Regular issues
Special Issues

Southeastern Naturalist
    SENA Home
    Range and Scope
    Board of Editors
    Editorial Workflow
    Publication Charges

Other EH Journals
    Northeastern Naturalist
    Caribbean Naturalist
    Neotropical Naturalist
    Urban Naturalist
    Eastern Paleontologist
    Journal of the North Atlantic
    Eastern Biologist

EH Natural History Home

Effects of Class-level Vegetation Characteristics on Nesting Success of Bewick’s Wrens
Sara E. Harrod and M. Clay Green

Southeastern Naturalist, Volume 17, Issue 3 (2018): 381–395

Full-text pdf (Accessible only to subscribers.To subscribe click here.)


Site by Bennett Web & Design Co.
Southeastern Naturalist 381 S.E. Harrod and M.C. Green 22001188 SOUTHEASTERN NATURALIST 1V7o(3l.) :1378,1 N–3o9. 53 Effects of Class-level Vegetation Characteristics on Nesting Success of Bewick’s Wrens Sara E. Harrod1,* and M. Clay Green2 Abstract - Understanding how habitat affects avian nest success is paramount, particularly when nest boxes are used. Once widespread, Thryomanes bewickii (Bewick’s Wren) are presently confined to south-central and western North America. Our objective was to determine which vegetation variables influenced Bewick’s Wren nesting success in nest boxes in central Texas. Between 2013 and 2014, we estimated nest success for 83 nests and measured vegetation variables surrounding nest boxes. Shrubland perimeter–area ratio negatively impacted nest success, whereas distance to nearest tree had a positive effect. Our results indicate that nests close to trees and in patches with large areas exposed to edge experienced lower nest success; these vegetation variables should be considered when using nest boxes to manage for Bewick’s Wrens. Introduction The mechanisms and factors that influence avian nest success have received considerable attention in the ornithological literature (Fu et al. 2016, Gjerdrum et al. 2005, P.A. Smith et al. 2007). Such variables often include abundance and density of predators (Benson et al. 2010), intra- and interspecific competition (Ingold 1989), brood parasitism (Patten et al. 2006), and habitat characteristics (Hudson and Bollinger 2013). Vegetation characteristics may be particularly important as landscapes become urbanized, fragmented, or otherwise altered as human populations continue to grow. Therefore, understanding the effects of vegetation on nesting success is of critical importance for assessing the long-term persistence of species (Chace and Walsh 2006, Herkert et al. 2003, Newmark and Stanley 201 1). Conservationists often use artificial cavities (i.e., nest boxes) as an inexpensive management tool to encourage nesting by avian species (Jackson et al. 2012, Rohrbaugh and Yahner 1997). Nest boxes are particularly important as a management tool in areas where trees and snags are not abundant (Goldingay and Stevens 2009, Petty et al. 1994). Nest boxes are essential to the nesting success of particular species; Progne subis L. (Purple Martins) are now almost entirely dependent on artificial nest-cavities (Jackson and Tate 1974), and nest box programs are partially credited with restoring once-declining Sialia sialis L. (Eastern Bluebird; Gowaty and Plissner 2015) and Purple Martin (Copley et al. 1999) populations. Although a growing body of evidence suggests that cavity-nesting birds select nest sites based on multiple macro- and microhabitat features (Jackson et al. 2012, Maziarz and 1Department of Biological Sciences, Arkansas State University, Jonesboro, AR 72401. 2Department of Biology, Texas State University, San Marcos, TX 78666. *Corresponding author - Manuscript Editor: Frank Moore Southeastern Naturalist S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 382 Wesołowski 2013, Winter and Faaborg 1999), nest boxes are often erected based on vegetation type alone and can create ecological traps (Jackson et al. 2011, Kight and Swaddle 2007). For example, consider a nest box placed in a woodland patch. The patch may be too small and therefore have too few resources to support parents and offspring, but if the patch is adjacent to discontinuous yet accessible patches, parents may be able to successfully rear young at the nest box. However, the breeding pair would potentially expend much time and energy foraging far from the nest site (Mazgajski and Rejt 2006) and risk considerable exposure to predators (Hinsley 2000). Additionally, in such a case, the fledglings may be forced to disperse long distances to reach optimal habitat patches (Jackson et al. 2011). In addition to vegetation type, habitat and vegetation characteristics are important factors for nest success (Milligan and Dickinson 2016). Within fragmented patches, nests near habitat edges may be more susceptible to depredation and brood parasitism (Deng and Gao 2005, Newmark and Stanley 2011, Patten et al. 2006). Sites with a high percentage of vegetation cover may conceal nests from predators and increase nest success (Gjerdrum et al. 2005, P.A. Smith et al. 2007). Other variables that may be important include nest height (Hudson and Bollinger 2013), vegetation species (Gjerdrum et al. 2005), basal area of live trees (Purcell and Verner 2008), and patch size (Herkert et al. 2003). Although each of these factors may not be considered or addressed, nest-box studies make field work less logistically