Marsh Wren Habitat Model
go to: USFWS Gulf of Maine Watershed Habitat Analysis
go to: Species Table
Feedback: We welcome your suggestions on improving this model!

Draft Date:
February 2002

Species:
Marsh wren, Cistothorus palustris

Use of Study Area Resources:
Reproduction. Marsh wrens breed from ..."central British Columbia across southern Canada to New Brunswick, south to southern California, northern Mexico, Texas, Gulf Coast, and Florida; local breeder in interior. WINTERS: coastal areas of the breeding range and in interior from southern U.S. to southern Mexico" (Hammerson 1994). However, on the eastern seaboard, marsh wrens winter only as far north as the mid-Atlantic states (DeGraff and Rudis 1986, Kroodsma and Verner 1997). Marsh wrens expanded into Maine and New Hampshire in the 1920's, but numbers have declined in the past 20 years (Kroodsma and Verner 1997). In New England they are sympatric with sedge wrens, but use deeper water habitats (Kroodsma and Verner 1997).

Habitat Requirements (Reproduction):
Cover:  Marsh wrens breed in large fresh and brackish marshes (DeGraff and Rudis 1986, Gutzwiller and Anderson 1987), having tall vegetation, such as cattails, bulrushes, or sedges, reeds, cordgrass, needlerush, or even mangroves (Gutzwiller and Anderson 1987). However, they typically avoid nesting in shrubby or forested wetland areas (Gutzwiller and Anderson 1987). Nests usually are built over water with depths of several to 90 cm, although they may nest in salt marsh flooded only during spring tides (Gutzwiller and Anderson 1987). In salt marshes marsh wrens prefer the tall form of cordgrass, which typically grows in mid-tidal depths, along the edges of channels (Kroodsma and Verner 1997).

Wetland locations are likely to provide abundant food sources, some protection from predation, and water used for bathing (Gutzwiller and Anderson 1987).  Nesting success may be greatest in marshes with relatively dense vegetation and deep water (Leonard and Picman 1987 in Hammerson 1994), although some open water seems desirable (Robbins 1994). Denser concentrations of emergent plants facilitate suspension of the nest over the water (Gutzwiller and Anderson 1987). Wetland drainage and filling has been associated with population declines; marsh wrens readily colonize restored or artificial wetlands (Kroodsma and Verner 1997).

Gibbs and Melvin (1990) examined a number of environmental parameters in Maine wetlands used by marsh wrens, and compared these to conditions for wetlands not used.  Significant differences ( P < .01) were found for: area (in ha) of surface water in broad-leaved emergent vegetation (e.g., cattails), area (in ha) of open surface water, area (in ha) covered by water over 0.1 m deep, linear distance (in m) around the edge of the littoral zone, and total area (in ha) of wetland within a 1 km radius. The first four indicate that larger lakes or marshes tend to be preferred to smaller ones; when expressed as a proportion for the specific water body (i.e., percent of surface water that was open or was in emergent vegetation), these were not significantly different for used and unused wetlands. The last parameter indicates that extensive wetlands or wetland complexes are used more than isolated wetlands.

Used wetlands had an average of 16.5 ha of broad-leaved emergent vegetation, averaged 13 ha of open water, a surface water area over 0.1 m deep of 93 ha, 5000 m of marsh edge, and had 116 ha of wetlands within 1 km (Gibbs and Melvin 1990).

Area: Marshes smaller that 0.4 ha are infrequently used by marsh wrens (Gutzwiller and Anderson 1987).

Foraging: Marsh wrens eat insects (predominantly bees, ants, beetles - including aquatic stages, also mosquitoes, bugs, flies, damselflies, etc.) and spiders) taken from marsh vegetation, the marsh floor, and by flycatching (Gutzwiller and Anderson 1987, (Kroodsma and Verner 1997)).

Eggs and nestlings are predated by rats, raccoons, snakes, and other birds; marsh wrens are ready to destroy the nests of conspecifics and of other marsh-nesting birds (Gutzwiller and Anderson 1987, Leonard and Picman 1987 in Hammerson 1994, Kroodsma and Verner 1997). Red-winged or yellow-headed blackbirds may displace marsh wrens into lower suitability wetlands (Gutzwiller and Anderson 1987).

Model:
The Breeding Bird Atlas (BBA) data for Maine, New Hampshire and Massachusetts (courtesy of the University of Vermont COOP Unit) were used to identify the range of the marsh wren within the study area.  BBA blocks in which marsh wrens were known to occur were used to select USDA Forest Service Ecological subunits (Keys et al. 1995); habitat mapping then was restricted to these areas.

