Northern Prairie Wildlife Research Center
In Ohio, bobolinks nest primarily in grassy hayfields and pastures, clover/alfalfa hayfields, wet prairies, and the grassy margins of marshes (Peterjohn and Rice 1991). Grassy hayfields appear to be the preferred nesting habitat throughout the species' range (Kendeigh 1941, Good and Dambach 1943, Bent 1958, Graber and Graber 1963, Speirs and Orenstein 1965, Kantrud 1981, Gavin 1984, Wootton et al. 1986, Bollinger et al. 1988, Gavin and Bollinger 1988, Bollinger and Gavin 1989, Sullivan 1992, Wootton and Bollinger 1992). Other habitats commonly used by bobolinks for nesting include meadows (Dambach and Good 1940; Bent 1958; Wittenberger 1978,1982; Gavin and Bollinger 1988; Jobanek 1994), old fields (Speirs and Orenstein 1965, Joyner 1978, Sample 1989), native and restored tallgrass prairie (Kendeigh 1941, Blankespoor 1980, Herkert 1991, Volkert 1992, Hartley 1994), pastures (Graber and Graber 1963, Speirs and Orenstein 1965, Wiens 1969, Sample 1989), and small grains (Bent 1958, Graber and Graber 1963, Wiens 1969, Sample 1989). Bobolinks also were reported nesting in annually-tilled cropland (Johnson and Schwartz 1993a), no-tillage cropland (Basore et al. 1986), reclaimed surface mines (Whitmore and Hall 1978, Peterjohn 1989), and Conservation Reserve Program (CRP) grasslands Johnson and Schwartz 1993a,b; Reynolds et al. 1994; Swanson et al. 1995).
Several studies associated bobolinks with tall, dense vegetation (Dambach and Good 1940, Bent 1958, Cody 1968, Wiens 1969, Herkert et al. 1993), whereas others reported a preference for vegetation of intermediate height and density (Tester and Marshall 1961, Rotenberry and Wiens 1980, Sample 1989, Bollinger et al. 1990, Bollinger and Gavin 1992, Herkert 1994a). In North Dakota, Renken and Dinsmore (1987) found a positive relationship between bobolink abundance and vegetation height-density and, in Illinois, Herkert (1994c) reported that bobolink density was positively correlated with mean herbaceous canopy and grass height. Wiens (1969) reported that vegetation height-density was higher in bobolink territories than in territories of any other grassland-nesting bird except the Henslow's sparrow. However, in Wisconsin, bobolink density was negatively correlated with maximum vegetation height and vegetation height-density (Sample 1989).
Bobolink density was positively correlated with percent herbaceous vegetation cover in Wisconsin (Sample 1989), percent forb cover in Illinois (Herkert 1994c), and percent grass cover on CRP grasslands in the north-central United States (Johnson and Schwartz 1993b). Wittenberger (1978, 1980) reported that territories of bachelor males had significantly lower percent forb cover than that of mated males and concluded that optimal bobolink nesting habitat in Oregon was characterized by a high forb:grass ratio. However, in New York, Wootton et al. (1986) found no difference in percent forb cover between bachelor and mated male territories, leading them to conclude that a high grass:forb ratio was representative of optimal bobolink habitat. A preference for grass-dominated communities by nesting bobolinks also was reported by Wiens (1973b), Bollinger et al. (1990), and Bollinger and Gavin (1992). Bobolink abundance was >15 times higher in old (>/=8 years since planting), sparsely-vegetated, grass-dominated hayfields compared to young, densely-vegetated, legume-dominated hayfields in New York (Bollinger et al.1990, Bollinger and Gavin 1992). Dry, sparsely-vegetated habitats (e.g., dry prairie and rowcrops) were avoided by bobolinks in Wisconsin (Sample 1989).
Bobolinks usually nest on the ground at the base of grass or forb clumps (Wiens 1969). Height of nest-site vegetation averaged 10 cm (range: 5-20 cm) in Ontario (Joyner 1978). Bobolink territories in Wisconsin were characterized by 76% grass cover, 22% forb cover, 2% bare ground, and 35-cm mean herbaceous canopy height (Wiens 1969). Mean vegetative characteristics of habitats used by nesting bobolinks in Wisconsin were 2% woody cover, 78% herbaceous vegetation cover, 16% litter cover, 5% bare ground, 63-cm maximum vegetation height, and 22-cm vegetation height-density (Sample 1989). Woody cover in bobolink territories averaged 2% in North Dakota (Renken and Dinsmore 1987). Several studies reported that preferred bobolink nesting habitat included a well developed litter layer (Tester and Marshall 1961, Wiens 1969, Rotenberry and Wiens 1980, Huber and Steuter 1984, Sample 1989).
