2007 NORTHEASTERN NATURALIST 14(3):361–374 Seasonal Use of Recently Fenced Agricultural Riparian Habitat by Avifauna in Pennsylvania David G. Argent1,* and Roberta J. Zwier2 Abstract - Streambank fencing is increasingly used to exclude livestock from riparian corridors and to enhance biological communities. Our study examined vegetative change and avian-community use of recently fenced agricultural habitat. We conducted strip-transect surveys to census bird communities, line-transect and plot surveys to assess vegetation, and intensive nest monitoring to gauge use and reproductive success across 12 fenced riparian sites in southwestern Pennsylvania. Selected sites varied in age from 3 to 8 years since fencing and averaged 21 m in width. We found avian use was significantly greater in spring than in fall across our fenced sites. We determined that canopy cover, shrub cover, and herbaceous ground cover could predict various attributes of the avian community present within the fenced riparian areas. Our results also suggest that the avian community has greater species richness within sites containing greater habitat complexity, and that these sites are important breeding and nesting areas. Among the 145 nests monitored, 38% successfully fledged young. We found no differences in distance to corridor edge between successful nests and nests that failed. Our study confirms that riparian renovation efforts do have conservation value for both migratory and resident birds. Introduction As transition zones between aquatic and terrestrial systems, riparian areas often maintain diverse biological communities when compared to other portions of the landscape (Bub 2004). These areas can be important to both breeding and migratory birds (Bub 2004, Murray and Stauffer 1995). In arid regions of the United States, riparian habitats often support significantly higher species diversity than surrounding uplands because of their increased vegetative structure (Bub 2004). Disturbance to these transitional habitats, however, can have implications beyond the plant community, affecting both terrestrial and aquatic fauna. Grazing practices, which permit cattle unlimited access to streams, often provide a significant disturbance to these corridors. Structural vegetative diversity can be lost as cattle trample, rub against, graze, and browse streamside vegetation (Ammon and Stacey 1997, Kaufman and Krueger 1984). Changes in vegetative structure often result in shifts of the avian community by reducing the quality and availability of foraging habitat, cover, and nesting substrate (Taylor 1988). Moreover, nests in grazed pastures often have increased rates of failure because cattle trample or otherwise destroy nests or nestlings (Paine et al. 1996). Reduced vertical foliage may also 1Department of Biological and Environmental Sciences, California University of Pennsylvania, 250 University Avenue, California, PA 15419. 2AMEC Earth and Environmental, 285 Davidson Avenue, Suite 100, Somerset, NJ 08873. *Corresponding author - argent@cup.edu. 362 Northeastern Naturalist Vol. 14, No. 3 make birds more vulnerable to predation because of reductions in cover (Ammon and Stacey 1997). Streambank fencing has been used to reduce effects of livestock on riparian areas. In arid western riparian systems, this relatively low-cost renovation practice has yielded positive responses by the vegetative and avian community after cattle have been excluded (Dobkin et al. 1998, Farley et al. 1994, Knopf et al. 1988, Taylor 1988). In the eastern United States, where water is typically not a limiting factor, the importance of riparian areas as avian habitat is not well understood (Bub 2004). In Pennsylvania, streambank fencing was first implemented in the mid-1980s on farms in the Chesapeake Bay watershed, primarily to address concerns for degraded water quality (Hafner and Brittingham 1993). During the mid-1990s, the US Fish and Wildlife Service expanded streambank fencing programs to areas outside of the Chesapeake Bay watershed, and it was during this time that renovation efforts accelerated in southwestern Pennsylvania. Few published studies from the northeastern United States exist that document the effects of livestock exclusion on avian and vegetative communities. Furthermore, those studies are somewhat contradictory (Hafner and Brittingham 1993, Popotnik and Giuliano 2000) and report responses over short durations, only 1–2 years after fencing has been installed. We sought to assess the response of avian and vegetation communities 3–8 years after cattle had been excluded from the riparian zones. Our specific objectives were to determine: (1) if vegetative community structure and avian community composition were related to the time since fencing was installed, (2) if differences existed in use of riparian corridors between