Demography and Populations

Age At First Breeding; Intervals Between Breeding

Probably begins breeding at 3 yr old, with attainment of full breeding plumage (Allen 1942, see Appearance: molts and plumages, below). Earliest breeding reported at about 2 yr old (Basic III plumage) in Tampa Bay area and Florida Bay, FL (Dunstan 1976, Bjork and Powell 1996, R. Paul pers. comm.). Normally 1 clutch/year.

Clutch

See Appendix 1 . Clutch size 1–5, usually 3–4 (mode 3); maximum 7, from Texas (Oberholser 1974). Means from different geographic locations overlap; most frequently reported near 3. Little information from Louisiana: 75 nests had 3–5 eggs each (Oberholser 1938). Clutch size varied significantly among years at Nueces Bay, TX (White et al. 1982).

Annual And Lifetime Reproductive Success

See Appendix 1 . Only 2 major studies from Nueces Bay, TX, and Florida Bay (White et al. 1982, Bjork and Powell 1994). Highly variable among years, from complete colony failure to mean of 2.5 young/nest, up to 2.8/successful nest at 21 d of age in Florida Bay (Bjork and Powell 1994). Values overlap among locations, but measure of success differs among studies. Typically, without colony abandonment, average of 1–2 young/nest produced.

From 1992–1993, and 1994–1998, poor reproduction in Florida Bay: complete failure at Tern Key over 4 yr; in 1 yr the second nest attempt was successful. At Duck Key, only 1 successful year; at Sandy Key, reproductive success better but variable (J. Lorenz pers. comm.). From 1987 to 1992, nest success usually high, except low value of Mayfield estimate in 1990–1991 because half of nests at Tern Key failed during incubation because of disturbance by boats (Bjork and Powell 1994; see Conservation and management: effects of human activity, below). Reproduction is probably limited mostly by food availability; deter-mined by extent of drying in feeding areas during the dry season that concentrates prey into depressions or pools of water, but this information is not conclusive. Annual reproductive success during 11 yr (1974–1975, 1977–1979, 1982–1984, 1986–1992) at Tern Key was negatively impacted by high rainfall and low temperature (≤11°C) during nestling period, but water recession rate and level in foraging habitat not significant (Bjork and Powell 1994; see also Frederick and Spalding 1994). Mechanism of rain and temperature effect on nestling mortality not determined; these effects are difficult to separate (see Causes of mortality, below). However, the colony failed during years with minimal drying shortly after reversals (sharp rise in water level in feeding areas from rain and discharge by water management); suggests sensitivity to rapid dispersal of prey. Poor reproduction at Tern Key during 1995–1997 with high rainfall and water levels, combined with emaciated carcasses of nestlings, suggests starvation from malnutrition because of dispersed prey (J. Lorenz pers. comm.; but see Disease and body parasites, below). Good reproduction in 1987–1992, during drier years, supports this model.

In Texas, annual reproductive success varied significantly among years because of differences in hatching success and nestling mortality (1980 << 1978 and 1979; White et al. 1982). Low success in 1980 due to high percentage of eggs that failed to hatch (42%) because of embryonic death during incubation (31% of unhatched eggs). Embryonic death occurred late in development, but environmental pollutants were not implicated. Egg failure much lower in 1978 and 1979 (21% and 12%, respectively). Differences in nestling mortality among years are most apparent during first 2 wk, accounting for almost all deaths in 1980 to about half in other years. Overall mortality during incubation and nestling periods is similar for 3 yr combined (27 and 32%, respectively), but mortality of nestlings by 2 wk of age (14–32%) is either less than or greater than mortality during incubation. Causes of embryonic death and nestling mortality are unknown (see Causes of mortality, below).

Clutch size did not affect hatching rate or nest success; percentage of nests that hatched ≥1 egg ranged from 80 to 100% for clutches of 1–5 (3 yr combined; White et al. 1982). Proportion of nests that raised ≥1 young to 2 wk of age was 87–100% for clutches of 1–5 (3 yr combined; White et al. 1982). No information on relative survival of nestlings versus order hatched but minor differences in weight gain among nestlings up to 19 d suggest that hatching order is not crucial (see Breeding: young birds, above).

No information on lifetime reproductive success. Unknown what proportion of population breeds in a given year.

Number Of Broods Normally Reared Per Season

Single-brooded; no evidence of multiple broods (White et al. 1982, Bjork and Powell 1994).

Proportion Of Total Females That Rear At Least One Brood To Nest-Leaving Or Independence

No information.

