Pandion haliaetus


Demography and Populations

Welcome to the Birds of North America Online!

You are currently viewing one of the free species accounts available in our complimentary tour of BNA. In this courtesy review, you can access all the life history articles and the multimedia galleries associated with this species.

For complete access to all species accounts, a subscription is required. Subscriptions are available for as little as $5 for 30 days of complete access! If you would like to subscribe to BNA, please visit the Cornell Lab of Ornithology E-Store or call us at 877-873-2626 (M-F, 8:00-4:00 ET).

Measures of Breeding Activity

Age at first breeding; intervals between breeding

Age at first breeding summarized in Poole 1989a. Based on a small sample (20 individuals identified as nesting and producing eggs for the first time) in an expanding population with abundant artificial nesting sites in southern New York and New England, Spitzer 1980 found 50% 3-yr-olds, 30% 4-yr-olds, and 20% 5-yr-olds. [Note “3-yr old” birds arrive from the overwintering grounds on their second trip north (see migration) as 2-yr-olds and turn 3 during that breeding season.] Larger sample (45 additional breeders) in this region revealed a similar pattern, with mean age at first breeding 3.6 yr (Poole 1984). Similar pattern also in Michigan, 1974–1987, with 55% of 82 known-age individuals first breeding at age 3 yr, 33% at 4 yr, 10% at 5 yr, and 2% at 6 yr (Postupalsky 1989b). By contrast, along the eastern shore of Chesapeake Bay, most individuals during the 1980s did not breed until 5–7 yr old (mean age at first breeding 5.7 yr; P. Spitzer in Poole 1989a), apparently because safe nest sites were limited there so young adults delayed breeding until they found an older individual in need of a mate. Most young Ospreys in a Michigan population bred for the first time with older birds presumably filling vacancies created by mortality during migration (Postupalsky 1989b). Such delayed breeding is probably the rule for most Osprey populations, especially as density of breeders increases.

Breeds annually; e.g., in New England, generally fledge young in July, lay eggs the following April. Even in non-migratory populations (e.g., southern Florida, Baja California), only 1 brood/yr (Ogden 1975, Judge 1983).


Few studies with accurate data; at a minimum, weekly nest checks needed for accuracy, throughout the laying period. Ranges from 1–4 eggs, with 3 the mode in most populations; 1 rare, probably an incomplete clutch, most likely in renests. In Lower Peninsula of Michigan, mean of 2.92 eggs, with 76.4% 3-egg clutches (n = 537 clutches, 1967–1987; Postupalsky 1989b). Westport, Massachusetts (1979–1984): mean 3.3 (n = 94 [range 2–4]; ca. 20–30% 4-egg clutches, depending on year; Poole 1989a).

In non-migratory populations, clutches generally smaller, although 3-egg clutches still the mode, and 4-egg clutches almost nonexistent; e.g., Baja California: 2.63 ± 0.58 SD (n = 86), with 5.8% 1-egg, 24.4% 2-egg, and 69.7% 3-egg (Castellanos et al., unpublished data); southern Florida, 1978 and 1979: 2.7 ± 0.6 SE (n = 22; Poole 1982a); Sonora, Mexico, 1992–1997: 72% 3-egg clutches, 25% 2-egg clutches, a few 1-egg and 4-egg clutches (n = 105; Cartron 2000).

Factors influencing clutch size remain unclear. A few tentative patterns seem evident, however: (1) Largest clutches in northeastern coastal U.S., 1980s—small but rapidly growing populations with ample food and nest sites available (new studies needed of these populations, now at higher densities). (2) Clear latitudinal trend, with southern populations laying significantly smaller clutches than northern ones (see Appendix 4 in Poole 1989a). (3) Fresh-water populations appear to have smaller clutches than salt-water populations at the same latitude, perhaps related to food quality and availability (e.g., Postupalsky 1989b). (4) Rates of courtship feeding not correlated with clutch size in southeastern Massachusetts (Poole 1985a). (5) Clear decline in clutch size with laying date; e.g., southeastern Massachusetts: 3.42 ± 0.10 SE (n = 31) in first week of colony laying versus 3.14 ± 0.14 SE (n = 14) ≥ 4 wk after first egg in colony (Table 3.1 in Poole 1984; see also Steeger and Ydenberg 1993); renestings always produce smaller clutches than first nestings.

