Saltmarsh Sparrow

Ammospiza caudacuta


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).

Saltmarsh Sparrow chick that has climbed out of its nest during tidal flooding.

~5 day old Saltmarsh Sparrow chick waiting out high tide above its flooded nest.

© Alyssa Borowske, Connecticut, United States, 24 August 2013

Measures of Breeding Activity

Age at First Breeding

Males and females are able to breed at 1 yr in age.

Range 2–6 eggs reported, with 3–5 most common, and a 4-egg modal clutch size throughout range (Table 4). As noted (see also Breeding: Eggs), 2-egg clutches were not confirmed as a complete clutch size during detailed tracking of egg-laying in New York (JSG, unpublished data), and may result from partial egg loss (195, 128; JSG, unpublished data). Clutch size varied among sites in a study of 23 populations from Maine to southern New Jersey (n = 860 nests), increasing with latitude, but with much variation unexplained (185; Table 4); clutch size distribution from the same study: 2 eggs (6.3%); 3 eggs (31%); 4 eggs (53%); and 5 eggs (9.7%) (n = 839 nests; 188), with a global mean of 3.65 eggs ± 0.74 SD (185), although these numbers are likely slight underestimates due to partial losses. In New York, clutch size distribution (using only clutch sizes found at end of egg laying or during incubation): 2 eggs (1%); 3 eggs (24%); 4 eggs (62%); and 5 eggs (13%) (n = 188 nests). Mean clutch size in New York was 3.9 eggs ± 0.60 SD (n = 187 nests; JSG, unpublished data), and similar to mean clutch size reported in Connecticut, 3.9 eggs ± 0.65 SD (n = 78 nests) (194). Seasonal variation in clutch size in New York is minimal until late July and early August when clutches with 3 eggs are most prevalent (JSG, unpublished data).

Annual and Lifetime Reproductive Success

Until recently, most estimates of life history parameters related to reproductive success were reported from local studies that focused on fledging rates and nest success, rather than on annual or seasonal productivity (200). In the case of Saltmarsh Sparrows, early studies were limited in scope, methods, and geography (reviewed in Greenlaw and Rising [127]; see also Field et al. [201]). Ruskin et al. (185, 193) and Field et al. (201) broke this pattern by examining productivity (and survival in the latter case) across most of the species’ global range using new methods. Some apparently contrasting results reviewed below between recent SHARP studies (see Acknowledgments) and earlier ones (1970s–1980s) in part may reflect a current species–threatening change in status from relatively healthy populations to those suffering now from accumulated impacts of habitat degradation and climate change (188; see Conservation and Management).

Nest success rates have been estimated in multiple studies, the largest and most recent of which (865 nests over 3 years, from 23 sites across seven states) found that daily nest survival ranged from 86–97% depending on site, equivalent to nest survival of 2–48% over the entire nest period (193). An expanded analysis using data from 2011–2015 found an average daily survival probability of 95% across the species range (95% confidence interval [CI]: 94.9–95.6), and reported 57% of nests fail to produce at least one fledgling (K. Ruskin, unpublished data). Rangewide, 29% of nests failed after flooding, 15% after depredation, and 13% to unknown causes (n = 1,416). Seasonal fecundity was relatively consistent within a site, but differed between sites as little as 1 km apart and did not vary systematically with latitude or proximity to the center of the species range (185).

Six earlier studies examined nest success and produced estimates that corrected for exposure, varying from 27–53%, equivalent to daily nest survival rates ranging from 94–97% (Table 5). Previous researchers also found that flooding and predation both contribute importantly to nest losses at various sites across the species range (79). Nest failure due to flooding increases with maximum tide height, extremity of rare flooding events, and decreases over the course of the breeding season, but is not linked to latitude (193).

Predation of nest contents increases at lower latitudes, and is the primary cause of nest failure at the southernmost sites studied (79, 193). In southern New Jersey, 29% of nests failed to predators, 16% to flooding, and 16% to unknown causes (46). In this study, depredation rates declined over the course of the summer and nest fates were similar to those for Seaside Sparrows at the same sites. In New York, 36% of nests (n = 230) were depredated and 20% failed to flooding (JSG). Predators are poorly known, but an artificial nest study in the same New Jersey marshes found that mammalian predators damaged clay eggs at a much higher rate than avian predators (R. A. Longenecker, unpublished data).

