TY - JOUR
AU - Swihart, Robert, K
AB - Abstract Many small mammal populations respond quickly to timber harvest aimed at oak (Quercus) regeneration, which alters microhabitat. We used mark-release–recapture data collected 6–8 years postharvest from the Hardwood Ecosystem Experiment in southern Indiana, United States, to model density and apparent survival of eastern chipmunks (Tamias striatus) and white-footed mice (Peromyscus leucopus) as a function of timber harvest treatments (shelterwood, clearcut, patch cut, and unharvested control). Density, estimated using spatial capture–recapture, increased for chipmunks in all types of harvest openings, but survival was unaffected by harvest. Chipmunk densities in unharvested forest matrix habitat averaged 58% and 71% lower relative to harvest openings and opening edges, respectively. White-footed mouse density was less responsive to timber harvest, but monthly survival rates were reduced by 13% in shelterwoods and 17% in patch cuts relative to control sites. Both rodent species tended to exhibit distance-dependent responses, with higher density of home-range centers near harvest boundaries relative to forest matrix. Structural complexity created at the edges of timber harvest openings can benefit rodents associated with edge habitat 6–8 years after harvest, presumably due to improved foraging efficiency and resource diversity. Cascading effects of rodent demographic responses are likely to affect predation and seed dispersal, which are critical trophic interactions in oak forest ecosystems. apparent survival, central hardwood forest, clearcut, patch cut, Peromyscus leucopus, population density, shelterwood, spatially explicit capture–recapture, Tamias striatus Suppression of natural disturbance regimes in oak–hickory (Quercus–Carya)–dominated forests of eastern North America has contributed to regeneration failure of oak, a keystone species group in the region (Fralish 2004). To promote conditions that favor oak regeneration, forest managers implement timber harvest and other forms of disturbance (McShea and Healy 2002). In the Central Hardwoods Forest region, harvesting methods frequently involve 1- to 5-ha patch cuts, larger clearcuts, and multistage shelterwood prescriptions. Disturbances induced by these harvesting activities alter abiotic (e.g., solar radiation, moisture levels) and biotic (e.g., increased herbaceous vegetation, coarse woody debris) conditions in the affected area (Ash 1995; Fredericksen et al. 1999; Boerner 2000; Zheng et al. 2000; Perry et al. 2004; Greenberg et al. 2006, 2007). Changes in habitat conditions on the forest floor can have short-term implications on other forest organisms. In particular, small mammals rely on microhabitat components such as coarse woody debris for travel and cover, herbaceous vegetation for cover and food resources, and leaf litter for foraging (Zollner and Crane 2003; Kaminski et al. 2007; Greenberg et al. 2011). Small rodents such as white-footed mice (Peromyscus leucopus) and eastern chipmunks (Tamias striatus) can be useful indicators of the ecological effects of harvest-based forest management. These and other rodents respond to changes in habitat resources at scales of stands (e.g., overstory composition, oak and hickory mast) and microsites (e.g., coarse woody debris, herbaceous vegetation, leaf litter; Miller and Getz 1977; Greenberg 2002; Zollner and Crane 2003; Kaminski et al. 2007). Demographic responses of rodents have implications for other facets of forest ecology, including seed predation (Whitaker and Hamilton 1998), seed and fungi dispersal (Maser et al. 1978), songbird nest success (Schmidt and Ostfeld 2003), and predator abundance (Whitaker and Hamilton 1998; Leonard et al. 2015; Olson et al. 2015). Prior studies that examined responses of small mammal populations to oak-focused silviculture generally have been limited to short-term results (≤5 years following harvest) from longitudinal studies (Kirkland et al. 1985, 1990; Fantz and Renken 2005; Kellner et al. 2013; Larsen et al. 2016) or comparisons of stands of varying ages and histories (Healy and Brooks 1988; Urban and Swihart 2011). An exception is a recent analysis of responses of Peromyscus to forest management in the Missouri Ozarks over a 20-year period (Gitzen et al. 2018). Moreover, response measures in past studies have relied on population size or an index presumably related to abundance. Comparisons that rely on population size implicitly assume that the area sampled is equivalent between the sites targeted for comparison. Population density, i.e., abundance/area, enables direct comparison of different silvicultural treatments within a study or variation in responses among studies, but past attempts to convert abundance to density usually relied on ad hoc approaches to estimate sampling area that failed to account for heterogeneity in captures arising from the spatial organization of animals or traps (Wilson and Anderson 1985; Parmenter et al. 2003). In contrast, spatially explicit capture–recapture models provide a statistically rigorous estimate of area sampled and thus estimates of density (Efford 2004; Borchers and Efford 2008; Royle et al. 2014). Even though estimates of density have advantages over abundance, they may not accurately reflect habitat quality (Van Horne 1983). Estimates of fitness parameters such as survival are desirable when inferring quality differences that stem from forest management. Statistical estimates of vital rates for rodents are relatively common in harvested coniferous forests (Manning and Edge 2004; Herbers and Klenner 2007; Ransome et al. 2009; Hodson et al. 2010; Sullivan and Sullivan 2017) and in coppiced oak woodlands (Gasperini et al. 2016), but comparable estimates are lacking in managed oak–hickory forests of the eastern United States. In this study, we estimated population density and survival of white-footed mice and eastern chipmunks 6–8 years postharvest to gain insight into effects of oak-centered silviculture on these important components of Central Hardwood Forest ecosystems. On the basis of prior work (Kellner et al. 2013; Gitzen et al. 2018), we predicted positive density responses of both species to timber harvest, with greatest effects along habitat edges created by harvesting. Materials and Methods Study area Our study was part of the Hardwood Ecosystem Experiment (HEE), a long-term, landscape-level experiment studying the ecological impacts of timber management for oak regeneration (Kalb and Mycroft 2013). The HEE is located in the Brown County Section of the Highland Rim Natural Region of Indiana in Morgan-Monroe and Yellowwood State Forests (Homoya et al. 1984). The overstory of HEE sites was dominated by oak (primarily Q. prinus, Q. alba, Q. velutina, and Q. rubra) and hickory species with an understory composed of sugar maple (Acer saccharum), American beech (Fagus grandifolia), and flowering dogwood (Cornus florida; Saunders and Arsenault 2013). The HEE is composed of nine research units divided evenly among uneven-aged management, even-aged management, or no harvest (Fig. 1A). Each uneven-aged unit received small patch clearcuts (0.4–2.0 ha) and single-tree selection in 2008 and 2009 (Kalb and Mycroft 2013; Fig. 1B). Within each even-aged unit, clearcuts (n = 2) and three-stage shelterwood harvests (n = 2), each approximately 4 ha, were created in 2008 (Fig. 1C). Shelterwood harvests received midstory removal cuts in 2008–2009, and the second stage of the harvest took place in October–November 2015 and November 2016. The second stage was an establishment cut in which basal area was reduced to 13.8–16.1 m2 ha−1. Overstory trees were thinned, retaining overstory oaks and hickories as seed sources and shelter for regenerating seedlings. Fig. 1. View largeDownload slide Map of the study area on the Hardwood Ecosystem Experiment in Morgan-Monroe and Yellowwood