difficult (i.e., less time is spent searching for nests) and are often valuable for evaluating nest-site–selection factors. Studies of cavity-nesting species often focus on a small subset of avian species, e.g., widespread species that frequently use nest boxes like Eastern Bluebirds and Tachycineta bicolor Vieillot (Tree Swallows), whereas fewer studies have occurred on opportunistic cavity-nesting species like Thryomanes bewickii Audubon (Bewick’s Wren). Bewick’s Wrens are small, secondary cavity-nesting passerines native to North America. Primarily insectivores, they regularly nest in a variety of sites, including natural cavities, nest boxes, exposed shelves in buildings, and underneath buildings (Bibbee 1947). Habitat preference varies by region, but Bewick’s Wrens typically breed in scrublands, thickets, chaparral, and woodlands (Kennedy and White 2013). Populations once extended across the US, but declined by 31.0% between 1970 and 2015 (Partners in Flight 2017). These declines were correlated with an increase in post-agricultural secondary succession (Odum and Johnston 1951) and range expansion of Troglodytes aedon Vieillot (House Wrens), which outcompete Bewick’s Wrens and destroy their nests and eggs (Kennedy and White 1996). Today, Bewick’s Wrens are effectively extirpated from the eastern US, and populations are confined to the southwest and Pacific coasts (Kennedy and White 2013, Taylor 2003). Despite this drastic decline in range, the primary literature for this species is scant, and little is known about their demographic parameters and which habitat factors affect nest success (but see Pogue and Schnell 1994). The 2017 Partners in Flight Assessment also predicted that Bewick’s Wren populations are likely to experience moderate to large population declines in the future, and a large portion of the population is likely to be affected by deterioration in breeding and non-breeding Southeastern Naturalist 383 S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 conditions (Partners in Flight 2017). Given these historical and future declines, as well as the existing knowledge gaps about the general ecology of the species, research on the use of nest boxes and the effects of vegetation characteristics on nest success are essential for managing Bewick’s Wren populations in the remainder of their range. The main objective of our study was to examine the influence of class-level vegetation characteristics on the nesting success of Bewick’s Wrens in central Texas. Although Bewick’s Wrens are abundant in this region (Kennedy and White 2013), no studies of these populations have been published. We hypothesized that the type and characteristics of vegetation surrounding nest boxes impact nest success. Based on previous studies of Bewick’s Wren habitat preferences and the effects of small, fragmented patches on nest success, we predicted that nesting success would be highest at nest boxes located within larger (i.e., contiguous) patches of shrublands and woodlands. Field-Site Description The Freeman Center (29°56'17.05"N, 98°0'30.24"W), located 10 km northwest of San Marcos, TX, comprises 1416 ha of mostly undeveloped land. Most of the anthropogenic disturbance is due to gravel roads and cattle grazing (~85 head). The landscape of the Center is typical of the eastern edge of the Edwards Plateau–Texas Hill Country, with drought-resistant vegetation that thrives in shallow, rocky soil. Common tree species include Juniperus ashei J. Buchholz (Ashe Juniper), Quercus fusiformis Small (Plateau Live Oak), Prosopis glandulosa Torr. (Honey Mesquite), and Acacia farnesiana L. Wight and Arn. (Huisache). Buchloe dactyloides Nutt Engelm. (Buffalograss), and Schizachyrium scoparium Michx (Little Bluestem) are also abundant (Barnes et al. 2000). The primary vegetation types are savannah, mesquite shrubland, and oak–juniper woodland. Methods In January 2013, we followed the Cornell Lab of Ornithology’s specifications (Cornell Lab of Ornithology 2009a) to build and establish 40 Eastern Bluebird boxes at the Freeman Center, (Fig. 1). Our reason for building the nest boxes was to monitor the nest success of Eastern Bluebirds; however, due to the similarities in nest-box dimensions for Eastern Bluebirds and Bewick’s Wrens (Cornell Lab of Ornithology 2009a, b), most nest boxes were occupied by wrens. To compensate for the extreme summer heat of the region, we used ~3-cm thick untreated red-cedar wood for insulation and heat shields and constructed ventilation slots to reduce heat stress on parents and young (Texas Bluebird Society 2002). We mounted nest boxes ~1.5 m from the ground on t-posts placed ~100 m from the road. We placed nest boxes alternately on the left and right side of the road ~100 m from each other to mitigate potential intraspecific territorial disputes. To avoid bias in our nest success estimates, we did not install predator guards on the poles or boxes. Southeastern Naturalist S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 384 Nest monitoring We conducted nest checks twice weekly from February through July of 2013 and 2014. Although we focused primarily on Bewick’s Wrens, we monitored other native cavity-nesting birds that used the nest boxes. We occasionally found wasps in nest boxes and manually removed them (i.e., without pesticide spray). During each check, we recorded the