The Condor 103:715–724
q The Cooper Ornithological Society 2001
FALL MIGRATION ROUTES, TIMING, AND WINTERING SITES OF
NORTH AMERICAN OSPREYS AS DETERMINED BY
SATELLITE TELEMETRY
MARK S. MARTELL1,4, CHARLES J. HENNY2, PETER E. NYE3
AND
MATTHEW J. SOLENSKY1
1
The Raptor Center at the University of Minnesota, 1920 Fitch Avenue, St. Paul, MN 55108
2USGS-Forest & Rangeland Ecosystem Science Center, 3200 SW Jefferson Way, Corvallis, OR 97331
3New York Department of Environmental Conservation Wildlife Research Center, Delmar, NY 12054
Abstract. Satellite telemetry was used to determine fall migratory movements of Ospreys
(Pandion haliaetus) breeding in the United States. Study areas were established along the lower
Columbia River between Oregon and Washington; in north-central Minnesota; on Shelter Island,
New York; and in southern New Jersey. Seventy-four adults (25 males, 49 females) were tracked
from 1995 through 1999. Migration routes differed among populations but not by sex. Western
Ospreys migrated through California and to a lesser degree other western states and wintered in
Mexico (88%), El Salvador (6%), and Honduras (6%) (25.98N to 13.08N and 108.38W to
87.38W). Minnesota Ospreys migrated along three routes: (1) through the Central U.S. and then
along the east coast of Mexico, (2) along the Mississippi River Valley, then across the Gulf of
Mexico, or (3) through the southeastern U.S., then across the Caribbean. East Coast birds migrated along the eastern seaboard of the U.S., through Florida, and across the Caribbean. Midwestern
birds wintered from Mexico south to Bolivia (22.358N to 13.648S, and 91.758W to 61.768W),
while East Coast birds wintered from Florida to as far south as Brazil (27.488N to 18.58S and
80.48W to 57.298W). Dates of departure from breeding areas differed significantly between sexes
and geographic regions, with females leaving earlier than males. Western birds traveled a shorter
distance than either midwestern or eastern Ospreys. Females traveled farther than males from
the same population, which resulted in females typically wintering south of males.
Key words: migration, Osprey, Pandion haliaetus, satellite telemetry, wintering sites.
Rutas de Migración Otoñales, Coordinación y Sitios de Invernada de Pandion haliaetus Determinados
por Telemetrı́a Satelital
Resumen. Se utilizó telemetrı́a satelital para determinar los movimientos de migración de
otoño de individuos de Pandion haliaetus que nidifican en los Estados Unidos. Las áreas de
estudio se establecieron a lo largo del Rı́o Columbia entre Oregon y Washington; en el centronorte de Minnesota; en la Isla Shelter, Nueva York; y en el sur de Nueva Jersey. Setenta y cuatro
adultos (25 machos, 49 hembras) fueron seguidos mediante telemetrı́a desde 1995 hasta 1999.
Las rutas de migración se diferenciaron entre poblaciones pero no entre sexos. Los individuos
de P. haliaetus del oeste, migraron a través de California y en menor grado a través de otros
estados del oeste e invernaron en México (88%), El Salvador (6%) y Honduras (6%) (25.98N a
13.08N y 108.38O a 87.38O). Las aves de Minnesota migraron a lo largo de tres rutas: (1) a
través del los E.E.U.U. centrales y luego a lo largo de la costa este de México, (2) a lo largo
del valle del Rı́o Mississippi y luego a través del Golfo de México, o (3) a través del sur de los
E.E.U.U. y luego a través del Caribe. Las aves de la costa este, migraron a lo largo de la costa
este de los E.E.U.U., por Florida y a través del Caribe. Las aves del medio-oeste, invernaron
desde México hacia el sur hasta Bolivia (22.358N a 13.648S, y 91.758O a 61.768O), mientras
que las aves de la costa este invernaron desde Florida hasta tan al sur como Brasil (27.488N a
18.58S y 80.48O a 57.298O). Las fechas de partida desde las áreas de nidificación difirieron
significativamente entre sexos y regiones geográficas, partiendo las hembras antes que los machos. Las aves del oeste viajaron distancias más cortas que las aves del medio-oeste y del este.
Considerando una misma población, las hembras viajaron más lejos que los machos, lo que
resultó en que las hembras invernaron tı́picamente más al sur que los machos.
INTRODUCTION
Migration routes and wintering localities of
North American Ospreys (Pandion haliaetus)
Manuscript received 3 October 2000; accepted 25
July 2001.
4 E-mail: marte006@umn.edu
have been previously described by various authors using information from band recoveries
(Henny and Van Velzen 1972, Poole and Agler
1987), migration count sites (Zalles and Bildstein 2000), and wintering areas (Saggese et al.
