded
catch in 13 years occurred in 2002 (Figure 1). Also, mean catch per unit effort was significantly
(P <0.025) lower in later years of the survey relative to the early 1990s (Table 4.2; Andres
analysis). Although not significant, differences between periods were in the same direction as
the 30-foot trawl survey for juvenile and young-of-the-year crabs in the 16-foot trawl and for all
crabs caught in the NMFS fall survey (Table 4.2; Andres analysis; Figure 2,3). Note that the
catch per unit effort for these latter surveys is very low. Horseshoe crab populations may now be
stable but are likely at lower levels than in the early 1990s, and possible decreases may be
apparent in all age classes of the population. Preliminary estimates from trawl surveys off of
Delaware Bay (extending approximately from Ocean City, Maryland, to Atlantic City, New
Jersey, and 22.2 km offshore) indicate a total population of 11,400,000 " 5,453,000 crabs (95%
confidence interval), of which about 2.7 million are spawning age females (Berkson and Hata,
unpublished data). The estimate of primiparus females ranges from 200,000 to 522,000 crabs.
This does not include any animals within the Delaware Bay or animals beyond 22.2 km, assumes
100% gear efficiency, and should therefore be considered a minimal estimate. Landings of
female horseshoe crabs for the states of Delaware, Maryland, and New Jersey in 2002 totaled
297,932 crabs, suggesting that the stock may be rebuilding as recruitment is exceeding landings
in this area (but the 95% confidence limit of the estimate includes the landings value).
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 29
5.0. ESTIMATES OF SHOREBIRD POPULATION SIZES AND TRENDS
5.1. Shorebird Population Sizes
5.1.1. Coarse continental estimates
Morrison et al. (2001) compiled published and unpublished counts of shorebirds, by season and
region, to generate coarse, flyway population estimates for North American-breeding shorebirds.
They used the maximum summation of counts within a region to determine population size. For
example, maximum counts of red knots at all sites on the Atlantic coast, during northward
migration, would be summed to produce an estimate of that flyway’s population. All regions
would then be summed to produce a continental estimate. These estimates were thought to be
the minimum population present during the late 1980s and early 1990s. The method would
likely only over-estimate population size, by counting individuals multiple times, if large
numbers of the same individuals would stop at a few sites within the same region. Each estimate
was assigned an accuracy (confidence) score which reflected quality and breadth of data used to
generate the estimate. Of the 8 species considered in this report, populations in eastern North
America range from 11,300 to 994,600 individuals (Table 5.1). Confidence in estimates for
these species ranged from low to moderate. The population of rufa red knots was estimated to be
170,000 (150,000 birds in eastern North America) in the late 1980s and was one of the smallest
populations of red knots known to occur throughout the world (Piersma and Davidson 1992).
5.1.2. Re-sighting banded red knots in the 1980s
Between 1980 and 1987, Harrington (2002*) and his colleagues marked red knots with color
bands in North and South America. Between springs of 1981 and 1990, Delaware Bay knots
were scanned for color bands. This information allowed for calculation of the frequency with
which birds of each band cohort (a group of birds banded at the same location in the same year)
were found. This number, in combination with an estimate of annual survivorship of knots and
known band cohort sizes, was used to estimate the population size as: [(number checked for
bands*estimated number alive) ) number of cohort birds found]. The estimated number alive is
the [(cohort size*( monthly survival rate* number of months since banding)]. The re-sighting
rate was calculated as: {[(number of cohort-marked birds found ) expected cohort number alive)
) number of birds checked for bands]*1000}. The expected cohort number alive was the
original number banded in the cohort reduced to adjust for an annual survivorship of 0.752
(details on model selection are not provided). A population estimate of red knots (rufa) made for
each year was based on band re-sighting ratios of knots banded in Massachusetts during fall and
re-sighted in New Jersey in spring, and on knots banded in New Jersey in spring and re-sighted
in New Jersey in spring. Before estimates were calculated, cohorts were removed if <5 banded
birds from a cohort were observed per 1,000 birds checked . This removed re-sighted cohorts
where the original banding cohort was small. In addition, 2 cohorts were removed (banded on
Delaware Bay in 1980 and 1981) where color band loss was a problem. Mean re-sighting rates
of knots banded in Massachusetts and re-sighted in New Jersey were compared to that of birds
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 30
both banded and re-sighted in New Jersey. The 2 groups were not significantly different (F-test,
P > 0.05) and were therefore combined. Separate population estimates were calculated for each
band cohort during each re-sighting year. For example, estimates, adjusted for survivorship,
were made for a cohort banded in Massachusetts in 1984 and re-sighted in New Jersey 10, 22,
34, 46, 58, and 70 months later.