State Forests, southern Indiana, United States. A) The distribution of the nine research units across the state forest properties. B) An enlargement of an uneven-aged unit and the distribution of patch cuts and grids on which live trapping of small mammals occurred. C) An enlargement of an even-aged unit and the distribution of clearcut and shelterwood harvest treatments as well as trapping grids. Fig. 1. View largeDownload slide Map of the study area on the Hardwood Ecosystem Experiment in Morgan-Monroe and Yellowwood State Forests, southern Indiana, United States. A) The distribution of the nine research units across the state forest properties. B) An enlargement of an uneven-aged unit and the distribution of patch cuts and grids on which live trapping of small mammals occurred. C) An enlargement of an even-aged unit and the distribution of clearcut and shelterwood harvest treatments as well as trapping grids. Data collection Trapping grids (Figs. 1B, C) were placed in HEE clearcuts (n = 6), shelterwoods (n = 6), 0.4-ha patch cuts (n = 8), 2-ha patch cuts (n = 6), and in unharvested control areas (n = 6). For assessment of survival, a random subset of 12 grids (two each in shelterwood and clearcut harvests, four each in patch cuts and unharvested control areas) was sampled in each of the five primary trapping occasions (June–August 2014, June–July 2015, October 2015, June–July 2016, and October 2016); logistical constraints precluded sampling at this frequency on all 32 grids. Sherman traps (n = 36; LFA folding trap, 7.62 cm × 8.89 cm × 22.86 cm; H.B. Sherman Traps, Inc., Tallahassee, Florida) were arranged in square or nearly square grids with 20-m spacing. In patch cuts, trap arrays were positioned so that traps occurred within the harvest interior, on the edge, and in the surrounding forest matrix. The proportion of a patch-cut grid in interior, edge, and matrix habitat averaged 0.36, 0.21, and 0.43, respectively, for 0.4-ha harvests and 0.76, 0.18, and 0.06, respectively, for 2-ha harvests. Locations of all trap sites were recorded with a GPS unit (Earthmate PN-40; Delorme, Yarmouth, Maine). Grids were sampled according to a robust design (Pollock 1982). All traps were prebaited for 3 days and then checked twice daily for 5 consecutive days in the summer and 3 consecutive days during fall sampling sessions. Traps were baited with sunflower seeds, rolled oats, and freezer-killed mealworms. Carded wool bedding material was placed in traps when overnight temperatures were predicted to be lower than 10°C, and traps were covered with leaf litter or an asphalt roofing shingle to provide shelter. Captured small mammals were transferred to handling bags and species identity, sex, reproductive condition, and body mass were recorded. Rodents were marked with a unique ear tag (Hasco Tag Company, Dayton, Kentucky). When trap disturbance by raccoons (Procyon lotor) or opossums (Didelphis virginiana) became problematic, 2–4 Tomahawk traps (Tomahawk Live Traps Co., Tomahawk, Wisconsin) were set nearby, baited with canned cat food. Captured nuisance animals were relocated at least 10 km away from any trapping grid. All trapping and handling of animals was conducted according to protocol #1112000404 approved by the Purdue Animal Care and Use Committee and in accordance with guidelines provided by Sikes et al. (2016). Data on mast availability were collected late August through December each year from 93 to 108 overstory black oak (Q. velutina) and white oak (Q. alba) trees (8–12 per research unit), using two 0.34-m2 elevated mast traps under each tree (for details, see Kellner et al. 2014). Analysis To estimate density (i.e., density of activity centers, Borchers and Efford 2008), trapping data for chipmunks and white-footed mice were converted into spatial encounter histories for spatial capture–recapture analysis in program R (R Development Core Team 2016), package ‘secr’ (Efford 2016). Spatial capture–recapture methods incorporate a detection function, which includes a baseline encounter probability (g0) and a scale parameter (σ) that decreases with increasing distance between an animal’s activity or home-range center and each trap location. Multisession models were used to stratify density estimates by sampling grid, and separate models were run for each primary occasion and species combination. To define the state space, buffers around the perimeter of each grid were set to 60 m for both white-footed mice and chipmunks. Plots of effective sampling area were examined for both species to ensure that the buffer distance did not affect density estimates (Royle et al. 2014: 131–132; Efford 2016). Covariates on g0 were included to capture variation in detection due to individual time or behavioral effects. The best models for g0 for each primary occasion and species combination were determined using Akaike’s information criterion corrected for small sample sizes (AICc) while holding σ and density constant. Predictor variables to compare with the null model were: time of each sampling occasion (t), time as a linear trend over occasions (T), learned response (b), and transient response (B). The best predictor variables were used in subsequent models for density. Activity centers probably vary nonuniformly with respect to harvests, so rasterized habitat masks were created as a series of points with 10-m spacing across the state space. We applied habitat covariates to these habitat masks by overlaying harvest boundaries on top of locational coordinates for small mammal trapping grids. Categorical covariates, including harvest opening (created by clearcut or patch cut), opening edge, shelterwood harvest, or matrix, were created to examine harvest effects on density. Shelterwood harvest was considered separately from the other harvest treatments because its first and second stages retained a forest canopy. Edge was defined as habitat within 5 m of a harvest boundary, roughly the distance outside of harvest boundaries at which oak regeneration can be expected for some forms of forest management (Greenler and Saunders 2019). Matrix was defined as forest beyond the edge boundary that was not subjected to the harvest treatments. To examine the effects of harvest openings on a continuous scale for clearcuts and patch cuts, standardized distances (mean = 0, standard deviation = 1) of each mask point from the harvest boundary were calculated and incorporated into the habitat mask. Models were fitted to allow density (individuals ha−1) to vary by a categorical designation of habitat type (i.e., harvest opening, opening edge, shelterwood, or matrix), as well as by standardized distance to harvest boundary. Analyses of apparent survival (hereafter, simply survival) were conducted separately for eastern chipmunks and white-footed mice using the robust design of package ‘RMark’ (Laake 2016) in program R to build candidate models in Program MARK (White and Burnham 1999). Probabilities of initial capture (p) and recapture (c) were modeled as equal and constant (p = c), and models incorporating variation due to primary occasion and secondary occasion (days or nights within a primary occasion) were compared with null models (constant). The best model for capture probability was selected with temporary emigration and survival parameters held constant. Candidate models for monthly survival (computed by assuming constant survival between primary intervals) included harvest treatment, time (primary occasion), and additive and interactive effects of time and treatment relative to null models (constant survival). In all analyses, estimated coefficients were judged significant if 95% confidence intervals did not include zero. Results We recorded 4,249 captures of 1,638 individuals of eight species over 23,205 trap nights between 2014 and 2016. Eastern chipmunks (2,804 captures and 781 individuals) and white-footed mice (1,157 