Important bobolink nestling foods were Lepidopteran and Dipteran larvae (Wiens 1969; Wittenberger 1978, 1980, 1982; Gavin 1984). Forbs exploited by bobolinks for their seeds during the nesting season included dandelion (Taraxacum officinale), cinquefoil (Potentilla spp.), common yarrow (Achillea millefolium), Canada thistle (Cirsium arvense), and dock (Rumex spp.) (Wittenberger 1978, 1980). Bobolinks foraged primarily in grassy cover 6-15 cm tall (Wiens 1969).
Bobolink nesting densities declined significantly after hayfields were abandoned (i.e., not mowed, grazed, burned, or otherwise managed) (Bollinger and Gavin 1992). Bollinger and Gavin (1992) considered brushy fields with > or equal to 25% woody cover unsuitable breeding habitat for bobolinks. They reported that mowing grasslands every 2-3 years during August prevented extensive encroachment of woody vegetation and maintained the relatively sparse, grass-dominated vegetation bobolinks prefer.
Bobolinks depend on new vegetative growth for nesting and foraging, so the presence of large amounts of residual vegetation in spring reduces habitat quality (Wittenberger 1978). Mowing outside the nesting season (after 1 Aug) maintained bobolink habitat in Oregon (Wittenberger 1978) and New York (Bollinger et al. 1990). Bobolink densities were higher on recently mowed than unmowed (for > or equal to 3 growing seasons) grasslands in Illinois (Herkert 1991). Hay-cropping has likely contributed to the decline of eastern bobolink populations. Bobolinks are affected by hay-cropping because they are late nesters (Wiens 1969, Stewart and Kantrud 1972) and nest extensively in hayfields throughout the eastern United States. Mowing- induced nest mortality times over the past 50 years because of the trend toward earlier cutting (Bollinger and Gavin 1992). About 16-23% of the annual productivity of hayfield-nesting bobolinks was destroyed by hay-cropping on a New York study area during 1984-86 (Bollinger et al. 1990). If such trends continue, agricultural habitats, especially hayfields might shift from optimum habitats to reproductive sinks, and bobolink populations will continue to decline in the eastern United States.
Light grazing maintained high-quality bobolink nesting habitat in Oregon (Wittenberger 1978), South Dakota (Huber and Steuter 1984), and North Dakota (Kantrud and Kologiski 1982). However, heavily-grazed grasslands, where most vegetation had been cropped to approximately ground level, were devoid of nesting bobolinks (Kantrud and Kologiski 1982, Bollinger and Gavin 1992). Because of their preference for tall, dense vegetation, Herkert et al. (1993) believed that no grazing would be most beneficial to bobolinks.
Tester and Marshall (1961) and Huber and Steuter (1984) reported that bobolinks were not found on recently burned grasslands because of the reduced litter and vegetation density. However, Herkert (1991, 1994a) found that bobolinks showed a significant preference for recently burned tallgrass prairie in Illinois and avoided grasslands >3 growing seasons post-fire.
Herkert (1991) categorized bobolinks as highly sensitive to habitat fragmentation. In Illinois, bobolinks were found most often on large (>50 ha) grasslands and rarely encountered on tracts <10 ha (Herkert 1991, Herkert et al. 1993). Herkert (1994c) reported that bobolink incidence increased with grassland area and reached 50% at 50 ha. Herkert (1991) considered 10-30 ha the bobolink's minimum area requirement (i.e., minimum amount of contiguous grassland habitat required before an area will be occupied by a species). Bobolink densities were significantly correlated with grassland area in New York (Gavin and Bollinger 1988) and Illinois (Herkert 1994a). Thirteen CRP grasslands occupied by bobolinks in Ohio averaged 12 ha and ranged from 5 to 26 ha (Oh. Div. Wildl., unpubl. data).