species using the nest box, stage of nest building, number of eggs, number of chicks, and the approximate developmental stage of chicks. We defined a nest as active from the day the first egg was laid to the day the last chick fledged or the nest failed. Based on the Bewick’s Wren’s life history, eggs were assumed to have been laid 1 day apart (Kennedy and White 2013). We defined success as at least 1 chick fledging from the nest and assumed the nest was successful if the nest was empty with no obvious signs of depredation (e.g., disturbed nest, feathers on the ground, dead chicks) and chicks were near the end of the nestling stage, ~14–17 days post-hatching (Bibbee 1947). Although snakes may depredate nestlings without disturbing nests, fledglings can be force-fledged at this stage and could potentially escape depredation. Failure was determined to be caused by depredation, parental abandonment, infertile eggs, or “nest hijacking” (i.e., another bird builds a new nest over the previous nest). House Wrens do not breed in central Figure 1. Location of the nest boxes at the Freeman Center, San Marcos, TX. Each point represents 1 nest box. Southeastern Naturalist 385 S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 Texas; thus, there was no competition for nest sites at our study area between Bewick’s and House Wrens (Johnson 2014). Common nest predators include Procyon lotor L. (Raccoon), Elaphe obsolete Say in James (Texas Ratsnake), Sciurus niger L. (Eastern Fox Squirrel) (Li and Martin 1991), and Solenopsis invicta Buren (Red Imported Fire Ant; S. Harrod, unpubl. data). Accipiter cooperii Bonaparte (Cooper’s Hawk) also prey on adults and inexperienced fledglings (Jackson et al. 2011, Miller 1941). If depredation occurred, we removed the remaining nest and eggs/dead chicks. Nests were presumed abandoned if neither parent was observed around the nest for an extended time (i.e., at least 2 weeks) or if eggs became cold during the incubation period. If parents were observed near the nest but the eggs failed to hatch within the approximate incubation time (~14–16 d; Kennedy and White 2013), the eggs were determined to be infertile. Nest hijacking was assumed if a new nest was found built on top of the previous occupant’s nest and the previous eggs or chicks were missing or buried beneath the new nest. In these instances, we deemed the previous nest a failure, and a new nesting attempt began with the first egg laid in the new n est. Vegetation characterization We generated ~0.10-ha (17.5-m radius) buffers around each nest box using ArcMap (ESRI 2011). Buffers of 0.04 ha are commonly used in assessing vegetation characteristics adjacent to nest sites (Hudson and Bollinger 2013, Martin 2014, Purcell and Verner 2008); thus, we feel that ~0.10-ha buffers accurately represented the immediate vegetation surrounding nest boxes. Additionally, this was the largest buffer size we could generate while preventing buffers from overlapping. We employed a geodatabase layer from Texas Parks and Wildlife to classify each vegetation type within the buffers (Elliott 2014, Texas Parks and Wildlife 2013). Due to the similarities of several vegetation types (e.g., Live Oak motte and woodland vs. Post Oak motte and woodland), we reclassified these vegetation types as woodland, shrubland, or grassland (Table 1). We used ArcMap to measure the distance to the nearest new vegetation type (i.e., edge) within each buffer and FRAGSTATS (McGarigal et al. 2012) to measure Table 1. List of vegetation types re-classified as woodland, shrubland, or grassland. Vegetation types originally classified by Texas Parks and Wildlife (Elliott 2014). Vegetation type Reclassification Live Oak motte and woodland Woodland Post Oak motte and woodland Woodland Deciduous oak/evergreen motte and woodland Woodland Ashe Juniper motte and woodland Woodland Oak/hardwood motte and woodland Woodland Riparian hardwood forest Woodland Riparian hardwood/Ashe Juniper forest Woodland Ashe Juniper/Live Oak shrubland Shrubland Mesquite shrubland Shrubland Savannah grassland Grassland Riparian herbaceous vegetation Grassland Southeastern Naturalist S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 386 the following class-level variables for each vegetation type: patch area weighted mean (hereafter “patch area”), perimeter–area ratio weighted mean (hereafter “perimeter–area ratio”), and patch density (Table 2). Although McGarigal et al. (2012) defined patches as “discrete areas of relatively homogeneous environmental conditions at a particular scale”, for the purpose of this study we defined patches as areas of a homogenous vegetation type. We chose the class level because this scale incorporates multiple patches (i.e., vegetation types). Bewick’s Wrens are gleaners and use trees for foraging (Miller 1941), and nest predators often use trees to reach nest boxes (Cornell Lab of Ornithology 2011); thus, we measured the distance from each nest box to the nearest tree above 2 m in height. Finally, we used a vegetation profile board (Nudds 1977) to measure the percent vertical vegetative cover in the 4 cardinal directions at 15 m from each nest box. These scores represent the density of the vegetation surrounding the nest boxes. The class-level variables patch area, perimeter- area ratio, and patch density were measured for each of the 3 habitat types