Abstract

Grassland birds have declined throughout North America. In the midwestern United States, reclaimed surface mines often provide expanses of contiguous grassland that support grassland birds. However, some reclaimed surface mines often experience severe woody vegetation encroachment, typically by invasive trees and shrubs, including black locust Robinia pseudoacacia, autumn olive Elaeagnus umbellata, and bush honeysuckle Lonicera spp. We conducted point-count surveys to investigate the effects of woody canopy cover and response to treatments of woody vegetation on the abundance of birds. Our treatments were a control, an herbicide application, and an herbicide application followed by cutting and shredding of standing dead woody vegetation. Estimated density of eastern meadowlark Sturnella magna, grasshopper sparrow Ammodramus savannarum, and Henslow's sparrow Centronyx henslowii was 670%, 958%, and 200%, respectively, greater on areas treated with herbicide and shredding and 279%, 666%, and 155%, respectively, greater on areas treated with herbicide-only when compared with control sites. When woody canopy cover increased from 0% to 20%, densities of eastern meadowlark, grasshopper sparrow, and Henslow's sparrow decreased by 83.9%, 74.9%, and 50.7%, respectively. Conversely, densities of eastern towhee Pipilo erythrophthalmus, prairie warbler Setophaga discolor, yellow-breasted chat Icteria virens, and yellow warbler Setophaga petechia increased 67.4%, 57.0%, 34.6%, and 117.7%, respectively, as estimated woody canopy coverage increased from 20% to 60%. Our results showed treating encroaching woody vegetation on reclaimed surface mines with herbicide and shredding increases available habitat used by grassland birds. Maintaining grasslands on reclaimed surface mines at ≤10% woody canopy coverage would be most beneficial to eastern meadowlarks, grasshopper sparrows, and Henslow's sparrows.

Introduction

Grassland birds have declined throughout North America (Vickery and Herkert 2001; Sauer et al. 2017). Primary reasons for the decline of grasslands include conversion to row-crop agriculture, intensifying agricultural practices, and woody encroachment (Matson et al. 1997; Vickery and Herkert 2001; Askins et al. 2007). Restoring and protecting grasslands throughout the midwestern United States provides opportunities to curtail declining grassland bird populations. One often overlooked source of grasslands in the midwestern and Appalachian regions of the United States is reclaimed surface mines (Scott and Lima 2004). Reclaimed surface mines can provide large blocks of contiguous grassland in areas that would otherwise be primarily forested or agricultural landscapes. Prior to mining activities, many areas were typically forested or composed of small farms that often had small grazed or mowed pastures. After mining activities are completed, mining companies are required to restore the land as part of the Surface Mining Control and Reclamation Act of 1977. However, efforts to return reclaimed surface mines to native forest are costly and often fail because of soil compaction and soil condition following mining activities (Brothers 1990; Scott and Lima 2004). Therefore, reclamation efforts following mining activity often include planting large areas to herbaceous grassland vegetation (Brothers 1990; Scott and Lima 2004).

Reclaimed surface mines are primarily restored with cool-season grasses and nonnative herbaceous vegetation; therefore, they may provide large blocks of contiguous habitat (e.g., 100–3,000 ha) for grassland birds such as bobolink Dolichonyx oryzivorus, dickcissel Spiza americana, eastern meadowlark Sturnella magna, grasshopper sparrow Ammodramus savannarum, Henslow's sparrow Centronyx henslowii, and savannah sparrow Passerculus sandwichensis (Brothers 1990; Bajema et al. 2001; DeVault et al. 2002; Ingold 2002; Scott et al. 2002; Mattice et al. 2005). Despite reclaimed surface mines being composed primarily of nonnative grasses and forbs, grassland bird communities and abundance on reclaimed surface mines are comparable to other grasslands in the midwestern United States (Bajema et al. 2001; DeVault et al. 2002; Scott et al. 2002). Additionally, nest survival and nest fidelity on reclaimed surface mines are similar to other grassland types (Monroe and Ritchison 2005; Galligan et al. 2006; Graves et al. 2010; Ingold et al. 2010; Stauffer et al. 2011; Ingold and Dooley 2013).

Similar to other grasslands in the midwestern United States, reclaimed surface mines suffer from woody encroachment, primarily from invasive trees and shrubs species such as autumn olive Elaeagnus umbellata, bush honeysuckle Lonicera spp., and black locust Robinia pseudoacacia. Dense stands of autumn olive were estimated to occupy 12.6% of reclaimed surface mines in Virginia (Oliphant et al. 2017). Within 9 y, woody vegetation density, primarily by black locust, increased 2.6-fold on Pennsylvania reclaimed surface mines (Hill and Diefenbach 2014). Furthermore, existing woody plants can enhance the rate of woody encroachment by providing additional perch sites for birds, which can spread seeds through defecation (Prather et al. 2017). Woody encroachment negatively affects grassland bird incidence, abundance, habitat selection, and nest survival (e.g., Bakker 2003; Grant et al. 2004; Graves et al. 2010; Hill and Diefenbach 2014; Thompson et al. 2014; Lautenbach et al. 2017). As woody encroachment occurs in prairies and other similar ecosystems, bird communities often shift from grassland species toward species associated with shrublands and early-successional forests (Coppedge et al. 2001; Chapman et al. 2004).

To increase habitat available to grassland birds on reclaimed surface mines, the Ohio Department of Natural Resources, Division of Wildlife (ODNR) began treating encroaching woody plants by applying herbicide and shredding standing dead vegetation. Three treatments were evaluated: a control where no treatment occurred (control), application of herbicide (herbicide-only), and application of herbicide followed by shredding to remove standing dead vegetation (herbicide and shredded). Our objectives were threefold. First, we quantified canopy cover before and after treatments at survey points to determine differences before and after treatments. Second, we evaluated different bird species' response to treatments to determine treatment effectiveness. Last, we evaluated bird responses to woody canopy cover to better inform future management efforts on reclaimed surface mines. We hypothesized grassland bird abundance would be greater in treated areas, with areas that received the herbicide and shredded treatment having the greatest abundance. Last, we hypothesized grassland bird abundance would be negatively affected as estimated woody canopy cover reached 30% (Grant et al. 2004).