captures and 695 individuals) were encountered most frequently, followed by short-tailed shrews (Blarina brevicauda, 261 and 142), pine voles (Microtus pinetorum, 13 and 10), southern flying squirrels (Glaucomys volans, 7 and 4), southeastern shrews (Sorex longirostris, 5 and 4), smoky shrews (S. fumeus, 1 and 1), and long-tailed weasels (Mustela frenata, 1 and 1). Mean mast production in 2013 and 2015 was similar and characterized by practically no production by white oak (0.06 and 0.38 m−2, respectively) and production by black oak (3.96 and 4.86 m−2, respectively) similar to the species’ average from 2006 to 2017 (Swihart et al. 2019). Comparatively, mean mast production was 3.2 times greater in 2014, primarily due to a large increase in production of white oak acorns (10.02 m−2 for white oak and 4.85 m−2 for black oak). Eastern chipmunk density had a consistently strong positive relationship with harvest openings and edges of openings relative to the forest matrix (Table 1). Density (mean ± SE) of chipmunks across all sampling occasions in opening edge (12.08 ± 1.89) and opening habitats (8.54 ± 0.97) was 3.4- and 2.4-fold greater, respectively, than in matrix habitat (3.54 ± 0.81) (Fig. 2). The effects of shelterwood harvest were positively and significantly related to density in both 2016 sampling periods (1.9 times greater than matrix sites in summer; 2.4 times greater in fall; Table 1). For white-footed mice, effects of harvest edge on density were consistently positive relative to matrix habitat and significant in two of the three occasions (Table 1, Fig. 3). In contrast to eastern chipmunks, mouse density did not differ appreciably between openings and matrix habitat for any occasion (Table 1, Fig. 3). However, density of mice in shelterwoods exceeded density in matrix habitat during both sampling occasions in 2015, with increases of 1.6- and 2.0-fold for summer and fall sampling occasions, respectively (Table 1). Table 1. Spatially explicit capture-recapture estimates of eastern chipmunk (Tamias striatus) and white-footed mouse (Peromyscus leucopus) mean density (ha−1) across all sites, and effects of habitat type corresponding to experimental harvest units in southern Indiana, United States. Density Habitat type Opening edge Opening Shelterwood Mean SE β CIL CIU β CIL CIU β CIL CIU T. striatus  Summer 14 1.39 0.32 2.08 1.14 3.02 1.30 0.60 2.01 −1.05 −2.80 0.70  Summer 15 6.20 0.51 0.90 0.41 1.39 0.53 0.29 0.76 0.08 −0.23 0.38  Fall 15 3.00 0.56 0.83 −0.80 2.46 1.18 0.68 1.68 0.54 −0.07 1.14  Summer 16 3.93 0.48 1.45 0.91 1.99 0.89 0.57 1.20 0.64 0.25 1.02  Fall 16 2.93 0.63 1.24 −0.06 2.53 1.19 0.66 1.72 0.88 0.30 1.47 P. leucopus  Summer 14 4.27 0.90 1.33 0.50 2.16 −1.01 −2.17 0.15 −0.12 −0.80 0.56  Summer 15 8.70 0.99 0.14 −0.93 1.20 0.13 −0.18 0.44 0.46 0.16 0.76  Fall 15 6.99 0.92 1.19 0.53 1.84 −0.04 −0.55 0.48 0.72 0.34 1.10  Summer 16 5.20 1.55 N/A N/A N/A −0.53 −1.38 0.31 0.11 −0.50 0.71 Density Habitat type Opening edge Opening Shelterwood Mean SE β CIL CIU β CIL CIU β CIL CIU T. striatus  Summer 14 1.39 0.32 2.08 1.14 3.02 1.30 0.60 2.01 −1.05 −2.80 0.70  Summer 15 6.20 0.51 0.90 0.41 1.39 0.53 0.29 0.76 0.08 −0.23 0.38  Fall 15 3.00 0.56 0.83 −0.80 2.46 1.18 0.68 1.68 0.54 −0.07 1.14  Summer 16 3.93 0.48 1.45 0.91 1.99 0.89 0.57 1.20 0.64 0.25 1.02  Fall 16 2.93 0.63 1.24 −0.06 2.53 1.19 0.66 1.72 0.88 0.30 1.47 P. leucopus  Summer 14 4.27 0.90 1.33 0.50 2.16 −1.01 −2.17 0.15 −0.12 −0.80 0.56  Summer 15 8.70 0.99 0.14 −0.93 1.20 0.13 −0.18 0.44 0.46 0.16 0.76  Fall 15 6.99 0.92 1.19 0.53 1.84 −0.04 −0.55 0.48 0.72 0.34 1.10  Summer 16 5.20 1.55 N/A N/A N/A −0.53 −1.38 0.31 0.11 −0.50 0.71 For each habitat effect coefficient (β), lower (CIL) and upper (CIU) bounds of the corresponding 95% confidence interval are provided. Forest matrix served as the reference level for habitat. Coefficients in bold and italicized type indicate significance at 0.05 and 0.10 levels, respectively. Captures were insufficient to model mouse density for fall 2016 or the effect of opening edge on mouse density in summer 2016. View Large Table 1. Spatially explicit capture-recapture estimates of eastern chipmunk (Tamias striatus) and white-footed mouse (Peromyscus leucopus) mean density (ha−1) across all sites, and effects of habitat type corresponding to experimental harvest units in southern Indiana, United States. Density Habitat type Opening edge Opening Shelterwood Mean SE β CIL CIU β CIL CIU β CIL CIU T. striatus  Summer 14 1.39 0.32 2.08 1.14 3.02 1.30 0.60 2.01 −1.05 −2.80 0.70  Summer 15 6.20 0.51 0.90 0.41 1.39 0.53 0.29 0.76 0.08 −0.23 0.38  Fall 15 3.00 0.56 0.83 −0.80 2.46 1.18 0.68 1.68 0.54 −0.07 1.14  Summer 16 3.93 0.48 1.45 0.91 1.99 0.89 0.57 1.20 0.64 0.25 1.02  Fall 16 2.93 0.63 1.24 −0.06 2.53 1.19 0.66 1.72 0.88 0.30 1.47 P. leucopus  Summer 14 4.27 0.90 1.33 0.50 2.16 −1.01 −2.17 0.15 −0.12 −0.80 0.56  Summer 15 8.70 0.99 0.14 −0.93 1.20 0.13 −0.18 0.44 0.46 0.16 0.76  Fall 15 6.99 0.92 1.19 0.53 1.84 −0.04 −0.55 0.48 0.72 0.34 1.10  Summer 16 5.20 1.55 N/A N/A N/A −0.53 −1.38 0.31 0.11 −0.50 0.71 Density Habitat type Opening edge Opening Shelterwood Mean SE β CIL CIU β CIL CIU β CIL CIU T. striatus  Summer 14 1.39 0.32 2.08 1.14 3.02 1.30 0.60 2.01 −1.05 −2.80 0.70  Summer 15 6.20 0.51 0.90 0.41 1.39 0.53 0.29 0.76 0.08 −0.23 0.38  Fall 15 3.00 0.56 0.83 −0.80 2.46 1.18 0.68 1.68 0.54 −0.07 1.14  Summer 16 3.93 0.48 1.45 0.91 1.99 0.89 0.57 1.20 0.64 0.25 1.02  Fall 16 2.93 0.63 1.24 −0.06 2.53 1.19 0.66 1.72 0.88 0.30 1.47 P. leucopus  Summer 14 4.27 0.90 1.33 0.50 2.16 −1.01 −2.17 0.15 −0.12 −0.80 0.56  Summer 15 8.70 0.99 0.14 −0.93 1.20 0.13 −0.18 0.44 0.46 0.16 0.76  Fall 15 6.99 0.92 1.19 0.53 1.84 −0.04 −0.55 0.48 0.72 0.34 1.10  Summer 16 5.20 1.55 N/A N/A N/A −0.53 −1.38 0.31 0.11 −0.50 0.71 For each habitat effect coefficient (β), lower (CIL) and upper (CIU) bounds of the corresponding 95% confidence interval are provided. Forest matrix served as the reference level for habitat. Coefficients in bold and italicized type indicate significance at 0.05 and 0.10 levels, respectively. Captures were insufficient to model mouse density for fall 2016 or the effect of opening edge on mouse density in summer 2016. View Large Fig. 2. View largeDownload slide Eastern chipmunk (Tamias striatus) density (individuals/ha) estimates and 95% confidence intervals computed with spatially explicit capture–recapture models incorporating effects of habitat type in experimental harvest units in southern Indiana, 2014–2016. To facilitate comparison with other studies, minimum number alive (MNA) values are pooled across all habitat types and standardized by number of 1-ha grids sampled during each primary occasion to yield an ad hoc density estimate. Fig. 2. View largeDownload slide Eastern chipmunk (Tamias striatus) density (individuals/ha) estimates and 95% confidence intervals computed with spatially explicit capture–recapture models incorporating effects of habitat type in experimental harvest units in southern Indiana, 2014–2016. To facilitate comparison with other studies, minimum number alive (MNA) values are pooled across all habitat types and standardized by number of 1-ha grids sampled during each primary occasion to yield an ad hoc density estimate. Fig. 3. View largeDownload slide White-footed mouse (Peromyscus leucopus) density (individuals/ha) estimates and 95% confidence intervals computed with spatially explicit capture–recapture models incorporating effects of habitat type in experimental harvest units in southern Indiana, 2014–2016. To facilitate comparison with other studies, minimum number alive (MNA) values are pooled across all habitat types and standardized by number of 1-ha grids sampled during each primary occasion to yield an ad hoc density estimate. The fall 2016 sampling occasion produced too few captures to analyze density with spatially explicit capture–recapture