Protonotaria citrea

Length 14 cm. Relatively large, plump, short-tailed, and long-billed compared to other Parulidae. Eyes are large, dark, and prominent. Male's head and underparts are golden yellow, fading to white undertail coverts; wings are blue-gray, without wing bars; blue-gray tail has large white patches. Female is duller, head less golden. See Kowalski (1986) for information on weights, measurements, and methods of aging.

VOCALIZATIONS: Song is a series of loud, ringing "zweet" notes (NGS 1987).

NEST: cup-shaped hollow of mosses, rootlets, twigs, and leaves, smoothly lined with fine grasses, leaf stems, and feathers; inside diameter 5.1 cm, depth 3.8 cm.

EGGS: average 18.47 x 14.55 mm; oval to short-oval; shell is smooth, somewhat glossy, creamy, boldly and liberally spotted and blotched with brown.

Distinguished from yellow warbler (DENDROICA PETECHIA) by blue-gray wings and uppertail feathers; distinguished from blue-winged warbler (VERMIVORA PINUS) by lack of a black eye line.

BREEDING: Mature deciduous floodplain, river, and swamp forests; wet lowland forest. Primary habitats are almost always near standing water; swamps that are somewhat open with scattered dead stumps are preferred. Bottomland forests and extensive willow thickets near lakes or ponds are also quite suitable. Requires dense underbrush along streambanks (Bushman and Therres 1988). Nests in cavity (natural, old woodpecker hole, bird box, etc.), in snag or living tree, often or always near or over water, at average height of 1.5-3 m (range 0.9-9.8 m); male selects territory, nest site, places some material before female's spring arrival. May be limited by the number of available nesting cavities. See Blem and Blem (1994) for information on composition and microclimate of nests in Virginia. This is the only eastern warbler that nests in tree cavities or other crannies (the only other cavity-nesting wood warbler (Parulidae) is Lucy's warbler [VERMIVORA LUCIAE]). Petit and Petit (1988) provided the first record of a prothonotary warbler using an open-cup nest built by another species (red-winged blackbird [AGELAIUS PHOENICEUS]).

Hamel et al. (1982) provided details on habitat use and suitability in Virginia, North Carolina, South Carolina, Georgia, and Florida. The following five vegetation types used by this species in five physiographic provinces (Piedmont, Sandhills, Inner Coastal Plain, Outer Coastal Plain, and Coastline) are listed in order of decreasing suitability: oak-gum-cypress is suitable at the sapling-poletimber successional stage, but optimal at the late successional sawtimber stage; both bay swamp-pocosin and elm-ash-cottonwood are only marginal at the sapling-poletimber stage and suitable at the sawtimber stage; southern mixed mesic hardwoods are only marginal in both the sapling-poletimber and sawtimber stages; and pond-pine pocosin is marginal at the late successional sawtimber stage only. In all cases, shrubs, midstory canopy and dead trees or limbs are used for all activities (foraging, nesting, perching, roosting, and singing), and snags must have a dbh of at least 15 cm.

Kahl et al. (1985) described habitat around song perches in Missouri as typically level terrain, with a small number of woody stems (< 2.5 cm dbh; < 2800 per ha, never > 4200), short ground vegetation (< 0.20 m, never > 0.36), and a high canopy (16-40 m, never < 12). Other important features included few dead stems (2.5-9.9 cm dbh; < 200 per ha, never > 250) and an intermediate subcanopy closure (30-80%, never < 10 or > 90).

NON-BREEDING: In migration, habitat includes dry woodland, scrub, thickets, and mangroves (AOU 1983). Commonly mist-netted in citrus groves in Belize (Mills 1989). Individuals were mist-netted in second-growth marshes, mangrove scrub, and seasonally flooded low forest in the northeastern coastal region of the Yucatan Peninsula (Ornat and Greenberg 1990). Usually occurs near water and most typically in mangroves in Puerto Rico (Raffaele 1989), though Faaborg and Arendt (1992) reported mist-net captures in dry forest in southwestern Puerto Rico. Records from Colombia include mangroves, freshwater swamps, coastal shrubs, and along the Rio Frio River in the edge of the foothills. The birds usually occur near water, but were also historically noted "in yellow, acacia-like trees on the border of stump land and dry forest, far from water" (Bent 1953).

POPULATION DENSITY: Published information on bird densities from breeding bird censuses in the southeastern U.S. between 1947 and 1979 were summarized by Hamel et al. (1982): mean (se) density was listed as 14.9 (3.9) pairs per 40 ha, with a density range of 6-28 pairs per 40 ha. Accurate measures of population densities from censusing techniques are rare for this species because few studies of this nature occur in swamp forests. Encouragingly, however, there are two studies of cypress-tupelo wetlands that provide similar results: Mitchell and Lancia (1990) found prothonotary warblers to be the most abundant bird species of their studies in South Carolina; the mean number of detections per 25 m radius 10-min point count ranged from 0.11 in a clearcut to 2.68 in the forest interior. Similar results were obtained by R. Sallabanks (unpubl. data) in a study of bottomland hardwood forests along the Roanoke River in North Carolina; the mean number of detections per unlimited radius 10-min point count ranged from 0.73 in levee forest to 2.21 in large patches of cypress-tupelo swamp forest. Again, the prothonotary warbler was the most common species censused. Whitcomb et al. (1981) found 40 males per sq km in their Maryland study area and Fowler and Fowler (1985) report 0.90 birds per 0.8 km segments of the Duck River in middle Tennessee.

TERRITORY SIZE: Territory size seems variable. Mean territory size was 1.48 ha in Michigan (Walkinshaw 1953), 0.50 ha in southern Illinois (Kleen 1973), and 0.48 ha in Tennessee (Petit 1989). Lefebvre et al. (1994) considered this species to be nonterritorial in winter in northeastern Venezuela mangroves. May roost communally during nonbreeding season (Warkentin and Morton 1995).

SITE FIDELITY: In Venezuela, exhibited some fidelity to wintering sites (Lefebvre et al. 1994).

In the mid-Atlantic region, nesting occurs from late April to late June, with a peak from mid-May to mid-June (see Bushman and Therres 1988). Petit (1989) found the mean initiation dates for late nests to be 7 June in 1985, and 13 June in 1987. Clutch size is 3-8 (usually 4-6) (Walkinshaw 1941, 1953; Petit 1989; Blem and Blem 1992). Incubation, by female, typically lasts 12-14 days. Young are tended by both parents, leave nest at 10-11 days. Typically, individual females produce two broods per year, sometimes one, especially in the north. Petit (1989) found most females to lay two clutches, but Blem and Blem (1992) never found the same female incubating two sets of eggs in any single year.

Reproductive success in natural cavities (e.g., Walkinshaw 1941) appears to be significantly lower than in nest-boxes (e.g., Petit 1989). Walkinshaw (1941) reported a hatching success of 38% in Michigan and 61% in Tennessee; Petit (1989) reported a hatching success of 84% and a low nestling mortality of 11%, compared to Walkinshaw's (1941) 33% in Michigan. The difference in nest success appears to be most attributable to regional differences in competition from house wrens (TROGLODYTES AEDON). Life history accounts were provided by Brewster (1878), Loucks (1894), Meyer and Nevius (1943), Bent (1953), Walkinshaw (1953), and Petit (1989).