Study Site

Tri-Valley Wildlife Area (6,146 ha) and Woodbury Wildlife Area (7,789 ha) are managed for the benefit of wildlife in Muskingum and Coshocton counties, Ohio, respectively (ODNR, Office of Real Estate and Lands Management 2019; Figure 1). Approximately 50% of each Wildlife Area was strip-mined for coal prior to reclamation efforts resulting in large, reclaimed grasslands, which ranged in size from 306 ha to 1,200 ha. Mining activity continued on Tri-Valley and Woodbury Wildlife Areas until 2005 and 2000, respectively. However, mine reclamation efforts on treated and surveyed areas as part of this effort were concluded between 1985 and 1990 on both Wildlife Areas.

Figure 1.

Location of Woodbury (North) and Tri-Valley (South) Wildlife Areas in Coshocton and Muskingum Counties, Ohio, where bird surveys were conducted during 2018 to evaluate response to treatments to control woody plant encroachment on reclaimed surface mines.

Figure 1.

Location of Woodbury (North) and Tri-Valley (South) Wildlife Areas in Coshocton and Muskingum Counties, Ohio, where bird surveys were conducted during 2018 to evaluate response to treatments to control woody plant encroachment on reclaimed surface mines.

Vegetation on each Wildlife Area was similar, with both areas dominated by herbaceous species present in the reclaimed grasslands on both areas, including tall fescue Schedonorus arundinaceus, smooth brome Bromus inermis, Kentucky blue grass Poa pratensis, orchard grass Dactylis glomerata, timothy grass Phleum pratense, sericea lespedeza Lespedeza cuneata, clovers Trifolium spp., sweet clovers Melilotus spp., and common teasel Dipsacus laciniatus. There are also areas on each Wildlife Area that were planted to native warm-season grasses by ODNR following acquisition of each Wildlife Area. Big bluestem Andropogon gerardii, indiangrass Sorghastrum nutans, and switchgrass Panicum virgatum were the dominant species established in plantings. The reclaimed surface mines experienced extensive woody encroachment from nonnative shrub species, primarily autumn olive, black locust, and bush honeysuckle.

Methods

Field methods

From 2014 to 2017, we aerially applied the herbicides metsulfuron methyl, triclopyr, and picloram at both Tri-Valley and Woodbury Wildlife Areas. Some areas needed additional herbicide applications to kill woody plants growing in the understory because woody plants in the canopy prevented herbicide from reaching understory plants. Upon completion of aerial herbicide application, we removed standing dead woody vegetation through mechanical cutting and shredding or hand cut on a portion of areas treated with herbicide over a period of several months. All treated areas were mapped by hand and digitized. We treated 157.9 ha with the herbicide and shredding treatment and 160.3 ha with the herbicide-only at Woodbury Wildlife Area. At Tri-Valley Wildlife Area, we treated 351.7 ha using the herbicide and shredding treatment and 231.0 ha using the herbicide-only treatment.

We used point-count surveys to identify bird use of treated and untreated grasslands. Bird surveys began 1–3 y after treatments were initiated. We were unable to implement treatments all in 1 y because of the extensive area and time limitations associated with treatment implementation; therefore, we pooled all treated areas regardless of time-since-treatment to ensure we had an adequate sample size of bird detections. Point-count survey locations were randomly distributed within each treatment type using a generalized random tessellation stratified, spatially balanced sampling design. All point-count locations were a minimum of 150 m from another sampling point to reduce the possibility of double-counting birds. On average the minimum distance between nearest points was 304 m, with only six point-count locations being <200 m from another survey point. Survey locations were oversampled (i.e., we selected 12 extra points/treatment), to maintain spatially balanced survey locations in case a point-count location needed to be removed and replaced. As an example, we removed one point-count location that was in a pond and replaced it with one of the oversampled points within the same treatment.

We conducted point-count surveys between 15 May and 15 June in 2018 at the survey locations. We sampled 28 point-count locations within each treatment (i.e., control, herbicide-only, and herbicide and shredded) at Tri-Valley and Woodbury Wildlife Areas. Each point-count consisted of a 1-min waiting period after arriving at each point-count location followed by a 6-min point-count survey. We began point-count surveys 30 min before sunrise and concluded by 3 h after sunrise, but not when winds exceeded 20 km/h or when there was precipitation. We surveyed each location twice during the study period. During each visit, we recorded all individuals of 20 focal bird species. Our 20 focal bird species included obligate grassland birds (bobolink, dickcissel, eastern meadowlark, grasshopper sparrow, Henslow's sparrow, savannah sparrow, sedge wren Cistothorus platensis, vesper sparrow Pooecetes gramineus, and upland sandpiper Bartramia longicauda), facultative grassland birds (common yellowthroat Geothlypis trichas, field sparrow Spizella pusilla, northern bobwhite Colinus virginianus, ring-necked pheasant Phasianus colchicus), and shrubland or early successional forest birds (brown thrasher Toxostoma rufum, eastern towhee Pipilo erythrophthalmus, gray catbird Dumetella carolinensis, prairie warbler Setophaga discolor, willow flycatcher Empidonax traillii, yellow-breasted chat Icteria virens, and yellow warbler Setophaga petechia). All five observers were trained to identify these 20 focal bird species by sight and sound. We selected focal bird species prior to beginning point-count surveys to ensure observers focused observation efforts on appropriate species. Upon detection of an individual, we recorded species, sex, distance and direction from observer, time of detection, and type of detection (sight or sound). We used laser rangefinders to ensure distances were measured accurately.