1996). These studies show a widespread distribution of Osprey wintering sites and migration
[715]
716
MARK S. MARTELL
ET AL.
routes. Winter band recoveries from Ospreys
that nest along the East Coast of North America
have come mostly from South America and to a
lesser degree the West Indies (Poole and Agler
1987). Band recoveries of birds that breed in the
midwestern U.S. have been from South and
Central America (Niemuth 1991). Band recoveries from western U.S. breeding birds indicate
they winter in Mexico and Central America
(Melquist et al. 1978, Johnson and Melquist
1991).
Avian migration studies have been enhanced
by the recent availability of small, satellitetracked radios known as platform transmitter terminals (PTTs; Seegar et al. 1996) that have allowed the plotting of migration routes and wintering localities of a variety of birds, including
Ospreys (Kjellen et al. 1997, Martell et al.
1998).
Here we report on fall migratory movements
of Ospreys breeding in North America as determined by satellite telemetry. Our objectives
were (1) to describe routes, distance, and timing
of fall migration; (2) to identify wintering localities and arrival dates; and (3) to determine if
migration varies by sex or geographic origin.
METHODS
STUDY AREAS
Ospreys were monitored from study areas chosen to represent the eastern, midwestern, and
western United States breeding populations
(Henny 1983). All study areas were between 468
and 388 north latitude. Birds in the eastern U.S.
were trapped on The Nature Conservancy’s
Mashomack Preserve and private land on Shelter Island, Suffolk County, New York (on the
eastern end of Long Island). Ospreys have been
banded and monitored in this area since the early
1900s (Puleston 1972). Additional eastern U.S.
birds came from The Wetlands Institute Preserve, Cape May County; the Maurice River,
Cumberland County; and Salem County, all in
New Jersey.
In the Midwest, Ospreys from a reintroduced
population in the Minneapolis-St. Paul, Minnesota, area (Martell 1995) as well as from the
forested north-central region around Brainerd,
Crow Wing County, Minnesota, were used.
Western U.S. Ospreys came from the lower Columbia River (river miles 76–286) bordering
Oregon and Washington, and the Willamette
Valley in western Oregon. Here the migration
study was integrated into ongoing population
and contaminant research (Henny and Kaiser
1996).
CAPTURE
Adult birds were trapped on their nesting territories using either a noose carpet placed over the
nest, or a modified dho-ghaza, using mist nets
with a Great Horned Owl (Bubo virginianus) as
a lure (Bloom 1987). We determined the sex of
each bird through a combination of plumage,
size, and behavior at the nest (Poole 1989). All
birds were banded with a U.S. Fish and Wildlife
Service band on one leg and a black alphanumeric coded band on the other leg. Wherever
possible, subsequent nest success for each
trapped individual was monitored along with
number of chicks fledged.
TELEMETRY
Two types of PTTs weighing from 30–35 grams
(Microwave Telemetry Inc., Columbia, Maryland) were deployed. One type, used in 78 deployments, was battery powered and pre-programmed to turn on for 8-hr periods during
varying cycles through the year. This conserved
battery power, extending the effective life of the
units. Most typical was a cycle that turned the
PTT on for 8 hr and then off for 48–72 hr during
the anticipated fall and spring migrations and an
8/240-hr cycle during summer and winter. Cycles were adjusted over the course of the study
to maximize data collection during migratory
periods. Solar-charged units were used on 12
Ospreys in 1999 and were set to transmit for 10
hr followed by 31 hr off. All units were placed
on the backs of birds using Teflon ribbon (Bally
Ribbon, Bally, Pennsylvania) in a standard backpack configuration (Dunstan 1972, Kenward
1987). Capture, handling, and attachment protocols were reviewed and approved by the Institutional Animal Use and Care Committee at
the University of Minnesota.
Birds were located using NOAA satellites
with onboard tracking equipment operated by
Service Argos Inc. (Landover, Maryland). Locations in latitude and longitude decimal degrees, date, time, location error, and other data
were received from ARGOS within 24 hr of satellite contact with a bird. Location error was reported by Service Argos as one of six ‘‘location
classes’’ (LC): LC3 5 ,150 m, LC2 5 150–
OSPREY MIGRATION DETERMINED BY SATELLITE TELEMETRY
TABLE 1.
717
North American Ospreys trapped and fitted with satellite tracked telemetry from 1995–1999.
Individuals tagged in
Male
Female
Total
2 Years
3 Years
11
10
4
25
16
15
18
49
27
25
22
74
2
3
3
3
1
0
East
Midwest
West
Total
350 m, LC1 5 350–1000 m, LC0 . 1000 m,
LCA and B 5 no location accuracy (Service
ARGOS 1996).
DATA ANALYSIS
Location data were analyzed, entered into databases, and then plotted using ArcView PC-GIS
system (Environmental Systems Research Institute Inc., Redlands, California). The most accurate reading during any transmission period,
as determined by the LC, was considered ‘‘best
of day’’ and used in data calculations unless otherwise noted. Distances were calculated between
best-of-day locations for each segment over the
entire route of migration; the total migration distance was the sum of these segments. The total
time of migration was calculated in days using
the best estimate of departure and arrival dates.