Can live up to ≤ 7 yr, on basis of 1 band recovery (Telfair and Swepston 1987). One individual that was collected as an egg from the wild but raised in captivity was still alive at 29 yr (L. Hudson pers. comm.).

No information on age-specific survival rates after fledging. Postfledging survival of radio-tagged juveniles from 5–7 wk after leaving natal colony was 22% (n = 18; Powell and Bjork 1990; see Range, below). Band recoveries are insufficient for survival estimates (1.3%, 16 recoveries, 910 nestlings banded 1955–1983; Telfair and Swepston 1987).

Diseases

Caused by parasites and toxins; demographic effects are poorly understood in wading birds (Frederick and Spalding 1994). Disease accounted for 20% of nestling deaths in Florida Bay; agent not determined (n = 110 necropsies, 1987–1992; Bjork and Powell 1994, M. Spalding pers. comm.). Helminth parasites (internal) have been well described, but their significance as pathogens has not (Sepulveda et al. 1994). Trematodes (Trematoda) and nematodes (Nematoda) are the most common helminths in Florida Bay; 61% of birds examined had ≥1 species of trematodes, 79% nematodes, 46% cestodes (Cestoda), and 26% acanthocephalans (Acanthocephala; n = 136 [128 nestlings, 7 adults and juveniles, 1 unknown]). Roseate Spoonbill is the only known definitive host for Ascocotyle chandleri, the most common trematode found. Two helminths found on the Roseate are known to be severe pathogens in other avian hosts: the trematode Renicola ralli and the nematode Contracaecum multipapillatum . The latter seems to do well in wet conditions in Florida; may be important source of nestling mortality during wet years. One nestling collected from Florida Bay was infected in its ventriculus with immature individual of the nematode Eustrongylides sp., probably ignotus (Nematoda: Dioctophymidae, n = 128 nestlings; Spalding and Forrester 1993, Sepulveda et al. 1994). Infestations of E. ignotus have severe impacts on reproductive success in herons and egrets; associated with colony abandonment in central Florida (Frederick and Spalding 1994). Presence of Eustrongylides sp. and the nematode Cosmocephalus obvelatus suggests that emaciation may be caused by parasites rather than by malnutrition from low food availability, but malnutrition may also predispose individuals to infection (Frederick and Spalding 1994, Sepulveda et al. 1994).

Body Parasites

Seed tick or large mite burrowed in skin of nestlings up to 6 d old; as many as 50 found on 1 specimen (no species identification), San Antonio Bay, TX (Allen 1942). Caused colony abandonment and losses of young and eggs estimated at 98%, including opportunistic predation by grackles (see Behavior: predation, above, and Appendix 1). Quiescent deutonymphs of 5 species of mites—Hypoderatidae: Phalacrodectes wartoni Fain, Neottialges ajajae, Neottialges montagui, P. parvus, Hypodectes propus bubulci —found in subcutaneous adipose tissue of 2 specimens (Fain 1966, 1967, in Pence and Spalding 1996).

Important sources of mortality for all age groups outside of breeding season are unknown.

Exposure

In Florida Bay, cold, wet conditions re-duce reproductive success significantly; malnutrition (food limitation), exposure, disease are all potential factors (see Disease and body parasites, above). Rain and low temperatures reduce availability of fish by diluting available (not absolute) density or by re-ducing fish activity, respectively (Frederick and Loftus 1993, Bjork and Powell 1994; see Measures of breeding activity: annual and lifetime reproductive success, above). Rain and cold may also kill nestlings directly (Bjork and Powell 1996). No data on thermal tolerance of nestlings. Data comparing colony attendance patterns of parents (i.e., time away presumably feeding, versus incubating or brooding), amount of food brought back, and diet to water level, prey abundance and dispersion, and causes of nestling death during good- and bad-weather years are needed. Malnutrition accounted for 21% of nestling deaths during dry period (n = 110 necropsies, 1987–1992; Bjork and Powell 1994). Mechanical losses from storms caused 30 and 92% loss of nests, eggs, and young in Texas and Florida Bay, respectively (Allen 1942). In Texas, average 13% of eggs disappeared or rolled out of nests (n = 468 eggs at 154 nests; White et al. 1982).

Timing of bad-weather events probably is important for parental tenacity (Bjork and Powell 1994). Unseasonably cold weather in Jan 1973 probably resulted in desertion in Florida Bay (Ogden 1978b). Dry freeze in Dec 1989 during late incubation-hatching stage had little effect; low food demands and strong brooding behavior at this stage increased tenacity (Bjork and Powell 1994).