spring and fall seasons, and (3) if avian nesting was successful within fenced riparian sites. Methods We conducted our study in the West Pike Run, Fishpot Run, and Maple Creek watersheds within the Ohio River drainage basin in Washington County, PA during 2002. These watersheds, although heavily forested, do support some residential and light industry; however, the predominant landuse disturbance is cattle grazing for beef production. Row crops are a secondary agricultural practice. We selected 12 study sites on first- and second-order streams within pastures where riparian fencing had been installed 3 to 8 years earlier to exclude cattle from the stream corridor. We delineated the boundary of each fenced riparian corridor using a Trimble® ProXR portable global positioning system (GPS). Coordinates were later incorporated into ArcView® 3.2 to determine site dimensions (length, width, and area). The length of the riparian corridor at each site varied, depending upon the linear length of stream that traversed the cattle pasture, whereas the width of each corridor was largely a function of the willingness of landowners to concede pasture. Vegetation characteristics at each site varied depending on factors such as time that had passed since fencing was installed, grazing intensity prior to cattle exclusion, site topography, and site history (Zwier 2003). 2007 D.G. Argent and R.J. Zwier 363 Vegetation community To determine shrub cover, we established one 200-m central transect parallel to and centered on the axis of the fenced riparian corridor. The starting point for each 200-m transect was determined randomly by tossing a plastic stake within the corridor while looking down to the ground. Twenty vegetation transects were established at 10-m intervals, centered within the corridor, and running perpendicular to the corridor axis (Dobkin et al. 1998). If the corridor was less than 10-m wide at the vegetation transect, the entire width of the corridor was evaluated excluding the portion of maintained vegetation under the fence. We determined percent shrub cover as length of the perpendicular transect crossed by live shrubs per total length of perpendicular transect at each site (Dobkin et al. 1998). Maximum height of each shrub intersected by a transect was measured using a pole marked at 1-cm intervals. We defined shrubs as woody species >0.5-m tall but <3.0-m tall. Woody species 􀂕3.0-m tall were considered part of the overstory. Canopy cover was determined every 5 m using a Geographic Resource Solutions densitometer along the entire length of each corridor. Percent canopy cover was then expressed as the number of trees present per number of sampling points along each transect. We measured percent ground cover and percent cover by forbs, grasses, and bare ground within twenty 0.25-m2 quadrats located within the fenced riparian corridor at each site. Groundcover plots were placed 1 m from the end of each perpendicular transect to either the left or right side of the corridor. Subsequent plots alternated from the left to the right side of the corridor. If a quadrat landed in the stream, the ground cover plot was moved to the opposite side of the perpendicular transect. We measured the maximum height of groundcover in the center of each plot using a pole marked at 1-cm intervals. Avian community We used the strip-transect survey method (Conner and Dickson 1980) to determine bird use of the fenced riparian corridors, from March to June (spring) and September to December (fall). We chose the strip-transect survey method over distance-sampling or point-survey techniques because it allowed us to cover a larger area more quickly and because this technique increased our chances of encounter for birds present at each site (Conner and Dickson 1980, Ralph et al. 1993). When conducting a survey, we noted the species observed, the number observed, their position relative to the fencing (i.e., inside or outside), and their direction (to reduce the probability of double counting). Because birds commonly moved between the fenced riparian corridor and adjacent pastures during surveys, we counted birds as inside or outside the riparian fencing based upon their point of origin. Birds outside the fencing that flew into a corridor, reacting to our movement, were not counted. However, low flying birds (<25 m) that were following or foraging over or adjacent to the fenced corridor were included in our survey. Surveys were performed eight times per season at each site within 30 minutes of sunrise and were terminated by 10:00 am EST, giving a total of 364 Northeastern Naturalist Vol. 14, No. 3 16 site visits during the two seasons for each fenced area. Successive surveys were separated by at least 7 days. We did not conduct surveys in rain, fog, or high winds (>25 km/hr). We searched each study