Annual and lifetime reproductive success

Annual Success. Considerable research on this topic; focus here is on selected studies since 1985, emphasizing different habitats. See Poole 1989a for summary of breeding success in populations studied earlier

In North Carolina (n = about 40 nests in each of 2 yr; Hagan 1986): Most significant reproductive loss when nestlings 2–3 wk of age, period of peak growth; e.g., dropped from about 2.3 young/nest at hatching to 0.9–1.0/nest near fledging; sibling aggression common and males had long commutes for food, suggesting nestling starvation.

In southern Florida (Florida Bay and Keys; Bowman et al. 1989): young/active nest = 0.56 ± 0.8 SE (n = 34) in Florida Bay, 1.21 ± 1.0 SE (n = 19) in the Keys, a significant difference apparently related to food availability; Florida Bay is a more disturbed ecosystem with food less accessible than along Atlantic Coast, where Ospreys from the Florida Keys forage. Likewise only 38% of active Florida Bay nests successful versus 63% of nests in Florida Keys. See also Poole 1982a for Florida Bay Ospreys.

In southern New Jersey: Delaware Bay versus Atlantic Coast nests (Steidl et al. 1991a). Coastal nests had a lower failure rate (21% vs. 50%) and overall eggs hatched more successfully (69% vs. 50%); food likely a factor in difference, even though authors suspected contaminant effects due to dredging activity in Delaware Bay (Steidl et al. 1991a).

In southeastern British Columbia: 2 neighboring regions, differing in prey base and nest sites (Steeger et al. 1992). Despite easier foraging, larger and more nutritious prey, and abundant artificial nest sites in one region, no difference in breeding success detected between the 2. Apparently neither colony was limited energetically by weather or foraging distance.

In Sonora, Mexico, 1992–1997: mean annual productivity 0.67 fledged young/occupied nest (breeders plus nonbreeders), 0.83/active nest (breeders only; range 0.18–1.3 over 6 yr; Cartron 2000). Considerable local variation in success; years/areas with poor success had more late breeders than others, but overall no clear explanation for large annual variation in productivity. No difference in success of early versus late pairs, unusual for this species. Feeding rates not monitored. Overall more eggs than chicks lost, but chick death rates high in poor years (47–73% loss).

See Population Regulation for link between reproductive rates and population dynamics.

Advantages of early breeding: more young produced and much better rate of survival of those young (Poole 1984, Poole 1989a). In British Columbia, however, delayed nesting produced fewer young but “quality” (as measured by growth rate and number of fault bars in rectrices) not seen to decline; mostly loss of eggs, early versus late (16% [41] vs. 33% [39]); no difference early/late in loss of chicks, but postfledging survival not measured (Steeger and Ydenberg 1993).

Weather a significant influence on breeding success; overall, fewer young in years with heavy rainfall, especially during late incubation and early nestling stage, when young are vulnerable to chilling; loss of eggs greatest during rainy periods (Poole 1984, Johnson et al. 2008f). Not clear how this relates to male provisioning rates, but see Food Habits: feeding, above.

In St. Mary’s River, northern Lake Huron, steady decrease in reproductive success 1985–1992; 1.5 young/occupied nest to 1.1 (3-yr means) with increase in population (Ewins et al. 1995a). Study of recovering population in Estonia showed why this might be so: Poorer quality nest sites (farther from food; fewer young raised) were settled as population increased; initial settlers had better sites and continued to raise large broods (Lohmus 2001b).

Since these studies (most after 2002), others have documented breeding success in a few key locations; selected highlights follow:

(1) In a 15-yr study at Hg-contaminated lakes in northern California (1991–2006), breeding success averaged 1.44 young/active nest, with a significant increase in productivity in the latter half of the study, as the population grew from 7 to 31 nests (Anderson et al. 2008c).

(2) Along the lower Columbia River, mean productivity (1997–1998) was 1.67 young/active nest, despite high contaminant levels in some regions (Henny et al. 2004).

(3) At 5 lake sites across southern British Columbia, mean number of young/occupied territory varied from 0.8 to 2.1, reflecting differences in mean food delivery rates at each of these sites (Morrissey et al. 2004c). This study also showed how differences in food availability swamped any effects that contaminants might have on the breeding success of this population.

