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 Birds of North America. 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.
Already a subscriber? Sign In
Measures of Breeding Activity
Age At First Breeding; Intervals Between Breeding
Annual breeding. Age varies, depending on territory availability, the latter influenced by floater com-petition and breeder turnover. Female tends to breed year earlier than male (Cade and Fyfe 1978, Ratcliffe 1993). Breeder age distribution skewed to younger individuals in depleted or expanding populations or when not otherwise density limited (Tordoff and Redig 1997). Mean recruitment age of 16 males in an increasing population in Nunavut was 4 yr (range 2-8), 3 yr for 4 females (range 3-5; Johnstone 1998). In midwestern population, also increasing, most individuals of both sexes bred at 2 yr, but twice as many females bred at 1 yr and 10 times more males than females first bred at >3 yr (Tordoff and Redig 1997). Yearling female more frequent as pair member than yearling male, although both sexes have bred successfully as yearlings (Wendt and Septon 1991).
In captivity, mean age of first breeding of 21 females was 3.4 yr ± 1.15 SD (range 2-5; Clum 1995a). Of 22 wild-taken males in captivity, 5 first copulated at 3 yr old, 9 at 4 yr, 3 at 5 yr, 2 at 6 yr, and 3 at 7 yr (mode 4 yr, mean 4.2 yr); of 30 females, 2 first laid at 2 yr, 10 at 3 yr, 12 at 4 yr, 4 at 5 yr, 1 at 6 yr, and 1 at 9 yr (mode 4 yr, mean 3.9 yr; Cade and Fyfe 1978).
See Breeding: eggs, above.
Annual And Lifetime Reproductive Success
Annual Success. Best characterized by recent data (Appendix 1). Breeding success (percentage of nests fledging ≥1 young) and productivity (number of fledged young/territorial pair) varied greatly from region to region and year to year in decades of 1970s-1990s as populations recovered from effects of pesticides on reproduction. Prior to 1980s, declining or greatly diminished populations generally characterized by depressed annual productivity rates of <1.0 to <0.5 fledglings/territorial pair (Cade et al. 1988, Ratcliffe 1993); continued in Colorado into 1980s (Appendix 1), but after 1984, in association with massive reintroductions, breeding population dramatically increased there and annual productivity from 1985 to 1998 ranged from 1.2 to 1.9 young/pair (mean 1.6, n = 395 pair-yr; G. Craig and J. H. Enderson in Mesta 1999). Similar but less severe pattern in California (Linthicum and Walton 1992); by 1993-1997, annual productivity there ranged from 1.4 to 1.7 young/pair (mean 1.6, n = 356 pair-yr; Mesta 1999).
Other factors influencing annual productivity include: (1) egg and chick mortality from cold, wet, and late spring weather, a major factor (Cade and White 1971, Court et al. 1988b, Mearns and Newton 1988, Ratcliffe 1993); (2) local yearly variation in prey abundance (Court et al. 1988b); (3) regional differences in overall prey availability (Ratcliffe 1993); in 6 eco-regions of ne. U.S. from 1984 to 1996, productivity varied from 1.12 to 1.81 young/pair and nesting success from 59 to 93% (n = 332 pair-yr), apparently related to prey availability (Corser et al. 1999); (4) predation/disease: not quantified for any population but can be locally significant; e.g., Great Horned Owls prevented at least 8 pairs from establishing successful nests on cliffs of upper Mississippi (Cade et al. 1989, Tordoff and Redig 1997); owl and Golden Eagle depredations often main cause of deaths among hacked young in reintroduction programs (Barclay and Cade 1983, Cade et al. 1988).