Timing of nest initiation relative to the monthly peak tides is a common determinant of nest success (116, 117), highlighting sensitivity to both timing and amplitude of flooding. No evidence, however, that individuals delay first nests in May until after spring tides (176); rather synchronized failure causes synchrony of renesting. Most studies have found that other aspects of nest site selection, including height and composition of surrounding vegetation, are unrelated to nest survival (117, 190, 67), although nest elevation relative to tide height appears to matter (186). Results of many studies on flood failure and nest height may suffer from confounding effects of depredation, which may not be closely related to nest height in the face of a combination of ground and aerial predators.

Females will renest after failure and sometimes raise two broods per year, but frequency not well known. Observed rates of total number of nests initiated/female averaged 1.5 in 1981, 2.3 in 1982 in Rhode Island (176). Rates of clutch initiation in New York (1977–1980) varied from 0.39 to 0.59 clutches/d in an unmodified, mostly low marsh (150). The daily breeding initiation rate across 23 sites was 0.04, but differences between studies make it difficult to compare reported rates. Females also were estimated to wait an average of 11–15 d after completion of a clutch before renesting, with birds more likely to renest later in the season at more northern sites (185). In Rhode Island, females continued to feed fledged young to independence for 15–20 d (176), so the preceding average range evidently included post-fledging loss of broods. Although sample sizes were small, preliminary analysis suggests that renesting can occur as soon as 1–2 d after a nest failure (185, and longer (15–20 d) after a brood fledges and successfully survives to independence (176, 188; JSG). In Rhode Island in the 1980s, females averaged at least 1.0 successful nests/year (based on small sample size) (176), somewhat higher than the estimate of 0.46 ± 0.17 SD (min–max across sites = 0.09–0.78) successful nests/year at sites across the species range in the early 2010s (185). Mean number of young fledged from successful nests in New York was 3.0 ± 0.10 SE (range 1–5, n = 109) (JSG, unpublished data). Model-estimated minimum number of nesting attempts/season varied regionally from 1.06 to 1.36 and increased with latitude across most of the species’ range (185). Mean estimated annual fecundity was 1.27 ± 0.45 SD (range 0.26–2.29) young/female/yr across multiple sites in the early 2010s (185), which compares poorly with historical estimates from New York of 2.65 young/female/yr (1980) to 5.25 young/female/yr (1978) (150). How well historical estimates from single sites can be compared to the more recent rangewide estimates is unclear, but the magnitude of the differences suggests a reduction in reproductive success over the past 30–40 years (see Conservation and Management).

Unknown rates of post-fledging brood loss create uncertainty about evaluations of seasonal success based solely on egg and nestling stages of breeding (200). Of 101 nestling Saltmarsh Sparrows banded in New York in 1978, a late summer program of netting (August–September) found that at least 36% of them survived to independence (150).

No estimates of recruitment rates or lifetime reproductive success.

Life Span and Survivorship


Nest flooding caused by high spring or storm-driven tides and depredation are primary causes of nestling mortality throughout species’ range and are highly dependent on maximum tide height during the nesting period (79, 188; see Causes of Mortality). Nestling “climbing” behavior benefits fledging success in the face of tidal floods because older, stronger nestlings that are partially mobile but still too young to fledge are known to climb into the grass above rising water (See ML90419981) (39, 186). Probability that young survived the nestling period in New York averaged higher (0.56) than egg success to hatching (0.45), but difference was significant in only 1 yr (127; WP, unpublished data). Mean survival rates for offspring while in the nest (egg and chick stages combined) was 0.27 (95% CI: 0.19–0.39), based on 796 nests across species range during early 2010s (SHARP, unpublished data).


Little information on first year survival. At a site in Rhode Island, mean apparent survival of chicks from hatch until their second year was estimated as 0.11% (95% CI: 0.08–0.16), less than half the estimate for adults (38). Of 101 nestlings banded at a site in New York, 36% were recaptured before September of the same year, providing a conservative estimate of immediate post-fledging survival by ignoring the potential for dispersal and non-detection (150). Comparable information is unavailable for other sites making it hard to assess whether these estimates are representative, or what role emigration plays.