(4) In south-central Ontario (Kawartha Lakes), 1991–2001, success varied with nesting substrate: 1.48 (stump nests; n = 96), 1.16 (tree nests; n = 16), and 0.73 (artificial sites; n = 148) fledglings/occupied nest (Martin et al. 2005a).

(5) In west-central Saskatchewan, 1975–2002, breeding success was most affected by water levels: in low-water (drought) areas ( Houston et al. 2010).

(6) In northern Idaho and northeastern Washington, 1970–2005, mean brood size was 2.1 chicks/successful nest (n = 1,008), with significantly smaller broods in years with severe rainstorms (≥ 1.3 cm rain) during the mid to late nesting season (Johnson et al. 2008f).

(7) In Chesapeake Bay, 2000–2001, Rattner et al. 2004synthesis of historical (1960s to early 1990s) information from several field sites throughout the Bay shows a collective increase in reproductive rate (young/active pair) from less than 0.8 in the 1960s to more than 1.2 by the mid-1980s followed by a reduction to below 1.0 in the late 1980s—the reduction apparently mediated by food supply (Watts and Paxton 2007).

See also Bierregaard et al. 2014a for a summary of Osprey breeding success in Long Island, New York and southern New England, 1970-2013.

Lifetime. Data for an inland (Michigan) fresh-water population over a 15+ yr period (Postupalsky 1989b); 17 females, 23 males (known age) followed; only such study to date on this species. Key findings: Individuals differed greatly in production of fledglings, but mean production and variance did not differ between males and females; 22% of females that laid eggs, and 12% of males with egg-laying mates, produced no young during their lifetime; of females that did raise young, the number fledged ranged from 1 to 29, males 1 to 18; 16% of females produced 50% of total young, 28% raised 74%; among males, 24% raised 51% of young, 42% raised 76%; lifetime production of fledglings was correlated (positively and highly significantly) with longevity in both sexes—e.g., all individuals that produced ≥ 10 fledglings lived ≥ 8 yr, although some long-lived individuals produced no more fledglings than did other individuals that lived only 3–7 yr; numbers of young raised to fledging were correlated with numbers subsequently recruited to the local breeding population; no more than 13–14% of eggs laid, and no more than 23–27% of fledglings, contributed to the next generation of fledgling Ospreys; mortality before breeding age prevented most individuals (70.5%) from contributing fledglings to the next generation. Thus longevity and individual ability appear to drive lifetime breeding success in Ospreys, and these 2 factors appear to be often, but not always, related. Long-term study ongoing in Scotland should provide comparable results. One female bred through her 27th year and fledged 50 young (R. Dennis, personal communication)

Number of broods normally reared per season

A single brood is reared in both resident and migratory populations.

Proportion of total females that rear at least one brood to nest-leaving or independence

Number of failed breeders varies greatly from year to year within populations, as well as among populations. Food availability and especially weather have major impacts, but few data. Five to 50% of occupied nests in Sonora, Mexico, successfully fledged ≥ 1 young (6-yr study); generally egg loss greater than chick loss (n = 25–44 nests/yr; Cartron 2000). In 18-yr study in southern New England, average nest failure rate 26.3% (10.1 to 57.4%; ROB).

Nonbreeders: 9.6% of 469 pairs in Lower Michigan, 1974–1987; mostly young adults, first attempts at nesting; proportion increased somewhat with increased density of breeders (thus fewer nest sites available; Postupalsky 1989b). In southeastern Massachusetts, 10–15%/yr; generally all young individuals that arrive late, after others in population are well into incubation (AFP). In Mexico, higher percentage of nonbreeding population at southern latitudes: 29% of 126 nests along west coast of Baja California (A. Castellanos, personal communication); 12–45% in Sonora (Cartron 2000).

Life Span and Survivorship

A fairly long-lived species. Oldest known North American individuals reported to date: 25-yr-old male (Spitzer 1980); 23-yr-old female (AFP); 20+ yr-old female; all were still breeding (Postupalsky 1989b). Of 1,742 band recoveries from 1985 to 2014, 3 individuals were 23 yrs old and 12 were 20–22 yrs old. One female in Scotland bred until she was 27 yrs old (R. Dennis, personal communication), although very few individuals survive to this age. Among 17 female breeders in a Michigan population, for example, 47% were under 6 yrs old, 41% 7–11 yrs, and only 12% were older than 12 yrs; among 23 breeding males, 39% were under 6 yrs old, 52% 7–11 yrs, and only 9% were older than 12 yrs (Postupalsky 1989b). Similar data from a coastal Atlantic population in southeastern Massachusetts (Poole 1984): 54% of 46 breeding females were 3–5 yrs old; 17% 6–8 yrs; 28% were older than 9 yrs old with most (20%) of these being under 15 yrs old. Both of these were expanding populations and migratory; age structure may differ in populations limited by food and nest sites and in non-migratory populations.