Annual population size estimates during 9 years between 1981 and 1990 ranged from 59,215 "
16,085 (1990) to 212,885 " 49,575 (1981). Ranges of the annual standard deviations of these
estimates was 20–40% of their corresponding annual population estimate (Table 5.2) The overall
mean of 28 separate estimates was 143,680 " 13,579 (SE). There was no population trend
evident among the yearly estimates (R2 = 0.003, P = 0.74). Note, however, that little is known
of the size and annual variation of the non-breeding (presumably sub-adult) population, which
evidently remains in South America and the southeastern U. S. during the northern summer.
Some unknown portion of sub-adults visit Delaware Bay each spring. Finally, little is known of
how the size of the non-breeding population relates to the size of the breeding population or to
annual variation of breeding production. The mean re-sighting estimate of population size of
knots in eastern North America in the 1980s was similar to the coarse estimate (see Table 5.1)
generated by Morrison et al. (2001).
5.1.3. Red knot band re-sighting in South America
González et al. (2001*) color-banded 107 red knots in Rio Grande, Tierra del Fuego, Argentina,
in December 2000 and used re-sighting information from there and Bahia de San Antonio (1,450
km to the north), in early 2001, to estimate the population size of red knots wintering in southern
South America. Scans of Rio Grande-banded birds re-sighted at San Antonio gave an estimate
of the entire population wintering south of San Antonio stopover (in Rio Grande and Bahía
Lomas) as 31,800 (95% confidence interval = 26,850–37,850). Scans of San Antonio-banded
birds re-sighted at either site gave an estimate of 37,600 individuals, which was likely an
estimate of the southern South American wintering population. This estimate corresponds fairly
well with estimates from aerial surveys made during the same period (see section 5.2.1). If the
current population of knots wintering in southern South America is about 30,000 individuals, and
the population wintering in northern South America is about 15,000 birds (A. J. Baker, personal
communication), then the total population of rufa red knots (. 45,000 birds) is probably
substantially lower than late 1980s levels. Maximum counts on spring aerial surveys in
Delaware Bay (see section 5.2.2) from 2000–2002 were lower than this estimated value (Table
5.4).
5.2. Shorebird Population Trends
5.2.1. Aerial surveys of red knots in South America
Aerial surveys, usually with fixed-winged aircraft, were conducted along the southern South
America coastline during the boreal winter 1982–86 (Morrison and Ross 1989a,b). The
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 31
Argentine coast was surveyed in 1982 and Tierra del Fuego was flown in 1985. Flights, at high
tide when possible, were made at an altitude of 50–80 m and 160 km/hour. The flight line was
selected to survey the most important marine-influenced habitats and was usually 50 m offshore.
Shorebirds were identified to species (except for small Calidris sandpipers) unless conditions or
size of flocks prevented a reasonable assessment. Along the Atlantic coast of South America,
red knots (n = 76,392 birds) were distributed among Tierra del Fuego (69.7%), the Argentine
Patagonian coast (18.7%), northern Brazil (10.9%), and western Venezuela (0.7%). In Tierra del
Fuego, the most important site was Bahia Lomas where 41,700 knots were counted (54.6% of all
observations). Aerial surveys of the same shorelines of Tierra del Fuego were repeated with the
same methods and same observers in 2000–2002. Counts of red knots made in Bahia Lomas,
and for the entirety of Tierra del Fuego, in 2000 tended to be similar to counts made in 1982/85
(Table 5.3). However, substantial decreases in knot counts, relative to 2000, occurred in 2001
and 2002 (Table 5.3). A complete survey of Tierra del Fuego and the Patagonia coast in 2002
indicated that knots did not re-distribute themselves at sites north of Tierra del Fuego (Table
5.3). Numbers from Tierra del Fuego in 2003, although analysis is incomplete, suggest a slight
increase from 2002 levels (R. I. G. Morrison, personal communication). Lack of a longer time
series precludes a thorough analysis of this dataset. Ground counts and re-sighting information
suggests that knot numbers at San Antonio declined from >20,000 in 1996 to 15,000 in 1997–
1998 and further to 8,500 ("500) in 2001. Because of relative stability on wintering grounds,
continued surveys of southern South America could provide important information on knot
population change.
