Petra Bohall Wood

Are traditional methods of determining nest predators and nest fates reliable? An experiment with Wood Thrushes (Hylocichla mustelina) using miniature video cameras

Abstract We used miniature infrared video cameras to monitor Wood Thrush (Hylocichla mustelina) nests during 1998–2000. We documented nest predators and examined whether evidence at nests can be used to predict predator identities and nest fates. Fifty-six nests were monitored; 26 failed, with 3 abandoned and 23 depredated. We predicted predator class (avian, mammalian, snake) prior to review of video footage and were incorrect 57% of the time. Birds and mammals were underrepresented whereas snakes were over-represented in our predictions. We documented ≥9 nest-predator species, with the southern flying squirrel (Glaucomys volans) taking the most nests (n = 8). During 2000, we predicted fate (fledge or fail) of 27 nests; 23 were classified correctly. Traditional methods of monitoring nests appear to be effective for classifying success or failure of nests, but ineffective at classifying nest predators.

Roost tree selection by northern long-eared bat (Myotis septentrionalis) maternity colonies in an industrial forest of the central Appalachian mountains

As part of a bat ecology research effort in managed forests, we tracked seven lactating northern long-eared bats (Myotis septentrionalis Trouessart) to 12 roost trees used by maternity colonies ranging in size from 11 to 65 individuals in the central Appalachians of West Virginia. Lactating females switched among roosts on average every 5 days. Roosts were located in nine tree species, with 11 roosts in tree cavities and 1 under exfoliating bark. Eight of the roosts were located in snags; the other four were located in cavities in living trees. Black locust (Robinia pseudo-acacia L.) snags were selected as a roost sites more than expected based on their occurrence in the forests we surveyed. Only 3 of the 24 habitat characteristics differed between actual roost trees and randomly located cavity trees. Structurally, actual roosts tended to be closer to neighboring trees than did suitable roosts in randomly located cavity trees. Tree species was an important factor that influenced roost tree selection, with black locusts being used significantly more than expected based on their abundance across the managed forest landscape we studied. All northern long-eared bat roosts were located in or below the forest canopy which suggests that solar exposure may not be as critical for selection of roost trees by maternity colonies of this species as opposed to other species of tree-roosting bat at more northern latitudes in North America. Although many previous studies have demonstrated the importance of old-growth habitat in providing roosting habitat for tree-roosting bats, use of roosting habitat and characteristics of roost sites have been little documented within intensively managed forests. Our research indicates that in managed, Allegheny hardwood-northern hardwood forests of the central Appalachians, short-lived, early successional species such as black locust may provide suitable roosting habitat for northern long-eared bats, as well as other species of bats. Published by Elsevier Science B.V.

CERULEAN WARBLER (DENDROICA CERULEA) MICROHABITAT AND LANDSCAPE-LEVEL HABITAT CHARACTERISTICS IN SOUTHERN WEST VIRGINIA

Abstract The Cerulean Warbler (Dendroica cerulea) is a species of conservation concern in eastern North America, where declines in its population have been documented over the past several decades. A high proportion of the population occurs in forested areas of southern West Virginia, where it may be threatened by loss and degradation of forested habitat from mountaintop mining and valley fill (MTMVF). We examined, from a landscape perspective, the effects of forest fragmentation (in particular, effects of fragment size and response to edges) on Cerulean Warblers, using territory mapping techniques and geographic information system (GIS) technology in portions of four counties in southwestern West Virginia. We quantified landscape characteristics from digitized aerial photographs and measured microhabitat characteristics on spot-mapping plots. Territory density of Cerulean Warblers was 4.6 territories per 10 ha in intact forest and 0.7 territories per 10 ha in fragmented forest. The best habitat model included both landscape and microhabitat variables and indicated that territory density increased with increasing snag density, percentage of canopy cover >6–12 m and >24 m in height, and distance from mine edge. Models for predicting microhabitat use at the territory level were weak, indicating that microhabitat characteristics of territories were similar to habitat available on spot-mapping plots. The species did not appear to avoid internal edges, such as natural canopy gaps and open-canopy or partially open-canopy roads. Territory placement on ridges was greater than expected, and in bottomlands (ravines) and midslopes less than expected, given availability. Fifty percent of all territories were on ridges. Preference for ridges suggests that MTMVF may have a greater effect on Cerulean Warbler populations than other sources of forest fragmentation, given that ridges are removed in MTMVF. Our data indicate that Cerulean Warblers are negatively affected by mountaintop mining from loss of forested habitat, particularly ridgetops, and from degradation of remaining forests, as evidenced by lower territory density in fragmented forests and lower territory density closer to mine edges.