Annual survival rates first estimated by (Henny and Wight 1969) from recoveries of individuals banded in New York and New Jersey. Recoveries of birds “found dead” and combined recoveries of birds “found dead” and “shot” yielded maximum and minimum estimates, respectively, of survival rates: 48.5–42.7% for the first year of life and 83.8–81.5% for each year thereafter. (Values based on birds “found dead” are believed to be most valid.) Data from annual returns of individually color-banded Ospreys have refined these estimates somewhat: 85% annual survival of adults ≥3 yr old over a 15-yr period in Michigan (Postupalsky 1989b); 83–90% for adults in Massachusetts and Chesapeake Bay, depending on age (highest mortality in age category 10–13 yr, a period when reproductive effort is highest; Table 8.1 in (Poole 1989a).

Spitzer 1980, estimating survivorship for a stable Osprey population—one in which breeding rates needed to balance mortality were known, also found about 85% annual survival of adults ≥ 2 yrs old, with about 60% survival in first year of life. See discussion of mortality in (Poole 1989a), which can be summarized as follows: “. . . each [age] cohort is quickly whittled down to a handful of survivors. On average, out of 100 young fledged in any year, 37 will be alive 4 years after fledging, 17 eight years after, and only 6–8 twelve years after.” To understand how mortality rates influence population dynamics, see Population regulation, below. In Sweden, slightly lower rates than in U.S. (Eriksson and Wallin 1994).

Disease and Body Parasites


Avian cholera found in 4 adult Ospreys during an outbreak among waterfowl on Chesapeake Bay in Maryland and Virginia; mode of transmission may have been use of bones and carcasses of waterfowl to line nests (Hindman et al. 1997). In Florida, 1 Osprey was diagnosed with pneumonia (Deem et al. 1998) and 4 with aspergillosis (S. Little, personal communication),


Endoparasites. In 17 Ospreys from the U.S., 28 species of helminths (17 trematodes, 3 cestodes, 7 nematodes, 1 acanthocephalan) were documented, although infection intensities were low and no lesions attributed to the parasites (Kinsella et al. 1996). Trematode Renicola lari found in an Osprey from Alberta, the most northerly report of this avian parasite (Kennedy and Frelier 1984). Paradilepis rugovaginosus and P. simoni, cyclophyllid tapeworms, thought to be specialists in Ospreys only (Kinsella et al. 1996), although 2 other species, Cladotaenia cylindracea and Cyclustera ibisae, have also been found in Ospreys (Lacina and Bird 2000). A report of 12 additional species of helminths (8 trematodes, 2 cestodes, 1 nematode, 1 acanthocephalan) found in Ospreys outside of North America (Lacina and Bird 2000), including Ligula intestinalis, found only in Ospreys.

Haemoproteus found in 7 of 23 adult and 4 of 84 nestlings, leucocytozoans in 0 of 23 adults and 1 of 84 nestlings examined at the Raptor Center at the University of Minnesota (MSM).

Ectoparasites. Ospreys harbor feather mites of the genus Analloptes (Xolalgidae) and Bonnetella (Avenzoariidae; restricted to Ospreys), and skin mites Myialges caulotoon (Philips 2000). Bonnetella fusca found on 85% of nestlings and all adults (n = 9) examined in Ontario; Analloptes sp. found on chicks from all 9 nests checked (Miller et al. 1997b). Kurodaia haliaeeti (Phthiraptera: Menoponidae; restricted to Ospreys) found on 40% of nestlings examined (8 of 20) in nests around Great Lakes; no hematozoa found (Miller et al. 1997b).

Causes of Mortality


No data. Large, hardy birds; not likely to die easily due to exposure, although starvation undoubtedly kills some (especially young postfledging and during migration). During migration, probably also vulnerable to hurricanes, which may kill significant numbers in some years (but no data).