Predation

Raccoon is probably the most important predator during breeding; limits availability of colony locations and causes colony abandonment (Allen 1942, Frederick and Spalding 1994, R. Paul pers. comm.). In Texas, trapping raccoons and excluding predators at certain colony sites are necessary measures to attract and maintain presence of nesting birds (M. Lange pers. comm., R. Spear pers. comm.). At raccoon-free locations in Florida Bay, predation accounted for 35% of nestling deaths (n = 110 necropsies, 1987–1992; Bjork and Powell 1994). Turkey Vultures are primarily responsible (see Behavior: predation, above); predation is a problem mostly during periods of food shortage, when adults may be away from colonies longer or may be less protective at nests (J. Ogden pers. comm.). Fire ants are responsible for low hatching success in some cases in Texas (Chaney et al. 1978). Distrubance by humans also a factor (see Conservation and management: effects of human activity, below).

Competition With Other Species

No information on mortality resulting from contests over resources.

Initial Dispersal From Natal Site

See Distribution: the Americas, above.

In Texas, sightings of color-marked juveniles south of natal colony (4 band recoveries in Mexico in winter) indicate eventual movement south to Gulf Coast of Mexico; but 1 recovery north and inland (5 total, n = 910 banded plus 622 color-marked, 1955–1983; Telfair and Swepston 1987); see Migration, above.

In Florida Bay, juveniles disperse north fairly rapidly to mainland coast and interior (Robertson et al. 1983, Powell and Bjork 1990); 5 traveled large distances northwest, mostly ≥50 km along Gulf Coast 3–5 d after colony departure; 1 individual reached J. N. (Ding) Darling National Wildlife Refuge, 180 km north, in 5 d (n = 18 radio-tagged individuals; Powell and Bjork 1990). Patagium-marked juveniles traveled longer distances: 2 birds traveled ≥400 km along Atlantic Coast to Vero Beach and Merritt I. National Wildlife Refuge, and 1 bird 320 km along Gulf Coast to Manatee Co. by midsummer (7 juvenile sightings, n = 162 marked; Robertson et al. 1983). Two sightings in Nov and Feb indicate that first-year birds spend winter on coast of southern peninsular Florida. No information on microhabitat of located individuals, group composition, or distances dispersed between fledging and first breeding.

Fidelity To Breeding Site And Winter Home Range

No data.

Dispersal From Breeding Site Or Colony

One observation; an adult renested in Everglades in spring after it had an unsuccessful winter season in Florida Bay—a distance of about 110 km (Frederick and Towles 1995, Bjork and Powell 1996). Renesting probably occurs to Tampa Bay as well, since peaks in number of breeders in Tampa Bay coincide with poor breeding conditions in Florida Bay (R. Paul pers. comm., see Population status: historical trends; recent trends, below).

Home Range

While breeding in Florida Bay, averages 12 km from colony to feeding areas in some years (see Food Habits: feeding above).

Numbers

See Appendix 2 . Probably less abundant in U.S. than in Neotropics (core of range), but few data available, especially from South America, where described as common to abundant in certain regions (Hancock et al. 1992; see Distribution, above). Comparisons between Middle America and s. Florida show roughly similar densities during dry season. In Usumacinta Delta of Mexico, 0.00 bird/km²in 1979 mid-dry season, but 0.08 bird/km²in 1977 late dry season (Ogden et al. 1988). In Atlantic coastal wetlands bordering Honduras and Nicaragua, 0.09 bird/km²and 0.16 bird/km², respectively in 1994 and 1992 (Frederick et al. 1997). In s. Florida, mean monthly densities, 0.1–2.2 birds/km²in 1988–1992 (Bjork and Powell 1994).

U.S. breeding population is about 5,500 pairs (1992, 1996, 1998 combined, depending on region). Surveys outside of U.S. are not comprehensive. In Mexico, 3,230 pairs on Gulf Coast (1971); mean 35.5 and 182.5 pairs on upper Gulf Coast (Rio Bravo south to Tampico, Tampico south to Veracruz, respectively, 1973–1976; Sprunt and Knoder 1980). Mean 155 nests ± 17 SD at Sian Ka’an Biosphere Reserve, Yucatán Peninsula, about 10% of Mexican population (1982–1986; Ornat-Lopez and Ramo 1992). About 500 pairs at Usumacinta Delta (1971–1979; Ogden et al. 1988); 18–50 nests at 2 Pacific Coast locations (1972, 1979; Knoder et al. 1980). In Atlantic coastal wetlands of Honduras and Nicaragua, about 175 and 24 nests, respectively (Frederick et al. 1997). On Pajaros and San Pablo Is. in Costa Rica, 600 pairs; on coast of Suriname, 10 pairs (Luthin 1984). Small numbers (20–30 pairs) bred in n. Belize in 1990 (A. Poole pers. comm.).