Difference in productivity at individual territor-ies within local population is notable characteristic of Peregrines. At Rankin Inlet, Nunavut, at regularly occupied (high-quality) sites, productivity over 14 yr averaged 1.4 young; at infrequently occupied (poor quality) sites, 0.8 young/pair (Johnstone 1998; see also Mearns and Newton 1988 for similar findings in Scotland). In coastal British Columbia, half of all nestlings produced by 21% of nesting pairs, one-quarter by just 9% (R. W. Nelson pers. comm.). In Massachusetts, 1935-1947, pairs at 6 superior eyries (based on physical characteristics) produced >76 fledglings, while pairs at 8 inferior sites produced only 27 young (Hagar 1969). In Scotland, both cliff height and accessibility of nest ledge positively correlated with number of young produced (Mearns and Newton 1988). Also, immature females on average have reduced reproductive performance (Mearns and Newton 1984) and often fail to lay eggs (Ratcliffe 1993). Generally, <50% of breeders produce >70% of young in local populations.
Lifetime Reproduction. Few data; needs study. Partial lifetime reproductive success reported as follows: (1) male on Langara I., British Columbia, raised 22 young in 7 yr; female, 18 young in 8 yr, but females consistently producing more young reported to have had lower survival than birds producing fewer young/yr (Nelson Nelson 1988b, Nelson 1990). (2) Female on Yukon River, AK, raised 3-4 young/yr for 7 yr (White et al. 1995b). (3) falcon on Sun Life building in Montreal, Quebec, produced 22 young over 17 yr with 3 different males, close to lifetime record (she lost clutches to egg breakage in 3 yr after 1947 and had no place to make scrape in first 2 yr of occupancy, so effective breeding career was 12 yr; Hall 1955). (4) In midwest U.S., of breeders that died early, 31 males with average age of 4.0 yr at death raised 143 young (4.6 per male); 18 females dying at average age 2.7 yr fledged 75 young (4.2 per female); 9-yr-old female produced 25 of those young before being killed by another adult female; 31 males still living at average age 4.6 yr had already produced 238 young (7.7 per male) and 36 living females with average age 5.1 yr had fledged 355 young (9.9 per female); 4 of those females aged 9, 9, 9, and 7 yr fledged 87 young (25% of total; Tordoff and Redig 1997). (5) At Rankin Inlet, falcons at frequently occupied eyries had mean breeding life span (i.e., site tenure within study area) of 2.7 yr (range 1-5, n = 39) for male and 2.9 yr (range 1-5, n = 65) for female, with mean lifetime production of 4.7 young (range 1-11, n = 65); at infrequently occupied eyries mean breeding life span for male was 2.0 yr (range 1-4, n = 9), for female 2.2 yr (range 1-4, n = 15), with mean lifetime production of 3.0 young (range 1-10, n = 18; Johnstone 1998). Difference in perceived breeding careers in Arctic and Midwest may reflect differences in site tenure rather than mortality; most Midwest breeders held territories year round, whereas arctic falcons must annually reclaim them, a difference possibly increased by annual variations in suitability of arctic nesting territories (see Population regulation, below).
Reproductive Success in Captivity. Female age has significant effect (p = 0.0001) on all measures of reproductive success: clutch size, fertility, hatchability, brood size, nestling survivability, and number of fledglings; in all but 1 case (nestling survival), data best fit quadradic model, reflecting initial increase in performance followed by decrease with age; females that retained mates throughout lifetimes had higher annual fertility, hatchability, and brood sizes at hatching and fledging than females changing mates ≥1 times, suggesting selection in nature should favor lifetime monogamy; changing breeding locations of mated pairs had no influence on breeding performance. Similarity of reproductive patterns between wild and captive birds, not limited by access to mates, nest sites, and food or subject to environmental hazards, suggests age-related changes in reproduction are not necessarily resource-limited; in the absence of resource limitation in captivity, experience of the pair is primary factor determining reproductive success, but benefits of increasing experience are eventually offset by senescence (Clum 1995a).
Life Span and Survivorship
Maximum longevity records for banded birds range from 16 to 20 yr. Nestling female banded on Colville River, AK, in 1981, trapped alive and released in California, Sep 2000 may be oldest, >19 yr (T. Swem, B. Walton pers. comm.); male hacked at Cobb I., VA, in 1978 found dead at James Bay Bridge, VA, in Apr 1995, <17 yr (J. Barclay pers. comm.); also famous Sun Life female in Montreal disappeared after >18 yr. In captivity, few live beyond 20 yr; maximum up to about 25 yr (TJC).