Departure dates were calculated using the median date between the last signal from the breeding area and the first signal on migration. When
there was more than a 10-day gap between these
signals we did not calculate departure date. Similarly, date of arrival on the wintering area was
calculated as the median date between the last
signal on migration, and the first signal on the
wintering area unless the difference was greater
than 10 days. Mean distance traveled per day
was calculated using total migration distance divided by the number of days spent on migration.
Wintering locations for each bird were determined by averaging all locations with a LC of
1, 2, or 3 from the ‘‘wintering’’ period.
STATISTICAL ANALYSES
One-way ANOVA was used to test for group
differences, with LSD test for significant differences between means. Significance was set at P
, 0.05. Statistical tests were done using STATISTIX software (Analytical Software, Tallahassee, FL). The results presented are means 6 SD
unless otherwise noted.
Total
deployments
35
30
25
90
RESULTS
THE MARKED POPULATION
Seventy-four adult Ospreys (25 males, 49 females) were trapped and tagged from 1995
through 1999 (Table 1). All were trapped at their
nests with the exception of a 1999 New York
bird whose nesting status was unknown to us.
Eight individuals were tracked in two different
years, and four were tracked in three different
years resulting in a total of 90 PTT deployments
(two in 1995, 16 in 1996, 29 in 1997, 21 in
1998, and 22 in 1999) (Table 1). More females
than males were tagged, by design in the West
and as a result of trapping bias in the East and
Midwest. Both members of 15 nesting pairs
were tagged, one in the West, four in the East,
and 10 in the Midwest, including one Minnesota
pair that was tagged in 2 successive years (1997
and 1998).
Of 8211 locations from the breeding, fall, and
winter seasons, 33% were classified as LC 5 1,
2, or 3. Due to limited battery life, radio failure,
and Osprey mortality, the amount of data collected varied among birds. 54 PTTs (60%) transmitted fall and winter data, 16 PTTs (18%) provided only fall data, 16 PTTs (18%) failed before fall migration, while 4 PTTs (4%) transmitted into a second fall after attachment.
FALL MIGRATION
Migration routes. Fall migration routes differed
substantially among the nesting populations
studied. Midwest Ospreys showed some overlap
with East Coast Ospreys south of 308N and with
West Coast birds south of the U.S.–Mexico border. East and West Coast birds showed no overlap of migratory routes (Fig. 1). Individual Ospreys followed the same routes south in different
years in all cases (n 5 12).
East Coast Ospreys (n 5 20) took similar
routes south with only minor deviations in the
northern hemisphere (Fig. 1). Ospreys nesting
718
MARK S. MARTELL
ET AL.
FIGURE 1. Fall migration routes of North American nesting Ospreys as determined by satellite telemetry.
Dashed lines indicate movements of only one bird.
on eastern Long Island began their migration by
flying west along the island before moving
south. Most crossed the Delaware and Chesapeake Bays, staying east of the Appalachian
Mountains while continuing in a broad front
through the eastern half of Virginia and the Carolinas. Their route narrowed over the Georgia
coast into Florida. In Florida, birds moved both
along the Atlantic coast and across the peninsula
to the Gulf Coast near Sanibel Island.
Migrating East Coast Ospreys crossed the
Florida Keys to the north coast of Cuba, 100 to
250 km west of Havana. They moved southeast
through Cuba, crossing over to Hispaniola, then
flew approximately 600 km across the Caribbean
to South America making landfall between Santa Marta, Colombia, and Lake Maracaibo, Venezuela. The main migration front then moved to
the southeast through Venezuela and into Brazil,
with final destinations variable.
Midwest-nesting Ospreys (n 5 27) moved
along one of three routes (Fig. 1). The first route
(Central America) took them south through central Iowa and Missouri, then southwest through
Oklahoma and Texas to the Gulf Coast of Mexico. Migration continued along the coast over
Veracruz, Mexico, across the state of Chiapas,
Mexico, down the west coast of Honduras, El
Salvador, and Nicaragua, crossing to the Caribbean coast of Costa Rica and Panama and into
Colombia. The second route (Gulf of Mexico)
followed the Mississippi River Valley south,
across the Gulf of Mexico to the Yucatan Peninsula, and then south through Central America
and into Colombia. The third route (Caribbean
Islands) went southeast through Illinois, across
Kentucky, Tennessee, Alabama, and Georgia,
and into Florida. These birds continued south
along the Gulf Coast to the Florida Keys, then
followed the route described earlier for East
Coast birds across the Caribbean Islands (Fig.
1). One Midwest male varied from the norm in
1997 by moving from Cuba to Jamaica before
continuing to South America.
Midwestern males and females used all three
routes; the land route through Central America
was used by 59% of all Midwest birds (9 males,
7 females), followed by the Caribbean route (2
OSPREY MIGRATION DETERMINED BY SATELLITE TELEMETRY
TABLE 2.