within the buffer; together with the other 3 variables—distance to edge, distance to nearest tree, and average vegetation cover— resulted in 12 measured vegetation variables per buffer (Table 2). Statistical analyses We employed program R (R Core Team 2015) to calculate daily survival probability and nest success for each breeding season, using the logistic exposure method (package MASS; Ripley et al. 2010), a generalized linear model with a binomial-response distribution (i.e., 0 = failure, 1 = success) and a logit-link function reflecting variability in length of nest exposure (i.e., number of days between nest visits) (Shaffer 2004). Bewick’s Wrens often build multiple dummy nests (Kennedy and White 2013), so we only considered nesting attempts in which at least 1 egg was laid. We included nest boxes as a random variable in our generalized linear mixedeffect models (package 'lme4'; Bates et al. 2015) to account for pseudoreplication Table 2. Description of vegetation measurements taken around each box. Note that for patch area weighted mean, perimeter–area ratio weighted mean, and patch density, we obtained estimates for each vegetation type (woodland, shrubland, and grassland) present in the buffer. Variable Description Unit Patch area weighted mean Measures area of each patch, with larger patches Hectares assigned higher weights Perimeter–area ratio weighted Measures degree of complexity of patches and N/A mean amount of patch area exposed to edge Patch density Total number of patches divided by total landscape Patches/ area 100 ha Distance to edge Distance from nest box to next vegetation type Meters Distance to nearest tree Distance from nest box to the nearest tree or shrub Meters ≥2 m in height Average vegetation cover at 15 m Average percent vegetation cover in each cardinal N/A direction 15 m from nest box Southeastern Naturalist 387 S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 (Hurlbert 1984), although since no Bewick’s Wrens were banded, we could not account for pseudoreplication among individuals. To determine if our nest success estimates were affected by this lack of accountability, we compared our original estimates of annual nesting success (i.e., using all nests in a season) to nest success estimates made using only the first nest in a box for each season. Prior to further analyses, we z-transformed our vegetation variables. We ran forward model selection using AICc (Akaike’s information criterion corrected for small sample sizes; Burnham and Anderson 2002) to determine which model best explained variation in nest success. We considered models with ΔAICc values of ≤2 to be equivalent, and used their variables used to build additive models. The additive models only included uncorrelated variables. We chose the final model based on having a ΔAICc value of ≤2 and a high (≥0.95) AICc weight (wi). Results Daily survival probability and nest success During the 2013 and 2014 nesting seasons, we recorded 136 nesting attempts. Bewick’s Wrens accounted for 83 of these attempts, but we also found Baeolophus atricristatus Cassin (Black-crested Titmouse; n = 26), Eastern Bluebird (n = 18), Myiarchus cinerascens Lawrence (Ash-throated Flycatcher; n = 7), and Poecile carolinensis Audubon (Carolina Chickadee, n = 2) nests. Bewick’s Wrens occupied 75.0% and 65.0% of nest boxes in 2013 and 2014, respectively, whereas other species occupied 30.0% and 72.5% of nest boxes, respectively, each year. Nesting activity began as early as February and persisted until late June/early July. Average clutch size (for nests with eggs that hatched) for 2013 (n = 38) and 2014 (n = 19) was 5.97 ± 0.81 SD and 5.95 ± 0.85 SD, respectively. We recorded the earliest first-egg date on 23 February 2013, although most clutches began mid-late March. The latest first-egg date was recorded on 15 June 2013. Probabilities of daily survival for Bewick’s Wre remained high for both years (99.2% and 98.4%, respectively), although overall nest success declined from 77.1% ± 6.58% SE in 2013 to 58.9% ± 8.78% SE in 2014. Across both years, 32.5% of all Bewick’s Wren nests failed; of these, 22.2% were due to abandonment, 25.9% due to nest hijacking, and 51.9% due to depredation. Our inability to control for all sources of pseudoreplication does not seem to have affected these results, as nest success estimates calculated using only the first nest in each box for each season were not significantly different from our original estimates (Fig. 2). Effects of vegetation on nesting success Our first model-selection analysis revealed that the model that included shrubland perimeter–area ratio was equivalent to the model that included distance to nearest tree (Table 3). These variables were weakly correlated (r = -0.015, P = 0.93) and were combined into an additive model. When compared to our null and retained single-covariate models, the additive model was best supported by our data (wi = 0.98). Using the original (i.e., not z-transformed) covariates, the final model showed that an increase in shrubland perimeter–area ratio caused a negative effect Southeastern Naturalist S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 388 (-0.001 ± 0.0002 SE; P < 0.001) on nest success (Fig. 3a), whereas increasing distance between nest box and nearest tree had a positive effect (0.95 ± 0.32 SE; P less than 0.001) on nest