First-year survival not well known but now generally assumed to be 40-50% of fledglings (see Ratcliffe 1993 for higher estimates in Britain). No reliable estimate of first-year survivorship in any North American population; estimated minimum of 23% in midwestern F. p. anatum based on resightings of marked birds (Tordoff and Redig 1997), but actual survival must be higher. Beebe (Beebe 1960) proposed that survival among first-year F. p. pealei was low due to harsh maritime environment.
Minimum breeder survival in F. p. pealei estimated at 63% for female, 74% for male, based on observed turnover rates (using sketches of physical differences of individuals); females producing large broods were more frequently replaced (Nelson Nelson 1988b, Nelson 1990). Data on turnover rates of banded falcons corrected for breeder dispersal indicated 91% survival for breeding female Scottish Peregrines (89% for both sexes combined; Mearns and Newton 1984). Based on photo-graphs of individual facial markings (primarily), minimum breeder (F. p. anatum) survival (turnover cor-rected for off-site encounters) calculated at 84% (n = 57) in Colorado during 1981-1985 (Enderson and Craig 1988). Based on corrected turnover rate of banded females (n = 40), minimum survival 77% on Yukon River, AK, 1981-1984 (Ambrose and Riddle 1988). Minimum of 81% annual survival estimated for breeding female F. p. tundrius and 85% for males, based on a marked population in Canadian Arctic (Court et al. 1989). In same region, Johnstone (Johnstone 1998) estimated annual survival (from turnover) of 74% for male (80% if adjusted for known breeder dispersals) and 69% for female (76% adjusted). Using capture-recapture method, Gould and Fuller (Gould and Fuller 1995) estimated minimum annual female breeder survival of F. p. tundrius in Greenland 1983-1991 at 79%. Based on band identifications and voice-printing, Telford (l996) estimated 18% turnover of females at 11 eyries (possible range 16.7 to 25.0%) in ne. U.S. from 1989 to 1991, or 82% minimum survival. From 1987 to 1995 in midwest U.S., territorial adult survival of males based on 115 male territory-yr varied annually from 71 to 89% (average 79%), and for females based on 136 female territory-yr range was 80 to 100% (average 93%), highest reported for any population; difference between sexes highly significant (Tordoff and Redig 1997).
Theoretically, where adult survival = 70%/yr, assuming no change with age, median adult life (after second year) is about 2 yr (mean 2.8 yr), at 10 yr 3% of cohort still alive, and about 0.1% survive to 20 yr; for 80% rate, median >3 yr (mean 4.5), 11% alive at 10 yr, 1.0% at 20 yr; for 90% rate, median >6 yr (mean, 9.5), 35% alive at 10 yr, 12% at 20 yr. Arctic migrants fall in range of 70-80% minimum adult survival; resident, temperate zone birds, in range of 80-90%, with possible exception of falcons on Langara I., British Columbia (Nelson Nelson 1988b, Nelson 1990). Known population growth rates in recent years and well-known pro-ductivity rates indicate true adult survival rates for migrants likely fall in range of 80-85% and for residents in range of 85-90% (see Population regulation, below).
Disease and Body Parasites
Peregrines harbor numerous organisms: avian pox (Poxvirus avium), Newcastle disease, herpesvirus, various mycotic infections, strigeid trematodes, air sac nematodes (Serratospiculum amaculata; particularly common in falcons of arctic origin), malaria (Plasmodium relictum), tapeworms, various bacterial organisms of at least 13 kinds (e.g., Proteus, Escherichia, Enterobacter, Staphylococcus, Streptococcus, Pasterella, Pseudomonas, Proteus, Salmonella). Have suffered mortality from Clostridium botulinum Type C and Trichomonas gallinae acquired secondarily from prey (Coues 1874a, Fowler 1986b). Various ectoparasites (e.g., hippoboscid flies and mites). Also 4 species of Phthiraptera ("chewing lice"; Colpocephalum zerafae, Degeeriella rufa, Laemobothrion tinnunculus, and Nosopon lucidum; D. Clayton pers. comm.), the Siphonaptera (Ceratophyllus garei), and dipterans (Icosta nigra, Ornithoctona erythrocephala; Wheeler and Threlfall 1989).
