Northern Prairie Wildlife Research Center
Swamp Sparrow nests were most abundant in vegetation types where loosestrife was dominant (Table 3). Reinert and Golet (1986) determined that breeding Swamp Sparrows principally required shallow standing water, low (<1.5 m) dense cover, and elevated song-posts, similar to our loosestrife-dominated sites. Swamp Sparrows constructed nests using fine-stemmed sedges and grasses anchored in persistent loosestrife stalks. We also observed Mallard, Blue-winged Teal, Virginia Rail, and Red-winged Blackbird nests at our loosestrife-dominated sites, and found American Bittern, Sedge Wren, Yellow Warbler, Common Yellowthroat, and American Goldfinch breeding based on our criteria. Pied-billed Grebe (Rawinski and Malecki 1984), Least Bittern (Swift et al. 1988), Red-winged Blackbird (Rawinski and Malecki 1984), and American Goldfinch (Kiviat 1996) were observed nesting in loosestrife habitats previous to this study.
Rawinski and Malecki (1984) observed that Marsh Wrens preferred cattail habitats, but Red-winged Blackbirds preferred loosestrife habitats. We also found that nesting Marsh Wrens used cattail habitats, but we observed Red-winged Blackbird nests most frequently in scrub-shrub zones (Table 5). Inconsistencies in vegetation type, period, and year effects (i.e., significant three-way interaction) on Sedge Wren abundance may reflect this species' variable breeding site selection (Table 5). Burns (1982) observed that Sedge Wrens show little site fidelity; this characteristic may be due to the ephemeral nature of wet meadow habitats (Kroodsma and Verner 1978). We believe that Sedge Wren abundance may decline as loosestrife increases in wet meadow canopies. We observed higher areal cover of loosestrife at the wet meadow/loosestrife site compared with the wet meadow/scrub-shrub/loosestrife site and Sedge Wren abundance was significantly higher in two of four sampling periods at the site with less loosestrife (Table 5).
The avian diversity of loosestrife dominated habitats was lower on average than that of other wetland habitats that we surveyed, indicating uneven distributions of fewer species. We found higher avian densities in loosestrife-dominated habitats compared to other vegetation types, although Swamp Sparrows comprised the majority of overall density in loosestrife habitats. Swamp Sparrows accounted for 59% of the overall wet meadow density. Swamp Sparrow densities reported in other studies ranged up to 8.78 individuals/ha (Mowbray 1997) and are considerably lower than our densities in several vegetation types. We observed a significant increase in Swamp Sparrow density between mid- and late periods, which may be explained, in part, by the addition of juveniles from early nests (Peck and James 1987, Beaver 1991, Mowbray 1997). Swamp Sparrows prefer open wetlands of sedges, grasses (i.e., wet meadow), and cattail during the breeding season (Beaver 1991, Mowbray 1997). Principally, loosestrife occurs in the wet meadow, strand, and emergent portions of a typical wetland profile, which are the areas where Swamp Sparrows reach their highest abundance (Beaver 1991, Mowbray 1997).
Nesting female and young Swamp Sparrows satisfy their high protein requirements by consuming invertebrates. Wetherbee (1968) determined that 88% of Swamp Sparrow diets during spring and early summer consisted of insects. Arroll (1995) found that aquatic invertebrate abundance in loosestrife in central Washington was similar to that in cattail and bulrush. Arroll (1995) found only nine statistically significant results in 111 individual comparisons of aquatic invertebrates associated with macrophyte stems (using stem vacuum), sediment (using sediment core), and the water column (using activity traps). Of the four statistically different comparisons involving loosestrife, two showed higher Diptera and Ostracoda abundance in cattail compared with loosestrife, and two showed higher Copepod abundance in loosestrife compared with cattail (Arroll 1995). Thus, invertebrate food items during the breeding season do not appear limiting in loosestrife habitat, although quantitative data from the Northeast are needed.
Loosestrife is an anathema to wetland managers because it often replaces seed-producing mudflat species managed to attract waterfowl. Water level manipulations such as early season drawdowns encourage loosestrife establishment (Thompson 1989). Loosestrife forms dense stands that are difficult for some bird species to negotiate and this may be especially true for larger birds such as waterfowl or species that walk on the ground such as bitterns and rails. Our study demonstrates that loosestrife may provide suitable habitat for some passerines.
Many researchers have observed that habitat diversity leads to faunal diversity in wetlands (Weller and Spatcher 1965, Weller and Fredrickson 1974, Weller 1978, Kantrud and Stewart 1984, Burger 1985). The highest avian density, diversity, and productivity in marshes occurs where emergent vegetation is interspersed 1:1 with open water (Weller and Spatcher 1965, Weller and Fredrickson 1974, Fredrickson and Reid 1988). Wetland managers manipulate vegetative interspersion in marshes using artificial drawdowns, muskrat management, and other means (Fredrickson and Reid 1988). Kaminski and Prince (1981) observed increased waterfowl density and diversity coincident with increased abundance, biomass, and diversity of macroinvertebrates in manipulated emergent wetland habitat. Our loosestrife sites contained few openings. We suspect that manipulated loosestrife habitat (to create interspersion) could result in higher bird diversity.
Loosestrife was widespread in Saginaw Bay coastal wetlands and dominated canopies at several sites. Although diversity was low, loosestrife provided nesting and brood rearing habitat to birds in Saginaw Bay wetlands where alternative habitat choices were available. Some species, such as Marsh Wren, may be disadvantaged as loosestrife displaces other plant forms (e.g., cattail and bulrush). Swamp Sparrows may prefer loosestrife habitat where nest-building materials (fine-stemmed grasses and sedges) are available. We conclude that avian use of loosestrife warrants further quantitative investigation because avian use may be higher than is commonly believed.