success (Fig. 3b). Figure 2. Comparison of nest success estimates and 95% confidence intervals using all nests (black lines) and only the first nests per box (gray lines). Table 3. Summary of model parameters with corresponding K, AICc, ΔAICc, and wi values from our AICc forward model selection. Model K AICc ΔAICc wi Null vs. single-covariate models Shrubland perimeter–area ratio weighted mean 3 199.2 0.00 0.34 Distance to nearest tree 3 199.2 0.02 0.34 Shrubland patch density 3 201.7 2.47 0.10 Average vegetation cover at 15 m 3 202.3 3.06 0.07 Woodland patch density 3 203.8 4.63 0.03 Combined average vegetation cover 3 204.3 5.10 0.03 Shrubland patch area weighted mean 3 204.9 5.69 0.02 Null 2 205.1 5.90 0.02 Woodland perimeter–area ratio weighted mean 3 206.4 7.18 0.01 Woodland patch area weighted mean 3 206.5 7.33 0.01 Average vegetation cover at 35 m 3 206.9 7.72 0.01 Distance to edge 3 207.0 7.77 0.01 Grassland perimeter–area ratio weighted mean 3 207.0 7.77 0.01 Grassland patch area weighted mean 3 207.0 7.81 0.01 Grassland patch density 3 207.1 7.92 0.01 Additive vs.null and retained single-covariate models Shrubland perimeter–area ratio weighted mean + Distance to nearest tree 4 189.9 0.00 0.98 Shrubland perimeter–area ratio weighted mean 3 199.2 9.31 0.02 Distance to nearest tree 3 199.2 9.33 0.01 Null 2 205.1 15.2 less than 0.01 Southeastern Naturalist 389 S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 Figure 3. Regression coefficients (with 95% confidence intervals) for effects of (a) shrubland perimeter–area ratio and (b) distance to nearest tree on nesting success of Bewick’s Wren. Southeastern Naturalist S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 390 Discussion The results of our AICc stepwise model selection indicate that shrubland perimeter– area ratio and distance to nearest tree were the most important predictors of Bewick’s Wren nest success at the Freeman Center. Although we were surprised that the model that included shrubland patch area received low support (ΔAICc = 5.69, wi = 0.02), we were unsurprised to find that 1 of our shrubland variables was an important predictor because Bewick’s Wrens readily nest in shrubland vegetation (Kennedy and White 2013). We hypothesize that increasing shrubland perimeter– area ratio values caused a small decline in nest success because high perimeter–area ratio values are indicative of complex vegetation patches with less interior vegetation and more edge (Helzer and Jelinski 1999). Shrubland perimeter–area ratio could therefore be viewed as a proxy for shrubland patch area, suggesting that patch area and shape are important for Bewick’s Wrens. We also hypothesized that the positive effect of increasing distance to nearest tree on nest success was because nest boxes placed closer to trees were more accessible to arboreal predators such as Texas Ratsnakes, Eastern Fox Squirrels, and Raccoons. Although we tried to avoid placing our nest boxes directly against trees, we were constricted by our study design (i.e. alternative left/right assignment of nest box location) and features of the Freeman Center (e.g., barbed wire fencing), and we placed some nest boxes closer to trees (i.e., within 3 m) than others, which likely allowed predators to more easily access nest boxes and depredate the nest. Although prominent organizations such as the Cornell Lab of Ornithology (Cornell Lab of Ornithology 2011) recommend placing nest boxes on poles away from trees, it is still common practice to hang nest boxes from tree limbs, or more commonly, to secure them against the trunk. The results of our study add further evidence against this practice. All of our boxes were identical; thus, we do not believe that the differing occupancy rates among boxes were an artifact of their design, but rather of the vegetation types and characteristics. It is highly likely that males and females re-nested multiple times within and between years, but because individuals were not banded, we were unable to account for this source of pseudoreplication in our model. However, nest success estimates calculated using all nests and only the first nest per box per year were not significantly different (Fig. 2). Although there are no published studies of the effects of vegetation characteristics on nesting success for Bewick’s Wrens in central Texas, Magill et al. (2003) reported on the effects of habitat and vegetation characteristics on nest success of Bewick’s Wrens in Texas’ Rolling Plains Ecoregion. The authors of that study found no effect of any of the variables they examined on Bewick’s Wren nest success, though this may be due to small sample size (n = 17) and use of predator guards. A similar study by Smith et al. (2007) in eastern New Mexico found that grass cover negatively affected nest success by increasing depredation risk, whereas distance from a levee (i.e., habitat edge) had a positive effect on nest success. The authors hypothesized that the positive effect of distance from a levee/edge acted in the same way as natural habitat edges act on nest depredation rates (i.e., nests near habitat Southeastern Naturalist 391 S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 edges are more susceptible to nest predators; Cox et al. 2012, Fletcher 2005). Grass cover was positively associated with depredation, and therefore we hypothesized that high