Fledglings at cliffs may be killed prior to independence by other raptors, especially Great Horned Owls and Golden Eagles, occasionally by mammalian predators, and also suffer disease and accidents. Urban fledglings may have greater variety of postfledging fatalities than fledglings in natural landscapes; deaths primarily from collisions with automobiles, windows, and other human-made objects, and drowning after falling from bridges (Cade and Bird 1990, Sweeney et al. 1997). Collisions also affect older age classes; in nonurban environments, face a variety of human-related hazards, including electrocution by utility lines, wire and fence collisions, shooting, and airplane strikes (Barclay and Cade 1983, Santa Cruz Predatory Bird Research Group unpubl.). In total of 455 recorded fatalities among midwestern Peregrines, 78 caused by collision with buildings, 50 by vehicle collisions, 33 by miscellaneous accidents, 28 by disease and starvation, 19 by adult Peregrines, 15 by predators, 10 by shooting, and 10 by storms and lightning and lesser numbers of other causes (Tordoff et al. 2000).
Territorial rivalry, including actions on winter territories, between adults may result in death, increasingly so as breeding density and floaters in-crease (Herbert and Herbert 1965, Tordoff and Redig 1999b, R. E. Ambrose and P. Pyle pers. comm.).
Initial Dispersal From Natal Site
In reintroduced eastern population, natal dispersal of 29 females ranged from 0 to 752 km, with 18 (62%) >100 km; for 13 males, 0-1,117 km, with 8 (62%) >100 km (Barclay 1995). Female generally disperses farther than male from natal localities to breed. In upper Midwest, 67 females moved on average 345 km and 73 males, 174 km; but direction was random (Tordoff and Redig 1997); 75% of females nested 355 km and 75% of males 170 km from natal areas (Septon et al. 1996). In Alaska, average distances were 121 km for 20 females and 69 km for 66 males (Ambrose and Riddle 1988). Of 1,702 nestlings banded from 583 broods in Greenland, only 35 males and 7 females were recruited as breeders back into study area; natal dispersal of 21 males was 28.1 km, and for 6 females, 27.1 km. One female banded in 1990 found breeding in 1997 about 690 km from natal site (Restani and Mattox 2000). Females had a higher mortality rate (12 males to 33 females recovered), suggesting that proportionately fewer natal females were available for return and recruitment.
Fidelity To Breeding Site And Winter Home Range
High degree of nest-site fidelity among F. p. tundrius in n. Canada, where only 1 of 26 resighted females and none of 16 males moved in following year (Court 1986). In Alaska, only 2 of 40 resighted females moved to new territories in different drainage (Ambrose and Riddle 1988). In Colorado, only 3 of 48 resighted (F. p. anatum) moved to other territories in following year (Enderson and Craig 1988). In reintroduced midwest population, of 241 territory-yr through 1995, only 10 involved adults moving to new territory: 7 left territories where breeding was unsuccessful or mate disappeared, 4 where bird left successful site for better one, and 3 associated with eviction by fighting (Tordoff and Redig 1997). Female more likely to switch breeding sites than is male. In Porto Alegre, Brazil, during boreal winter, birds (unbanded but identified by morphological traits) on same territory several years (J. L. B. Albuquerque pers. comm.); banded resi-dent California falcons show great territory fidelity in winter (B. Walton pers. comm.). Fidelity to territory (eyrie) probably accounts for durability of pair bond (Cade 1960, Tordoff and Redig 1997).
Defined here as hunting range beyond defended boundary around eyrie. Extent of home range movements by territorial pairs during breeding reflects prey density (Ratcliffe 1993). With radio telemetry in Colorado, Enderson and Craig (Enderson and Craig 1997) estimated mean home range during nesting of 358-1,508 km2for 2 adult males and 3 adult females. In relatively prey-rich habitat of Scotland, as little as 117 km2(Mearns 1985). In Channel Is., CA, breeders with radio transmitters foraged primarily within a 5 km of breeding cliffs (WGH); also on Queen Charlotte, I., British Columbia (Beebe 1960, Nelson 1977b). Of hunting flights of breeding female within home range in n. California, 47% >1 km; for male 65% >1 km; hunted on average 5 km (range 3-8) from eyrie (Enderson and Kirven 1983).