Gutzwiller and Anderson (1987) described a habitat suitability index model for the marsh wren, which defined habitat on the basis of vegetation structure (cattails, bullrush, cordgrass preferred to sedges or shrubs), density (the greater the canopy cover the better), water depth (over 10 cm mean depth preferred), and percentage of woody vegetation (the less the better).  Suitable conditions for marsh wrens observed by Gibbs and Melvin (1990) were listed above. The remotely-sensed data available for the study area do not match the specific model variables for either study, and so those variables were interpreted as favoring the cover classes 'palustrine emergent marsh' and the fringing (tall form of cordgrass) portion of 'estuarine emergent marsh'. Fringing estuarine marsh was mapped by selecting only 'regularly flooded' salt marsh (a National Wetlands Inventory modifier), and adding estuarine emergent vegetation within 30 m of channels.

Areas with these habitat types adjoining upland or open water were assumed to have inadequate water depths or vegetation density, respectively, and had their scores reduced by half (0.5). Heterogenous marshes (cells within 30 m of shrub or forest vegetation) also had their scores reduced by half (0.5). Marshes with less than 0.5 ha of emergent vegetation were eliminated. Marshes > 16 ha were scored higher (1.0) than those of intermediate size (0.5).

NWI Designations (wetlands only)	Cover Types	Cover Suitability (0 - 1 scale)
	Upland deciduous forest
	Upland coniferous forest
	Upland mixed forest
	Grassland
	Upland scrub/shrub
	Cultivated
	Developed
	Bare ground
PEM, L2EM	Lake/pond, emergent vegetation	1.0*
PFOcon	Palustrine forest, conifer
PFOdec	Palustrine forest, deciduous
PSSdec	Palustrine scrub shrub, deciduous
PSScon	Palustrine scrub shrub, conifer
PAB, L2AB	Lake/pond, aquatic vegetation
L1UB, PUB	Lake/pond, unconsolidated bottom
L2US	Lake, unconsolidated shore
L2RS	Lake, rocky shore
R1UB	Riverine subtidal unconsolidated
Rper	Riverine perennial
E1AB	Estuarine subtidal vegetated
E1UB	Estuarine subtidal unconsolidated bottom
E2AB	Estuarine intertidal algae
E2EM	Estuarine intertidal emergent	1.0*,**
E2RS, R1RS	Estuarine, tidal river rocky shore
E2SS	Estuarine intertidal shrub
E2US	Estuarine intertidal unconsolidated shore
M1AB	Marine subtidal vegetated
M1UB	Marine subtidal unconsolidated bottom
M2AB	Marine intertidal algae
M2RS	Marine intertidal rocky shore
M2US	Marine intertidal unconsolidated shore
NOTES	* score reduced if area, edge condition unsuitable ** tall cordgrass, typically along tidal channels

Model testing: Marsh wren occurrences along Breeding Bird Survey routes were initially used to test the habitat map from the above model. We compared the presence of habitat near these sites to that for a random set of 591 non- open water points within the species range. Only 4 BBS sites had documented marsh wren use; all had mapped habitat. On the other hand, only 106 sites out of the 591 randomly distributed sites had habitat. Birds occurred in areas mapped as having habitat more frequently than would be expected by chance (Chi-square significant at < .00002).

A data set of 118 marsh wren occurrences in southern and central Maine later was obtained from Tom Hodgman (Maine Department of Inland Fisheries and Wildlife). These were results of 1998, 1999, and 2000 field surveys of inland and coastal marshes. Mapped habitat intersected 93 of the 118 sites, in comparison with 28 of 590 equivalent randomly chosen non- open water sites in the model's range, a highly significant difference.

Sources:

DeGraaf, R.M. and D.D. Rudis. 1986. New England Wildlife: Habitat, Natural History and Distribution. USDA Technical Report NE-108.

Gibbs, J.P and S.M. Melvin. 1990. An assessment of wading birds and other wetlands avifauna and their habitats in Maine. Report to Endangered and Nongame Wildlife Grants Program, Maine Department of Inland Fisheries and Wildlife, Bangor, ME. 61 pp.

Gutzwiller, K.J., and S.H. Anderson. 1987. Habitat suitability index models: marsh wren. U.S. Fish Wildl. Serv. Biol. Rep. 82(10.139). 13 pp.

Hammerson, G. 1994. Cistothorus palustris - (Wilson, 1810) MARSH WREN. NatureServe Explorer: An online encyclopedia of life [web application]. 2001. Version 1.6 . Arlington, Virginia, USA: NatureServe. Available: http://www.natureserve.org/explorer. (Accessed: February 8, 2002 ).

Keys, J.E., Jr., J.C. Carpenter, S. Hooks, F. Koenig, W.H. McNab, W. Russell and W. Smith. 1995. Ecological units of the eastern United States - first approximation (map and booklet of map unit tables), USDA Forest Service. Atlanta, GA.

Kroodsma, D.E. and J.V. Verner. 1997. Marsh wren, Cistothorus palustris. In A. Poole and F. Gill, (eds.) The Birds of North America, No. 308. The Academy of Natural Sciences, Philadelphia, and The American Ornithologists' Union, Washington, D.C.

Robbins, G.C. 1994. Marsh Wrens (Cistothorus palustris). Pg. 206 in Atlas of Breeding Birds in New Hampshire (C. Foss, ed.). Arcadia, Dover, NH. 414pp.