methods. Fig. 3. View largeDownload slide White-footed mouse (Peromyscus leucopus) density (individuals/ha) estimates and 95% confidence intervals computed with spatially explicit capture–recapture models incorporating effects of habitat type in experimental harvest units in southern Indiana, 2014–2016. To facilitate comparison with other studies, minimum number alive (MNA) values are pooled across all habitat types and standardized by number of 1-ha grids sampled during each primary occasion to yield an ad hoc density estimate. The fall 2016 sampling occasion produced too few captures to analyze density with spatially explicit capture–recapture methods. When habitat and distance from edge were considered simultaneously, there was no discernible effect of distance on eastern chipmunks except in summer 2016, when density of activity centers increased with distance into harvest areas and decreased at greater distances into forest matrix (Table 2). Density of mouse activity centers exhibited consistently positive effects of distance into the opening and consistently negative effects of distance into the forest matrix, although only the matrix × distance effect for summer 2014 was significant (Table 2). Table 2. Estimates of eastern chipmunk (Tamias striatus) and white-footed mouse (Peromyscus leucopus) mean density (ha−1) in clearcut and patch-cut harvest areas, and effects of habitat type (β opening), distance from harvest boundary into forest matrix (β distance into forest), and distance from harvest boundary into the harvest opening (β distance into opening) in southern Indiana, United States. Density Opening Distance into forest Distance into opening Mean SE β CIL CIU β CIL CIU β CIL CIU T. striatus  Summer 15 39.10 72.90 0.26 −0.09 0.62 −0.02 −0.50 0.45 −0.22 −0.77 0.34  Fall 15 2.74 5.33 1.15 0.36 1.93 −0.01 −1.20 1.18 −0.20 −1.55 1.15  Summer 16 2.67 2.31 0.35 −0.14 0.83 −0.62 −1.16 −0.08 0.68 0.04 1.31  Fall 16 6.43 6.23 0.61 −0.03 1.25 0.14 −0.66 0.93 −0.24 −1.25 0.77 P. leucopus  Summer 14 0.92 4.57 −0.78 −2.21 0.66 −1.24 −2.41 −0.09 1.40 −0.37 3.16  Summer 15 12.1 2.99 0.30 −0.21 0.82 −0.29 −0.98 0.41 0.32 −0.52 1.15  Fall 15 7.80 6.65 −0.24 −0.77 0.30 −0.23 −0.82 0.36 0.23 −0.57 1.03 Density Opening Distance into forest Distance into opening Mean SE β CIL CIU β CIL CIU β CIL CIU T. striatus  Summer 15 39.10 72.90 0.26 −0.09 0.62 −0.02 −0.50 0.45 −0.22 −0.77 0.34  Fall 15 2.74 5.33 1.15 0.36 1.93 −0.01 −1.20 1.18 −0.20 −1.55 1.15  Summer 16 2.67 2.31 0.35 −0.14 0.83 −0.62 −1.16 −0.08 0.68 0.04 1.31  Fall 16 6.43 6.23 0.61 −0.03 1.25 0.14 −0.66 0.93 −0.24 −1.25 0.77 P. leucopus  Summer 14 0.92 4.57 −0.78 −2.21 0.66 −1.24 −2.41 −0.09 1.40 −0.37 3.16  Summer 15 12.1 2.99 0.30 −0.21 0.82 −0.29 −0.98 0.41 0.32 −0.52 1.15  Fall 15 7.80 6.65 −0.24 −0.77 0.30 −0.23 −0.82 0.36 0.23 −0.57 1.03 The reference level is forest matrix habitat. For each effect coefficient, bounds of the corresponding 95% confidence interval are provided. Coefficients in bold and italicized type indicate significance at 0.05 and 0.10 levels, respectively. Too few mice were captured in harvest areas in 2016 to model a distance-to-edge effect. View Large Table 2. Estimates of eastern chipmunk (Tamias striatus) and white-footed mouse (Peromyscus leucopus) mean density (ha−1) in clearcut and patch-cut harvest areas, and effects of habitat type (β opening), distance from harvest boundary into forest matrix (β distance into forest), and distance from harvest boundary into the harvest opening (β distance into opening) in southern Indiana, United States. Density Opening Distance into forest Distance into opening Mean SE β CIL CIU β CIL CIU β CIL CIU T. striatus  Summer 15 39.10 72.90 0.26 −0.09 0.62 −0.02 −0.50 0.45 −0.22 −0.77 0.34  Fall 15 2.74 5.33 1.15 0.36 1.93 −0.01 −1.20 1.18 −0.20 −1.55 1.15  Summer 16 2.67 2.31 0.35 −0.14 0.83 −0.62 −1.16 −0.08 0.68 0.04 1.31  Fall 16 6.43 6.23 0.61 −0.03 1.25 0.14 −0.66 0.93 −0.24 −1.25 0.77 P. leucopus  Summer 14 0.92 4.57 −0.78 −2.21 0.66 −1.24 −2.41 −0.09 1.40 −0.37 3.16  Summer 15 12.1 2.99 0.30 −0.21 0.82 −0.29 −0.98 0.41 0.32 −0.52 1.15  Fall 15 7.80 6.65 −0.24 −0.77 0.30 −0.23 −0.82 0.36 0.23 −0.57 1.03 Density Opening Distance into forest Distance into opening Mean SE β CIL CIU β CIL CIU β CIL CIU T. striatus  Summer 15 39.10 72.90 0.26 −0.09 0.62 −0.02 −0.50 0.45 −0.22 −0.77 0.34  Fall 15 2.74 5.33 1.15 0.36 1.93 −0.01 −1.20 1.18 −0.20 −1.55 1.15  Summer 16 2.67 2.31 0.35 −0.14 0.83 −0.62 −1.16 −0.08 0.68 0.04 1.31  Fall 16 6.43 6.23 0.61 −0.03 1.25 0.14 −0.66 0.93 −0.24 −1.25 0.77 P. leucopus  Summer 14 0.92 4.57 −0.78 −2.21 0.66 −1.24 −2.41 −0.09 1.40 −0.37 3.16  Summer 15 12.1 2.99 0.30 −0.21 0.82 −0.29 −0.98 0.41 0.32 −0.52 1.15  Fall 15 7.80 6.65 −0.24 −0.77 0.30 −0.23 −0.82 0.36 0.23 −0.57 1.03 The reference level is forest matrix habitat. For each effect coefficient, bounds of the corresponding 95% confidence interval are provided. Coefficients in bold and italicized type indicate significance at 0.05 and 0.10 levels, respectively. Too few mice were captured in harvest areas in 2016 to model a distance-to-edge effect. View Large The top-ranked AICc model for survival of eastern chipmunks included only an effect of time (primary occasion), with capture probability varying among secondary occasions ( Appendix I). Harvest treatment did not appear to alter survival (Fig. 4A). Relative to the summer 2014 to summer 2015 period, monthly survival was significantly higher from fall 2015 to summer 2016 but significantly lower during the summer–fall periods in 2015 and 2016 (Table 3). The top-ranked model for survival of white-footed mice was an additive combination of treatment and time effects, with capture probability varying among primary occasions ( Appendix I). Monthly survival probabilities (±SE) ranged from 0.44 ± 0.08 to 0.86 ± 0.03 and were lower in harvested sites than in control sites, especially in patch cut and shelterwood harvest areas (Table 3, Fig. 4B). Survival of white-footed mice also was significantly lower from summer to fall 2016 relative to the summer 2014 to summer 2015 reference period (Table 3). No reduction in survival of either species was noted in the overwinter time interval for which there was a poor mast crop (fall 2015–summer 2016). Table 3. Parameter estimates from top robust design models of apparent survival (ϕ) for rodents captured in experimental timber harvest units in southern Indiana, United States. Parameter β SE Lower Upper T. striatus  ϕ(Intercept) 1.96 0.16 1.65 2.27  ϕ(Time: Summer 15–Fall 15) −0.67 0.20 −1.05 −0.29  ϕ(Time: Fall 15–Summer 16) 0.86 0.22 0.42 1.30  ϕ(Time: Summer 16–Fall 16) −0.80 0.20 −1.19 −0.41 P. leucopus  ϕ(Intercept) 1.60 0.22 1.17 2.04  ϕ(Clearcut) −0.30 0.24 −0.78 0.18  ϕ(Shelterwood) −0.53 0.24 −0.99 −0.06  ϕ(Patchcut) −0.66 0.22 −1.10 −0.23  ϕ(Time: summer 15–fall 15) 0.21 0.25 −0.29 0.70  ϕ(Time: fall 15–summer 16) −0.19 0.25 −0.67 0.30  ϕ(Time: summer 16–fall 16) −1.20 0.34 −1.87 −0.53 Parameter β SE Lower Upper T. striatus  ϕ(Intercept) 1.96 0.16 1.65 2.27  ϕ(Time: Summer 15–Fall 15) −0.67 0.20 −1.05 −0.29  ϕ(Time: Fall 15–Summer 16) 0.86 0.22 0.42 1.30  ϕ(Time: Summer 16–Fall 16) −0.80 0.20 −1.19 −0.41 P. leucopus  ϕ(Intercept) 1.60 0.22 1.17 2.04  ϕ(Clearcut) −0.30 0.24 −0.78 0.18  ϕ(Shelterwood) −0.53 0.24 −0.99 −0.06  ϕ(Patchcut) −0.66 0.22 −1.10 −0.23  ϕ(Time: summer 15–fall 15) 0.21 0.25 −0.29 0.70  ϕ(Time: fall 15–summer 16) −0.19 0.25 −0.67 0.30  ϕ(Time: summer 16–fall 16) −1.20 0.34 −1.87 −0.53 The best model for eastern chipmunks (Tamias striatus) included time-specific apparent survival. The best model for white-footed mice (Peromyscus leucopus) included harvest treatment and time effects. Reference levels were control treatment and the summer 2014–summer 2015 time interval. For each parameter estimate, bounds of the corresponding 95% confidence interval are provided. Coefficients in bold