Most comprehensive study included 21 breeding sites between Maine and New Jersey during 2010–2014. Mean annual adult survival rates were 0.44 (95% [Bayesian] credible interval: 0.37–0.52) and 0.49 (0.42–0.56) for females and males respectively (201). Site-level differences were found, but were not predicted well by large scale landscape variables such as latitude and marsh size. There also were no strong differences among years, although the timespan of the study was short, limiting the scope to assess temporal variation. An earlier study in Rhode Island, found no difference in apparent survival between sexes but did find differences among years, ranging from 0.27 to 0.66 with an overall mean of 0.40 (38). Estimated annual adult survival of birds captured at overwintering sites in North Carolina during 2006–2010 was 0.52 ± 0.12 SE (97). Estimates of minimum annual adult survival based on return rates from 2 cohorts banded between 1967 and 1977 averaged higher than more recent estimates (females: 0.53, 0.63; males: 0.55, 0.60) (150), although this might reflect local conditions at the single study site where this earlier work was done.

Comparison of Saltmarsh and Seaside sparrows nesting together at the same sites in New Jersey during 2011–2015 showed that the latter had substantially higher apparent adult survival (46). In contrast, an earlier New York study concluded that survival rates of the 2 species in an unaltered low marsh were similar (150). Reasons for these differences are unknown. Weekly survival estimates from sites on the breeding (Connecticut) and wintering (South Carolina) grounds during 2010–2013 did not differ and were sufficiently high (> 0.98 for both seasons, with no differences between sexes) as to suggest that most mortality occurs during migratory periods (202).

Maximum observed lifespan is 10 yr for males and 6 yr for females (127).

Disease and Body Parasites

Little information and no systematic study. Based on observations from bird banders throughout species’ range, there were no confirmed incidences of foot pox or other specific diseases (A. Kocek, R. Longenecker, W. Post, J. Walsh, personal communications; CSE). Occasional, unexplained, injuries include: missing digits, sometimes linked to inflammation as a nestling, followed by blackening and loss (a wing was affected in similar fashion in one chick); inflammation and sores on the abdomen and face (the latter sometimes leading to eye loss); protruding mandible; undiagnosed growths (e.g., a mass that filled and protruded from the furcular area); sunken or missing eyes; and nestlings without feathers (B. Benvenuti, L. Cline, A. Kocek, K. Ruskin, J. Walsh, personal communications).

Ectoparasite loads on adults are variable. Most contemporary banders consulted report seeing chewing lice, ticks, and mites on Saltmarsh Sparrows only occasionally (CSE). Hill (113) examined 12 birds in Massachusetts and found lice Phylopterus subflavescens (Phylopteridae: Phthiraptera) on 4 individuals and an unidentified tick (Acari) on 1 bird. In contrast, Post and Enders (203) found lice on 22 of 51 Saltmarsh Sparrows (43%) that were carefully examined in a New York salt marsh. The mean number of lice/bird was 26.2 (range 15–35), appeared to be highest in June, and was lower in Saltmarsh Sparrow than in Seaside Sparrow, perhaps because the former has a spring molt.

Examination of 23 nests from Connecticut showed Acari (ticks and mites) to be present in all nests (median count = 9, range 2–202) and to comprise about half the total invertebrate fauna of the nests (196). Nests that failed due to flooding had fewer Acari than those with other fates.

No evidence of endoparasites in adults, but inspections were limited to blood smears from 12 individuals (113).