Spatial and statistical methods

We estimated woody plant canopy cover within 200 m of each point-count location using the Tree Cover Mapping Tool (Cotillon and Mathis 2016) in ArcGIS 10.3.1 (ESRI, Inc., Redlands, CA) using 1-m-resolution imagery collected in 2013 (pretreatment) and 2017 (posttreatment) through the National Agriculture Imagery Program during each respective growing season. Prior to treatments, woody vegetation was distributed relatively homogenously throughout the reclaimed surface mines on each Wildlife Area. We selected 200 m to evaluate woody canopy cover because 96.0% of bird observations occurred within 200 m of point-count locations. Once canopy coverage was estimated for each point-count location, we calculated an average estimate of canopy coverage for each treatment type. We used two separate one-way analysis of variance (ANOVA) tests to identify differences in canopy coverage among treatments and before and after treatments occurred. If an ANOVA test indicated that there was a difference among treatment means, we conducted a post hoc Tukey honest significant difference (HSD) test.

Following our analysis of woody canopy cover, we evaluated the response of bird species to the treatments. We used the extended hierarchical distance sampling framework to estimate density of bird species (Royle et al. 2004) in the package ‘unmarked' and used the ‘gdistsamp' function (Fiske and Chandler 2011, 2014). The ‘unmarked' package uses the delta method to estimate variance (Powell 2007). We right-truncated the distance data at 5% to exclude outliers and achieve a better fit for detection models (Buckland et al. 2001). We binned detections for all species into 25-m distances because ‘unmarked' requires binning of distances. We selected distance bins with a preliminary viewing of histograms of observation distances, and 25 m was logical for all species surveyed. We modeled detection functions using hazard rate, half-normal, exponential, and uniform functions for each species with the top-ranked detection model selected for each species. We included focal species that exceeded the minimum number of detections required to estimate densities for most species (≥40; Burnham et al. 1980; Buckland et al. 1993, 2001).

We used Akaike's Information Criterion (AIC) to rank models and identify the most parsimonious model for each species for detection (Burnham and Anderson 1998). We selected models using model weights and difference in AIC value. We were interested in assessing response by birds to our treatments, so all models included treatment for the abundance term. We modeled detection as a constant and with univariate models containing calendar day, time after first light, and wind speed. We did not include observer because all five observers were trained and tested on the focal species prior to beginning point-count surveys.

To identify trends between estimated woody canopy cover and different bird species, we used the top-ranked detection model from above and we added the estimated canopy coverage in 2017. We considered a model significant if the 85% confidence intervals for the beta value for estimated percent woody canopy cover for the model did not overlap zero (Arnold 2010). We used the ‘predict' function in ‘unmarked' to estimate species densities relative to woody canopy cover. We conducted all statistical analyses in the R Statistical Environment (version 3.1.2; R Core Team 2014).

Results

Woody canopy coverage was estimated within 200 m at all 84 survey locations using aerial imagery for both 2013 and 2017. Prior to treatments beginning in 2013, there was no difference in average estimated woody canopy coverage between control areas and treatments at the survey points (F2, 83 = 0.20, P = 0.82; Table 1). As expected, there was a difference in average estimated canopy coverage following the treatments (F2, 83 = 23.44, P < 0.001; Table 1). Following treatments, estimated woody canopy coverage decreased 70.4%, on average, for survey points that were treated with herbicide and shredded to remove standing dead vegetation (Table 1). Estimated woody canopy coverage decreased 28.5%, on average, for survey points that were treated with herbicide-only (Table 1). In the untreated control plots, estimated percent woody canopy coverage increased 22.4% on average at survey points (Table 1). The Tukey HSD test revealed that average estimated woody canopy coverage was 0.70- and 2.8-fold greater at the untreated survey points than at survey points treated with herbicide-only and herbicide and shredded, respectively. Additionally, the Tukey HSD test indicated that estimated woody canopy coverage was 1.2-fold greater at survey points treated with herbicide-only than at survey points receiving the herbicide and shredded treatment.

Table 1.

Estimated percent woody canopy coverage within 200 m (656 ft) of point-count locations at Woodbury and Tri-Valley Wildlife Areas in Coshocton and Muskingum counties, Ohio during 2013 and 2017 before (2013) and after (2017) treatments. Treatments were an untreated control (Control), an aerial application of herbicide (Herbicide-only), and an aerial application of herbicide followed by cutting and shredding of standing dead woody vegetation (Herbicide and shredded). Table includes mean, standard error (SE), and lower and upper 95% confidence limits (LCL and UCL, respectively).

Estimated percent woody canopy coverage within 200 m (656 ft) of point-count locations at Woodbury and Tri-Valley Wildlife Areas in Coshocton and Muskingum counties, Ohio during 2013 and 2017 before (2013) and after (2017) treatments. Treatments were an untreated control (Control), an aerial application of herbicide (Herbicide-only), and an aerial application of herbicide followed by cutting and shredding of standing dead woody vegetation (Herbicide and shredded). Table includes mean, standard error (SE), and lower and upper 95% confidence limits (LCL and UCL, respectively).
Estimated percent woody canopy coverage within 200 m (656 ft) of point-count locations at Woodbury and Tri-Valley Wildlife Areas in Coshocton and Muskingum counties, Ohio during 2013 and 2017 before (2013) and after (2017) treatments. Treatments were an untreated control (Control), an aerial application of herbicide (Herbicide-only), and an aerial application of herbicide followed by cutting and shredding of standing dead woody vegetation (Herbicide and shredded). Table includes mean, standard error (SE), and lower and upper 95% confidence limits (LCL and UCL, respectively).

We conducted 168 point-counts at our 84 survey locations in 2018 and detected 1,434 individuals of our focal bird species (Table 2; Data S1, Supplemental material). There were either no detections or too few detections of nine focal bird species, and we excluded those species from analyses (Table 2). The hazard-rate detection function was the highest ranked model for eastern meadowlark, field sparrow, grasshopper sparrow, prairie warbler, and willow flycatcher. Half-normal detection functions fit the data best for common yellowthroat, eastern towhee, Henslow's sparrow, yellow-breasted chat, and yellow warbler. The uniform and exponential detection functions were not good fits for any bird species tested. Only single covariate models were used to estimate effects on detection.