Fall departure and winter arrival dates of satellite tracked Ospreys by sex and region.
Median
departure
date
East Coast
Female
Male
Midwest
Female
Male
West Coast
Female
Male
719
Range of
departure dates
n
Median arrival
date
Range of
arrival dates
n
19 Aug
7 Sep
12 July–9 Sep
27 Aug–3 Oct
22
8
11 Sep
8 Oct
30 Aug–3 Oct
3 Sep–18 Nov
15
7
31 Aug
20 Sep
1 Aug–12 Sep
27 Aug–1 Oct
12
13
2 Oct
10 Oct
7 Sep–11 Oct
21 Sep–5 Nov
8
11
9 Sep
1 Sep
28 Aug–24 Sep
—
14
1
19 Sep
22 Sep
7 Sep–14 Oct
—
14
1
males, 5 females; 26% of Midwest birds), and
the Gulf of Mexico route (3 males, 1 female;
15%). In only two of seven Midwest pairs did
the male and female follow the same routes
south, and these were on different schedules.
West Coast Ospreys (n 5 15) moved south
along a broad front through the western United
States (Fig. 1). Eleven Ospreys nesting on the
west side of the Cascade mountain range traveled directly south along the foothills of the Cascade and Sierra Nevada mountain ranges, then
through central California to the west coast of
Mexico. Two birds nesting on the west side of
the Cascades and three birds nesting on the east
side of the range headed in a more dispersed
southeasterly direction through Nevada, Utah,
and Arizona, with one bird moving as far east
as eastern Utah and New Mexico. These routes
took them to both the Pacific and Gulf coasts of
Mexico. Most birds wintered in coastal Mexico,
while some continued into Central America as
far south as El Salvador.
Departure dates. Dates of departure from
breeding areas occurred over a 10-week period,
ranging from 12 July to 1 October (Table 2).
Departure dates differed significantly by geographic origin and sex. Significant differences
were found among geographic areas when both
males and females were included in the analysis
(F2, 67 5 5.8, P , 0.001), and when only females
were analyzed (F2, 45 5 5.1, P 5 0.01). The median departure date of 19 August for East Coast
females (n 5 22) was 12 and 21 days earlier
than the median departure dates for Midwest (n
5 12) and West Coast (n 5 14) females, respectively (Table 2). Differences in departure
dates between the East and West Coasts were
significant (t 5 2.0, P 5 0.05). No significant
differences were found between the Midwest
birds and either the East or West Coast birds.
Median departure dates for females were significantly earlier than for males in the East and
Midwest (F1, 53 5 16.2, P , 0.001; the sample
size of males in the West was too small for analysis). Within mated pairs, females departed before their mates in all cases (n 5 11). The mean
difference in departure dates within mated pairs
was 22 days (range 7 to 39 days). Departure
dates of individual Ospreys in different years
varied by 1–16 days (n 5 14). Nest failure did
not affect departure date within geographic or
sex groups.
Distance. Total distance traveled by individual birds (n 5 49) varied from 1890 km for an
East Coast male that in 1999 traveled from Shelter Island, New York, to Ft. Pierce, Florida, in
5 days, to 8720 km for an East Coast female
that traveled from Shelter Island to the Pantanal
area of Brazil in 38 days. Mean distance traveled
by Ospreys from different study areas was significantly different (F2, 48 5 6.1, P , 0.001) with
West Coast birds (3824 6 862 km, n 5 16) travelling less distance than either East Coast (5134
6 1964 km, n 5 20) or Midwest (5872 6 1780
km, n 5 16) Ospreys. Females showed an even
greater difference (F2, 29 5 15.3, P , 0.001) with
West Coast females traveling less distance (3784
6 705 km, n 5 13, range 2713–5008 km) than
females from the East Coast (6015 6 1297 km,
n 5 11, range 3762–8720 km) or Midwest (6623
6 1855 km, n 5 8, range 3031–8405 km). We
found no significant difference between distances traveled by Midwest vs. East Coast Ospreys.
Males from the East Coast (4057 6 2165 km, n
5 9) and Midwest (5122 6 1435 km, n 5 8)
traveled significantly less distance than females
720
MARK S. MARTELL
FIGURE 2.
ET AL.
Wintering locations of North American Ospreys as determined by satellite telemetry.
(F1, 34 5 9.0, P 5 0.05). Too few West Coast
males (4082 km, n 5 2) were available for comparisons.
The number of days spent on fall migration
ranged from 5 days for an East Coast male that
wintered in Florida, to 68 days for an East Coast
male that spent 20 days in Cuba before wintering in central Brazil. A significant difference
was found in the number of days traveled among
Ospreys from different study areas (F2, 48 5 8.4,
P , 0.001). West Coast birds traveled fewer
days (13 6 4, n 5 13) than East Coast (31 6
16, n 5 22) or Midwest Ospreys (26 6 11, n 5
17; t 5 2.0, P 5 0.05). We found no significant
difference in number of days traveled between
males and females in the East (31.0 days, n 5
6 for males, 31.3 days, n 5 15 for females) and
Midwest (23 days, n 5 10 for males, 28.4 days,
n 5 7 for females). No significant difference was
found for between-year differences in the same
individuals.