5.2.2. Spring aerial surveys in Delaware Bay
To determine shorebird use in Delaware Bay, weekly aerial surveys of the entire shoreline of
Delaware Bay have been conducted, since 1986, by 2 constant observers, and 1 recorder, in a
Cessna 172 (see Clark et al. 1993). Flights, at a height of 30 m above the shoreline, started at
Cape May 3 hours after high tide, headed north along the New Jersey coast to the mouth of the
Delaware River, and then turned south along Delaware’s shoreline to end at Cape Henlopen.
Because little information exists on species-specific turnover rates, the maximum counts
obtained during a single flight are used to determine changes in numbers in Delaware Bay.
Yearly maximum counts are provided in Table 5.4. Using this method, Niles et al. (2003*)
found that the maximum annual counts of red knots differed among recent years (1998–2002;
Kruskal-Wallis, P2 = 19.26, df = 5, P = 0.002). A decrease in maximum red knot counts was
marginally significant (P = 0.068) from 1997 to 2002 (Andres analysis; Kendall’s nonparametric
concordance test; Hollander and Wolfe 1973:185–199). The mean of maximum knot counts,
however, did not differ between 1986–1996 and 1997–2002 periods (Table 5.5). No other
species showed consistent declines, but maximum counts of dunlins and dowitchers have
increased significantly in recent years (Table 5.5). Because of their unknown relationship to real
population size, maximum aerial survey counts are not be useful to determine population change.
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 32
5.2.3. International and Maritime Shorebird Surveys
Bart et al. (2003*) used data from the Maritime Shorebird Survey (MSS) and the complementary
International Shorebird Survey (ISS) data to assess trends in migrant shorebird numbers along
the north Atlantic coast (from Georgia to Newfoundland). The primary purpose of these surveys
is to document abundance and distribution of migrant shorebirds. Volunteers visit sites every
10–14 days, when shorebirds are present in the site’s region, and count all shorebirds. ISS
guidelines ask that counts (or estimates) of all shorebird species be made once each third month
(once between the 1st and 10th, once between the 11th and 20th, and once after the 20th) during
spring (1 April–10 June) and fall (10 July–31 October) migration. Migration periods were
defined for each species by determining the 20th and 80th percentiles of the cumulative
distribution of spring and fall periods. A linear model was used to determine site-specific rates
of change in shorebird numbers, for sites that had >3 visits, and were combined to determine an
average rate of change. Only species that were observed at $8 sites were included in the
analysis, and highly significant outliers (residual P < 0.005) were removed from the analysis.
Morrison and Hicklin (2001*) independently used average counts at “paired” Canadian Maritime
sites to make comparisons between decades (1970s, 1980s, and 1990s). They reported the sign
of the difference (negative or positive) and the significance (P-value) of the difference. Bart et
al. (2003*) found that knot counts declined, but not significantly (P > 0.1), at a rate of
1.65%/year in eastern North America. Sanderling, semipalmated sandpiper, and least sandpiper
all decreased at significant rates (.4–7%, P <0.05) in the ISS/MSS analysis (Table 5.6). Red
knots and semipalmated sandpipers were the only species that showed consistent, negatives
changes among time periods and analysis methods (Table 5.6). In a previous analysis of ISS
data, sanderlings had decreased substantially (Howe et al. 1989). P. Hicklin (unpublished data)
has found a shift in the distribution of bill lengths of semipalmated sandpipers captured while
migrating through the Bay of Fundy, Canada. Proportionally fewer long-billed birds, those from
the most eastern population that use Delaware Bay in the spring, have been captured in recent
years.
5.2.4. Quebec migration checklists
Since 1950, opportunistic information has been collected from daily checklists of volunteer
birders in Quebec. These records have been computerized and were used by Aubry and Cotter
(2001*) to assess the population trends of fall-migrating shorebirds in the province. They used
the frequency of occurrence of shorebird species occurrence on checklists, from 1976 to 1998, to
determine if reporting rates changed through time. From this analysis, significant decreases were
found in reporting frequencies for ruddy turnstones, red knots, and semipalmated sandpipers
(Table 5.6). Decreases in the latter 2 species are consistent with ISS/MSS analyses.