font are significant at the 0.05 level. View Large Table 3. Parameter estimates from top robust design models of apparent survival (ϕ) for rodents captured in experimental timber harvest units in southern Indiana, United States. Parameter β SE Lower Upper T. striatus  ϕ(Intercept) 1.96 0.16 1.65 2.27  ϕ(Time: Summer 15–Fall 15) −0.67 0.20 −1.05 −0.29  ϕ(Time: Fall 15–Summer 16) 0.86 0.22 0.42 1.30  ϕ(Time: Summer 16–Fall 16) −0.80 0.20 −1.19 −0.41 P. leucopus  ϕ(Intercept) 1.60 0.22 1.17 2.04  ϕ(Clearcut) −0.30 0.24 −0.78 0.18  ϕ(Shelterwood) −0.53 0.24 −0.99 −0.06  ϕ(Patchcut) −0.66 0.22 −1.10 −0.23  ϕ(Time: summer 15–fall 15) 0.21 0.25 −0.29 0.70  ϕ(Time: fall 15–summer 16) −0.19 0.25 −0.67 0.30  ϕ(Time: summer 16–fall 16) −1.20 0.34 −1.87 −0.53 Parameter β SE Lower Upper T. striatus  ϕ(Intercept) 1.96 0.16 1.65 2.27  ϕ(Time: Summer 15–Fall 15) −0.67 0.20 −1.05 −0.29  ϕ(Time: Fall 15–Summer 16) 0.86 0.22 0.42 1.30  ϕ(Time: Summer 16–Fall 16) −0.80 0.20 −1.19 −0.41 P. leucopus  ϕ(Intercept) 1.60 0.22 1.17 2.04  ϕ(Clearcut) −0.30 0.24 −0.78 0.18  ϕ(Shelterwood) −0.53 0.24 −0.99 −0.06  ϕ(Patchcut) −0.66 0.22 −1.10 −0.23  ϕ(Time: summer 15–fall 15) 0.21 0.25 −0.29 0.70  ϕ(Time: fall 15–summer 16) −0.19 0.25 −0.67 0.30  ϕ(Time: summer 16–fall 16) −1.20 0.34 −1.87 −0.53 The best model for eastern chipmunks (Tamias striatus) included time-specific apparent survival. The best model for white-footed mice (Peromyscus leucopus) included harvest treatment and time effects. Reference levels were control treatment and the summer 2014–summer 2015 time interval. For each parameter estimate, bounds of the corresponding 95% confidence interval are provided. Coefficients in bold font are significant at the 0.05 level. View Large Fig. 4. View largeDownload slide Monthly estimates of apparent survival and 95% confidence intervals for A) eastern chipmunks (Tamias striatus) and B) white-footed mice (Peromyscus leucopus) in experimental timber harvest units of the Hardwood Ecosystem Experiment, southern Indiana. The top-ranked model for chipmunks included an effect of time only, whereas the top-ranked model for white-footed mice included an effect of time and harvest treatment (Table 3). Fig. 4. View largeDownload slide Monthly estimates of apparent survival and 95% confidence intervals for A) eastern chipmunks (Tamias striatus) and B) white-footed mice (Peromyscus leucopus) in experimental timber harvest units of the Hardwood Ecosystem Experiment, southern Indiana. The top-ranked model for chipmunks included an effect of time only, whereas the top-ranked model for white-footed mice included an effect of time and harvest treatment (Table 3). Discussion Rodent populations in southern Indiana exhibited detectable responses to regeneration openings relative to forest matrix sites 6–8 years after harvest. Moreover, interspecific differences in response to harvest were evident. Eastern chipmunk density was greater in harvest openings and along edges of openings, consistent with findings 1–3 years following harvest at the HEE (Kellner et al. 2013), analysis of a 5-decade chronosequence in another oak-dominated forest of southern Indiana (Urban and Swihart 2011), and a study of 3-year-old clearcut oak forest in Pennsylvania (Kirkland et al. 1985). Increased cover provided by dense early successional vegetation, reduced basal area of overstory trees, and coarse woody debris remaining after harvest probably contribute to habitat suitability of harvest openings for eastern chipmunks. Chipmunk responses to shelterwood harvest reinforced this assessment. In 2014 and 2015, densities in shelterwood treatments that had only received the midstory removal preparatory cut did not differ from control sites, comparable with the findings of Kellner et al. (2013). However, after the second stage of the shelterwood harvest in winter 2015 removed canopy trees and reduced basal area, chipmunk density increased significantly in 2016. Moreover, high chipmunk densities in harvest areas were attained without a decrease in per capita survival rate. Collectively, these results suggest that early successional habitat created by clearcuts, patch cuts, and shelterwood harvests after stage 2 provided high-quality food and cover for eastern chipmunks. Few other studies have paired estimates of density and vital rates for forest rodents. However, in a similar study in oak woodlands of central Italy, the density and survival of yellow-necked mice (Apodemus flavicollis) and wood mice (A. sylvaticus) 5–10 years after coppicing exhibited positive responses that were attributed to carrying capacity enhanced by increased protective cover and adequate food resources (Gasperini et al. 2016). In contrast to chipmunks, density of white-footed mice tended not to differ in treatment and control sites 6–8 years after harvest. Kellner et al. (2013) observed a 32% decline in abundance of mice at 4-ha clearcuts during the 3 years immediately following harvest. When examined at the scale of the harvest opening, we saw no evidence for area-dependent negative effects of harvest on mouse density, which others have ascribed to an affinity of white-footed mice for edge habitat (Buckner and Shure 1985; Kellner et al. 2013). It seems likely that in our mesic forests, 6–8 years postharvest is sufficient time for vegetative succession in openings to blur the distinction between opening interiors and edges from the perspective of a mouse population. In the more xeric oak–hickory–shortleaf pine (Pinus echinata) forests of the Missouri Ozarks, Peromyscus abundance increased an estimated 73% on average at the landscape scale; elevated population levels persisted 13–14 years postharvest (Gitzen et al. 2018), consistent with a positive effect of habitat modification and edge creation. In general, reliance on abundance or density to infer habitat quality warrants caution as social interactions, site tenacity, or temporal variation in environmental conditions may produce local densities that mask or even conflict with actual effects of management on habitat quality (Van Horne 1983; Martineau et al. 2016). For instance, density of Myodes glareolus responded positively to recent coppicing of oak, but survival did not vary with forest management (Gasperini et al. 2016). Abundance of M. gapperi also was greater in recently logged boreal forests relative to old growth, but greater use did not reflect higher habitat quality; rather, it was attributed to interference competition that resulted in higher numbers of juveniles forced into harvested areas by negative social interactions (Martineau et al. 2016). In our study, per capita survival of mice was negatively impacted by patch cuts and shelterwood harvests relative to controls, which suggests that mouse populations may incur survival costs from these management systems 6–8 years postharvest. Since survival of eastern chipmunks (diurnal) was not affected by harvest treatment, nocturnal predators that feed frequently on white-footed mice (nocturnal), such as owls, may be important mortality factors for these mouse populations. Eastern screech-owls (Megascops asio), in particular, are associated with edge habitats during the nonbreeding season (Sparks et al. 1994), commonly incorporate mice into their diets (Ritchison and Cavanagh 1992), and thus may contribute to decreased survival of mice in small patch or shelterwood cuts with a large proportion of edge (see also Widén 1994). Moreover, edge species such as white-footed mice occur in greater densities in habitats dominated by edge (Table 1; Nupp and Swihart 1996, 1998; Anderson et al. 2003; Wilder and Meikle 2005, 2006), which may induce behavioral responses by predators such as owls (Schmidt and Ostfeld 2008). Edge creation is a prominent feature of timber harvest. Surprisingly, the extent of edge effects on rodent populations in eastern deciduous forests has received little