Nestlings are vulnerable to extended periods of wind and rain, with brood sizes reduced in years with extended storms – likely owing to a combination of reduced foraging by males and exposure of the chicks (Reese 1977, Johnson et al. 2008f, AFP). In addition, nest failure (abandonment) during incubation is higher in seasons with above average rainfall (AFP).


Predators on adults include Great Horned Owls, at least one case of a Peregrine Falcon (Falco peregrinus), and probably caiman; predators on nestlings include Bald Eagles and raccoons (see Behavior: Predation, above).

Competition with Other Species

Little information. Competes with Bald Eagle for nest sites (see Ogden 1975), but not likely to cause mortality. Kleptoparasitism by Bald Eagles may result in fatal injuries as Ospreys try to evade pursuing eagles (R. Munquia, personal communication).

Interactions with humans

Shooting, collisions with vehicles or human-made structures including wind farms, electrocution on power lines (see Conservation and Management, below).


Natal Dispersal

Summarized in Poole 1989a. When returning to breed for the first time, males are more faithful than females to natal sites. This constitutes remarkably good “homing,” considering the great distances traveled in migration, the time (18 mo) that young are on distant overwintering grounds, and often dramatically different routes taken on the first return migration compared to the first trip south (ROB; see Migration, above). Of 39 females and 33 males hatched in southern New England 1970–1978, all males and 80% of females had first breeding sites within 50 km of their natal sites (Fig. 8.1 in Poole 1989a). In north-central Lower Peninsula of Michigan, mean distance dispersed from fledging to first breeding site: males 14.5 km (n = 37); females 38 km (n = 31; Postupalsky 1989a); in both populations, these should be considered minimums as some individuals, especially females, likely move outside the study area. In addition, these were expanding populations with access to abundant artificial nest sites. By contrast, in boreal forests of Finland where nest sites are more limited, mean distance between natal site and nesting site was 47 km ± 8 SD (maximum 145 km) for males, and 135 km ± 16 (maximum 303 km) for females (Saurola 1995).

Fidelity To Breeding Site And Winter Home Range

Remarkably few published data regarding breeding site, but generally high fidelity; e.g., over 80% of 150 breeding pairs monitored in southeastern New England returned to breeding sites year to year (AFP). Similar data for Michigan (S. Postupalsky, personal communication). Note that these studies tracked populations with stable, abundant, artificial nest sites; others probably differ.

All North American breeders tracked by satellite telemetry through at least 2 migration cycles (n = 31) returned to overwintering areas used in a previous year (Martell et al. 2001a, ROB). In western Africa, 6 of 9 overwintering individuals captured during 1979 were resighted in the same area during 1980; thus between overwintering periods, individuals show strong fidelity to local regions (Prevost 1982).

Of 5 satellite-tagged juveniles that completed 2 or more migration cycles, 3 returned to overwinter where they settled after their first migration south; 1 overwintered 130 km from where it had overwintered during the migration, and 1 alternated for 3 yrs between its original overwintering area in Venezuela and a location in Cuba discovered on the bird’s first trip north (ROB).

Home Range

Defined here as distance covered by breeding adults; e.g., distances traveled during hunting forays. Few data. About 14 km in North Carolina nesting colony (Hagan 1986). Average maximum distance traveled by 26 adult males during breeding season (both inland in New Hampshire and in coastal southeastern Massachusetts) was 19.4 km (± 7.08, range 8–41.8; ROB). On overwintering grounds in western Africa, individuals moved up to 10 km from daytime feeding grounds to night roosts (Prevost 1982).

Population Status

See Poole 1989a and references therein for details up to 1985 when world population (early 1980s) was estimated at 24,000–31,000 breeding pairs, with about 7,500–8,000 of these in U.S. (contiguous 48 states), about 800 pairs in the Baja Peninsula of Mexico, and roughly 10,000–12,000 in Alaska and Canada. We focus here on estimates and trends since 1990. Overall, very few precise counts, but significant increases in breeding numbers estimated since 1990, up 50–100% in many areas, reflecting availability of nesting platforms and other artificial nesting sites, ability of this species to habituate to human activity, and broad diet.