Regardless of abundance trends with latitude, species apparently constitutes small proportion of wading-bird population throughout most of range, except possibly in parts of South America. Numbers are low in Mexico compared to species such as Wood Stork or White Ibis (Ogden et al. 1988, Ornat-Lopez and Ramo 1992). Roseates constitute 1.2–1.6% of all wading birds seen on aerial surveys in Nicaragua and Honduras in 1992 and 1994, respectively (Frederick et al. 1997). For U.S., see Portnoy 1977, Lange 1990, Martin and Lester 1991, and Runde 1991 .

Regional Comparisons In The Us

See Appendix 2 . Texas and Louisiana have largest breeding populations; 1996 estimates were 2,901 pairs and 1,505 pairs, respectively. State estimates for Florida about 1,100 pairs (1992 and 1998 totals). Largest nesting concentration in Texas is on upper coast, above 28°50’N (Lange 1990). Estimates suggest that Louisiana has highest mean colony size in U.S. (Martin 1991)—about 200 pairs in 1976–1996; mean range for Texas coast is 44–198 pairs (1973–1990). Ability to support large colonies in Louisiana suggests high-quality habitat adjacent to colony locations (Martin 1991). Mean colony size in Texas and Florida reflects in part spatial constraint of small islands suitable for nesting. Both Texas and Louisiana have maintained large colonies, exceeding 450 pairs; in Florida, only 1 colony located in Florida Bay has had >450 pairs in 2 yr since 1935 (Powell et al. 1989, Bjork and Powell 1994). Florida Bay has largest number of breeders in the state (about 90%) and the largest number of colonies in the state since mid-1930s (Bjork and Powell 1996; see below).

Historical Trends

See Distribution: historical changes, above. Abundance cannot be reconstructed quantitatively prior to early 1930s; records are mostly descriptive and not comprehensive (Allen 1942, Ogden 1978a, Chapman 1982). Regardless, abundance in all 3 states clearly declined during plume-hunting era, which was well under way by the 1870s. Abundance in U.S. was lowest in 1890–1919 (Allen 1942). Recovery most rapid in Texas.

Florida. Historical population trends have been best described for Florida; anecdotal reports indicate that this species was once abundant in s. Florida (Allen 1942), even relative to today. Formerly very large colonies existed on both coasts, south of Cape Romano and Pelican I. in Indian River (Bryant 1859 and Scott 1889, both in Bent 1926). Despite prohibition of plume-hunting (1910 law), and protection of wading-bird colonies (mostly interior southern mainland), Roseates continued to decline into 1930s because still heavily exploited for food in coastal nesting areas (Powell et al. 1989). Probably no more than 200 pairs nested in Florida by the early 1930s; population dropped to historical low of 15–16 pairs by 1941 (Allen 1942, Ogden 1978a, Powell et al. 1989). In contrast, recovery of spoonbills began in Texas at this time (see below), and in Florida other species that were affected by plume-hunting but nested farther in the interior (e.g., Great and Snowy egrets, White Ibis, Wood Stork) recovered rapidly by 1930s because of effective protection (Ogden 1978a, 1994, Powell et al. 1989). Slow initial recovery of spoonbills in Florida Bay through 1950s was due primarily to continued human disturbance (food, many incidental activities), until Florida Bay was added to Everglades National Park in 1960 (J. Ogden pers. comm.). But failure to recover in former nesting habitat (interior of mainland), once protection was achieved, suggests little immigration or recruitment of breeders from outside, with eventual increases throughout state due to production in Florida Bay alone; see discussion of recent trends, below.

Texas. Little information. Breeding population was extirpated by 1895. Large flocks of chiefly immature birds (likely from Mexico) still frequented southern coast up to 1891 (Rhoads 1892 in Allen 1942), but declined thereafter as well. Unclear when upward trend began, but increasing numbers were noted in flocks by 1920 (T. G. Pearson and W. L. Finleys in Allen 1942). Breeding was reestablished in 1923 (Allen 1942); earliest data from 1938 (379 nests at 4 colony sites; Sprunt 1939). By 1941, about 830 pairs existed at 5 colonies from Sabine Pass to Rio Grande (Allen 1942). Relatively rapid recovery in Texas probably was due to immigration from Mexico, protection in Audubon sanctuaries (nesting, feeding, other areas of concentration), and reduction in hunting for meat (Allen 1942).