site for nests during the breeding season (late April–early August; Brauning 1992). Telescopic inspection mirrors were used to locate and to monitor nests discovered in the canopy. Upon detection, nests were identified with flagging 3 m away to facilitate later visits. We followed nest-monitoring procedures outlined by Geupel (1993) and Ralph et al. (1993). We checked nests from as great a distance as possible, approached the nest from different directions on subsequent visits, wore rubber boots, and used a circular path to reduce the potential that our activities might lead to nest predation. Disturbed vegetation surrounding the nest was teased back into position once the nest had been checked. We recorded species that occupied the nest, condition of the nest, and number of eggs, nestlings, and fledglings present during each visit. We recorded plant height, plant species, nest height, and distance and direction of the nest from the nearest corridor edge after the nesting effort was completed. Nests were recorded as successful, failed, or undetermined. We considered nests successful if they had not been destroyed and we observed at least one fledgling, adults exhibiting defensive behavior, or signs of fledging (e.g., guano on nest edge). Nest success was listed as undetermined if the nest remained intact, but no fledglings, parents, or signs of fledging were observed by the estimated fledge date. We recorded failed nests as preyed upon, failed due to weather, or unknown. We also noted if brood parasitism by Molothrus ater (Boddaert) (Brown-headed Cowbird) occurred. Data analyses We used Spearman’s rank correlation to evaluate interactions among various characteristics of the vegetation community and analysis of variance (ANOVA) to determine if vegetation and site dimensions differed across renovation years. Species richness and abundance were tabulated for each site during each season. Sturnus vulgaris Linnaeus (European Starling) and large, transient flocks of blackbirds including Agelaius phoeniceus Linnaeus (Redwinged Blackbird) and Quiscalus quiscala Linnaeus (Common Grackle) were not included in the analyses of the bird communities because they were difficult to accurately count in the field, thus potentially biasing the interactions we were attempting to evaluate (Conner and Dickson 1980). Because the area searched at each site varied, we scaled our bird counts by various measures of effort yielding relative abundance (# counted/corridor area) and relative species richness (# of species counted/corridor area). Sites were then compared using ANOVA to determine if differences existed seasonally across renovation years for each avian-community metric. We used regression analyses to evaluate the relationships between avian communities and vegetation to determine the relative importance of each vegetation-community variable. We classified birds into seasonality guilds (Brooks and Croonquist 1990) to assess community composition across the 12 study sites. Seasonalityguild data were used to classify migrants, migratory transients, and resident 2007 D.G. Argent and R.J. Zwier 365 populations. Migratory transients included short-distance migrants and species that winter but do not breed in Pennsylvania, while migrants included long-distance (Neotropical) migrants and species that breed in Pennsylvania (Brauning 1992, Santner et al. 1992). Seasonality-guild data were evaluated using t-tests to determine if the proportion of migrants, migratory transients, and residents differed between spring and fall. Nest density was determined by calculating the number of active nests divided by the area searched. Because sample sizes of nests for individual species were low, we combined nest data for comparison. Nest density was then compared to renovation year using ANOVA for the 12 study sites. T-tests were used to determine if differences existed among proportion of successful nests versus failed nests in relation to distance to edge and nest height. All statistical analyses were performed using Minitab (2000) statistical software, and data were arc-sine square-root transformed where percentage data were analyzed. Results Vegetation community Individual riparian corridors varied in size from 0.24–2.58 ha, and had an average length of 435 m and an average width of 21 m (Zwier 2003). Dominant plant species encountered included Plantanus occidentalis L. (American sycamore), Salix spp. (willow), Alnus spp. (alder), Rosa multiflora Thunb. ex Murr (multiflora rose), Cornus spp. (dogwood), and a variety of grasses and forbs (Zwier 2003). Canopy and shrub cover averaged 40% and 31% across all sites, respectively (Table 1). Mean herbaceous cover among the sites was 76%. Several of the vegetation variables measured were correlated with one another, but renovation year was not related to any of