Mean distance traveled per day by individuals
varied from 111 km day21 for an East Coast
male that spent 10 days in Cuba followed by 14
days in the Dominican Republic (while sitting
out Hurricane Georges), to 380 km day21 for a
West Coast female that wintered near Tampico,
Mexico. A significant difference was found in
distance traveled per day between Ospreys from
different study areas (F2, 49 5 6.3, P , 0.001).
West Coast Ospreys (296 6 55 km day21, n 5
14) traveled farther per day than either Midwest
(230 6 61 km day21, n 5 17) or East Coast
Ospreys (214 6 81 km day21, n 5 21; t 5 2.0,
P , 0.05). No significant difference in the distance traveled per day was found between males
and females for the East and Midwest.
WINTERING LOCATIONS
Ospreys from this study wintered in 11 countries
from the United States (Florida) and Mexico (Sinaloa) in the north, to Brazil, Bolivia, and Peru
in the south (Fig. 2, Table 3). The most southerly
wintering bird was an East Coast female that in
1996 wintered in the Pantanal area shared by
Bolivia and Brazil, at 18.498S, 57.428W. Midwest birds (n 5 16) wintered from 22.48N to
13.78S, and 91.78W to 61.88W in eight countries
(Table 3) ranging from Mexico to Brazil. Sixtynine percent of the Midwest birds wintered in
South America, the remainder in Central America and Mexico. East Coast birds (n 5 19) wintered from 27.58N to 18.58S and 80.48W to
57.38W. East Coast Ospreys wintered mainly in
northern South America (89%), with two wintering in Cuba and one in Florida. West Coast
birds (n 5 15) showed the tightest distribution,
wintering from 25.98N to 13.08N and 108.38W
to 87.38W. Fourteen West Coast Ospreys
(87.5%) wintered in Mexico, but none wintered
OSPREY MIGRATION DETERMINED BY SATELLITE TELEMETRY
721
TABLE 3. Wintering localities by country of Ospreys from three North American breeding populations (East
Coast, Midwest, and West Coast). Individuals at sites in more than one year counted only once.
Wintering area
United States
Mexico
Cuba
El Salvador
Honduras
Nicaragua
Colombia
Venezuela
Brazil
Peru
Bolivia
Total
Range of
latitude
27.48N
25.9–15.18N
21.9–22.38N
13.38N
13.08N
13.0–11.78N
10.7–8.88N
10.9–8.38N
2.5–18.48S
3.8–11.78S
11.1–13.68S
Range of
longitude (8W)
80.4
108.3–91.3
83.1–79.3
88.1
87.3
87.3–85.8
75.0–74.5
71.4–62.7
73.8–49.6
74.3–70.2
66.6–61.7
in Baja California or along the northern twothirds of the Sea of Cortez where a substantial
resident breeding population lives (Henny and
Anderson 1979). One bird wintered in El Salvador and one in Honduras. There was overlap
in winter ranges of Midwest and East Coast, and
Midwest and West Coast Ospreys, but not of
West Coast and East Coast Ospreys (Fig. 2).
East Coast and Midwest females tended to
winter farther south than males from the same
regions. This tendency was most pronounced in
Midwest birds, where 63% of females (n 5 8)
wintered south of all males (n 5 8). The area
between latitudes 88N and 98N seemed to define
this split, with 66% of the East Coast and Midwest males (n 5 15) wintering north of 98N and
68% of the females (n 5 19) wintering south of
88N. We did not have a sample of West Coast
males large enough for comparison. In 7 of 9
pairs, the female wintered south of the male. All
individuals (n 5 9) tracked in more than one
year returned to the same wintering area.
Arrival dates (Table 2) on wintering areas
ranged from 30 August to 18 November. Arrival
dates for birds from different geographic areas
varied significantly (F2, 53 5 9.6, P , 0.001) with
Midwest Ospreys arriving later than either East
or West Coast birds. East Coast and Midwest
females arrived on their wintering areas significantly earlier than males (F1, 38 5 17.6, P ,
0.001). From the East Coast, 60% of females
and only 25% of males arrived by the median
arrival date for that population. In the Midwest,
75% of females and 45% of males arrived by
the median arrival date for that population.
Within pairs, females tended to arrive on their
East Coast
Midwest
West Coast
Total
1
—
2
—
—
—
2
4
8
1
1
19
—
2
1
—
—
2
3
1
2
2
3
16
—
13
—
1
1
—
—
—
—
—
—
15
1
15
3
1
1
2
5
5
10
3
4
50
wintering areas before their mates (recall that
females always departed before males). In six
pairs, the female arrived an average of 37 days
(range 20–70 days) ahead of its mate, while in
three pairs the male arrived an average of 9 days
(range 2–20 days) before the female.