Home-range Size and Habitat Used by the Northern Myotis (Myotis septentrionalis)

Abstract We examined home range size and habitat use of nine female northern myotis (Myotis septentrionalis) within an intensively managed forest in the central Appalachians of West Virginia. Using the 95% adaptive kernel method, we calculated a mean home range of 65 ha. Northern myotis used recent diameter-limit harvests and road corridors more than expected based on availability of these habitats. Intact forest stands and more open deferment harvested stands were used less than expected based on the availability of these habitats, although intact forest stands still constituted the overall majority of habitat used. Partial timber harvests that leave a relatively closed canopy appear to promote or improve northern myotis foraging habitat in heavily forested landscapes. However, the long-term ecological impacts on bats and other biota from this silviculturally unacceptable practice are unclear.

Movements of allegheny woodrats in relation to timber harvesting

The Allegheny woodrat (Neotoma magister) occurs in the Appalachian Mountains, forming colonies in rock outcrops, cliffs, and caves. Populations on the northern and western peripheries of the range have experienced drastic declines in the past 20-30 years. Dependence upon rock outcrops makes Allegheny woodrats vulnerable to land-use practices that alter habitats surrounding colonies. To examine the impacts of timber harvesting on Allegheny woodrat behavior, we radiotracked 37 adults during summer 1998 and 1999 in clearcut, diameter-limit, and intact forest stands in the central Appalachians of West Virginia. Home range size and foraging movements generally,were greatest at diameter-limit sites and smallest in intact forests in 1998, following a poor mast crop. We detected no differences among harvest methods in 1999 when mast was abundant. We believe that when hard mast was scarce, woodrats increased foraging movements and home range size to locate mast or sufficient alternative foods. Additionally, woodrats used clearcut and adjacent forested areas in proportion to availability. Our results suggested that clearcutting has minimal impact on woodrat movements, home range, and habitat use if sufficient intact forest is retained adjacent to colonies. Harvesting methods that selectively remove important mast-producing species may represent the greatest disturbance to Allegheny woodrats from forest management.

Cerulean Warbler Abundance and Occurrence Relative to Large-Scale Edge and Habitat Characteristics

Abstract We examined Cerulean Warbler (Dendroica cerulea) abundance and occurrence in southwestern West Virginia, where the coal-mining technique of mountaintop removal mining–valley fill converts large contiguous tracts of deciduous forest to forest patches surrounded by early successional habitats. Our study objectives were to quantify abundance and occurrence of Cerulean Warblers relative to (1) distance from the edge of extensive reclaimed grasslands and (2) habitat structure and landscape characteristics. Cerulean Warbler abundance increased with distance from the edge and edge effects extended 340 m into the forest. Percent occurrence did not vary with distance from mine edge, suggesting a degree of tolerance to the extensive edge occurring at the interface of forest and reclaimed lands. Abundance and occurrence were greater on ridges and midslopes than in bottomlands; consequently, disturbances such as mountaintop mining in which ridges are removed may have a greater impact on populations compared to other sources of fragmentation where ridges are not disturbed. Models based on the information-theoretic approach indicated that Cerulean Warblers were more likely to be present in productive sites on northwest to southeast facing slopes, upper slope positions (midslope to ridgetop), and forests with low sapling density. Cerulean Warbler abundance was positively associated with more productive sites, higher snag density, large blocks of mature deciduous forest, and low amounts of edge in the landscape. In addition to outright loss of forested habitat, mountaintop mining-valley fill alters the spatial configuration of forested habitats, creating edge and area effects that negatively affect Cerulean Warbler abundance and occurrence in the reclaimed mine landscape.