Winter ranges of 7 telemetered Peregrines in coastal Texas, most overlapping, were 20-28 km in diameter; adults more sedentary than first-year birds (Enderson et al. 1995b).
Lower densities in North America compared to Europe (Ratcliffe 1993); among Holarctic and Nearctic falconiforms, only Gyrfalcon and prob-ably Prairie Falcon have smaller total populations in America (Clum and Cade 1994, Steenhof 1998). In late 1990s, North American breeding populations increasing at 5-10%/yr (Enderson et al. 1995a, Mesta 1999). For F. p. anatum, known number of occupied eyries in 1997 was 301 in Alaska; 347 in Canada; 329 in Washington, Oregon, and California; 529 in the Rocky Mtns. and sw. U.S; and 205 hybrid- anatum from the Mississippi River to Atlantic Coast (Mesta 1999); additional 170 pairs estimated for Mexico and Baja California (Enderson et al. 1995a). Total estimated population, about 2,500-3,000 pairs (based on Enderson et al. 1995a and Mesta 1999).
For F. p. tundrius, known number of occupied eyries was 104 in Greenland (estimated 400-500 pairs, Falk and Møller 1988), 279 in arctic Canada (estimated at 1,500-2,000 pairs, G. S. Court, U. Banasch pers. comm.), and 158 in n. Alaska (estimated 400-500 pairs, T. Swem, B. Ritchie pers. comm.; compiled from Cade et al. 1988, Enderson et al. 1995a). Estimated total: 2,300-3,000 pairs.
Estimates based on counts of migrants in 1980 suggested total boreal population (tundrius and northern anatum) ranged between 6,700 and 13,000 pairs (Cade et al. 1988); in autumn 2000, >2,000 migrants counted in Florida Keys (J. Larrabee, J. En-derson pers. comm.), and 4-5 migrants estimated at each of >4,000 offshore oil rigs in Gulf of Mexico during Oct (A. Wormington, fide Pendergrass 2000, R. W. Russell pers. comm.).
For F. p. pealei, known number of occupied eyries in 1990s was 271 in Alaska, 77 in British Columbia, 17-20 in Washington, and about 5-10 in Oregon (Cade et al. 1988, Enderson et al. 1995a, Wilson et al. 2000b, C. M. Anderson pers. comm., D. Fenske pers. comm.), although not clear that all nesters on Washington/Oregon coast are pealei; estimated total nesting pairs, 850-1,000. Total continental breeding population estimated at 8,000-10,000 pairs at end of twentieth cen-tury and still increasing. Small proportion of increase represents increased effort to locate nesting pairs and to count migrants during past 30 yr (see below). If there are 8,000 breeding pairs, total population at end of breeding season, including immatures and floaters, could well be in range of 40,000-50,000 individuals (see Population regulation, below).
Generally wide-ranging but sparsely distributed. Limited first by distribution of suitable nest locations and further by territorial spacing of pairs (Hunt 1998b), the latter influenced by food availability (Hunt 1988, Ratcliffe 1993). Spacing of nesting pairs highly variable by region, but in some areas, superabundant food (colonial nesting seabirds) and few ground predators associated with greatly reduced spatial requirements for nesting; further affected by territorial spacing of pairs (Hunt Hunt 1988, Hunt 1998b). Yearly territory reoccupancy rates about 85-90% in many regions (Enderson et al. 1995a), but lower in some (e.g., Arctic, where annual variations in prey and weather modify yearly suitability of some territories; see Johnstone 1998).
Under favorable conditions, local density can reach 1 pair/10-20 km2or higher, but 1 pair/100 to >1,000 km2more typical for North America, in contrast to higher densities often found in Europe (Ratcliffe 1993). Density often best expressed as pairs/linear distance apart rather than pairs/area occupied, because breeding dispersion tends to be either dendritic along rivers and tributaries, linear along coastlines, or perimetric around islands and lakes.