attention. Explicitly modeling the relationship of density of activity centers with distance from harvest boundaries allows a more flexible definition of edge habitat and a more refined examination of rodent responses to edge. Distance-dependent density relationships were observed for eastern chipmunks in summer 2016; density of activity centers in the forest matrix declined with increasing distance from edge, whereas in the harvest openings it increased with distance from edge (Table 2). White-footed mice showed similar distance-dependent density in matrix habitat in summer 2014, and for harvest openings in all primary occasions, although standard errors were large. Consistent with the wide ecological tolerance of both white-footed mice and eastern chipmunks (Mahan and Yahner 1998; Swihart et al. 2003), the observed distance-dependent density relationship suggests that they are operating as edge species in this system. Structural complexity providing cover and food resources is increased in edge habitat (Anderson et al. 2003), which provides multiple benefits to prey species such as mice and chipmunks. The increased growth of tree seedlings 6–8 years following harvest may have contributed to structural complexity or otherwise modified the opening habitat in a way that made it more attractive to white-footed mice. The demographic responses we observed probably have cascading effects on trophic interactions generally, and specifically on the degree to which seed-caching rodents function as seed dispersers or seed predators in managed forests (Schupp et al. 2010; Kellner et al. 2016; Yu et al. 2017). For example, in oak-dominated forests of the eastern United States, eastern chipmunks and white-footed mice are important predators and dispersers of oak acorns (McShea and Healy 2002). Chipmunks are primarily larder hoarders, so increased local chipmunk abundance in the short- (1–3 years) and mid-term (5–10 years) following harvest may result in reduced acorn survival and thus reduced oak regeneration success in these localities. On the other hand, more rodents on the edges of harvest areas may increase the chance that acorns are dispersed from the surrounding area to the edges, where intermediate light conditions promote oak seedling growth and survival (Kellner and Swihart 2016; Greenler and Saunders 2019). The potential impacts of harvest on trophic interactions emphasize the need for researchers and land managers to monitor multiple components of managed forest ecosystems, both for conservation and to improve management outcomes (Kellner and Swihart 2017). Acknowledgments We thank S. Haulton, M. Saunders, and P. Zollner who provided invaluable input on study design, analysis, and earlier versions of the manuscript. We also thank the two anonymous reviewers for helpful comments on the manuscript, and members of the Swihart and Saunders labs for helpful feedback and support in the field. We are grateful to A. Meier, P. Ma, C. Owings, and J. Riegel for facilitating and coordinating data collection efforts with the help of numerous technicians. This paper is a contribution of the Hardwood Ecosystem Experiment, a partnership of the Indiana Department of Natural Resources, Purdue University, Ball State University, Indiana State University, Drake University, Indiana University Pennsylvania, Indiana University, and The Nature Conservancy. Funding for this project was provided by the Indiana Division of Forestry and the Purdue University Department of Forestry and Natural Resources. Literature Cited Anderson , C. S. , A. B. Cady , and D. B. Meikle . 2003 . Effects of vegetation structure and edge habitat on the density and distribution of white-footed mice in small and large forest patches . Canadian Journal of Zoology 81 : 897 – 904 . Google Scholar Crossref Search ADS WorldCat Ash , A. N . 1995 . Effects of clear-cutting on litter parameters in the Southern Blue Ridge Mountains . Castanea 60 : 89 – 97 . WorldCat Boerner , R. E. J . 2000 . Effects of fire on the ecology of the forest floor and soil of central hardwood forests. Pp. 56 – 63 in Fire, people, and the central hardwoods landscape ( Yaussy , D. A. , ed.). General Technical Report NE-274. U.S. Forest Service , Newtown Square, Pennsylvania. Google Preview WorldCat COPAC Borchers , D. L., and M. G. Efford . 2008 . Spatially explicit maximum likelihood methods for capture-recapture studies . Biometrics 64 : 377 – 385 . Google Scholar Crossref Search ADS PubMed WorldCat Buckner , C. A. , and D. J. Shure . 1985 . The response of Peromyscus to forest opening size in the southern Appalachian mountains . Journal of Mammalogy 66 : 299 – 307 . Google Scholar Crossref Search ADS WorldCat Efford , M. G . 2004 . Density estimation in live-trapping studies . Oikos 106 : 598 – 610 . Google Scholar Crossref Search ADS WorldCat Efford , M. G . 2016 . secr: Spatially explicit capture-recapture models. R package version 2.10.4 . https://CRAN.R-project.org/package=secr Fantz , D. K. , and R. B. Renken . 2005 . Short-term landscape-scale effects of forest management on Peromyscus spp. Mice within Missouri Ozark forests . Wildlife Society Bulletin 33 : 293 – 301 . Google Scholar Crossref Search ADS WorldCat Fralish , J. S . 2004 . The keystone role of oak and hickory in the Central Hardwood Forest. Pp. 78 – 87 in Upland oak ecology symposium: history, current conditions, and sustainability ( M. A. Spetich , ed.). U.S. Forest Service , General Technical Report SRS-73, Asheville, North Carolina. Google Preview WorldCat COPAC Fredericksen , T. S. , et al. 1999 . Short-term understory plant community responses to timber-harvesting intensity on non-industrial private forestlands in Pennsylvania . Forest Ecology and Management 116 : 129 – 139 . Google Scholar Crossref Search ADS WorldCat Gasperini , S. , A. Mortelliti , P. Bartolommei , A. Bonacchi , E. Manzo , and R. Cozzolino . 2016 . Effects of forest management on density and survival in three forest rodent species . Forest Ecology and Management 382 : 151 – 160 . Google Scholar Crossref Search ADS WorldCat Gitzen , R. A. , et al. 2018 . Peromyscus responses to alternative forest management systems in the Missouri Ozarks, USA . Forest Ecology and Management 429 : 558 – 569 . Google Scholar Crossref Search ADS WorldCat Greenberg , C. H . 2002 . Response of white-footed mice (Peromyscus leucopus) to coarse woody debris and microsite use in southern Appalachian treefall gaps . Forest Ecology and Management 164 : 57 – 66 . Google Scholar Crossref Search ADS WorldCat Greenberg , C. H. , S. Miller , and T. A. Waldrop . 2007 . Short-term response of shrews to prescribed fire and mechanical fuel reduction in a southern Appalachian upland hardwood forest . Forest Ecology and Management 243 : 231 – 236 . Google Scholar Crossref Search ADS WorldCat Greenberg , C. H. , D. L. Otis and T. A. Waldrop . 2006 . Response of white-footed mice (Peromyscus leucopus) to fire and fire surrogate fuel reduction treatments in a southern Appalachian hardwood forest . Forest Ecology and Management 234 : 355 – 362 . Google Scholar Crossref Search ADS WorldCat Greenberg , C. H. , R. W. Perry , C. A. Harper , D. J. Levey , and J. M. McCord . 2011 . The role of young, recently disturbed upland hardwood forest as high quality food patches. Pp. 121 – 141 in Sustaining young forest communities: ecology and management of early successional habitats in the Central Hardwood Region, USA ( C. H. Greenberg , B. S. Collins , and F. R. Thompson , III , eds). Springer , New York . Google Preview WorldCat COPAC Greenler , S. M. , and M. R. Saunders . 2019 . Short-term, spatial regeneration patterns following expanding group shelterwood harvests and prescribed fire in the Central Hardwood Region . Forest Ecology and Management 432 : 1053 – 1063 . Google Scholar Crossref Search ADS WorldCat Healy , W. M. , and R. T. Brooks . 