Table 2.

Total number of detections (Total) and detections by treatment type by bird species recorded during surveys at Woodbury and Tri-Valley Wildlife Areas in Ohio during 2018. Treatments were an untreated control (Control), an aerial application of herbicide (Herbicide-only), and an aerial application of herbicide followed by cutting and shredding of standing dead vegetation (Herbicide and shredded).

Total number of detections (Total) and detections by treatment type by bird species recorded during surveys at Woodbury and Tri-Valley Wildlife Areas in Ohio during 2018. Treatments were an untreated control (Control), an aerial application of herbicide (Herbicide-only), and an aerial application of herbicide followed by cutting and shredding of standing dead vegetation (Herbicide and shredded).
Total number of detections (Total) and detections by treatment type by bird species recorded during surveys at Woodbury and Tri-Valley Wildlife Areas in Ohio during 2018. Treatments were an untreated control (Control), an aerial application of herbicide (Herbicide-only), and an aerial application of herbicide followed by cutting and shredding of standing dead vegetation (Herbicide and shredded).

We found the density of eastern meadowlark, grasshopper sparrow, and Henslow's sparrows increased following treatments and achieved greatest densities in areas treated with herbicide and shredding (Table 3). On average, estimated density on herbicide and shredded treatments for eastern meadowlark, grasshopper sparrow, and Henslow's sparrow was 670%, 958%, and 200%, respectively, greater than untreated control sites (Table 3). Average estimated density on herbicide-only treatments for eastern meadowlark, grasshopper sparrow, and Henslow's sparrow was 279%, 666%, and 155%, respectively, greater than untreated control sites (Table 3). In contrast, the estimated densities of eastern towhee, prairie warbler, and yellow-breasted chat were lower in treatments when compared with the control (Table 3). For instance, prairie warbler density was 78% and 65% lower at sites treated with both herbicide and shredding and herbicide-only, respectively. Treatments had little effect on estimated density for common yellowthroat, field sparrow, willow flycatcher, and yellow warbler (Table 3).

Table 3.

Estimated density (Density, birds/ha) with standard error (SE), and lower and upper 95% confidence limits (LCL and UCL, respectively), for 11 bird species following 3 different treatments on reclaimed surface mines on Woodbury and Tri-Valley Wildlife Areas in Coshocton and Muskingum counties, Ohio, during 2018. Treatments included an untreated control, an aerial application of herbicide (herbicide-only), and an aerial application of herbicide followed by cutting and shredding of standing dead woody vegetation (Herbicide and shredded).

Estimated density (Density, birds/ha) with standard error (SE), and lower and upper 95% confidence limits (LCL and UCL, respectively), for 11 bird species following 3 different treatments on reclaimed surface mines on Woodbury and Tri-Valley Wildlife Areas in Coshocton and Muskingum counties, Ohio, during 2018. Treatments included an untreated control, an aerial application of herbicide (herbicide-only), and an aerial application of herbicide followed by cutting and shredding of standing dead woody vegetation (Herbicide and shredded).
Estimated density (Density, birds/ha) with standard error (SE), and lower and upper 95% confidence limits (LCL and UCL, respectively), for 11 bird species following 3 different treatments on reclaimed surface mines on Woodbury and Tri-Valley Wildlife Areas in Coshocton and Muskingum counties, Ohio, during 2018. Treatments included an untreated control, an aerial application of herbicide (herbicide-only), and an aerial application of herbicide followed by cutting and shredding of standing dead woody vegetation (Herbicide and shredded).

Estimated density decreased as estimated woody canopy cover increased at survey points for eastern meadowlarks, grasshopper sparrow, and Henslow's sparrow (Figure 2). Estimated density of eastern meadowlarks, grasshopper sparrow, and Henslow's sparrow decreased 83.9%, 74.9%, and 50.7%, respectively, when estimated woody cover increased from 0% to 20%. In contrast, birds associated with early successional forests and shrublands increased as woody canopy coverage increased (Figure 2). Estimated density increased by 67.4%, 57.0%, 34.6%, and 117.7%, for eastern towhee, prairie warbler, yellow-breasted chat, and yellow warbler, respectively, as woody cover increased from 20% to 60%. Estimated woody canopy cover was not a good predictor for abundance of common yellowthroat or field sparrow.

Figure 2.

Estimated density (with 95% confidence limits in gray) for common yellowthroat Geothlypis trichas, eastern meadowlark Sturnella magna, eastern towhee Pipilo erythrophthalmus, gray catbird Dumetella carolinensis, grasshopper sparrow Ammodramus savannarum, Henslow's sparrow Centronyx henslowii, prairie warbler Setophaga discolor, yellow-breasted chat Icteria virens, and yellow warbler Setophaga petechia as a function of percent woody cover within 200 m of survey locations on Woodbury and Tri-Valley Wildlife Areas in Coshocton and Muskingum Counties, Ohio, during 2018. Only species significant at the 85% confidence level are shown. Note differences in y-axis scales.

Figure 2.

Estimated density (with 95% confidence limits in gray) for common yellowthroat Geothlypis trichas, eastern meadowlark Sturnella magna, eastern towhee Pipilo erythrophthalmus, gray catbird Dumetella carolinensis, grasshopper sparrow Ammodramus savannarum, Henslow's sparrow Centronyx henslowii, prairie warbler Setophaga discolor, yellow-breasted chat Icteria virens, and yellow warbler Setophaga petechia as a function of percent woody cover within 200 m of survey locations on Woodbury and Tri-Valley Wildlife Areas in Coshocton and Muskingum Counties, Ohio, during 2018. Only species significant at the 85% confidence level are shown. Note differences in y-axis scales.