The Overlooked Terrestrial Impacts of Mountaintop Mining

Ecological research on mountaintop mining has been focused on aquatic impacts because the overburden (i.e., the mountaintop) is disposed of in nearby valleys, which leads to a wide range of water-quality impacts on streams. There are also numerous impacts on the terrestrial environment from mountaintop mining that have been largely overlooked, even though they are no less wide ranging, severe, and multifaceted. We review the impacts of mountaintop mining on the terrestrial environment by exploring six broad themes: (1) the loss of topographic complexity, (2) forest loss and fragmentation, (3) forest succession and soil loss, (4) forest loss and carbon sequestration, (5) biodiversity, and (6) human health and well-being.

MICROSATELLITE DNA ANALYSIS OF POPULATION STRUCTURE IN ALLEGHENY WOODRATS (NEOTOMA MAGISTER)

Abstract Eleven polymorphic microsatellite DNA markers were screened to investigate patterns and processes of genetic variation in Allegheny woodrats at 2 spatial scales, geographically distinct populations and subpopulations within populations. The microsatellite markers detected considerable genetic variation and diversity, with an average heterozygosity of 62.0% (range 25.7–75.0%) and an average of 10.4 alleles per locus (range 5–19). Subpopulations conformed to Hardy–Weinberg expectations in 98.9% of comparisons, and allele frequency differed among subpopulations in 99.3% of comparisons. All statistical tests indicated significant genetic subdivision at the population and subpopulation levels. Theoretical estimates of gene flow were low among subpopulations, suggesting that effective dispersal is limited among subpopulations separated by as little as 3 km. A significant range-wide relationship between geographic and genetic distance at both scales suggests isolation by distance as a mechanism for the observed differentiation. The relationship between genetic distance and geographic distance among the 19 subpopulations within the central Appalachians was significant but weak. Results suggest that Allegheny woodrats associated with clusters of geographically proximate rock outcrops may function as population units. We recommend that individual rock outcrops or groups of geographically proximate outcrops be considered as population units for management.

Emulating Natural Disturbances for Declining Late-Successional Species: A Case Study of the Consequences for Cerulean Warblers (Setophaga cerulea)

Forest cover in the eastern United States has increased over the past century and while some late-successional species have benefited from this process as expected, others have experienced population declines. These declines may be in part related to contemporary reductions in small-scale forest interior disturbances such as fire, windthrow, and treefalls. To mitigate the negative impacts of disturbance alteration and suppression on some late-successional species, strategies that emulate natural disturbance regimes are often advocated, but large-scale evaluations of these practices are rare. Here, we assessed the consequences of experimental disturbance (using partial timber harvest) on a severely declining late-successional species, the cerulean warbler (Setophaga cerulea), across the core of its breeding range in the Appalachian Mountains. We measured numerical (density), physiological (body condition), and demographic (age structure and reproduction) responses to three levels of disturbance and explored the potential impacts of disturbance on source-sink dynamics. Breeding densities of warblers increased one to four years after all canopy disturbances (vs. controls) and males occupying territories on treatment plots were in better condition than those on control plots. However, these beneficial effects of disturbance did not correspond to improvements in reproduction; nest success was lower on all treatment plots than on control plots in the southern region and marginally lower on light disturbance plots in the northern region. Our data suggest that only habitats in the southern region acted as sources, and interior disturbances in this region have the potential to create ecological traps at a local scale, but sources when viewed at broader scales. Thus, cerulean warblers would likely benefit from management that strikes a landscape-level balance between emulating natural disturbances in order to attract individuals into areas where current structure is inappropriate, and limiting anthropogenic disturbance in forests that already possess appropriate structural attributes in order to maintain maximum productivity.

Cerulean warbler use of regenerated clearcut and two-age harvests

Abstract We examined use of 2 silvicultural treatments (clearcut and two-age harvests), 15–18 years post-harvest by cerulean warblers (Dendroica cerulea) in mixed mesophytic and northern hardwood forests of the Allegheny Mountain region in West Virginia. Cerulean warbler abundance and occurrence were greater in 70–80-year-old mature forests than in 15–18-year-old clearcuts. Although abundance did not differ statistically between clearcut and two-age treatments, it was almost 5 times greater in the two-age treatments, likely because they provided a more complex canopy structure. Abundance of cerulean warblers in unharvested periphery stands adjacent to clearcut and two-age harvests was similar to that in unharvested control stands, suggesting that small harvests within mature forest do not negatively impact cerulean warbler abundance in the remaining forest, only within the clearcut harvests themselves.