Use of a Reclaimed Stripmine by Grassland Nesting Birds in East-Central Ohio

During the 1997-1998 breeding seasons the author examined the nesting success of grassland birds on plots that were mowed prior to the onset of nesting and on unmowed plots on a 3,700 ha reclaimed stripmine in east-central Ohio. Grasshopper sparrows (Ammodramus savannarum), eastern meadowlarks (Sturnella magnci), red-winged blackbirds (Agelaius pboeniceus^), Henslow's sparrows (A. henslowiO, and bobolinks (Dolichonyx oryzivorus^ were the most abundant nesting species on the reclaimed stripmine. No short-eared owl (Asio JTammeus}, Henslow's sparrow, bobolink, or mallard (Anas platyrhynchos) nests were located on the mowed plots. Significantly more nests of all species combined (P <0.05) were found on the unmowed plots, and pairs using mowed plots tended to initiate nesting later in the season. A Mayfield analysis suggests that grasshopper sparrows had the greatest overall nesting success on the reclaimed plots (46%), followed by red-winged blackbirds (30%) and eastern meadowlarks (30%). During both years combined, nests on mowed plots suffered slightly higher predation rates (47%) than did nests on unmowed (39%) and control (38%) plots (P >0.05). These data suggest that early season mowing is detrimental to some grassland bird species on this reclaimed stripmine since it precludes early nesting; however, it appears that Henslow's, savannah, and grasshopper sparrows, and other uncommon or sporadic grassland breeders in Ohio, are benefiting from this expansive, reclaimed surface mine. OHIO J SCI 102 (3):56-62, 2002 INTRODUCTION Surface mining for coal in the Appalachian region of the eastern United States has resulted in the transformation of thousands of hectares of natural areas to significantly altered ecosystems frequently dominated by exotic herbaceous and woody plant species (Whitmore and Hall 1978; Whitmore 1980; Wray and others 1982; Brenner 1984; Brenner and others 1984). These areas often lack the species diversity and ecosystem stability that existed prior to reclamation efforts (Wray and others 1982; Brenner 1985, 1990). Concomitant with the conversion of surface-mined areas in the eastern US from forests to grasslands has been a steady conversion of native tallgrass prairies and, to a lesser extent, hayfields and pastures to agricultural fields (Troutman and others 1979; Farris and Cole 1981; Bollinger and Gavin 1989; Hands and others 1989; Smith 1992; Herkert 1994a). This in turn has led to the steady decline of several grassland bird species in the central and eastern US (Robbins and others 1986; Herkert 1991, 1994a, 1997; Knopf 1994; Swanson 1996; Helzer and Jelinski 1999; Bajema and others 2001). For example, populations of grasshopper sparrows (Ammodramus savannarum), eastern meadowlarks (Sturnella magna), Henslow's sparrows (A. henslowii), and bobolinks (Dolichonyx oryzivorus) have declined by as much as 84%, 74%, 70%, and 59%, respectively, from 1966 through 1991 in Ohio (Swanson 1996). Thus, although reclaimed stripmines are artificial ecosystems, they provide a potential refuge for a variety of grassland birds whose populations are declining and might not othermanuscript received 1 June 2001 and in revised form 26 December 2001 (#01-13). wise be able to survive in the area (see Whitmore and Hall 1978; Whitmore 1980, 1981; Anderson and others 1994; Swanson 1996; Bajema and others 2001). Although populations of grasshopper and Henslow's sparrows have declined steadily in the glaciated portions of Ohio, their numbers have increased in the southeastern, unglaciated counties where nesting habitat has become increasingly common during the past 25 years on reclaimed stripmines (Peterjohn 1989; Peterjohn and Rice 1991). Eastern meadowlark populations in Ohio are most robust on reclaimed stripmines in the southeastern portion of the state (Peterjohn and Rice 1991). Bobolinks are not as common in southeastern Ohio, but they frequently use reclaimed surface mines for nesting (Whitmore and Hall 1978; Peterjohn 1989; Peterjohn and Rice 1991). If islands of reclaimed land are going to provide suitable nesting habitat that will help to maintain sustainable populations of grassland birds, the size of such tracts, as well as a variety of extrinsic factors that influence grassland habitat structure (for example, mowing, haycropping, burning, and grazing) should be considered. The size of a grassland tract has been shown to play a key role in determining which bird species are