Discussion

Our results corroborate estimates of grassland birds on reclaimed surface mines. Additionally, our results showed that eastern meadowlark, grasshopper sparrow, and Henslow's sparrow abundance decreased with increasing woody canopy coverage on Ohio reclaimed surface mines. Furthermore, eastern meadowlark, grasshopper sparrow, and Henslow's sparrow densities were greater on reclaimed surface mines in areas that received either the herbicide-only treatment or the herbicide and shredding treatment than control areas. However, shrubland bird densities—such as those of the eastern towhee, prairie warbler, and yellow-breasted chat—were lower in areas that received the herbicide and shredding treatment than in our untreated control sites. Last, estimated woody canopy cover increased by 22% at control sites in 4 y, indicating that if left untreated, invasive or aggressive trees and shrubs will continue to expand on Ohio reclaimed surface mines.

Results from our study affirm reclaimed surface mines in Ohio continue to provide habitat for eastern meadowlark, grasshopper sparrow, and Henslow's sparrow, similar to other studies from Ohio (e.g., Ingold 2002; Graves et al. 2010; Ingold and Dooley 2013). Our results are similar to other studies on reclaimed surface mines in the midwestern and Appalachian regions of the United States that recorded populations of breeding grassland birds (e.g., Bajema et al. 2001; DeVault et al. 2002; Scott et al. 2002; Mattice et al. 2005). Following reclamation efforts, reclaimed surface mines often do not support woody plants as a result of soil compaction and condition (Brothers 1990; Scott and Lima 2004; Cavender et al. 2014); however, many reclaimed mines experience woody encroachment as time-since-reclamation increases (e.g., Graves et al. 2010; Stauffer et al. 2011; Hill and Diefenbach 2014; Oliphant et al. 2017; this study).

Our results definitively show that the density of eastern meadowlark, grasshopper sparrow, and Henslow's sparrow decrease with increasing woody canopy coverage on Ohio reclaimed surface mines. These results are similar to other studies that show incidence, occupancy, or abundance of grassland birds decrease as woodland, woody plant density, or tree cover increased (Grant et al. 2004; Hill and Diefenbach 2014; Thompson et al. 2014; Tack et al. 2017). When woody vegetation exceeded 10–20% canopy coverage, we found reduced abundance of eastern meadowlark, grasshopper sparrow, and Henslow's sparrow, contrary to our hypothesis that 30% woody canopy coverage was the threshold where abundance becomes affected. Ideally, grasslands on reclaimed surface mines would be maintained between 0% and 5% woody canopy coverage to maximize the amount of habitat available to birds such as the eastern meadowlark, grasshopper sparrow, and Henslow's sparrow. However, in cases where that goal is not practical, maintaining reclaimed surface mine areas with ≤10% woody canopy coverage would still maintain adequate habitat for grassland birds.

Following both treatments, eastern meadowlark, grasshopper sparrow, and Henslow's sparrow used areas on reclaimed surface mines that formerly were dominated by woody plants. Woody vegetation encroachment reduced occupancy and abundance of grasshopper sparrows and Henslow's sparrows on reclaimed surface mines (Hill and Diefenbach 2014). Our results indicate efforts to remove invasive woody plants, such as autumn olive, on reclaimed surface mines benefit eastern meadowlarks, grasshopper sparrows, and Henslow's sparrows in Ohio. However, other research suggests there may be a delay in the benefit to grassland birds following treatment. For example, Thompson et al. (2016) observed a weak positive and delayed response by grassland birds following woody vegetation removal. Similarly, sagebrush-steppe birds increased in density as time-since-removal increased in areas where invasive conifers were removed (Holmes et al. 2017). Delayed response by grassland birds is likely a result of recovering herbaceous vegetation following removal of woody plants (Thompson et al. 2016).

Density of individuals does not necessarily translate to increased habitat quality (Van Horne 1983; Vickery et al. 1992), it is important to consider other factors as well. Woody vegetation removal increased nest density of bobolinks, eastern meadowlarks, grasshopper sparrows, and Henslow's sparrows, in addition to increasing available habitat (Ellison et al. 2013; Hill and Diefenbach 2013). However, woody vegetation removal did not increase nest survival, indicating that the primary benefit to grassland birds is obtained by restoring or creating available habitat (Ellison et al. 2013; Hill and Diefenbach 2013). Dickcissels, eastern meadowlarks, grasshopper sparrows, and Henslow's sparrows nesting on reclaimed surface mines typically have similar reproduction to other grassland habitats (Monroe and Ritchison 2005; Galligan et al. 2006; Graves et al. 2010; Hill and Diefenbach 2013; Stauffer et al. 2011). This indicates that reclaimed surface mines likely provide similar reproductive benefits to more traditional grasslands.

Predictably, we found some shrubland birds—such as eastern towhees, prairie warblers, and yellow-breasted chats—decreased with woody vegetation removal. Similarly, shrubland and woodland birds decreased following woody vegetation removal in Minnesota (Thompson et al. 2016). Removal of Tamarix ramosissima from riparian corridors in grasslands altered bird communities and reduced occupancy of shrubland and forest birds, while grassland birds were exclusively located in these areas (Cable et al. 2015; Raynor et al. 2017). Conversely, as grasslands succumb to woody encroachment bird communities shift toward shrubland bird species (Coppedge et al. 2001; Chapman et al. 2004). Anytime vegetation is substantially altered, there will likely be changes in bird communities.

Estimated woody canopy increased an average of 22% within 4 y (an average of 5.5% increase each year) in our untreated control sites. Species responsible for increases were primarily autumn olive, black locust, and bush honeysuckle. Without management of grasslands on reclaimed surface mines, these areas will likely continue to experience continued woody plant expansion by autumn olive, black locust, and bush honeysuckle. Spread of these species is likely further exacerbated by providing perch sites for birds, likely dispersing more seeds of these invasive shrubs and trees (Prather et al. 2017).