Louisiana. Record very spotty. Breeding numbers from 1915 to mid-1940s ranged from about 10 to 150 pairs at a few colonies in southwest (Sprunt 1939, Allen 1942, Portnoy 1977). Not clear when recovery began, but significant numbers not reported until 1970s (see Appendix 2).

Recent Trends

See Distribution: historical changes, above, and Appendix 2 .

Florida. Breeding population in Florida Bay from 1940 (15 pairs) through the 1970s grew exponentially, peaking in 1978 at 1,254 breeding pairs at 18 colony sites (Ogden 1978a, Powell et al. 1989). Overall population size has fluctuated since 1970s for reasons that are unclear; numbers down 64% in 1984 compared to peak in 1978, but increased by 50% in 1991–1992 compared to previous 5-yr average (Powell et al. 1989, Bjork and Powell 1996). This increase is attributed to good reproductive success from 1987 to 1992, given recruitment at maturity in 3 yr (Bjork and Powell 1994). Tampa Bay region is the most significant breeding population on mainland since range expansion in 1975 (7 nests; Dunstan 1976). Range expanded when Florida Bay population was peaking, probably as result of production of young and recruitment within Florida (Robertson et al. 1983). Population has doubled since 1992—from 79 pairs to peak of 185 pairs in 1998; most pairs (≥90%) at Alafia Bank every year, fewer pairs (≤14) at Washburn Sanctuary and Tarpon Key (R. Paul pers. comm.). Since range expansion to Kennedy Space Center in 1987 with 1 nest, population there peaked in 1992 at 28 nests, but declined to 5 nests in 1998 (R. Smith and D. Breininger pers. comm.). Roseate Spoonbill has not regained historical breeding abundance on coast of southern peninsula and interior; 12–15 pairs at Water Conservation Area 3A of Everglades from 1992–1998; recent increase to about 50 pairs in 1999 (P. Frederick pers. comm.). Six nests found at Corkscrew Swamp Sanctuary in 1999 (R. Paul pers. comm.). Reoccupation of summer range on Atlantic Coast and interior in 1970s coincided with increases in breeding population in Florida Bay (Robertson et al. 1983, Bjork and Powell 1996).

Texas. Systematic surveys began in 1973 (Blacklock et al. 1978). Trend analysis suggests a decline from 1973 through mid-1980 on all sections of coast: from 3,000 pairs in 1973 to mean of 1,124 pairs (range 1,004–1,256) in 1984–1987 (Lange 1990). Local population in Galveston Bay Estuary declined significantly from 1973 to 1990: mean annual abundance 686 pairs ± 279 SD (r = –0.50, p = 0.03; Gawlik et al. 1998). State-wide estimates for 1991–1996 have fluctuated, making interpretation difficult (mean 1,924 pairs ± 823 SD, range 1,033–2,901, not including 1993); the fact that the highest value occurred in 1996 may indicate upward trend for this period (Texas Colonial Waterbird Census 1973–1996).

Louisiana. Winter and breeding population sizes increased during 1966–1989; annual change of 17.5% from Christmas Bird Count data and 36.4% from Breeding Bird Survey (BBS) data (Fleury and Sherry 1995). These increases may be due to immigration from Texas, at least during 1980s, when numbers declined (–16.9%, p ≤ 0.05; Texas BBS; see above), as well as to locally high reproductive success and recruitment of new breeders because of large area of high-quality foraging habitat (coastal marsh) in Louisiana (Martin 1991, Fleury and Sherry 1995).

Few data. Food availability and extent and quality of foraging habitat probably are important factors limiting population growth, especially reproduction (Martin 1991, Bjork and Powell 1994, Fleury and Sherry 1995, Gawlik et al. 1998; see Measures of breeding activity, above; Population status, above; and Conservation and management: effects of human activity, below). Suitable nesting habitat is a factor; in Texas, for example, creation of dredge-material islands since 1946 has contributed to population increases (Chaney et al. 1978, Chapman 1982). Other factors limiting re-production include exposure, predation, outbreaks of infestations, and disturbance from humans (see Disease and body parasites, above; Causes of mortality, above; and Conservation and management: effects of human activity, below). Postfledging survival poorly known (see Life span and survivorship, above). Roseate Spoonbill can colonize new areas fairly easily, but little in-formation on distances dispersed between breeding attempts, or between natal site and first breeding site, so immigration and emigration rates among populations are unknown (see Range, above).