Causes of Mortality

Tidal flooding and depredation are primary causes of mortality of eggs and nestlings (see above), with depredation relatively more important farther south in the species range (193). Greenberg et al. (79) suggested that losses to these causes could be compensatory, but data from sites across the species range provide little evidence for this (193). Tidal flooding of nests generally caused by high spring tides, and so highly predictable and typically synchronous with monthly lunar cycle (116, 117), but also occurs when wind—which is less predictable—causes high storm surge (186). Because most nest flooding is tidal, duration typically < 2 h and both eggs and chicks can survive flooding events (39). Egg mortality typically occurs when eggs are washed from the nest cup. Chick mortality is a direct function of peak tide height (188) and appears more likely for younger nestlings. Eggs may remain viable after some inundation (186), although hatching success may be reduced in some cases (67). Flooding and depredation are typically lethal to the entire nest contents, but partial egg or nestling losses sometimes occur (39, 186). Little direct information on which species depredate eggs and nestlings, or their relative importance, but confirmed or strongly suspected predators include snakes (garter snake), birds (Mallard [Anas platyrhynchos], ciconiid waders, Northern Harrier, gulls, corvids, blackbirds), and mammals (Coyote [Canis latrans], Raccoon [Procyon lotor], cats, rodents) (B. Benvenuti, R. Longenecker, S. Roberts, K. Ruskin, G. Shriver, J. Walsh, personal communications; JSG, CSE). Eggs destroyed by small peck hole found not infrequently in marshes without wrens or other likely suspects, raising the unconfirmed suspicion that other Saltmarsh Sparrow might be responsible (K. Ruskin, personal communication). Other causes include nest abandonment during egg laying, infertility and egg inviability due to embryo death (in a New York study, this occurred in 9.6% of nests containing eggs that hatched; and for 2.2% of total eggs in nests that survived through hatching; JSG), heavy infestation of ectoparasites (1–2 individual nestlings in a nest in New York), starvation (4.8% of nests in one New York study), and nest-tipping (3 cases in New York) (JSG).

Limited information on causes of adult mortality (see also Behavior: Predation). Adult female mortality during nocturnal predatory attacks on nests occurs (New York, JSG, unpublished data); predators unknown, but circumstantial evidence in New York cases suggests Brown Rat (Rattus norvegicus) (JSG). Occasional reports of sharp-tailed sparrows from Northern Harrier pellets (113), including a banded adult Saltmarsh Sparrow from New York (184). Birds are occasionally taken by Osprey (Pandion haliaetus; C. Field, personal communication), Red-tailed Hawk (Buteo jamaicensis; B. Benvenuti, personal communication), and presumably other raptors.

Females were found to have poorer body condition (assessed by mass relative to size) than males during both breeding season and winter, despite carrying more visible fat, suggesting possible carryover effects of uniparental care (202). Nonetheless, no clear link between body condition and survival (202). Little known about other factors that cause adult mortality.202202


Initial Dispersal from Natal Site

During first day following fledging, females and fledglings remain near nest, and females usually quite conspicuous while delivering food to fledglings. Later, fledglings disperse relative to one another and move farther from nest, and adults and young become more cryptic and difficult to locate without telemetry (135). In a Connecticut study, telemetered females generally found within 100 m of nest during first 3 weeks after chicks fledge, suggesting chicks also within this area until independent (135). In New York, 6 fledglings remained within 200 m of nest for about 28 d, after which the distance between nest and point of first detection was: males, mean 282 m (range 27–554, n = 15); females, mean 214 m (range 15–646, n = 13) (WP, unpublished data).

High level of philopatry, with 12% return rate for juveniles in one study, the majority of which were males (38). Limited and patchy distribution of habitat likely contributes to high rate compared to most passerines.

For yearling returnees to natal marsh in New York, mean distance between hatching location and first capture point during recruitment year was: males, 285.9 m ± 178.3 SD (range 27–554, n = 15); females, mean 283 m ± 312.0 SD (range 25–1,266, n = 15). In subsequent years, males were first found an average of 271 m (range 127–556, n = 9) from their hatching sites; females, mean 285 m (range 65–808, n = 4) (WP, unpublished data).

Fidelity to Breeding Site and Overwintering Home Range

High fidelity to nesting areas, with 35% return rates for both sexes, which, after accounting for mortality and non-detection rates, suggests that most adults return to prior breeding site (38). Of > 7,500 birds banded during 2010–2015, few movements detected between study plots, even when only a few km apart within the same marsh system (SHARP, unpublished data). Of 31 birds that moved, only 8 traveled between marshes, with longest detected movements of ~60 km. Between-plot movements more likely for males (25 of 31 cases) and as likely to occur within the same summer as between summers (15 of 31 cases). 4 cases involved birds that were recaptured at their original plot following a movement within the same marsh system, suggest that birds of both sexes may wander quite far while breeding.