Management Implications

Reclaimed surface mines in the midwestern and Appalachian regions of the United States provide suitable habitat for grassland birds, such as the eastern meadowlark, grasshopper sparrow, and Henslow's sparrow. However, woody encroachment on reclaimed surface mines often reduces density of grassland birds. Management efforts to remove large stands of existing invasive shrubs are expensive. Estimated costs for woody vegetation treatments in this study were US $840.14/ha for the aerial herbicide application and an additional US $1,235.50/ha for removal of woody vegetation through cutting and shredding (ODNR, unpublished data). Treating invasive trees and shrubs with herbicide at a young age will likely reduce costs by removing the need to cut and shred standing dead vegetation. Cutting and shredding encroaching woody vegetation alone is not an effective technique for controlling black locust, autumn olive, and bush honeysuckle because it often stimulates growth (ODNR, unpublished data). Similarly, prescribed fire, while more economical (US $148.26/ha), is not an effective method of controlling autumn olive, black locust, and bush honeysuckle, often stimulating rhizomous stems or seed germination of these invasive shrubs (Anderson and Brown 1980; Szafoni 1991; Dibble et al. 2008; but see Emery et al. 2011). We believe removal of woody plants from grasslands on reclaimed surface mines will be most successful if managers use additional follow-up treatments to maximize the return on investment following initial treatments (Thompson et al. 2016). Spot-treating regenerating shrubs in areas that received treatments earlier may be an effective control method and reduce the need for future costly aerial herbicide application and shredding of standing dead woody vegetation.

Supplemental Material

Please note: The Journal of Fish and Wildlife Management is not responsible for the content or functionality of any supplemental material. Queries should be directed to the corresponding author for the article.

Data S1. Species observations for point-counts conducted between 15 May and 15 June 2018 at Woodbury and Tri-Valley Wildlife Areas in Coshocton and Muskingum counties, Ohio, respectively, following an herbicide application (herbicide-only), herbicide followed by mechanical shredding of dead woody vegetation (herbicide and shredded), and a control. The observations are in Excel format with two tabs (labeled “detections” and “site covariates”). The “detections” tab includes a column for wildlife area (Area), point identification (Point), species of bird (using the four-letter codes from the American Ornithological Society checklist to birds of North and Middle America), sex, whether the bird was a flyover (1) or not (0), visit, distance in meters, and how the bird was first detected (0 if seen first, 1 if heard first). The “site covariates” tab includes a point identification (PointID), Wildlife Area (Area), treatment (Control, HerbicideShredded, and HerbicideOnly), estimated woody cover in 2013 and 2017 (WoodyCover2013 and WoodyCover2017, respectively), wind for each visit using the Beaufort scale, day of each visit (using the calendar day), and minutes after first light (e.g., 0.5 h before local sunrise) as a start time of the survey for each visit (minLight1, minLight2).

Found at DOI: https://doi.org/10.3996/062019-JFWM-053.S1 (78 KB XLSX).

Reference S1. Anderson RC, Brown LE. 1980. Influence of a prescribed burn on colonizing black locust. Pages 330–336 in Garrett HE, Cox GS, editors. Proceedings of the Third Central Hardwood Forestry Conference. Columbia: University of Missouri.

Found at DOI: https://doi.org/10.3996/062019-JFWM-053.S2 (1.02 MB PDF).

Reference S2. Cotillon S, Mathis, M. 2016. Tree cover mapping tool—documentation and user manual (ver. 1.0, March 2016): U.S. Geological Survey Open-File Report 2016–1067.

Found at DOI: https://doi.org/10.3996/062019-JFWM-053.S3 (5.33 MB PDF); also available at https://doi.org/10.3133/ofr20161067.

Reference S3. Dibble AC, Zouhar K, Smith JK. 2008. Fire and nonnative invasive plants in the Northeast bioregion. Pages 61–90 in Zouhar K, Smith JK, Sutherland S, Brooks ML, editors. Wildland fire in ecosystems: fire and nonnative invasive plants. General Technical Report RMRS-GTR-42-vol. 6. Ogden, Utah: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station.

Found at DOI: https://doi.org/10.3996/062019-JFWM-053.S4 (1.38 MB PDF).

Reference S4. Mattice JA, Brauning DW, Diefenbach DR. 2005. Abundance of grassland songbirds on reclaimed surface mines in western Pennsylvania. Pages 504–510 in Ralph CJ, Rich TD, editors. Bird conservation implementation and integration in the Americas: proceedings of the Third International Partners in Flight Conference. U.S. Department of Agriculture Forest Service General Technical Report PSW-GTR-191.

Found at DOI: https://doi.org/10.3996/062019-JFWM-053.S5 (412 KB PDF).

Reference S5. [ODNR] Ohio Department of Natural Resources, Office of Real Estate and Lands Management. 2019. Land inventory summary 2019. Columbus: Ohio Department of Natural Resources.

Found at DOI: https://doi.org/10.3996/062019-JFWM-053.S6 (2.99 MB PDF).

Acknowledgments

We thank R. Burke, M. Butler, L. Fendrick, and H. Houser for assistance with data collection. We are indebted to J. Barber, R. Kalinen, S. Murphy, and E. Robinson for assisting with treatment implementation. We thank C. Dennison, the Associate Editor, and two anonymous reviewers for providing comments that improved this manuscript. Our research was funded in part by a grant through the Federal Aid in Wildlife Restoration Act awarded to Ohio Department of Natural Resources, Division of Wildlife. Pheasants Forever provided additional funding for the project.