the vegetation variables measured. Avian community Over the course of our study, we counted 4516 birds inside the fenced corridors. Species richness ranged from 35 to 55 between spring and fall among individual sites. We encountered a greater relative species richness and relative abundance of birds during spring than in fall, with the exception of sites fenced in 1999 (Fig. 1). Sites fenced in 1996 maintained higher relative abundance and relative species richness than did all other renovation years, even with comparably smaller corridor areas (Fig. 1). More resident species were found during fall than in spring, whereas more migrants were found in spring than in fall (Fig. 2). However, no relationship was evident between renovation year and presence of any specific seasonality guild (P = 0.80). Table 1. Range of vegetation site characteristics for riparian sites. Renovation Canopy % Shrub % Ground % Mean shrub Mean ground year cover cover cover height (cm) cover height (cm) 1996 25–71 24–40 51–97 148–216 49–101 1997 9–88 12–44 62–98 128–188 38–91 1998 41–67 32–54 44–67 173–181 48–74 1999 2–33 7–33 69–99 161–163 64–88 366 Northeastern Naturalist Vol. 14, No. 3 Habitat relationships Because no relationships were evident a priori across renovation years for measures of avian abundance, richness, and habitat, and because site dimensions were not different across renovation years, we plotted a number of habitat variables against avian population measures, regardless of renovation year. Avian species richness and abundance peaked in response to certain levels and combinations of canopy cover, shrub cover, and herbaceous cover (Figs. 3, 4). Sites containing approximately 60% canopy cover, 35% shrub cover, and 70% ground cover yielded the highest species richness and abundance. However, we did not observe a relationship between avian communities and measures of vegetative height. Nesting success Over the course of our study, we located 145 nests of 17 different species (Table 2). Of these nests, 55 (38%) successfully fledged at least one nestling, 48 (33%) were preyed upon, 20 (14%) failed due to weather or flooding, and Figure 1. Mean relative species richness (A) and abundance (B) in comparison to mean corridor area across renovation year. 2007 D.G. Argent and R.J. Zwier 367 22 (15%) had undetermined fates. None of the nests monitored were parasitized by Brown-headed Cowbirds even though we did detect this species five times during our surveys. Nest density was not related to renovation year among the 12 study sites (P = 0.90). Fifty-four percent of nests we monitored were located in multiflora rose, while 22% were located in nest boxes designed to attract Sialia sialis (Linnaeus) (Eastern Bluebird). Twenty percent of nests monitored were in other shrub or tree species, including Crataegus spp. (hawthorne), Rubus spp. (blackberry or raspberry), and Lonicera spp. (honeysuckle). Four percent of nests we monitored were in herbaceous vegetation or on the ground. The majority of nests (96%) were between 0.25 and 3.0 m above the ground, and 3% were on the ground. The remaining 1% of nests monitored were > 3.0 m above the ground, in trees or shrubs. No difference (P = 0.20) was detected between Figure 2. Seasonality guild comparison of avifauna by season (A = spring and B = fall). 368 Northeastern Naturalist Vol. 14, No. 3 Figure 3. Avian species richness compared with proportional vegetation measures. Spring = diamond, fall = square. 2007 D.G. Argent and R.J. Zwier 369 height of successful nests (mean = 129 cm, SE = 11) and height of failed nests (mean = 105 cm, SE = 11). Also, no difference (P = 0.05) was detected between Figure 4. Avian abundance compared with proportional vegetation measures. Spring = diamond, fall = square. 370 Northeastern Naturalist Vol. 14, No. 3 distance of successful nests from the corridor edge (mean = 6.1 m, SE = 1.5) and distance of failed nests from the corridor edge (mean = 7.5 m, SE = 1.3). Discussion Our study was designed to assess seasonal use by avian communities of recently fenced riparian areas in agricultural sites. Even though we found no differences among sites with respect to renovation year (time since fencing), we have demonstrated that avifauna use these habitats seasonally and with differing frequency between spring and fall. Perhaps more important is the use of these habitats as breeding/nesting grounds and the relationships of avian richness and density to various measures of vegetative cover. Vegetation community We expected the revegetation process to follow an early successional sequence: annual weeds and herbaceous perennials would dominate more recently fenced sites, whereas shrubs and saplings would begin to colonize the older sites (Bergon et al. 1990). These changes should have been detected as obvious relationships between