Maximum known density reported from range of F. p. pealei on Langara I., where superabundance of colonial nesting alcids and numerous nest sites (cliffs) reduce spatial requirements: Locally in Cloak Bay (linear shoreline about 1.7 km and area of about 520 ha) in 1952-1958, 5-8 pairs nested (Beebe 1960); for entire 42-km perimeter of island, estimates of 20-35 pairs (Brooks 1926b, Beebe 1960), but <10 pairs in recent decades (see beyond). High densities also in Aleutians: Amchitka I., mean of 18.6 pairs along 193 km of shoreline (1 pair/10.3 km); 7 other Aleutian islands ranged from 1 pair/4.8 km to 1 pair/13.0 km (White 1975a). Elsewhere in range of pealei, densities vary from 1 pair/10.0 km to 1 pair/80.0 km of coastline (Ambrose et al. 1988).
Relatively high densities also occur in arctic mainland regions. Fyfe (Fyfe 1969) considered 1 pair/52 km2average for optimum nesting habitat in Canadian Arctic and 1 pair/259 km2for areas of limited nesting habitat. Along 293 km of Colville River, AK, in 1990s, density varied between 1 pair/3.2 and 3.8 km (51-61 occupied eyries, data from T. Swem). At Rankin Inlet, Nunavut, northwest corner of Hudson Bay, in area of 450 km2, nesting pairs varied from 17 in 1981 to 28 in 1993 (1 pair/16.0-26.5 km2); distances between pairs 0.7 to 9.8 km (Court et al. 1988b, Swem and Ambrose 1994). Density in other regions of Northwest Territories ranged from 1/97 km2to 320 km2in early 1980s (Bromley 1988). In 1985, Mattox and Seegar (Mattox and Seegar 1988) found 40 occupied cliffs (38 pairs) in about 2,500 km2area of Søndre Strømfjord region, w. Greenland (1 site/63 km2); in s. Greenland, 10-13 pairs found in land area of 2,636 km2between 1981 and 1985 (1 pair/203 to 264 km2; Falk and Møller 1988).
In taiga zones: along 265 km of central Yukon River, AK, 25-46 pairs nested between 1988 and 1998 (1 pair/5.8-11.1 km), and along 375 km of Tanana River, AK, 12-33 pairs nested in same period (1 pair/11.4-31.3 km; Ambrose et al. 1988, Mesta 1999); along about 1,440 km of Mackenzie River, Northwest Territories, 85 pairs nested in 1990 (1 pair/18.1 km; Enderson et al. 1995a).
Former densities in e. U.S.: (1) Delaware Valley of New Jersey, 39 known eyries in about 55,980 km2but maximum of 21 occupied in 1940s (1 pair/about 2, 666 km2; Rice 1969a); (2) in about 25,900 km2around New York City prior to 1940s, 19 pairs (1 pair/1,363 km2; Hickey and Anderson 1969); (3) 28 known eyries in Vermont before 1940 (1 site/858 km2; Berger and Mueller 1969).