Any use of trade, product, website, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

References

References
Anderson
RC,
Brown
LE.
1980
.
Influence of a prescribed burn on colonizing black locust
.
Pages
330
336
in
HE,
Garrett
Cox
GS,
editors.
Proceedings of the Third Central Hardwood Forestry Conference
.
Columbia
:
University of Missouri
(see Supplemental Material, Reference S1).
Arnold
TW.
2010
.
Uninformative parameters and model selection using Akaike's Information Criterion
.
Journal of Wildlife Management
74
:
1175
1178
.
Askins
RA,
Chávez-Ramírez
F,
Dale
BC,
Haas
CA,
Herkert
JR,
Knopf
FL,
Vickery
PD.
2007
.
Conservation of grassland birds in North America: understanding ecological processes in different regions
.
Ornithological Monographs
64
:
1
46
.
Bajema
RA,
DeVault
TL,
Scott
PE,
Lima
SL.
2001
.
Reclaimed coal mine grasslands and their significance for Henslow's sparrows in the American Midwest
.
Auk
118
:
422
431
.
Bakker,
K.K.
2003
.
A synthesis of the effect of woody vegetation on grassland nesting birds
.
Proceedings of the South Dakota Academy of Science
82
:
119
141
.
Brothers
TS.
1990
.
Surface mine grasslands
.
Geographical Review
80
:
209
225
.
Buckland
ST,
Anderson
DR,
Burnham
KP,
Laake
JL.
1993
.
Distance sampling: estimating abundance of biological populations
.
London
:
Chapman and Hall
.
Buckland
ST,
Anderson
DR,
Burnham
KP,
Laake
JL,
Borchers
DL,
Thomas
L.
2001
.
Introduction to distance sampling: estimating abundance of biological populations
.
Oxford, UK
:
Oxford University Press
.
Burnham
KP,
Anderson
DR.
1998
.
Model selection and inference: a practical information-theoretic approach
.
New York
:
Springer
.
Burnham
KP,
Anderson
DR,
Laake
JL.
1980
.
Estimation of density from line transect sampling of biological populations
.
Wildlife Monographs
72
.
Cable
TT,
Fick
WH,
Raynor
EJ.
2015
.
The potential impacts of saltcedar eradication (Tamarix sp.) on the birds of the Cimarron National Grassland
.
Transactions of the Kansas Academy of Science
118
:
41
47
.
Cavender
N,
Byrd
S,
Bechtoldt
CL,
Bauman
JM.
2014
.
Vegetation communities of a coal reclamation site in southeastern Ohio
.
Northeastern Naturalist
21
:
31
46
.
Chapman
RN,
Engle
DM,
Masters
RE,
Leslie
DM.
2004
.
Tree invasion constrains the influence of herbaceous structure in grassland bird habitats
.
Ecoscience
11
:
55
63
.
Coppedge
BR,
Engle
DM,
Masters
RE,
Gregory
MS.
2001
.
Avian response to landscape change in fragmented southern Great Plains grasslands
.
Ecological Applications
11
:
47
59
.
Cotillon
S,
Mathis,
M.
2016
.
Tree cover mapping tool—documentation and user manual (ver. 1.0, March 2016)
:
U.S. Geological Survey Open-File Report 2016–1067
(see Supplemental Material, Reference S2).
DeVault
TL,
Scott
PE,
Bajema
RA,
Lima
SL.
2002
.
Breeding bird communities of reclaimed coal-mine grasslands in the American Midwest
.
Journal of Field Ornthology
73
:
268
275
.
Dibble
AC,
Zouhar
K,
Smith
JK.
2008
.
Fire and nonnative invasive plants in the Northeast bioregion
.
Pages
61
90
in
Zouhar
K,
Smith
JK,
Sutherland
S,
Brooks
ML,
editors.
Wildland fire in ecosystems: fire and nonnative invasive plants. General Technical Report RMRS-GTR-42-vol
.
6. Ogden, Utah: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station
(see Supplemental Material, Reference S3).
Ellison
KS,
Ribic
CA,
Sample
DW,
Fawcett
MJ,
Dadismann
JD.
2013
.
Impacts of tree rows on grassland birds and potential nest predators: a removal experiment
.
PLoS One
8
:
e59151
.
Emery
SM,
Uwimbabazi
J,
Flory
SL.
2011
.
Fire intensity effects on seed germination of native and invasive Eastern deciduous forest understory plants
.
Forest Ecology and Management
261
:
1401
1408
.
Fiske
IJ,
Chandler
RB.
2011
.
Unmarked: an R package for fitting hierarchical models of wildlife occurrence and abundance
.
Journal of Statistical Software
43
:
1
23
.
Fiske
IJ,
Chandler
RB.
2014
.
Overview of Unmarked: an R package for the analysis of data from unmarked animals. A vignette to unmarked
.
CRAN. Available: https://CRAN.R-project.org/package=unmarked (December 2019).
Galligan
EW,
DeVault
TL,
Lima
SL.
2006
.
Nesting success of grassland and savanna birds on reclaimed surface coal mines of the Midwestern United States
.
Wilson Journal of Ornithology
118
:
537
546
.
Grant
TA,
Madden
E,
Berkey
GB.
2004
.
Tree and shrub invasion in northern mixed-grass prairie: implications for breeding grassland birds
.
Wildlife Society Bulletin
32
:
807
818
.
Graves
BM,
Rodewald
AD,
Hull
SD.
2010
.
Influence of woody vegetation on grassland birds within reclaimed surface mines
.
Wilson Journal Ornithology
122
:
646
654
.
Hill
JM,
Diefenbach
DR.
2013
.
Experimental removal of woody vegetation does not increase nesting success or fledgling production in two grassland sparrows (Ammodramus) in Pennsylvania
.
Auk
130
:
764
773
.
Hill
JM,
Diefenbach
DR.
2014
.
Occupancy patterns of regionally declining grassland sparrow populations in a forested Pennsylvania landscape
.
Conservation Biology
28
:
735
744
.