In 1997, Houghton and Rymon 1997 compiled estimates of Osprey population sizes in contiguous U.S. (Alaska excluded). They contacted researchers or state wildlife agencies for best estimates of population sizes, as of 1994 breeding season. (For 3 states, data were based on earlier censuses.) For some states, especially those with very few pairs of Ospreys, fairly accurate estimates were available; for others, best estimates were made. In states with the largest populations (Florida, Maine, Virginia, Maryland—all with estimated populations over 1,000 pairs), where accurate censuses were not feasible, population estimates encompassed broad ranges (e.g., Florida’s 2,500–3,000 pairs). Total number of pairs estimated ranged from 12,772 to 15,602, 50% to nearly 100% above an estimated total of 8,000 in 1981.

In 2001, ROB conducted a similar polling of state agencies, natural-heritage programs, and raptor biologists. Numbers were available for 43 states. Of missing states, none had a population estimated at over 100 pairs in 1994 censuses. As with previous studies, numbers from larger states with large Osprey populations are all educated guesses. With widespread increases in Osprey populations across the country, few wildlife and natural resource agencies made any attempts at monitoring Osprey populations in late 1990s, but rather devoted their efforts to rare or declining species. For most states, estimates for 2000 or 2001 breeding seasons based in large part on qualitative assessments of relative population growth and correction factors applied to 1994 data.

Using 1994 data for 5 missing states, we estimated 2001 Osprey population in contiguous continental U.S. at 16,000–19,000 pairs, representing an increase of roughly 25% over 1994 numbers. There were 17 states in 1981 and 8 states in 1994 with no known pairs. In 2001, only 4 states (Oklahoma, North Dakota, Nebraska, and Kansas) had no known nesting pairs; every other state, with the exception of Michigan, reported Osprey populations increasing, often conspicuously.

Since the 2001 summary, populations have continued to increase and expand (see Distribution, above) and statewide monitoring efforts have, with the notable exception of New England, continued to wane, so a reliable estimate of the continental population size has become ever more difficult.

No recent data for Canada; very rough estimate of 10,000–12,000 pairs during late 1980s (Poole 1989a). Estimated 500 pairs for Manitoba in 2000 (G. Holland, unpublished data). In Great Lakes basin, 1988–1993, about 900 occupied nests (about 600 inland); 300 within 5 km of the lakes (Ewins 1996). Numbers have certainly increased since then, but no reliable estimate is available. Similarly, there are no systematic surveys or census data available for the central plains, where breeding birds are widely and very sparsely dispersed, nor for the Rocky Mountains and Pacific coast, where densities are higher, but nowhere near the densities seen on the east coast (see Distribution for further details).

Our best estimates of population levels come from the East Coast. Currently (2015), rough estimates for Maine: 1500 pairs (C. DeSorbo, personal communication). The southern New England and Long Island, New York population now exceeds pre-DDT levels (> 1,200 pairs) (Bierregaard et al. 2014a, Bierregaard et al. 2014b), with Vermont and New Hampshire close to 275 pairs (C. Martin and J. Gobeille, personal communications). No estimates for non-maritime New York State, but recent robust growth in the east along Lake Champlain and west of the Adirondacks spreading across the state (52 pairs on Cayuga Lake alone, C. Cornell, personal communication). New Jersey has at least 420 pairs (Clark and Wurst 2014). No data for Delaware Bay, but perhaps 9,000 pairs in the greater Chesapeake Bay region (B. Watts, personal communication). Surprisingly low densities around Pamlico Sound and Albemarle Sound in North Carolina, but more common along the coast (200 pairs within 25 km of Paris Island on the South Carolina coast; B. Jennings, personal communication) and widely distributed across inland lakes and reservoirs in North Carolina and South Carolina (70 pairs at Lake Norman in 2015; T. Gestwicki, personal communication), over 200 pairs around Lakes Marion and Moultrie in 2000 (T. Murphy, personal communication).

Florida population, which has certainly grown from the estimated 2,500-3,000 pairs in 1994 (Houghton and Rymon 1997), is robust, with the exception of Monroe County where it is a species of special concern (myfwc.com/wildlifehabitats/imperiled/profiles/birds/osprey). It is estimated that 45% of the cell towers in Florida now support Osprey nests (Gryta and Monga 2012), and the two densest breeding colonies known in the world occur in southeastern Florida: Blue Cypress Lake (298 nests along the 20 km shoreline in 2015; www.pelicanislandaudubon.org) and Lake Istokpoga (over 300 nests in 2012; McMillian 2013). Along the Gulf coast, an estimated 400 pairs in Pinellas County and 250 pairs in Hillsborough County (B. Walker, personal communication).