1988 . Small mammal abundance in northern hardwood stands in West Virginia . Journal of Wildlife Management 52 : 491 – 496 . Google Scholar Crossref Search ADS WorldCat Herbers , J. , and W. Klenner . 2007 . Effects of logging pattern and intensity on squirrel demography . Journal of Wildlife Management 71 : 2655 – 2663 . Google Scholar Crossref Search ADS WorldCat Hodson , J. , D. Fortin , M. L. Leblanc , and L. Bélanger . 2010 . An appraisal of the fitness consequences of forest disturbance for wildlife using habitat selection theory . Oecologia 164 : 73 – 86 . Google Scholar Crossref Search ADS PubMed WorldCat Homoya , M. A. , D. B. Abrell , J. A. Aldrich , and T. W. Post . 1984 . The natural regions of Indiana . Proceedings of the Indiana Academy of Science 94 : 254 – 268 . WorldCat Kalb , R. A. , and C. J. Mycroft . 2013 . The Hardwood Ecosystem Experiment: goals, design, and implementation. Pp. 36 – 59 in The Hardwood Ecosystem Experiment: a framework for studying responses to forest management ( R. K. Swihart , M. R. Saunders , R. A. Kalb , S. Haulton , and C. H. Michler , eds.). General Technical Report NRS-P-108. U.S. Forest Service, Newtown Square, Pennsylvania . Google Preview WorldCat COPAC Kaminski , J. A. , M. L. Davis , M. Kelly , and P. D. Keyser . 2007 . Disturbance effects on small mammal species in a managed Appalachian Forest . American Midland Naturalist 157 : 385 – 397 . Google Scholar Crossref Search ADS WorldCat Kellner , K. F. , N. I. Lichti , and R. K. Swihart . 2016 . Midstory removal reduces effectiveness of oak (Quercus) acorn dispersal by small mammals in the Central Hardwood Region . Forest Ecology and Management 375 : 182 – 190 . Google Scholar Crossref Search ADS WorldCat Kellner , K. F. , J. K. Riegel , and R. K. Swihart . 2014 . Effects of silvicultural disturbance on acorn infestation and removal . New Forests 45 : 265 – 281 . Google Scholar Crossref Search ADS WorldCat Kellner , K. F. , and R. K. Swihart . 2016 . Timber harvest and drought interact to impact oak seedling growth and survival in the Central Hardwood Forest . Ecosphere 7 : 10 . Google Scholar Crossref Search ADS WorldCat Kellner , K. F., and R. K. Swihart . 2017 . Simulation of oak early life history and interactions with disturbance via an individual-based model, SOEL . PLoS One 12 : e0179643 . doi:10.1371/journal.pone.0179643. Google Scholar Crossref Search ADS PubMed WorldCat Kellner , K. F. , N. A. Urban , and R. K. Swihart . 2013 . Short-term responses of small mammals to timber harvest in the U.S. central hardwood forest region . Journal of Wildlife Management 77 : 1650 – 1663 . Google Scholar Crossref Search ADS WorldCat Kirkland , G. L. , Jr. 1990 . Patterns of initial small mammal community change after clearcutiing of temperate North American forests . Oikos . 59 : 313 – 320 . Google Scholar Crossref Search ADS WorldCat Kirkland , G. L. , Jr. , T. R. Johnston , Jr. , and P. F. Steblein . 1985 . Small mammal exploitation of a forest-clearcut interface . Acta Theriologica 30 : 211 – 218 . Google Scholar Crossref Search ADS WorldCat Laake , J. L . 2016 . RMark: an R interface for analysis of capture-recapture data with MARK. R package version 2.2.2 . http://CRAN.R-project.Org/package=RMark Larsen , A. L. , J. J. Jacquot , P. W. Keenlance , and H. L. Keough . 2016 . Effects of an ongoing oak savanna restoration on small mammals in Lower Michigan . Forest Ecology and Management 367 : 120 – 127 . Google Scholar Crossref Search ADS WorldCat Leonard , O. D. , et al. 2015 . Effect of variation in forest harvest intensity on winter occupancy of barred owls and eastern screech-owls in deciduous forests of the east-central United States . Journal of Field Ornithology 86 : 115 – 129 . Google Scholar Crossref Search ADS WorldCat Mahan , C. G. , and R. H. Yahner . 1998 . Lack of population response by eastern chipmunks (Tamias striatus) to forest fragmentation . American Midland Naturalist 140 : 382 – 386 . Google Scholar Crossref Search ADS WorldCat Manning , J. A. , and W. D. Edge . 2004 . Small mammal survival and downed wood at multiple scales in managed forests . Journal of Mammalogy 85 : 87 – 96 . Google Scholar Crossref Search ADS WorldCat Martineau , J. , D. Pothier , and D. Fortin . 2016 . Processes driving short-term temporal dynamics of small mammal distribution in human-disturbed environments . Oecologia 181 : 831 – 840 . Google Scholar Crossref Search ADS PubMed WorldCat Maser , C. , R. A. Nussbaum , and J. M. Trappe . 1978 . Fungal small mammal interrelationships with emphasis on Oregon coniferous forests . Ecology 59 : 799 – 809 . Google Scholar Crossref Search ADS WorldCat McShea , W. J. and W. M. Healy . 2002 . Oak forest ecosystems: ecology and management for wildlife . Johns Hopkins University Press , Baltimore, Maryland. Google Preview WorldCat COPAC Miller , D. , and L. Getz . 1977 . Factors influencing local distribution and species diversity of forest small mammals in New England . Canadian Journal of Zoology 55 : 806 – 814 . Google Scholar Crossref Search ADS WorldCat Nupp , T. E. , and R. K. Swihart . 1996 . Effect of forest patch area on population attributes of white-footed mice (Peromyscus leucopus) in fragmented landscapes . Canadian Journal of Zoology 74 : 467 – 472 . Google Scholar Crossref Search ADS WorldCat Nupp , T. E. , and R. K. Swihart . 1998 . Effects of forest fragmentation on populationattributes of white-footed mice and eastern chipmunks . Journal of Mammalogy 79 : 1234 – 1243 . Google Scholar Crossref Search ADS WorldCat Olson , Z. H. , MacGowan , B. J. , Hamilton , M. T. , Currylow , A. F. , and R. N. Williams . 2015 . Survival of timber rattlesnakes (Crotalus horridus): investigating individual, environmental, and ecological effects . Herpetologica 71 : 274 – 279 . Google Scholar Crossref Search ADS WorldCat Parmenter , R. R. , et al. 2003 . Small-mammal density estimation: a field comparison of grid-based vs. web-based density estimators . Ecological Monographs 73 : 1 – 26 . Google Scholar Crossref Search ADS WorldCat Perry , R. W. , P. A. Tappe , and D. G. Peitz . 2004 . Initial response of individual soft-mast producing plants to different forest regeneration strategies in the Ouachita Mountains. Pp. 60 – 70 in Ouachita and Ozark Mountains symposium: ecosystem management research ( J. M. Guldin , ed.). General Technical Report SRS-74. U.S. Forest Service . Google Preview WorldCat COPAC Pollock , K. H . 1982 . A capture-recapture design robust to unequal probability of capture . Journal of Wildlife Management 46 : 752 – 757 . Google Scholar Crossref Search ADS WorldCat Ransome , D. B. , P. M. F. Lindgren , M. J. Waterhouse , H. M. Armleder , and T. P. Sullivan . 2009 . Small-mammal response to group-selection silvicultural systems in Engelmann spruce-subalpine fir forests 14 years postharvest . Canadian Journal of Forest Research 39 : 1698 – 1708 . Google Scholar Crossref Search ADS WorldCat R Development Core Team . 2016 . R: A language and environment for statistical computing . R Foundation for Statistical Computing , Vienna, Austria . www.R-project.org/. Google Preview WorldCat COPAC Ritchison , G. , and P. M. Cavanagh . 1992 . Prey use by Eastern screech-owls: seasonal variation in central Kentucky and a review of previous studies . Journal of Raptor Research 26 : 66 – 73 . WorldCat Royle , J. A. , R. B. Chandler , R. Sollmann , and B. Gardner . 2014 . Spatial capture-recapture . Academic Press , Boston, Massachusetts . Google Preview WorldCat COPAC Saunders , M. R. , and J. E. Arseneault . 2013 . Pre-treatment analysis of woody vegetation composition and structure on the Hardwood Ecosystem Experiment research units. Pp. 96 – 125 in The Hardwood Ecosystem Experiment: a framework for studying responses to forest management ( R. K. Swihart , M. R. Saunders , R. A. Kalb , S. Haulton and C. H. Michler , eds.). General Technical Report, NRS-P-108. U.S. Forest Service, Newtown Square, Pennsylvania . Google Preview WorldCat COPAC Schmidt , K. A. , and R. S. Ostfeld . 