Individuals arrived on their wintering sites an
average of 10 days apart in different years (n 5
7, range 0–20 days). Mean difference in arrival
dates between years was 12.6 days (n 5 3) for
males, and 6 days (n 5 4) for females.
DISCUSSION
Osprey migration from North America has broad
spatial and temporal variation, as shown in this
study as well as from band recoveries (Henny
and Van Velzen 1972, Poole and Agler 1987).
This variability, or differential migration (Alerstam and Hedenstrom 1998), was found among
the geographic groups we studied as well as between the sexes.
It is notable that the breeding pairs of Ospreys
in this study did not migrate or winter together.
Females from the East Coast and the Midwest,
as a group and within each pair, departed on migration before the males, often leaving the
breeding territory before their young were independent. In Sweden, satellite-tagged females
at two nests left around the end of July while
males were still feeding the young (Kjellen et
al. 1997). Females’ earlier departure from the
breeding areas may be due to the greater distance they travel on migration (females tend to
winter south of males), or, possibly, to reduce
competition for food with their young on the
nesting territory.
722
MARK S. MARTELL
ET AL.
Female Ospreys also tended to migrate farther
than males from the same geographic group.
However, this difference was not seen between
groups: West Coast females travelled less distance and wintered north of many Midwest and
East Coast males. Wintering separation of the
sexes is found in other avian species (Cristol et
al. 1999), as well as in Ospreys in Africa, where
separation occurs along an east-west gradient
(Kjellen et al. 2001). Explanations for separation
include body size differences, dominance hierarchies, and facilitating the males’ quick return
in spring for territorial defense (Cristol et al.
1999).
Although we did not find any difference in
departure dates between birds whose nests had
failed and those that succeeded, it is interesting
to note that the earliest departure dates we recorded (mid-July) were from an East Coast female whose nest failed in both 1997 and 1998.
Other birds whose nests failed either stayed in
the general vicinity of their nesting territory, or
moved away from the area but did not begin
migration. Birds from Shelter Island, New York,
whose nests failed were found on eastern Long
Island, as well as in Connecticut and as far north
as Massachusetts. These areas were used for up
to a month before the birds left on migration. A
similar situation was described by Kjellen et al.
(1997) for a female Osprey from Sweden.
In addition to the effects that geographic origin and gender have on migration timing, routes,
and wintering areas, individuals within groups
also showed variations based on factors not obvious to us. Ospreys nesting close to each other
(on Shelter Island for example) wintered far
apart, as noted also for Swedish birds (Kjellen
et al. 1997). Individuals showed strong fidelity
to their wintering areas, similar to their fidelity
to breeding territories (Poole 1989). This breeding-site fidelity allows individuals to obtain all
of the advantages conferred by familiarity with
foraging and roosting areas; we suggest that
similar advantages come from fidelity to wintering areas and migration routes.
We documented Ospreys regularly making
long-distance water crossings over the Caribbean Sea and Gulf of Mexico, and shorter ones
across the Delaware and Chesapeake Bays.
Long water crossings are typically avoided by
most Falconiformes (Kerlinger 1985), although
European Ospreys also make long water crossings over the Mediterranean (Kjellen et al. 1997,
Hake et al. 2001). It is of interest that Ospreys
crossing the Caribbean used both Cuba and Hispaniola as ‘‘stepping stones,’’ with individual
birds spending up to 23 days in Cuba (Rodriquez
et al. 2001). However, we did not see birds continue island hopping through Puerto Rico and
smaller islands to the southeast which would
have reduced the need for a longer water crossing between Hispaniola and Venezuela.
Ospreys in this study were found across most
of the previously reported wintering range of
North American Ospreys (Poole and Agler
1987, Poole 1989). Our East Coast Ospreys wintered in the West Indies (11%) and South America (89%), similar to band recovery data showing 15.5% of East Coast birds (n 5 71) wintering in the West Indies (Henny and Van Velzen
1972). In this study however, a higher percentage of East Coast (55%) and Midwest Ospreys
(43%) wintered south of the equator than was
found in band recovery analysis (26% of East
Coast and 22% of Midwest band returns south
of the equator, Poole and Agler 1987). We suspect that band recovery analysis incorrectly classified some birds that died on migration in northern South America as wintering, and that fewer
bands are reported from more southerly areas.
Ospreys in this study made either a relatively
short, quick migration or a longer, slower journey. This ‘‘sprint’’ vs. ‘‘marathon’’ strategy is
reflected in the variability seen in departure
dates, distance traveled, and speed of travel. Ospreys from the West employed the sprint strategy while Midwest and Eastern birds typically
made marathon journeys. Western Ospreys traveled, on average, significantly less distance than
either Midwest or East Coast Ospreys; differences that were even greater when comparing
only females. West Coast Ospreys also spent
significantly fewer days on migration than either
Midwest or East Coast Ospreys, and traveled
significantly farther per day (296 km) than either
Midwest or East Coast Ospreys. The mean distances traveled per day in this study (111–380
km day21) are similar to findings for Swedish
Ospreys (Kjellen et al. 1997, 2001) that traveled
between 159 and 413 km day21.