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 33
6.0. IMPORTANCE OF DELAWARE BAY TO SHOREBIRD POPULATIONS
Harrington (2002*) compared 8 years of population estimates of red knots in the in the 1980s
(see section 5.1.2) to aerial surveys conducted during the same period (see section 5.2.2). A
consistent relationship between the maximum count and the population size would only exist if a
constant proportion of the spring migrating knot population uses Delaware Bay each year. No
significant relationship (r = 0.27, P = 0.51) existed between annual estimates of population size
determined from color-banding ratios and maximum counts from annual spring aerial surveys.
On average, the maximum aerial survey count represented 38% of the adult population size
estimates from the same 8 years and ranged from 14 to 77%. Therefore, Delaware Bay is likely
not used by a consistent proportion of the knots each year, and use varies considerably among
years. Note that the error for population estimates is relatively high (see Table 5.2).
Harrington (2002*) also used counts made between 1974 and 2000 by ISS cooperators to
compare numbers of shorebirds at Delaware Bay to other Atlantic coastal regions (see section
5.2.3). From these counts, the maximum value of all counts of each species from Atlantic
marine locations was determined for spring and fall migration periods. To compare Delaware
Bay to other Atlantic locations, maximum counts made during 17 years of aerial surveys of
Delaware Bay (see Clark et al. 1993) were divided by the sum of maximum counts made at sites
surveyed by the International Shorebird Surveys (ISS). There were 483 Atlantic coast locations
visited (13,987 surveys) during fall migration and 259 visited during spring (5,795 surveys); 19
of the locations visited during fall were on Delaware Bay. Maximum counts from these
Delaware Bay sites were summed to provide an overall index for the bay. Because of
duplication with aerial surveys, ISS counts made during spring at sites on Delaware Bay were
excluded from evaluation. Delaware Bay provides important habitat to some migrant shorebirds
during fall migration, but is particularly important in spring (Table 6.1). Aggregations in
Delaware Bay were greater during spring than fall across all species, and were dramatically so
for all species except dowitchers (Table 6.1). The difference between proportional use in spring
and fall might be attributable to the fact that there were data from aerial surveys of Delaware Bay
during the spring but not during the fall. However, locations covered by the ISS in the fall
included all of the well-known shorebird sites on Delaware Bay. The differing methodology
does not seem to explain the large seasonal differences. Even if the method did confound
interpretation of results, it could not explain the seasonal shifts of relative occurrence between
species. For example, turnstones, knots and sanderlings were virtually absent from Delaware
Bay during fall, whereas semipalmated sandpipers, dowitchers and dunlin were conspicuously
present during both seasons. Clearly, Delaware Bay is critical spring stopover for many
shorebirds, and >50% of the flyway populations of ruddy turnstones, red knots, and
semipalmated sandpipers may use Delaware Bay beaches. Reliable estimates of turnover rates
could show an increased importance of Delaware Bay to these species. The comparisons
described above are probably the most reliable, minimal estimate of use of Delaware Bay by
migrant shorebird populations.
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 34
7.0. HABITAT USE BY SHOREBIRDS AND HORSESHOE CRABS
7.1. Shorebird Use of Marine and Non-marine Habitats
Harrington (2002*) used counts made between 1974 and 2000 by ISS cooperators to compare
use of marine and non-marine sites along the Atlantic coast (see section 5.2.3). Survey sites
were classified as primarily either marine or non-marine habitats and the average number of
birds recorded during surveys was computed for northward and southward migration. Ruddy
turnstones, red knots, sanderlings, dunlins, and short-billed dowitchers were all more abundant in
marine than non-marine habitats during northward and southward migration (Table 7.1).
Semipalmated sandpipers were more abundant in marine habitats in fall, but were equally
abundant between marine and non-marine habitats during spring. Long-billed dowitchers and
least sandpipers were equally abundant in both habitat types during both seasons (Table 7.1).
7.2. Red Knot Habitat Use and Movements in Delaware Bay
Meyer et al. (ND*) radio-tagged red knots taken from cannon-net catches on New Jersey beaches
15–19 May 1997 (5 birds) and on New Jersey (30 birds) and Delaware (20 birds) beaches 2–21
May 1998. Telemetric searches for radio-tagged birds were conducted from the ground 16–30
May in 1997 and from the ground and air 3 May–9 June 1998. Pre-determined ground locations
were surveyed in New Jersey and Delaware; transmitter range averaged 1.6 km on the ground
and 8 km in the air. Habitat, home range (kernels), and behavior was measured for each bird.