2003 . Songbird populations in fluctuating environments: predator responses to pulsed resources . Ecology 84 : 406 – 415 . Google Scholar Crossref Search ADS WorldCat Schmidt , K. A., and R. S. Ostfeld . 2008 . Numerical and behavioral effects within a pulse-driven system: consequences for shared prey . Ecology 89 : 635 – 646 . Google Scholar Crossref Search ADS PubMed WorldCat Schupp , E. W. , P. Jordano , and J. M. Gómez . 2010 . Seed dispersal effectiveness revisited: a conceptual review . The New Phytologist 188 : 333 – 353 . Google Scholar Crossref Search ADS PubMed WorldCat Sikes , R. S. , and The Animal Care and Use Committee of the American Society of Mammalogists . 2016 . 2016 guidelines of the american society of mammalogists for the use of wild mammals in research and education . Journal of Mammalogy 97 : 663 – 688 . Google Scholar Crossref Search ADS PubMed WorldCat Sparks , E. J. , J. R. Belthoff , and G. Ritchison . 1994 . Habitat use by eastern screech-owls in central Kentucky . Journal of Field Ornithology 65 : 83 – 95 . WorldCat Sullivan , T. P. , and D. S. Sullivan . 2017 . Old-growth characteristics 20 years after thinning and repeated fertilization of lodgepole pine forest: tree growth, structural attributes, and red-backed voles . Forest Ecology and Management 391 : 207 – 220 . Google Scholar Crossref Search ADS WorldCat Swihart , R. K. , T. M. Gehring , M. B. Kolozsvary , and T. E. Nupp . 2003 . Responses of ‘resistant’ vertebrates to habitat loss and fragmentation: the importance of niche breadth and range boundaries . Diversity and Distributions 9 : 1 – 18 . Google Scholar Crossref Search ADS WorldCat Swihart , R. K. , J. Riegel , and K. Kellner . 2019 . Oak acorn production and predation. P. 17 in Hardwood Ecosystem Experiment: 2006–2016 ( C. Owings , ed.). Purdue Extension FNR-570-W, in press. Google Preview WorldCat COPAC Urban , N. A. , and R. K. Swihart . 2011 . Small mammal responses to forest management for oak regeneration in southern Indiana . Forest Ecology and Management 261 : 353 – 361 . Google Scholar Crossref Search ADS WorldCat Van Horne , B . 1983 . Density as a misleading indicator of habitat quality . Journal of Wildlife Management 47 : 893 – 901 . Google Scholar Crossref Search ADS WorldCat Whitaker , J. O. , Jr. and W. J. Hamilton , Jr. 1998 . Mammals of the Eastern United States , Cornell University Press . Ithaca, New York . Google Preview WorldCat COPAC White , G. C. and K. P. Burnham . 1999 . Program MARK: survival estimation from populations of marked animals . Bird Study 46 ( Suppl ): 120 – 138 . Google Scholar Crossref Search ADS WorldCat Widén , P . 1994 . Habitat quality for raptors: a field experiment . Journal of Avian Biology 25 : 219 – 223 . Google Scholar Crossref Search ADS WorldCat Wilder , S. M. , and D. B. Meikle . 2005 . Reproduction, foraging and the negative density–area relationship of a generalist rodent . Oecologia 54 : 129 – 134 . WorldCat Wilder , S. M. , and D. B. Meikle . 2006 . Variation in effects of fragmentation on the white-footed mouse (Peromyscus leucopus) during the breeding season . Journal of Mammalogy 87 : 117 – 123 . Google Scholar Crossref Search ADS WorldCat Wilson , K. R. and D. R. Anderson . 1985 . Evaluation of two density estimators of small mammal population size . Journal of Mammalogy . 66 : 13 – 21 . Google Scholar Crossref Search ADS WorldCat Yu , F. , X. Shi , X. Zhang , X. Yi , D. Wang , and J. Ma . 2017 . Effects of selective logging on rodent-mediated seed dispersal. 2017 . Forest Ecology and Management 406 : 147 – 154 . Google Scholar Crossref Search ADS WorldCat Zheng , D. , J. Chen , B. Song , M. Xu , P. Sneed , and R. Jensen . 2000 . Effects of silvicultural treatments on summer forest microclimate in southeastern Missouri Ozarks . Climate Research 15 : 45 – 59 . Google Scholar Crossref Search ADS WorldCat Zollner , P. A. , and K. J. Crane . 2003 . Influence of canopy closure and shrub coverage on travel along coarse woody debris by eastern chipmunks (Tamias striatus) . American Midland Naturalist 150 : 151 – 157 . Google Scholar Crossref Search ADS WorldCat Appendix I Robust design models estimating apparent survival (ϕ), temporary emigration (ϒ′ and ϒ″), and capture probability (p) of white-footed mice (Peromyscus leucopus) and eastern chipmunks (Tamias striatus) in experimental timber harvest units at the Hardwood Ecosystem Experiment in southern Indiana, United States, 2014–2016. Competing models are ranked by Akaike’s information criterion adjusted for small sample sizes (AICc). The parameter K indicates number of parameters in each model. Parameters specified as (~1) are constant. Time refers to primary sampling occasions and 2° represents secondary occasions. K ΔAICc Weight Deviance Models for eastern chipmunks  ϕ(~Time) p(~2°) ϒ″(~1)ϒ′(~1) 14 0.00 0.91 3,234.65  ϕ(~Treatment + Time) p(~2°) ϒ″(~1)ϒ′(~1) 17 4.76 0.09 3,233.29  ϕ(~Treatment*Time) p(~2°) ϒ″(~1)ϒ′(~1) 27 13.61 0.00 3,221.61  ϕ(~1) p(~2°) ϒ″(~1)ϒ′(~1) 11 34.58 0.00 3,275.32  ϕ(~Treatment) p(~2°) ϒ″(~1)ϒ′(~1) 13 128.6 0.00 3,365.26 Models for white-footed mice  ϕ(~Treatment + Time) p(~Time) ϒ″(~1)ϒ′(~1) 17 0.00 0.86 −471.64  ϕ(~Time) p(~Time) ϒ″(~1)ϒ′(~1) 14 4.53 0.09 −460.87  ϕ(~Treatment*Time) p(~Time) ϒ″(~1)ϒ′(~1) 27 5.82 0.05 −486.98  ϕ(~1) p(~Time) ϒ″(~1)ϒ′(~1) 7 192.56 0.00 −258.44  ϕ(~Treatment) p(~Time) ϒ″(~1)ϒ′(~1) 9 194.20 0.00 −260.90 K ΔAICc Weight Deviance Models for eastern chipmunks  ϕ(~Time) p(~2°) ϒ″(~1)ϒ′(~1) 14 0.00 0.91 3,234.65  ϕ(~Treatment + Time) p(~2°) ϒ″(~1)ϒ′(~1) 17 4.76 0.09 3,233.29  ϕ(~Treatment*Time) p(~2°) ϒ″(~1)ϒ′(~1) 27 13.61 0.00 3,221.61  ϕ(~1) p(~2°) ϒ″(~1)ϒ′(~1) 11 34.58 0.00 3,275.32  ϕ(~Treatment) p(~2°) ϒ″(~1)ϒ′(~1) 13 128.6 0.00 3,365.26 Models for white-footed mice  ϕ(~Treatment + Time) p(~Time) ϒ″(~1)ϒ′(~1) 17 0.00 0.86 −471.64  ϕ(~Time) p(~Time) ϒ″(~1)ϒ′(~1) 14 4.53 0.09 −460.87  ϕ(~Treatment*Time) p(~Time) ϒ″(~1)ϒ′(~1) 27 5.82 0.05 −486.98  ϕ(~1) p(~Time) ϒ″(~1)ϒ′(~1) 7 192.56 0.00 −258.44  ϕ(~Treatment) p(~Time) ϒ″(~1)ϒ′(~1) 9 194.20 0.00 −260.90 View Large K ΔAICc Weight Deviance Models for eastern chipmunks  ϕ(~Time) p(~2°) ϒ″(~1)ϒ′(~1) 14 0.00 0.91 3,234.65  ϕ(~Treatment + Time) p(~2°) ϒ″(~1)ϒ′(~1) 17 4.76 0.09 3,233.29  ϕ(~Treatment*Time) p(~2°) ϒ″(~1)ϒ′(~1) 27 13.61 0.00 3,221.61  ϕ(~1) p(~2°) ϒ″(~1)ϒ′(~1) 11 34.58 0.00 3,275.32  ϕ(~Treatment) p(~2°) ϒ″(~1)ϒ′(~1) 13 128.6 0.00 3,365.26 Models for white-footed mice  ϕ(~Treatment + Time) p(~Time) ϒ″(~1)ϒ′(~1) 17 0.00 0.86 −471.64  ϕ(~Time) p(~Time) ϒ″(~1)ϒ′(~1) 14 4.53 0.09 −460.87  ϕ(~Treatment*Time) p(~Time) ϒ″(~1)ϒ′(~1) 27 5.82 0.05 −486.98  ϕ(~1) p(~Time) ϒ″(~1)ϒ′(~1) 7 192.56 0.00 −258.44  ϕ(~Treatment) p(~Time) ϒ″(~1)ϒ′(~1) 9 194.20 0.00 −260.90 K ΔAICc Weight Deviance Models for eastern chipmunks  ϕ(~Time) p(~2°) ϒ″(~1)ϒ′(~1) 14 0.00 0.91 3,234.65  ϕ(~Treatment + Time) p(~2°) ϒ″(~1)ϒ′(~1) 17 4.76 0.09 3,233.29  ϕ(~Treatment*Time) p(~2°) ϒ″(~1)ϒ′(~1) 27 13.61 0.00 3,221.61  ϕ(~1) p(~2°) ϒ″(~1)ϒ′(~1) 11 34.58 0.00 3,275.32  ϕ(~Treatment) p(~2°) ϒ″(~1)ϒ′(~1) 13 128.6 0.00 3,365.26 Models for white-footed mice  ϕ(~Treatment + Time) p(~Time) ϒ″(~1)ϒ′(~1) 17 0.00 0.86 −471.64  ϕ(~Time) p(~Time) ϒ″(~1)ϒ′(~1) 14 4.53 0.09 −460.87  ϕ(~Treatment*Time) p(~Time) ϒ″(~1)ϒ′(~1) 27 5.82 0.05 −486.98  ϕ(~1) p(~Time) ϒ″(~1)ϒ′(~1) 7 192.56 0.00 −258.44  ϕ(~Treatment) p(~Time) ϒ″(~1)ϒ′(~1) 9 194.20 0.00 −260.90 View Large © 2019 American Society of Mammalogists, www.mammalogy.org This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Rodent population density and survival respond to disturbance induced by timber harvest
JO - Journal of Mammalogy
DO - 10.1093/jmammal/gyz094
DA - 2019-07-27
UR - https://www.deepdyve.com/lp/oxford-university-press/rodent-population-density-and-survival-respond-to-disturbance-induced-z2fKqYZ40b
SP - 1253
VL - 100
IS - 4
DP - DeepDyve
ER -