Some individuals from the Midwest and East
also utilized a sprint strategy rather than the
marathon strategy. Seven individuals, four from
the East (three males, one female) and three
from the Midwest (two males, one female) traveled less than 4000 km (similar to the distance
OSPREY MIGRATION DETERMINED BY SATELLITE TELEMETRY
covered by West Coast Ospreys). These birds
spent an average of 12 days on migration covering 327 km day21, data more similar to West
Coast Ospreys than the other Midwest and East
Coast birds.
Different migration strategies were also reflected in times of departure from breeding areas. While departure dates can certainly be affected by nesting phenology (Poole and Agler
1987); in this study we noted no differences in
egg laying dates among the study areas (all occurred between mid-April and mid-May). Thus,
nesting phenology does not seem to explain the
differences in departure dates among these populations. It is apparent that birds travelling greater distances departed earlier, perhaps to insure
timely arrival on the wintering area. West Coast
Ospreys in this study can apparently remain on
their breeding grounds later than East Coast
birds before making a quick, short flight to their
wintering areas.
The marathon strategy may result from the
need of long distance migrants to hunt during
migration. Alternatively, the sprint strategy may
be made possible when individuals are able to
utilize wintering grounds close to their nesting
areas. The energetic demands faced by Ospreys
on migration are not well known, but Candler
and Kennedy (1995) produced models showing
it possible for Ospreys migrating through the
western United States to complete their journeys
without feeding. Information on Ospreys foraging during migration is mostly anecdotal (Kerlinger 1985); however, fall reports of Osprey
numbers increasing at reservoirs and other sites
outside of the nesting range (Castrale et al.
1998) suggest that some regularly feed on migration. Swedish Ospreys are thought to feed at
stopover sites on migration (Kjellen et al. 2001).
Migratory behavior and wintering locations of
North American nesting Ospreys vary among
geographic regions and by sex. Thus, the strategies employed by any individual bird are influenced by its nesting origin and gender. However,
there are still differences among individuals
within any group. Competition for resources,
particularly food, on the wintering ground and
probably during migration, may ultimately shape
the strategies employed by individual birds.
ACKNOWLEDGMENTS
We would like to acknowledge Kathy Clark, New Jersey Endangered and Nongame Species Program; Mike
723
Scheibel, The Nature Conservancy’s Mashomack Preserve; and Tedor Whitman, Wetlands Institute of New
Jersey, whose assistance made this study possible.
Sandy Weisberg provided statistical advice, while Ray
Bentley, Jane Galetto, Bob Grove, Jim Kaiser, Bill
Lane, Emily Peters, and Ananda Wiegand provided
help in the field. Comments on earlier drafts by Alan
Poole, Paul Switzer, Harrison Tordoff and David Mehlman greatly improved the manuscript. Funds for this
project were provided by Canon USA, Douglas Dayton, Wallace Dayton, Dellwood Foundation, Special
Projects Foundation of the Big Game Club, New Jersey PSE&G, Minnesota Legislature (LCMR), International Foods and Fragrances, and the Larsen Foundation.
LITERATURE CITED
ALERSTAM, T., AND A. HEDENSTROM. 1998. The development of bird migration theory. Journal of Avian
Biology 29:343–369.
BLOOM, P. H. 1987. Capturing and handling raptors, p.
99–124. In B. A. Giron Pendleton, B. A. Millsap,
K. W. Cline, and D. M. Bird [EDS.], Raptor management techniques manual. National Wildlife
Federation, Washington, DC.
CANDLER, G. L., AND P. L. KENNEDY. 1995. Flight strategies of migrating Osprey: fasting vs. foraging.
Journal of Raptor Research 29:85–92.
CASTRALE, J. S., E. M. HOPKINS, AND C. E. KELLER.
1998. Atlas of breeding birds of Indiana. Indiana
Department of Natural Resources. Bloomington,
IN.
CRISTOL, D. A., M. B. BAKER, AND C. CARBONE. 1999.
Differential migration revisited, p. 33–88. In V.
Nolan and E. D. Ketterson [EDS.], Current Ornithology. Vol. 15. Plenum Publishers, New York.
DUNSTAN, T. C. 1972. Application of radio-telemetric
techniques to Osprey research, p. 47–50. In J. C.
Ogden [ED.], Transactions of the North American
Osprey Research Conference. USDI, National
Park Service. Transactions and Proceedings Series; No. 2. Washington, DC.
HAKE, M., N. KJELLEN, AND T. ALERSTAM. 2001. Satellite tracking of Swedish Ospreys Pandion haliaetus autumn migration routes and orientation.