time since renovation and the vegetation community variables of percent canopy cover, percent shrub cover, or percent herbaceous cover, but this did not occur. These relationships were likely not evident because habitat conditions at the time of renovation were not necessarily equal among sites and because we had relatively low sample sizes for each time period. At the time of fencing, some sites had an established canopy and/or shrub layer, but grazing pressure was so intense at other sites that little to no vegetation was present when riparian fencing was installed. Two of our 1996 renovation year sites were in close proximity to wooded areas, thus providing an additional source of cover for birds. Table 2. Number of nests located within renovated riparian corridors during summer 2002. Species Number of nests American Robin, Turdus migratorius 32 Red-winged Blackbird, Agelaius phoeniceus 23 Eastern Bluebird, Sialia sialis 19 Gray Catbird, Dumetella carolinensis 19 Song Sparrow, Melospiza melodia (Wilson) 8 Willow Flycatcher, Empidonax trailii 15 Yellow Warbler, Dendroica petechia 7 Northern Mockingbird, Mimus polyglottos (Linnaeus) 4 Northern Cardinal, Cardinalis cardinalis (Linnaeus) 2 Chipping Sparrow, Spizella passerina 3 Mourning Dove, Zenaida macroura (Linnaeus) 3 Brown Thrasher, Toxostoma rufum (Linnaeus) 2 Wood Duck, Aix sponsa (Linnaeus) 2 Carolina Wren, Thryothorus ludovicianus (Latham) 1 Tree Swallow, Tachycineta bicolor (Vieillot) 2 Canada Goose, Branta canadensis (Linnaeus) 1 Orchard Oriole, Icterus spurius (Linnaeus) 1 Wild Turkey, Meleagris gallopavo (Linnaeus) 1 2007 D.G. Argent and R.J. Zwier 371 Avian community The presence of birds across all sites varied, regardless of time since renovation. Factors such as food availability and weather conditions can affect the distribution of birds at local and regional scales (Hurlburt and Haskell 2003). Limited food resources could cause individuals to use one habitat type over another or migrate to areas where environmental conditions are more favorable. For example, the Pennsylvania Game Commission reported in a 2002 news release that with the exception of berry production (i.e., blackberry, raspberry, Prunus serotina Ehrh. [black cherry]), dogwood, Sassafras albidum (Nutt.) Nees (sassafras), and Phytolacca americana L. (American pokeweed), wildlife food supplies were considered poor during fall 2002. Vitis spp. (grapes), Malus spp. (apple), Pyrus spp. (pear), and some mast crops were in limited supply, presumably because of late frosts during spring 2002. Dogwood, raspberry, and blackberry were common shrub species at some of our sites. As a result, birds may have been attracted to the fenced riparian sites over adjacent pastures or other nearby habitats because the riparian corridors offered higher quantity and better quality food. Year-round and non-breeding-season resident birds may also alter their behavior because of low temperatures. To conserve energy and maintain body temperatures, wildlife will seek thermal cover provided by vegetation, and vegetated riparian habitats provide this function for many species, including birds (Kaufman and Krueger 1984). Several recent studies have addressed the value of riparian habitat to migratory and breeding birds. Haas (1995) and Machtans et al. (1996) concluded that birds prefer to use buffer strips or corridors for movement rather than move across areas of unsuitable habitat. Buffer strips were also found to be important to resident and migratory individuals who use these habitats for breeding, food, and cover (Farley et al. 1994, Haas 1995, Machtans et al. 1996). Our analysis of seasonality guilds indicated some differences in species richness do exist, especially when examining the migratory subset (Fig. 2). The most common Neotropical migrants nesting in the fenced riparian corridors were Empidonax trailii Audubon (Willow Flycatcher), Dumetella carolinensis Linnaeus (Gray Catbird), and Dendroica petechia Linnaeus (Yellow Warbler). The most common Neotropical migrant species found inside the riparian renovation sites during migration were Dendroica coronata Linnaeus (Yellow-rumped Warbler) and Spizella passerine Bechstein (Chipping Sparrow). Long-distance migrants, short-distance migrants, and non-breeding residents (e.g., Zonotrichia leucophrys Forster [White-crowned Sparrow]) were also found using the fenced riparian sites during the non-breeding season. The results of our study are consistent with Machtans et al. (1996) and Popotnik (1997) and suggest that the fenced riparian sites have important conservation value to both resident and migratory bird species, especially in southwestern Pennsylvania. Avian nesting density and success Avian nest density and nesting success can be affected by factors such as patch size. We did not find nest density to be related