Holmes
AL,
Maestas
JD,
Naugle
DE.
2017
.
Bird responses to removal of western juniper in sagebrush-steppe
.
Rangeland Ecology & Management
70
:
87
94
.
Ingold
DJ.
2002
.
Use of a reclaimed strip-mine by grassland nesting birds in east-central Ohio
.
Ohio Journal of Science
102
:
56
62
.
Ingold
DJ,
Dooley
JL.
2013
.
Nesting success of grassland and shrub-nesting birds on The Wilds, an Ohio reclaimed surface mine
.
Ohio Journal of Science
111
:
37
41
.
Ingold
DJ,
Dooley
JL,
Cavender
N.
2010
.
Nest-site fidelity in grassland birds on mowed versus unmowed areas on a reclaimed surface mine
.
Northeastern Naturalist
17
:
125
35
.
Lautenbach
JM,
Plumb
RT,
Robinson
SG,
Hagen
CA,
Haukos
DA,
Pitman
JC.
2017
.
Lesser prairie-chicken avoidance of trees in a grassland landscape
.
Rangeland Ecology & Management
70
:
78
86
.
Mattice
JA,
Brauning
DW,
Diefenbach
DR.
2005
.
Abundance of grassland songbirds on reclaimed surface mines in western Pennsylvania
.
Pages
504
510
in
CJ,
Ralph
Rich
TD,
editors.
Bird conservation implementation and integration in the Americas: proceedings of the Third International Partners in Flight Conference. U.S. Department of Agriculture Forest Service General Technical Report PSW-GTR-191
(see Supplemental Material, Reference S4).
Matson
PA,
Parton
WJ,
Power
AG,
Swift
MJ.
1997
.
Agricultural intensification and ecosystem properties
.
Science
277
:
504
509
.
Monroe
MS,
Ritchison
G.
2005
Breeding biology of Henslow's Sparrows on reclaimed coal mine grasslands in Kentucky
.
Journal of Field Ornithology
76
:
143
50
.
[ODNR] Ohio Department of Natural Resources, Division of Wildlife
.
No date
.
Management of invasive shrubs on reclaimed surface mines
.
[ODNR] Ohio Department of Natural Resources, Office of Real Estate and Lands Management
.
2019
.
Land inventory summary 2019
.
Columbus
:
Ohio Department of Natural Resources (see Supplemental Material, Reference S5)
.
Oliphant
AJ,
Wynne
RH,
Zipper
CE,
Ford
WM,
Donovan,
PF,
Li
J.
2017
.
Autumn olive (Elaeagnus umbellate) presence and proliferation on surface coal mines in eastern USA
.
Biological Invasions
19
:
179
195
.
Powell
LA.
2007
.
Approximating demographic parameters using the delta method: a reference for avian biologists
.
Condor
109
:
949
954
.
Prather
CM,
Huynh
A,
Pennings
SC.
2017
.
Woody structure facilitates invasion of woody plants by providing perches for birds
.
Ecology and Evolution
7
:
8032
8039
.
R Core Team
.
2014
.
R: a language and environment for statistical computing
.
Vienna
:
R Foundation for Statistical Computing
.
Available: http://www.R-project.org/ (December 2019).
Raynor
EJ,
Cable
TT,
Sandercock
BK.
2017
.
Effects of Tamarix removal on the community dynamics of riparian birds in a semiarid grassland
.
Restoration Ecology
25
:
778
787
.
Royle
JA,
Dawson
DK,
Bates
S.
2004
.
Modeling abundance effects in distance sampling
.
Ecology
85
:
1591
1597
.
Sauer
JR,
Niven
DK,
Hines
JE,
Ziolkowski
DJ,
Pardieck
KL,
Fallon
JE,
Link
WA.
2017
.
The North American Breeding Bird Survey, results and analysis 1966–2015. Version 2.07.2017
.
Laurel, Maryland
:
U.S. Geological Survey Patuxent Wildlife Research Center
.
Available: https://www.mbr-pwrc.usgs.gov/bbs/ (December 2019).
Scott
PE,
DeVault
TL,
Bajema
RA,
Lima
SL.
2002
.
Grassland vegetation and bird abundances on reclaimed Midwestern coal mines
.
Wildlife Society Bulletin
30
:
1006
1014
.
Scott
PE,
Lima
SL.
2004
.
Exotic grasslands on reclaimed Midwestern coal mines: an ornithological perspective
.
Weed Technology
18
:
1518
1521
.
Stauffer
GE,
Diefenbach
DR,
Marshall
MR,
Brauning
DW.
2011
.
Nest success of grassland sparrows on reclaimed surface mines
.
Journal of Wildlife Management
75
:
548
557
.
Szafoni
RE.
1991
.
Vegetation management guideline: autumn olive, Elaeagnus umbellate Thunb
.
Natural Area Journal
11
:
121
122
.
Tack
JD,
Quamen
FR,
Kelsey
K,
Naugle
DE.
2017
.
Doing more with less: removing trees in a prairie system improves the value of grasslands for obligate bird species
.
Journal of Environmental Management
198
:
163
169
.
Thompson
SJ,
Arnold
TW,
Amundson
CL.
2014
.
A multiscale assessment of tree avoidance by prairie birds
.
Condor
116
:
303
15
.
Thompson
SJ,
Arnold
TW,
Fieberg
J,
Granfors
DA,
Vacek
S,
Palaia
N.
2016
.
Grassland birds demonstrate delayed response to large-scale tree removal in central North America
.
Journal of Applied Ecology
53
:
284
294
.
Van Horne
B.
1983
.
Density as a misleading indicator of habitat quality
.
Journal of Wildlife Management
47
:
893
901
.
Vickery
PD,
Herkert
JR.
2001
.
Recent advances in grassland bird research: where do we go from here?
Auk
118
:
11
15
.
Vickery
PD,
Hunter
ML,
Wells
JV.
1992
.
Is density an indicator of breeding success?
Auk
109
:
706
710
.

Author notes

Citation: Lautenbach JM, Stricker N, Ervin M, Hershner A, Harris R, Smith C. 2020. Woody vegetation removal benefits grassland birds on reclaimed surface mines. Journal of Fish and Wildlife Management 11(1):89–98; e1944-687X. https://doi.org/10.3996/062019-JFWM-053