Journal of Avian Biology 32:47–56.
HENNY, C. J. 1983. Distribution and abundance of nesting Ospreys in the United States, p. 175–186. In
D. M. Bird [ED.], Biology and management of
Bald Eagles and Ospreys. Harpell Press, Ste. Anne
de. Bellevue, Quebec, Canada.
HENNY, C. J., AND D. W. ANDERSON. 1979. Osprey distribution, abundance, and status in western North
America: III. The Baja California and Gulf of California population. Bulletin of the Southern California Academy of Sciences. 78:89–106.
HENNY, C. J., AND J. L. KAISER. 1996. Osprey population increase along the Willamette River,
Oregon, and the role of utility structures, 1976–
93, p. 97–108. In D. M. Bird, D. E. Varland, and
J. J. Negro [EDS.], Raptors in human landscapes.
Academic Press, London.
HENNY, C. J., AND W. T. VAN VELZEN. 1972. Migration
patterns and wintering localities of American Os-
724
MARK S. MARTELL
ET AL.
preys. Journal of Wildlife Management 36:1133–
1141.
JOHNSON, D. R., AND W. E. MELQUIST. 1991. Wintering
distribution and dispersal of northern Idaho and
eastern Washington Ospreys. Journal of Field Ornithology 62:517–520.
KENWARD, R. 1987. Wildlife radio tagging: equipment,
field techniques and data analysis. Biological
Techniques Series, Academic Press, London.
KERLINGER, P. 1985. Water crossing behavior of raptors
during migration. Wilson Bulletin 97:109–113.
KJELLEN, N., M. HAKE, AND T. ALERSTAM. 1997. Strategies of two Ospreys Pandion haliaetus migrating
between Sweden and tropical Africa as revealed
by satellite tracking. Journal of Avian Biology 28:
15–23.
KJELLEN, N., M. HAKE, AND T. ALERSTAM. 2001. Timing and speed of migration in male, female and
juvenile Ospreys Pandion haliaetus between Sweden and Africa as revealed by field observations,
radar and satellite tracking. Journal of Avian Biology 32:57–67.
MARTELL, M. 1995. Osprey Pandion haliaetus reintroduction in Minnesota, USA. Vogelwelt 116:205–
207.
MARTELL, M. S., M. J. KENNEDY, C. J. HENNY, AND P.
E. NYE. 1998. Highway to the tropics: using satellite telemetry and the internet to track Ospreys
in the western hemisphere, p. 163–172. In Y. Leshem, E. Lachman, and P. Berthold [EDS.], Migrating birds know no boundaries: scientific and education aspects of migrating bird conservation,
August 31–September 6, 1997, International Center for the Study of Bird Migration at Latrun and
Israel Ornithology Center, Tel Aviv, Israel.
MELQUIST, W. E., D. R. JOHNSON, AND W. D. CARRIER.
1978. Migration patterns of northern Idaho and
eastern Washington Ospreys. Bird-Banding 49:
234–236.
NIEMUTH, N. D. 1991. Recoveries of Osprey banded in
Wisconsin. Passenger Pigeon 53:109–114.
POOLE, A. F. 1989. Ospreys: a natural and unnatural
history. Cambridge University Press, Cambridge,
UK.
POOLE, A. F., AND B. AGLER. 1987. Recoveries of Ospreys banded in the United States, 1914–84. Journal of Wildlife Management 51:148–155.
PULESTON, D. 1972. Osprey population studies on Gardiner’s Island, p. 95–99. In J. C. Ogden [ED.],
Transactions of the North American Osprey Research Conference. U.S. Park Service Transactions and Proceedings, Series No. 2. Washington,
DC.
RODRIQUEZ, F., M. MARTELL, P. NYE, AND K. L. BILDSTEIN. 2001. Osprey migration through Cuba, p.
107–117. In K. L. Bildstein, and D. Klem Jr.
[EDS.], Hawkwatching in the Americas. Hawk Migration Association of North America, North
Wales, PA.
SAGGESE, M. D., E. R. DE LUCCA, S. F. KRAPOVICKAS,
AND E. H. HAENE. 1996. Resencia del aguila pescadora (Pandion haliaetus) en Argentina y Uruguay. El Hornero 14:44–49.
SEEGAR, W. S., P. N. CUTCHIS, M. R. FULLER, J. H.
SUTER, V. BHATNAGAR, AND J. G. WALL. 1996. Fifteen years of satellite tracking, development and
application to wildlife research and conservation.
Johns Hopkins APL Technical Digest 17:401–
411.
SERVICE ARGOS. 1996. User’s manual. Service ARGOS, Landover, MD.
ZALLES, J. I., AND K. L. BILDSTEIN. 2000. Raptor watch:
a global directory of raptor migration sites.
BirdLife International, Cambridge, UK, and Hawk
Mountain Sanctuary Association, Kempton, PA.