Representative densities of known eyries in w. North America: (1) Pacific Coast of Baja California from Los Coronados Is. to Asuncion I., 42 known eyries prior to 1964 along about 960 km of coastline and offshore islands (including about 6 eyries on 10.4-km2Natividad I. (Lamb 1927b), 1 eyrie/approx. 23 km (Banks 1969, Porter et al. 1988). (2) Sierra Madre Oriental, Mexico, 6 pairs in circle with 25-km diameter, spaced 4.2-8.1 km apart (Hunter et al. 1988a). (3) Green River, Dinosaur National Monument, CO/UT (12-15 pairs in 77 km2; S. Petersburg pers. comm.). (4) Colorado River, Grand Canyon National Park, AZ, 71 breeding areas in 1988-1989 between Lees Ferry and Lake Mead, a 444-km river and canyon corridor; in 1,158 km2, 1 breeding area/16.3 km2. (5) For California, Sierra Nevada region, 49 historical eyries in 146,335 km2(about 1 site/3,000 km2); northern coast region, 49 eyries in 32,022 km2(1 site/654 km2); San Francisco Bay region, 22 eyries in 10,360 km2(1 site/471 km2), 17 occupied in 2001 (B. Walton pers. comm.); central coast region, 93 eyries in 34,965 km2(1 site/376 km2); south coast region, 31 eyries in 30,433 km2(1 site/982 km2; Thelander 1977); in 2001, Long Beach, CA, had urban/industrial population of 7 pairs in about 25 km2area (pair/3.6 km2) with mean nearest neighbor distance of 1.85 km (range 1.2-2.94). (6) Columbia River Palisades, 13 known eyries (about 1 site/7.0 km; R. M. Bond in Nelson 1969c). (7) Okanagan Lake and River, British Columbia, about 15 eyries, including Vaseux Lake with 3 pairs on 800-m cliff (1 site/9.7 km; data from A. Brooks in Nelson 1969c).
(See also Conservation and management, below). Unstudied prior to 1930s-1940s (Hickey 1942, Bond 1946). Sparse, tree-nesting population in Mississippi River basin mostly gone by early 1900s owing to loss of big trees, 4 sites in cypress swamps of Ten-nessee and n. Louisiana remained occupied into 1940s (Hickey and Anderson 1969). Hickey (Hickey 1942) estimated possible 10-18% decrease in occupancy of all known eyries (408) in e. North America by about 1940, but cautioned that permanent abandonment of nesting territories is difficult to determine (pair may move to undetected alternate sites); in Pennsylvania/New Jersey, 16 of 39 eyries abandoned prior to 1940 (Rice 1969a), possibly owing to increased depredations of Great Horned Owl (TJC). In w. North America, 14 (4.3%) of 328 eyries abandoned before 1940 owing to human influences (Bond 1946); 80-90% of Peregrine eyries abandoned in intermountain region of n. Utah, w. Wyoming, w. Montana, Idaho, Oregon, and Washington (>50 known eyries deserted, many taken over by Prairie Falcons) from 1940 to early 1950s, perhaps related to period of low precipitation and consequent biotic changes between 1920 and 1960 (Nelson 1969c).
Otherwise, continentwide population before World War II considered stable, and breeders characterized by traditional use of same eyries and territories over periods measured in decades and centuries (Hickey 1942, Bond 1946). Prior to DDT use, number of territories estimated at 7,000-10,000 in North America (Kiff 1988), plus about 450 in Greenland (Falk and Møller 1988), with 10-15% of total territories unoccupied in any given year (actual occupancy rate higher because use of alternate sites sometimes undetected).
Precipitous population crash in 1950s to mid-1970s owing to effects of DDT (DDE) on reproduction (thin eggshells and embryo fatalities, and possibly increased fatalities of adults from HEOD [dieldrin, aldrin] resi-dues in prey; Hickey 1969, Cade et al. 1988, Mesta 1999; see Conservation and management, below). Decline began in temperate parts of range and spread northward into arctic regions. By 1965-1970, breeding populations totally extirpated in U.S. and s. Canada east of Rocky Mtns. to Atlantic Coast and south of boreal forest, and greatly reduced (10-50% of pre-pesticides numbers) in different parts of western range; in northern regions, populations declined from about 1969 to about 1975 by 50-75% (or virtual extirpation locally; e.g., Tanana River, AK, north slope of Yukon Territory; Fyfe et al. 1976b, Cade et al. 1988). F. p. pealei little affected by pesticides but local populations (e.g., Langara I., British Columbia) declined in same period in relation to dramatic reduction in seabird populations associated with major changes in marine ecosystem (Nelson and Myres 1976) and possibly increased rat predation (Nelson 1990).