levels of grass cover provided habitat for snakes. Considering all species combined, nest box occupancy rates were high for both study years (85% and 100%, respectively). While we did not assess the prevalence of natural cavities and so cannot speculate on whether natural nest sites are limited, we hypothesize that the high rates of nest box occupancy were a result of limited access to natural cavities via interspecific competition and that cavities were a limiting factor for Bewick’s Wrens at the Freeman Center. The Center is home to 3 resident primary cavity-excavators—Picoides pubescens L. (Downy Woodpecker), Melanerpes aurifrons Wagler (Golden-fronted Woodpecker), and Picoides scalaris Wagler (Ladder-backed Woodpecker)—which appear to be abundant (S. Harrod, unpubl. data) and may outcompete Bewick’s Wrens for natural nest sites. Although capable of excavating their own cavities, primary cavity-nesters often reuse previously excavated cavities (Aitken et al. 2002, Wiebe et al. 2007). Advantages to this strategy include saving energy and allowing for earlier laying dates, larger clutches, and more re-nesting attempts (Wiebe et al. 2007). Bewick’s Wrens may also compete for natural cavities with non-avian species such as Raccoons, Didelphis virginianus Kerr (Opossums), and wasps (Magill et al. 2003). The addition of nest boxes may prove beneficial for Bewick’s Wrens because many competitors (e.g., Golden-fronted Woodpeckers, Raccoons) are too large to use nest boxes. Taylor (2003) found that adding nest boxes led to an increase in abundance and range expansion of Bewick’s Wrens, further supporting the hypothesis of limited access to natural cavities. Future studies should address the population densities of primary cavity-nesters, as well as the proportion of natural cavities available in a given area. We also noted 2 instances of intraspecific brood parasitism by Black-crested Titmice and Bewick’s Wrens. In both instances, 6 eggs were added to the clutch in the span of 2 and 3 days, respectively. Given that Bewick’s Wrens and Black-crested Titmice lay 1 egg per day (Kennedy and White 2013, Patten and Smith-Patten 2008), it is unlikely that these females laid 2–3 eggs per day. Intraspecific parasitism is common among waterfowl (Yom-Tov 2001), but rates are lower than those of interspecific parasitism, perhaps due to the difficulty of detection (MacWhirter 1989). However, intraspecific brood parasitism has been documented in common passerine species in multiple taxa (e.g., Eastern Bluebird, Sturnus vulgaris L. [European Starling], Purple Martin, Spinus tristis L. [American Goldfinch]; Yom-Tov 2001), and such occurrences may be indicative of limited nesting sites, particularly for secondary cavity-nesters that rely on pre-existing cavities (Lombardo et al. 1989, Zink 2000). We did not install predator exclusion devices in our study so as to better mimic natural cavities and not bias our estimates of nest success. Across the study period, 51.9% of Bewick's Wren nest failures were attributable to depredation. In the years following this study, nest depredation has increased dramatically (M.C. Green, unpubl. data). This increase may occur because as nest boxes age, predators improve Southeastern Naturalist S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 392 their search image and retain spatial memory of nest box locations (Sonerud 1993). Therefore, predator guards may be particularly important for managing nest-box trails over the long term. Bailey and Bonter (2017) found that nest boxes with predator guards were 6.70% more likely to successfully fledge young than nest boxes without predator guards. In managing for cavity-nesting passerines, nest boxes are often used in lieu of managing natural cavities (e.g., snags), and general vegetation type (i.e., grassland, shrubland) is often the only criterion considered when determining where to erect nest boxes. However, nest boxes are not a perfect substitute (Maziarz et al. 2017) because microclimate may differ significantly between natural and artificial cavities, which can affect nest success. Nest boxes should therefore be used in conjunction with natural-cavity management because the latter is a more sustainable and cost-effective measure (Maziarz et al. 2017). This study demonstrates that shrubland perimeter–area ratio and distance to nearest tree influence nest success for Bewick’s Wrens. When managing for Bewick’s Wrens, particularly in areas where populations are declining or sparse, wildlife managers should consider these variables when deciding where to erect nest boxes. Additionally, predator guards should be used in conjunction with nest boxes, and natural cavities should be managed to alleviate high competition for nest sites. Acknowledgments We thank Donald Huebner for allowing us access to the Freeman Center, the North American Bluebird Society and Texas Ornithological Society for their financial support, and Rebekah Rylander, Oscar Perez, Rachel Connally, and Neil and Emily Clark for assisting us with this work. Literature Cited Aitken, K.E.H., K.L. Wiebe, and K. Martin. 2002. Nest-site reuse patterns for a cavitynesting bird community in interior British Columbia. The Auk 119:391–402. Bailey, R.L., and D.N. Bonter. 2017. Predator guards on nest boxes improve nesting success of birds. Wildlife Society Bulletin 41:434–441. Barnes, P.W., S.