able to establish breeding populations in a given area (Herkert 1991; Herkert and others 1993; Vickery and others 1994; Bollinger 1995; Swanson 1996; Helzer and Jelinski 1999; Walk and Warner 1999; Bajema and others 2001). Unlike most of the unmined land in southeastern Ohio (most of which is farmland), large portions of reclaimed stripmines in this region are ungrazed, unburned, and unmowed, and ostensibly provide suitable nesting habitat for a variety of grassland birds including those species that are sensitive to habitat fragmentation (for example, Henslow's sparrow; see Swanson 1996). Such sites should OHIO JOURNAL OF SCIENCE I). I. INGOL1) 57 also provide a good arena to study the effects of mowing and hay-cropping on the reproductive success of grassland birds. Numerous studies have demonstrated that mowing and hay-cropping have had adverse affects on the breeding success of a variety of grassland bird species in the midwestern and northeastern US, particularly when the mowing and hay-cropping were done between May and August (Baskett and others 1980; Bollinger 1988; Frawley 1989; Sample 1989; Warner and Etter 1989; Bollinger and others 1990; Frawley and Best 1991; Herkert 1994a,b; Swanson 1996). Few researchers, however, have examined the effects of spring mowing (prior to 1 May) and fall mowing (after 1 August) on habitat use of grassland birds and recruitment of species to grasslands (but, see Herkert 1994a,b). Surprisingly few studies have focused on the reproductive success of grassland birds on reclaimed stripmines versus comparable unmined pasture or farmland. My objectives in this study were to: 1) examine and compare the habitat use, nesting density and reproductive success of grassland-nesting birds on a reclaimed stripmine compared to nearby unmined plots, and 2) test the effects of spring mowing (prior to 1 May) on the differential use of reclaimed grasslands by these birds. Since the grasslands on the reclaimed mine are more contiguous and expansive, I hypothesized that grassland bird diversity would be greater in this area than on the unmined plots, and that grassland birds would be more abundant and initiate nesting earlier on the reclaimed plots that had not been mowed prior to the on nesting compared to mowed plots. STUDY AREA This study was conducted from early May through late July 1997-1998 at the Wilds (International Center for the Preservation of Wild Animals), a 3,700 ha reclaimed stripmine at the intersection of Muskingum, Guernsey, and Noble counties in southeastern Ohio. In addition, data were collected on five unmined pastures (control sites) situated approximately 25 km north of the Wilds. The Wilds was initially stripmined by the Ohio Power Coal Company in the 1940s and 1950s. After a 10to 15-year hiatus, the land was again mined from 1969-1984 and subsequently converted to mostly hilly grasslands with scattered wood lots, drainage ponds and lakes, and a large wetland area. The dominant grasses and herbs on the Wilds include fescue (Festuca spp.), orchard grass (Dactyli glomeratd), timothy grass (Phelum pratense), rye grass (Lolium spp.), brome grass (Bromus inermis), Kentucky blue grass (Poa pratensis), redtop (Agrostis giganted) yellow sweet clover (Melilotus officinalis), and birdsfoot trefoil (Lotus corniculatus). The dominant shrubs and woody species included Lespedeza (Lespedeza serecia), autumn olive (Elaeagnus multiflora), black locusts (Robinia pseudoacacid) and pines (Pinus spp.). Plant species composition on the control sites overlapped to some degree with the Wilds study site. Two of the 5 control sites in 1997 and 3 of the 5 control sites in 1998 were located in hay fields dominated by orchard grass, timothy grass, rye grass, brome grass, fescue, blue grass, red clover (Trifolium repens), and alfalfa (Medicago saliva). The remaining control sites during both years were situated in unmanaged pastures (not mowed or grazed) characterized by a variety of herbaceous dicots including goldenrod (Solidago spp.), ironweecl (Vernonia spp.), mint species, Indian hemp (Apocynum cannabinum), and blackberry (Rubus allegbeniensis), as well as orchard grass, fescue, timothy, and blue grass. One of the control sites in 1997 was a low-lying, unmanaged pasture with some standing water. This plot was dominated by sedges (Carex spp.), goldenrod, Indian hemp, blackberry, poison ivy (Rhus radicans), blue grass, and a patch of cattails (Typba spp.). Because this plot was somewhat unique in its species composition, it was replaced