Mean minimum duration of stay (calculated as the difference between initial capture day and day
of last detection) in 1998 was 17 " 8 days (" SD, n = 47 birds) and ranged from 1 to 35 days.
Birds may have been present in the bay for an unknown number of days before capture. Radio-tagged
birds preferentially used the lower, rather than upper, Delaware Bay region (P2 = 317, df
= 4, P = 0.001). The greatest number of radio-tagged birds were located in New Jersey on 16
May, whereas the greatest number of birds was detected in Delaware in 23 May. Radio-tagged
birds were not distributed evenly among all beaches and marshes and were concentrated on a few
beaches throughout the bay (P2 $179, df $ 22, P = 0.001) and also within each state. Radio-tagged
red knots commonly crossed Delaware Bay; in 1998, 60% of radio-tagged knots made $1
bay crossing. The number of bay crossings an individual knot would make was independent of
initial weight, banding date, minimum duration stay, capture location, number of re-sightings, or
any interactions. Frequency of bay crossings increased at the end of the May. Knots moved on
average 27.4 km (SD = 16.8). Significantly more knots were located on beaches than in marshes
(P2 = 4,797, df = 1, P < 0.0001), and most knots were found on sandy beaches (79% of beach
detections).
7.3. Shorebird Habitat Use on Cape May Peninsula, New Jersey
Burger et al. (1997) chose representative (non-random) marshes and beaches along the Atlantic
and Delaware Bay coasts of New Jersey to determine shorebird numerical and behavioral use;
the magnitude of shorebird use was a consideration in selection. Scan samples of shorebirds (20
Delaware Bay Shorebird Assessment Report and Peer Review - June 2003 35
minutes) were made from 22 May to 4 June, 1991–92, at 2 Atlantic Ocean marshes and 1 (each)
marsh, mudflat, and beach along Delaware Bay. Surveys occurred during different tidal stages.
Scans were considered independent (with no to few replicates in space) and multiple regression
procedures (arc-sine transformations) were used to construct habitat models. Univariate tests
(Kruskal-Wallis) were used to determine significance of individual variables. Burger et al.
(1997) found that location, date, tide, time, species, and location-tide interaction were significant
in explaining differences in the proportion of shorebirds that were alert, feeding, or resting.
Shorebirds fed mainly on falling, low, and rising tides. More birds fed in marshes and on
mudflats than on beaches, and a higher proportion of birds fed during the middle of migration
than at the beginning or end. The mudflats had the highest number of birds and the greatest
proportion of feeding shorebirds. Location was the most important factor that explained
differences in feeding within species. Ruddy turnstones and red knots were found in greater than
expected proportions in Atlantic marshes. The greatest number of semipalmated sandpipers, red
knots, ruddy turnstones, and sanderlings foraged on a rising tide. They conclude that migrant
shorebirds use a mosiac of habitats on the Cape May Peninsula, and that habitat switching likely
occurs because of the need to feed.
7.4. Shorebird Beach Use in Delaware
Carter (2002*) used information opportunistically collected during field work to generate a
preliminary map of beaches that supported the greatest numbers of red knots and ruddy
turnstones in Delaware. Beach use was grouped into 4 categories: 1) extremely high use—large
flocks at all weather conditions, 2) high use—large flocks in mild weather conditions, 3)
moderate use—occasional large flocks intermittently, and 4) occasional use—some individuals,
not regular. These criteria were applied to a 77-km length of shoreline between Woodland
Beach and Cape Henlopen. Extremely high or high use beaches constituted 14% of the shoreline
for red knots and 19% of the shoreline for ruddy turnstones (Table 7.2). Knots may distribute
themselves among Delaware beaches in response (negatively) to on-shore wind speed. Carter
and Scarborough (2002*) found that when average winds were >6.4 km/hour (over a 24-hour
period measured at 5-second intervals), resultant wave heights deterred crab spawning and
shorebird feeding on Delaware beaches. Information from radio-tagged knots is consistent with
shorebird beach use data from Delaware and suggest that large aggregations of shorebirds are
concentrated on a relatively small amount of Delaware Bay shoreline. Delineation and
maintenance of high quality beach habitats for spaw