-Y. Liang, K.E. Jessup, L.E. Ruiseco, P.L. Phillips, and S.J. Reagan. 2000. Soils, Topography, and Vegetation of the Freeman Ranch. Texas State University Press, San Marcos, TX. 32 pp. Bates, D., M. Mächler, B.M. Bolker, and S.C. Walker. 2015. Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67:1–49. Benson, T.J., J.D. Brown, and J.C. Bednarz. 2010. Identifying predators clarifies predictors of nest success in a temperate passerine. Journal of Animal Ecology 79:225–234. Bibbee, P.C. 1947. The Bewick’s Wren (Thryomanes bewickii (Audubon)). Ph.D. Dissertation. Cornell University, Ithaca, NY. 542 pp. Burnham, K.P., and D.R. Anderson. 2002. Model Selection and Multimodel Inference: A Practical Information-theoretic Approach. Springer, New York, NY. 488 pp. Chace, J.F., and J.J. Walsh. 2006. Urban effects on native avifauna: A review. Landscape and Urban Planning 74:46–69. Copley, D., D. Fraser, and J.C. Finlay. 1999. Purple Martin, Progne subis: A British Columbian success story. The Canadian Field-Naturalist 113:226–229. Southeastern Naturalist 393 S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 Cornell Lab of Ornithology. 2009a. Nest box: Eastern, Western, and Mountain Bluebirds; Tree and Violet-green Swallows. Special technical report. Cornell University, Ithaca, NY. 1 p. Cornell Lab of Ornithology. 2009b. Nest box: House and Bewick’s Wrens; Black-capped, Mountain, Boreal, Chestnut-backed, and Carolina Chickadees; Juniper, Black-crested, Oak, Tufted, and Bridled Titmice; Pygmy, White-breasted, Red-breasted, and Brownheaded Nuthatches; Prothonotary and Lucy’s Warbler; and Flying Squirrel. Special Technical Report. Cornell University, Ithaca, NY. 1 p. Cornell Lab of Ornithology. 2011. NestWatch: All about birdhouses. Available online at Accessed 12 July 2017. Cox, W.A., F.R. Thompson, and J. Faaborg. 2012. Landscape forest-cover and edge effects on songbird-nest predation vary by nest predator. Landscape Ecology 27:659–669. Deng, W.-H., and W. Gao. 2005. Edge effects on nesting success of cavity-nesting birds in fragmented forests. Biological Conservation 126:363–370. Elliott, L. 2014. Descriptions of systems, mapping subsystems, and vegetation types for Texas. Special technical report. Texas Parks and Wildlife Department, Austin, TX. 182 pp. Environmental Systems Research Institute (ESRI). 2011. ArcMap, version 10.3.1. Redlands, CA. Fletcher, R.J. 2005. Multiple edge-effects and their implications in fragmented landscapes. Journal of Animal Ecology 74:342–352. Fu, Y., B. Chen, S.D. Dowell, and Z. Zhang. 2016. Nest predators, nest-site selection, and nest success of the Emei Shan Liocichla (Liocichla omeiensis), a vulnerable babbler endemic to southwestern China. Avian Research 7:1–6. Gjerdrum, C., C.S. Elphick, and M. Rubega. 2005. Nest-site selection and nesting success in saltmarsh breeding sparrows: The importance of nest habitat, and study-site differences. The Condor 107:849–862. Goldingay, R.L., and J.R. Stevens. 2009. Use of artificial tree hollows by Australian birds and bats. Wildlife Research 36:81–97. Gowaty, P.A., and J.H. Plissner. 2015. Eastern Bluebird (Sialia sialis). In Birds of North America. Available online at Accessed 15 June 2017. Helzer, C.J., and D.E. Jelinski. 1999. The relative importance of patch area and perimeter– area ratio to grassland breeding birds. Ecological Applications 9:1448–1458. Herkert, J.R., D.L. Reinking, D.A. Wiedenfeld, M. Winter, J.L. Zimmerman, W.E. Jensen, E.J. Finck, R.R. Koford, D.H. Wolfe, S.K. Sherrod, M.A. Jenkins, J. Faaborg, and S.K. Robinson. 2003. Effects of prairie fragmentation on the nest success of breeding birds in the midcontinental United States. Conservation Biology 17:587–594. Hinsley, S.A. 2000. The costs of multiple patch use by birds. Landscape Ecology 15:765–775. Hudson, N.C., and E.K. Bollinger. 2013. Nest success and nest-site selection of Red-headed Woodpeckers (Melanerpes erythrocephalus) in east-central Illinois. American Midland Naturalist 170:86–94. Hurlbert, S.H. 1984. Pseudoreplication and the design of ecological field experiments. Ecological Monographs 54:187–211. Ingold, D.J. 1989. Nest phenology and competition for nest sites among Red-headed and Red-bellied Woodpeckers and European Starlings. The Auk 106:209–217. Jackson, J.A., and J. Tate. 1974. An analysis of nest-box use by Purple Martins, House Sparrows, and starlings in eastern North America. The Wilson Bulletin 86:435–449. Southeastern Naturalist S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 394 Jackson, A.K., J.P. Froneberger, and D.A. Cristol. 2011. Postfledging survival of Eastern Bluebirds in an urbanized landscape. Journal of Wildlife Management 75:1082–1093. Jackson, A.K., J.P. Froneberger, and D.A. Cristol. 2012. Habitat near nest boxes correlated with fate of Eastern Bluebird fledglings in an urban landscape. Urban Ecosystems 16:367–376. Johnson, L.S. 2014. House Wren (Troglodytes aedon). In Birds of North America. Available online at Accessed 15 June 2017. Kennedy, E.D., and D.W. White. 1996. Interference competition from House Wrens as a factor in the decline of Bewick’s Wrens. Conservation Biology 10:281–284. Kennedy, E.D., and D.W. White. 2013. Bewick’s Wren (Thryomanes bewickii). In Birds of North America. Available online at bewwre. Accessed 15 Jun 2017. Kight, C.R., and J.P. Swaddle. 