with a hay field in the 1998 study. METHODS Data were collected on fifteen 210 x 90 m randomlychosen, rectangular plots (1.9 ha; 10 on the Wilds and 5 control sites). All 10 mined plots were bordered on at least three sides by more than 50 ha of additional contiguous grasslands with some scattered autumn olive trees. The amount of contiguous grassland surrounding the control plots ranged from 4.0 ha to approximately 20 ha, and was frequently bordered by forests and woodlands on more than one side. Five of the reclaimed mined plots (the same each year) were bush-hogged and the cut vegetation was left lying in the plots during mid-April of both years. The remaining five Wilds plots were left uncut during both years. Both hay field control sites in 1997 and all three hay field control sites in 1998 were cut and raked for hay in late May or early June. We actively searched for nests four to five days each week. Most nests were located by dragging rope (see Higgins and others 1969), although additional nests were found by randomly flushing birds from nests while walking or by following adults feeding nestlings. Each nest was marked with a flag 2.0 m north of the nest, and all nests were checked once or twice weekly. The reproductive success for each bird species was calculated using the Mayfield method (Mayfield 1961, 1975). A nest was considered active if at least one egg was present, and a nesting effort was considered successful if at least one nestling fledged. Since the nestlings of many ground-nesting, grassland birds leave the nest before they are capable of flying, I considered them to have fledged if they remained in the nest at least 8-10 days (see Wheelright and Rising 1993; Vickery 1996), and there were no signs of predation. A nest was considered lost to a predator if the eggs were removed or broken, or if the nest was disturbed and the nestlings disappeared from the nest during their first week. I considered a nest abandoned if I detected no activity during three consecutive visits, or if the eggs were cold or the nestlings were found dead in the nest (see Kershner and Bollinger 1996). I conducted strip transect censuses (see Emlen 1977; Conner and Dickson 1980) between 0600 and 1030 on each of the 15 plots once a week throughout the study (14 transect days in 1997 and 13 transect days in 1998). In order to avoid the potential for observer bias, I was 58 BIRDS ON RECLAIMED STRIPMINE VOL. 102 the only person to collect transect data. Transect lengths were 210 m; transect widths were 45 m on each side of the transect line on all plots. All birds that were seen or heard within the transect boundaries, as well as birds that flew into or out of the transect plot during the census, were counted. Birds that flew over the plot without landing were not counted. Overall relative abundance, pooled from all plots, was calculated using the transect data for each bird species; additionally, bird abundances for each species at the Wilds were determined for mowed and unmowed plots. I used /-tests to test for differences in the numbers of each bird species on mowed versus unmowed plots; however, in some instances in which the data were not normally distributed, I used Mann-Whitney Rank Sum Tests. Since my nest sample size for most species on the Wilds was rather small, I combined the total number of nests for all species in each year and used Chi-square tests to examine whether there were differences in the total number of nests on mowed versus unmowed plots. I used a contingency table Chi-square test to compare the number of nest starts for all species combined before 1 June in each year (an arbitrary date by which most grassland bird pairs had initiated their first clutch) on mowed versus unmowed plots and to the number of nests initiated after 1 June. I examined whether there were differences in the number of nests lost to predators on mowed compared to unmowed and control plots during both years using contingency table Chisquare tests. All statistical analyses were performed with Sigma Stat (version 2.03 on a PC computer). RESULTS Species Abundance Grasshopper sparrows, Henslow's sparrows, eastern meadowlarks, red-winged blackbirds (Agelaius phoeniceus), and bobolinks were the most abundant species on the Wilds during both 1997 and 1998 (Fig. 1). Savannah sparrows (Passerculus sandwichensis) and short-eared owls (Asio flammeus) were found exclusively on the reclaimed plots. Red-winged blackbirds, song sparrows (Melospiza melodia), eastern Meadowlarks, European