to the length, mean corridor width, or area of the sites we studied. These findings are consistent 372 Northeastern Naturalist Vol. 14, No. 3 with those of Popotnik (1997), who found no relationship between the width of fenced riparian corridors and avian nest density. We also did not detect any relationships between the vegetation community variables measured and nest density. Birds nesting in our study sites are likely responding to changes that are occurring in the vegetation community as evidenced by the regression plots comparing proportions of vegetative cover to species richness and abundance. This analysis also demonstrates the potential for management of riparian areas, if species richness and abundance are desired goals among resident and migratory birds. For example, areas containing excessively high or low canopy cover yield fewer birds and species than those of more intermediate values. Edge habitats can act as “ecological traps” for birds because predation rates tend to be higher along habitat edges (Gates and Gysel 1978), especially when edges are abrupt and human-created (Ratti and Reese 1988). Davidson and Knight (2001) found distance from patch edge to be a significant predictor of daily survival rate in Turdus migratorius Linnaeus (American Robin), but not in other species they studied. When we compared distance from corridor edge for fledged nests and failed nests, we found no differences. There are two alternative interpretations for this finding. First, one could conclude that the edge effect is not occurring within the renovation sites we monitored. This interpretation is supported by the findings of Morse and Robinson (1999), who found that distance from edge had no significant effect on predation rates. An alternate explanation is that because of their small size and generally narrow arrangement, our fenced riparian sites, in their entirety, could be functioning as edge habitat (Bayne and Hobson 1997, Davidson and Knight 2001). Many studies evaluating nest success in patchy habitats include documenting the occurrence of nest parasitism by Brown-headed Cowbirds (e.g., Davidson and Knight 2001, Gates and Gysel 1978, Morse and Robinson 1999, Suarez et al. 1997). In these studies, incidence of brood parasitism varied. Morse and Robinson (1999) found parasitism by Brown-headed Cowbirds to be highest in habitats along abrupt agricultural edges. During the course of our study, we observed five Brown-headed Cowbirds at two sites. We did not observe brood parasitism in any of the study nests we monitored; however, in July 2002 we observed an adult Yellow Warbler feeding a Brown-headed Cowbird fledgling approximately 10 m from a fenced riparian corridor. Renovation of riparian habitats of the eastern United States is a relatively new conservation practice. Resource managers recognize the value of riparian buffers for improved water quality, and management recommendations for riparian buffers typically focus on aquatic resources (Bub 2004). Our study suggests that birds are responding to the habitat conditions that have developed inside riparian renovation sites. Given their seasonal presence, it is apparent that birds are using these newly created habitats for breeding and as stopping points along their migratory pathways. The findings of Hafner and Brittingham (1993) and Farley et al. (1994) suggest that a moratorium on grazing must occur for more than 10 years before significant changes in the bird and vegetation communities can be documented. Earnst et al. (2004) 2007 D.G. Argent and R.J. Zwier 373 found significant increases in riparian bird abundance 10 years after grazing had ceased. Our study and others (Dobkin et al. 1998, Farley et al. 1994, Popotnik and Giuliano 2000) documented significant changes in avian abundance and species richness only 2–3 years after grazing was halted. Therefore, we recommend a future study with the objective of documenting the changes that occur in riparian renovation sites as they relate to the process of ecological succession that compares the avian and vegetation communities among control (grazed) sites, newly renovated sites (1–2 years post fencing), middleaged sites (5–10 years post fencing), and older sites (> 10 years post fencing). Acknowledgments This project was funded by grants from the R.K. Mellon Foundation and the Heinz Endowments. We thank the 12 property owners who allowed us to conduct surveys on their land. Thanks to the US Fish and Wildlife Service’s Partners for Fish and Wildlife Program and the Foundation for California University of Pennsylvania for technical and administrative support. We thank J. Baer, M. Benedetto, D. Carter, G. Hutchko, B. Kirr, B. Ludrosky, R. McCone, A. Ostrander, and K. Smyth for assistance with field surveys. Literature Cited Ammon, E.M., and P.B. Stacey. 