A comparison of the number of Ospreys counted at La Gran Piedra in the Sierra Maestre mountains of southeastern Cuba (Rodríguez-Santana et al. 2014) and the tracks of satellite-tagged Ospreys migrating through Cuba suggests that we may be substantially underestimating the population size from eastern and central North America (ROB). Further analysis of these data is needed.

Population Regulation

Well studied in this species, owing to solid data on breeding rates, survival of fledged young and adults, dispersal distances (natal and between breeding attempts), and age at first breeding—the key variables influencing avian population dynamics (Newton 1979c). See Poole 1989a (Chapter 8) and Postupalsky 1989b for details; these built upon earlier Osprey studies (Henny and Wight 1969, Spitzer et al. 1983).

A few key findings regarding dispersal (see Range): females disperse farther than males between sites of fledging and first breeding, as is typical of birds, but only rarely do Ospreys of either sex breed > 50 km from their natal sites; year-to-year fidelity to breeding locales appears to be even higher (few move > 10–15 km). Together these findings suggest that (1) growth or decline of a population is determined largely by local reproductive and survival rates and (2) Ospreys are slow to colonize new areas.

Precise data on changes in population size in a few well-studied locales, along with data on annual survival of breeders and younger birds (see Lifespan and Survivorship, above) and breeding rates in these populations, have allowed researchers to link these parameters and determine breeding rates needed for population stability (Spitzer et al. 1983, Poole 1989a, Postupalsky 1989b and references therein). About 0.8–0.9 young/active nest appears to be the break-even point in this species; field data show that New England Osprey populations declined significantly in years (1950s and 1960s) when reproductive rates were lower than this (owing to DDT contamination), but grew quickly when rates improved above this level in the 1970s and 1980s (Table 8.4 in Poole 1989a). During same period, Michigan Osprey populations achieved stability at almost precisely the same breeding rate (Postupalsky 1989b). Rate at which new breeders are recruited to a population, however, is also a function of the age at which they start breeding, which appears to vary among populations, and likely (over time) within populations as well (see Measures of breeding activity, above). Thus in a region like Chesapeake Bay, where Ospreys start breeding at a mean age of 5.7 yr, a breeding rate of 1.15 young/active nest appears needed to achieve population stability, assuming survival of adults and fidelity to breeding and natal sites is the same as in New England and Michigan; this estimate jumps to 1.3 young/active nest when mean age at first breeding is 6.7 yr (Poole 1989a).

One notes that nearly all studies of Osprey population dynamics have examined populations that were small relative to the resources (food, nest sites) available to them. Poole 1989b argued that quality and availability of nest sites are key factors in the dynamics of Osprey populations studied during the 1980s: reproductive success was significantly higher among pairs nesting on artificial sites (owing mostly to stability of such sites); and populations with the largest number of empty sites available to them were those that grew most quickly. Findings during the 1990s suggest that it may be time to modify this assessment, however: (1) Gardiners Island, where breeding numbers declined 20–30% during 1996–2001, apparently because fish in nearby waters were not consistently available (P. Spitzer, unpublished data); (2) Martha’s Vineyard, Massachusetts where number of active nests doubled every 4–5 yr during the 1980s, leveled off and even dropped during the 1990s, subsequently growing only slowly despite young/active nest rates well above the 0.8–0.9 threshold level despite abundant empty nest sites in apparently desirable locations (ROB); (3) Finland, where breeding numbers grew only about 1%/yr for 2 decades, despite reproductive success equal to that of U.S. and Scottish populations growing at 5–10%/yr (Saurola 1995). Despite much study, there is much to learn about Osprey population dynamics.

In an analysis of 425 species recorded on the North American Breeding Bird Survey for 2003–2013, Ospreys had the 18th highest rate of increase; Bald Eagles had the fourth highest rate of increase (Sauer et al. 2014c).

Recommended Citation

Bierregaard, R. O., A. F. Poole, M. S. Martell, P. Pyle, and M. A. Patten (2016). Osprey (Pandion haliaetus), version 2.0. In The Birds of North America (P. G. Rodewald, Editor). Cornell Lab of Ornithology, Ithaca, NY, USA. https://doi.org/10.2173/bna.683