Information on fidelity to wintering sites largely limited to captures at high tide roosts. Many birds recaptured at same roosts within and between years, suggesting high fidelity in at least some circumstances. Fewer returns (10%) at sites in southeastern North Carolina (97) than at sites in South Carolina (63%, Shaw 2012; 58%, WP, unpublished data), and Georgia (A. Given, unpublished data); differences in area of marsh from which a roost draws individuals, the availability of alternative roosts, or specifics of trapping effort might account for this difference. After accounting for expected annual mortality, South Carolina results suggest close to 100% fidelity (Shaw 2012). Few documented cases of movement between overwintering sites, even when only a few 100 m apart (Shaw 2012, 97).

Dispersal from Breeding Site

Despite high fidelity, movements between marshes do occur, and more commonly among males, but rarely involve sites more than few km apart (38). SHARP data (unpublished records from Connecticut and Maine) document 7 of ~3,000 birds banded in 2010–2015 as nestlings or fledglings moved at least 20 km between natal and breeding sites. Longer movements between breeding sites up to at least 35 km apart occur (38; SHARP, unpublished data), but few examples despite coordinated banding of thousands of birds over several years (SHARP, unpublished data). For context, far more movements between breeding and wintering sites have been detected (26) than movements of >20 km between breeding sites.

Home Range

Saltmarsh Sparrow home range sizes during breeding season are considerably larger than those reported for other sparrow species, likely due to lack of territorial behavior. Home ranges for both sexes overlap extensively (127, 129). Average core areas (9.6 ha ± 1.7 SE) and home ranges (52.9 ha ± 8.7 SE) of males larger than those for females (5.3 ha ± 1.4 SE and 27.8 ha ± 6.3 SE, respectively) during breeding season at Scarborough Marsh, Maine (n = 77 radio-marked birds). Core areas in coastal New York, based on intensive, marsh wide mapping of point locations of marked birds, were mean of 4.3 ha (range 3.0–5.7) for males and 1.1 ha (0.4–3.1) for females in largely low marsh habitat with high population density. For those individuals that maintained two core areas within a season in Maine, the distance between core areas was 159 m (± 18) for males and 78 m (± 13) for females. Female home ranges in Connecticut marshes during the 2–3 weeks after young fledge much smaller, as birds stay close to nest (mean: 0.51 m, min–max: 0.14–1.06 ha; n = 23 radio-marked birds; 135). Occasional movements of 2–3 km and back during breeding season (SHARP, unpublished data).

Little information on home range sizes during nonbreeding periods, although birds seemingly travel moderately long distances (perhaps km) to roost on elevated marsh islands during high tides.

Population Status


Based on point-count surveys, mean density across the breeding range estimated as 0.17 birds/ha (SE = 0.02) in 2011–2012, ranging from 0.02 birds/ha (SE = 0.01) in coastal Maine to 0.31 birds/ha (SE = 0.07) on Long Island, New York. Densities in southern New England and Long Island marshes were significantly higher than in Maine marshes, with all other regions having intermediate densities (204). Little information on densities during nonbreeding periods, although birds often congregate in irrigation ditches, or on elevated marsh islands during high tides.

Population estimates are available for most of the species’ breeding range (Table 6, excluding only western Chesapeake Bay) and suggest an estimated global population of 53,000 individuals (95% CI: 37,000–69,000), with the largest numbers of breeding birds in coastal New Jersey and Maryland (205, 41). Detailed information on abundance across the overwintering range is not available as there have been no comprehensive regional surveys.


The population is declining at an average of ~9% per year, based on a compilation of multiple historical data sets spanning 1998–2012 (74). Individual analyses of local data sets from several locations spanning the latitudinal range suggests that this pattern is consistent across the species range, although data from the southern half of the range are sparser than from New England. A separate analysis of nesting densities in Connecticut marshes showed a mean decline of 5% per year (206).

Population Regulation

No direct tests, but population models suggest that incidence of nest flooding has a large influence on population size and extinction risk (188). Population projections suggest that tidal flooding likely to cause extinction before habitat becomes limiting through loss. Little evidence for density dependence in current populations, or for a regulatory role at current population densities (C. Field, unpublished data).

Recommended Citation

Greenlaw, J. S., C. S. Elphick, W. Post, and J. D. Rising (2018). Saltmarsh Sparrow (Ammospiza caudacuta), version 2.1. In The Birds of North America (P. G. Rodewald, Editor). Cornell Lab of Ornithology, Ithaca, NY, USA.