1997. Avian nest success in relation to past grazing regimes in a montane riparian system. The Condor 99:7–11. Bayne, E.M., and K.A. Hobson. 1997. Comparing the effects of landscape fragmentation by forestry and agriculture on artificial nests. Conservation Biology 63:1418–1429. Bergon, M., J. Harper, and C. Townsend. 1990. Ecology, 2nd edition. Blackwell Scientific Publications, Boston, MA. Brauning, D.W. 1992. Atlas of breeding birds in Pennsylvania. University of Pittsburgh Press, Pittsburgh, PA. Brooks, R.P., and M.J. Croonquist. 1990. Research note: Wetland, habitat, and trophic response guilds for wildlife species in Pennsylvania. Journal of the Pennsylvania Academy of Sciences 64:93–102. Bub, B. 2004. Riparian and upland breeding-bird assemblages along headwater streams in Michigan’s Upper Peninsula. Journal of Wildlife Management 68(2):383–392. Conner, R.N., and J.G. Dickson. 1980. Strip transect sampling and analysis for avian habitat studies. Wildlife Society Bulletin 8:4–10. Davidson, A.S., and R.L. Knight. 2001. Avian nest success and community composition in a western riparian forest. Journal of Wildlife Management 62:334–344 Dobkin, D.S., A.C. Rich, and W.H. Pyle. 1998. Habitat and avifaunal recovery from livestock grazing in a riparian meadow system of the northwestern Great Basin. Conservation Biology 12:209–221. Earnst, S.L., J.A. Ballard, and D.S. Dobkin. 2004. Riparian songbird abundance a decade after cattle removal on Hart Mountain and Sheldon National Wildlife Refuges. USDA Forest Service, Boise, ID. General Technical Report PSWGTR- 191. Farley, G.H., L.M. Ellis, J.N. Stuart, and N.J. Scott, Jr. 1994. Avian species richness in different-aged stands of riparian forest along the Middle Rio Grande, New Mexico. Conservation Biology 8:1098–1108. 374 Northeastern Naturalist Vol. 14, No. 3 Gates, J.E., and L.W. Gysel. 1978. Avian nest dispersion and fledging success in field-forest ecotones. Ecology 59:871–883. Geupel, G.R. 1993. Nest-monitoring plots: Methods for locating nests and monitoring success. Journal of Field Ornithology 64:507–519. Haas, C.A. 1995. Dispersal and use of corridors by birds in wooded patches on an agricultural landscape. Conservation Biology 9:845–854. Hafner, C.L., and M.C. Brittingham. 1993. Evaluation of a stream-bank fencing program in Pennsylvania. Wildlife Society Bulletin 21:307–315. Hurlburt, A.H., and J.P Haskell. 2003. The effects of energy and seasonality on avian richness and community composition. American Naturalist 161:83–97. Kaufman, J.B., and W.C. Krueger. 1984. Livestock impacts on riparian ecosystems and streamside management implications: A review. Journal of Range Management 37:430–438. Knopf, F.L., J.A. Sedgewick, and R.W. Cannon. 1988. Guild structure of a riparian avifauna relative to seasonal cattle grazing. Journal of Wildlife Management 52:280–289. Machtans, C.S., M.A. Villard, and S.J. Hannon. 1996. Use of riparian buffer strips as movement corridors for forest birds. Conservation Biology 10:1366–1379. Minitab. 2000. Minitab statistical software, release 13.2. State College, PA. Morse, S.F., and S.K. Robinson. 1999. Nesting success of a Neotropical migrant in a multiple-use, forested landscape. Conservation Biology 13:327–337. Murray, N.L., and D.F. Stauffer. 1995. Nongame-bird use of habitat in central Appalachian riparian forests. Journal of Wildlife Management 59:78–88. Paine, L., D. Undersander, D. Sample, G. Bartelt, and T. Schatteman. 1996. Cattle trampling of simulated ground nests in rotationally grazed pastures. Journal of Range Management 49:294–300. Popotnik, G. 1997. Avian response to streambank and wetland fencing. M.Sc. Thesis. California University of Pennsylvania, California, PA. Popotnik, G.J., and W. M. Giuliano. 2000. Response of birds to grazing of riparian zones. Journal of Wildlife Management 64:976–982. Ralph, C., G. Geupel, P. Pyle, T. Martin, and D. DeSante. 1993. Handbook of field methods for monitoring landbirds. General Technical Report PSW-GTR-144. Pacific Southwest Research Station, US Department of Agriculture Forest Service, Albany, CA. Ratti, J.T., and K.P. Reese. 1988. Preliminary test of the ecological-trap hypothesis. Journal of Wildlife Management 53:484–491. Santner, S., D. Brauning, G. Schwalbe, and P. Schwalbe. 1992. Annotated list of birds of Pennsylvania. Pennsylvania Biological Survey, Harrisburg, PA. Contribution No. 4. Suarez, A.V., K.S. Pfennig, and S.K. Robinson. 1997. Nesting success of a disturbance- dependent songbird on different kinds of edges. Conservation Biology 11:928–935. Taylor, D.M. 1988. Effects of cattle grazing on passerine birds nesting in riparian habitat. Journal of Range Management 39:254–258. Zwier, R.J. 2003. Avian response to riparian restoration efforts in southwestern Pennsylvania. M.Sc. Thesis. California University of Pennsylvania, California, PA.