2007. Associations of anthropogenic activity and disturbance with fitness metrics of Eastern Bluebirds (Sialia sialis). Biological Conservation 138:189–197. Li, P., and T.E. Martin. 1991. Nest-site selection and nesting success of cavity-nesting birds in high-elevation forest drainages. The Auk 108:405–418. Lombardo, M.P., H.W. Power, P.C. Stouffer, L.C. Romagnano, and A.S. Hoffenberg. 1989. Egg removal and intraspecific brood parasitism in the European Starling (Sturnus vulgaris). Behavioral Ecology and Sociobiology 24:217–223. MacWhirter, R.B. 1989. On the rarity of intraspecific brood parasitism. The Condor 91:485–492. Magill, R.T., L.M. Smith, and J.D. Ray. 2003. Nest box use by cavity-nesting birds in riparian zones of the southern Great Plains. Texas Journal of Science 55:235–246. Martin, T.E. 2014. Consequences of habitat change and resource-selection specialization for population limitation in cavity-nesting birds. Journal of Applied Ecology 52:475–485. Mazgajski, T.D., and L. Rejt. 2006. The effect of forest-patch size on the breeding biology of the Great Spotted Woodpecker, Dendrocopos major. Annales Zoologici Fennici 43:211–220. Maziarz, M., and T. Wesołowski. 2013. Microclimate of tree cavities used by Great Tits (Parus major) in a primeval forest. Avian Biology Research 6:47–56. Maziarz, M., R.K. Broughton, and T. Wesołowski. 2017. Microclimate in tree cavities and nest-boxes: Implications for hole-nesting birds. Forest Ecology and Management 389:306–313. McGarigal, K., S.A. Cushman, and E. Ene. 2012. FRAGSTATS, version 4.2.597. University of Massachusetts, Amherst, MA. Miller, E.V. 1941. Behavior of the Bewick Wren. The Condor 43:81–99. Milligan, M.C., and J.L. Dickinson. 2016. Habitat quality and nest-box occupancy by five species of oak woodland birds. The Auk 133:429–438. Newmark, W.D., and T.R. Stanley. 2011. Habitat fragmentation reduces nest survival in an Afrotropical bird community in a biodiversity hotspot. Proceedings of the National Academy of Sciences of the United States of America 108:11,488–11,493. Nudds, T.D. 1977. Quantifying the vegetation structure of wildlife cover. Wildlife Society Bulletin 5:113–117. Odum, E.P., and D.W. Johnston. 1951. The House Wren breeding in Georgia: An analysis of range extension. Auk 68:356–367. Partners in Flight. 2017. Avian-conservation assessment database. Available online at http:// Accessed 10 December 2017. Southeastern Naturalist 395 S.E. Harrod and M.C. Green 2018 Vol. 17, No. 3 Patten, M.A., and B.D. Smith-Patten. 2008. Black-crested Titmouse (Baeolophus atricristatus). In Birds of North America. Available online at Account/bna/species/blctit4. Accessed 15 June 2017. Patten, M.A, E. Shochat, D.L. Reinking, D.H. Wolfe, and S.K. Sherrod. 2006. Habitat edge, land management, and rates of brood parasitism in tallgrass prairie. Ecological Applications 16:687–695. Petty, S.J., G. Shaw, and D.I.K. Anderson. 1994. Value of nest boxes for population studies and conservation of owls in coniferous forests in Britain. Journal of Raptor Research 28:134–142. Pogue, D.W., and G.D. Schnell. 1994. Habitat characterization of secondary cavity-nesting birds in Oklahoma. The Wilson Bulletin 106:203–226. Purcell, K.L., and J. Verner. 2008. Nest-site habitat of cavity-nesting birds at the San Joaquin Experimental Range. Special technical report. United States Department of Agriculture, Albany, CA. 13 pp. R Core Team. 2015. Program R, version 3.0.2. The R Foundation for Statistical Computing, Vienna, Austria. Ripley, B., B. Venables, D.M. Bates, K. Hornik, A. Gebhardt, and D. Firth. 2010. Package “MASS.” Available online at Accessed Accessed 5 January 2018. Rohrbaugh, R.W., Jr., and R.H. Yahner. 1997. Effects of macrohabitat and microhabitat on nest-box use and nesting success of American Kestrels. The Wilson Bulletin 109:410–423. Shaffer, T.L. 2004. A unified approach to analyzing nest success. The Auk 121:526–540. Smith, D.M., J.F. Kelley, and D.M. Finch. 2007. Avian nest-box selection and nest success in burned and unburned southwestern riparian forest. Journal of Wildlife Management 71:411–421. Smith, P.A., H.G. Gilchrist, and J.N.M. Smith. 2007. Effects of nest habitat, food, and parental behavior on shorebird nest success. The Condor 109:15–31. Sonerud, G.A. 1993. Reduced predation by nest box relocation: Differential effect on Tengmalm’s Owl nests and artificial nests. Ornis Scandinavica 24:249–253. Taylor, R.V. 2003. Factors influencing expansion of the breeding distribution of Bewick’s Wren into riparian forests of the Rio Grande in central New Mexico. Southwestern Naturalist 48:373–382. Texas Bluebird Society. 2002. Texas Bluebird Society nestbox plans. Available online at Accessed 30 September 2012. Texas Parks and Wildlife. 2013. Ecological mapping systems. Available online at https:// Accessed 14 May 2014. Wiebe, K.L., W.D. Koenig, and K. Martin. 2007. Costs and benefits of nest reuse versus excavation in cavity-nesting birds. Annales Zoologici Fennici 44:209–217. Winter, M., and J. Faaborg. 1999. Patterns of area sensitivity in grassland-nesting birds. Conservation Biology 13:1424–1436. Yom-Tov, Y. 2001. An updated list and some comments on the occurrence of intraspecific nest parasitism in birds. Ibis 143:133–143. Zink, A.G. 2000. The evolution of intraspecific brood parasitism in birds and insects. The American Naturalist 155:395–405.