Population stability or increase locally noted by late 1970s (Fyfe et al. 1976b, White et al. 1990). After restrictions on use of DDT in Canada in 1970 and in U.S. in 1972 (37 FR 13369) and HEOD compounds in 1974 (39 FR 37246), number of nesting pairs increased rapidly in 1980s (Cade et al. 1988). Aided by release of >6,000 young produced or reared in captivity in s. Canada and coterminous U.S. (Holdroyd and Banasch 1996, Cade 2000), by 1990s continentwide numbers approached pre-DDT level, although distribution of nesting pairs was different, with some former areas still depleted, others with higher density than historically known, and some formerly unused habitats (urban and industrialized areas and coastal salt marshes) occupied for first time (Enderson et al. 1995a, Cade et al. 1997, Mesta 1999). Increase in population continuing into twenty-first century (e.g., Tordoff and Redig 2001).
Healthy populations stable overall for several reasons related to interplay among natality, mortality, dispersal, and territoriality. Many floaters of both sexes, excluded by territory holders, buffer breeding segment by quick replacement of lost breeders (Ratcliffe 1993; see Johnstone 1998, for results of first removal experiment with Peregrines). Occurrence of floaters indicates that serviceable breeding locations and cohorts from them strictly limited, a constraint bounding populations in state of equilibrium (Hunt Hunt 1988, Hunt 1998b). Number of pairs varies little annually because properties of the nesting situation are primarily physiographic, including those influencing prey availability and suitability of nest sites, and vacancies are immediately filled by floaters.
Some populations may respond to climatic cycles influencing prey numbers over long term, and nest success may be reduced at arctic territories some years owing to bad weather and unpredictable prey numbers (Bradley et al. 1997); but these factors less influential in temperate and tropical regions. If vital rates are high, density feedback may adjust equilibrium population size through floater pressure, impacting breeder survival and possibly nest success (Monneret 1987, Haller 1996a, Hunt 1998b). Food competition among nonbreeders may not regulate their numbers because (1) "survival habitat" without cliffs is vast relative to breeding habitat and (2) extreme variety of potential prey dampens population-scale impact of Peregrine depredations. Populations far more sensitive to changes in adult survival than juvenile or subadult survival or reproduction. In hypothetical population producing 1.5 young/territorial pair, where annual adult survival = 0.83, subadults = 0.70, juveniles = 0.50, there would be 1 floater/breeding pair at equilibrium (Moffat's equilibrium model; Hunt 1998b). Reducing floaters to zero (i.e., placing population at decline threshold), requires 11% reduction in adult survival. Same result requires 33% reduction in either subadult or juvenile survival or reproduction. Although both natal and breeder dispersal are known (see Range, above), exact role of emigration and immigration not yet determined; usually assumed to cancel each other, but some populations may be sources, others sinks.
Several models based on conventional calculations of growth rate (lambda) derived from Leslie matrix or life-table statistics have been applied to Peregrines (Grier and Barclay in Cade et al. 1988, Wootton and Bell 1992). Moffat's equilibrium model (Hunt 1998b) offers best framework for understanding natural regulation of falcon populations at environmental carrying capacity, because it is based on spatially imposed limits to productivity and calculation of floater-to-breeder ratios at equilibrium rather than growth rate.
According to Moffat's equilibrium model (Hunt 1998b), populations with adult mortality of 30% can remain stable or produce slight surplus of floaters only when juvenile mortality is <50%, unrealistically low, and productivity is 2.0 young/pair. Thus, some estimates of adult mortality for highly migratory, arctic populations must be too high, since they are known to have numerous floaters and have generally increased by factor of 2-4 in last 25 yr. Similarly, 37% female mortality on Langara I., British Columbia, with productivity of 2.31 young/pair would require juvenile mortality to be no more than 49% just to maintain breeder stability with no surplus floaters. Since floaters are reported to be common, estimates of adult mortality based on uncorrected turnover must be too high, owing either to breeder dispersal or misidentification of individual falcons.
The classic, stable Peregrine population is basically resident, with annual adult mortality ranging from 10 to 15% and productivity of 1.0-2.0 young/ pair, maintaining significant numbers of floaters at all com-binations of juvenile mortality and productivity except when juvenile mortality reaches 70% and productivity is 1.0 young/pair. Breeding population is highly buffered by floater-to-breeder ratios commonly in range of 1:1 to 2:1.