The University of Maine DigitalCommons@UMaine

Honors College

Spring 2014

Effect of Forest Vegetation on Nest-Site Selection by Spruce Grouse Across Two Spatial Scales

Nathaniel Scott Parkhill University of Maine - Main, [email protected]

Follow this and additional works at: https://digitalcommons.library.umaine.edu/honors

Part of the Environmental Health and Protection Commons, Environmental Indicators and Impact Assessment Commons, and the Environmental Monitoring Commons

Recommended Citation Parkhill, Nathaniel Scott, "Effect of Forest Vegetation on Nest-Site Selection by Spruce Grouse Across Two Spatial Scales" (2014). Honors College. 153. https://digitalcommons.library.umaine.edu/honors/153

This Honors Thesis is brought to you for free and open access by DigitalCommons@UMaine. It has been accepted for inclusion in Honors College by an authorized administrator of DigitalCommons@UMaine. For more information, please contact [email protected].

EFFECT OF FOREST VEGETATION ON NEST-SITE SELECTION BY SPRUCE

GROUSE ACROSS TWO SPATIAL SCALES

by

N. Scott Parkhill

A Thesis Submitted in Partial Fulfillment of the Requirements for a Degree with Honors (Wildlife Ecology)

The Honors College University of Maine May 2014

Advisory Committee: Dr. Daniel J. Harrison, Professor and Chair, Department of Wildlife, Fisheries, and Conservation Biology Dr. Margaret O. Killinger, Rezendes Preceptor for the Arts Dr. Erik J. Blomberg, Assistant Professor of Wildlife Ecology Lindsay C. N. Seward, Instructor of Wildlife Ecology Dr. Daniel W. Linden, Post-Doctoral Associate of Wildlife Ecology 2

ABSTRACT

The spruce grouse (Falcipennis canadensis) is a gallinaceous bird that is threatened or endangered throughout much of the southeastern limit of its range. Generally associated with short-needled conifer forests like those characteristic of northern Maine, this species may be particularly sensitive to recent changes in timber harvesting practices.

I examined nest-site selection to better understand spruce grouse habitat associations in northern Maine. In the summer of 2013, I located the nests of 12 female spruce grouse in commercially-managed forests of north-central Maine. I measured vegetation characteristics at nests and at sampling points 30 meters from nests, as well as points randomly distributed throughout the stand where a nest was located. I examined differences in characteristics at sites used for nesting and sites available across within- patch and patch-scales. Logistic regression revealed that at within-patch scale, sites with higher lateral cover were selected for nesting. At the patch-scale, lower tree density and lower basal area of live trees, but higher lateral cover and greater recess height were associated with sites selected for nesting. These scale-dependent differences suggest that high concealment is selected by nesting hens, but that small forest-gap structure is also selected for by hens in the stand surrounding their nest. My results indicate that nesting spruce grouse select for gaps within dense forest structure which provide a combination of nest-level lateral cover, overhead canopy cover, and nearby trees for escape cover by adults. iii

ACKNOWLEDGEMENTS

I would like to thank those who made this work possible through their generous assistance, including field work in the fly-ridden interior of the North Maine Woods, and spanning the world of statistical analysis programs.

Technical Help:

Stephen Dunham

Andrew Parnas

Sheryn Olson

Dan Stitch

Harrison Lab vegetation field crews 2012-2013

Committee Members:

Dr. Daniel Harrison (Advisor)

Dr. Mimi Killinger

Dr. Erik Blomberg

Ms. Lindsay Seward

Dr. Dan Linden

Partners:

University of Maine Department of Wildlife Ecology

Maine Cooperative Forestry Research Unit

Maine Agricultural and Forest Experiment Station iv

TABLE OF CONTENTS

INTRODUCTION…..………………………..………………………….………….1

METHODS….…….……………….……………………………………….……....3

RESULTS….………….…………….………………………………………….…6

DISCUSSION………….….………………………………………………………7

FIGURES AND TABLES…….….…....…………………………………..….…..13

LITERATURE CITED….…..……………………………………………….…....16

AUTHOR’S BIOGRAPHY…….…..………………………………….………….19

1

INTRODUCTION

Nest placement is widely considered the most importance decision made by ground nesting birds because nest predation is a major limit to reproductive success and adult female survival (Bergerud and Gratson 1988, Martin and Roper 1988, Martin 1992,

Tirpak et al. 2006, Roper et al. 2010). Nest predation accounts for the greatest loss in fitness aside from adult mortality and the importance of nest-site selection depends on the number of nesting opportunities over the individual’s lifespan (Bergerud and Gratson

1988). Selection for a nest-site that reduces the probability of nest predation affects fitness given that predators cause 80% of nest failures (Newton 1998, Li and Martin

1991). Nest-site selection in many ground nesting birds is determined by vegetative characteristics, demonstrating that structure must be considered to understand nest placement (Pietz and Tester 1982, Redmond et al. 1982, D’Eon 1997, Watters et al. 2002,

Tirpak et al. 2006, Anich et al. 2013, Fuller et al. 2013, Lovell et al. 2013, Seibold et al.

2013). Thus, identifying the forest structural characteristics selected for by ground nesting birds gives natural resource managers the ability to target conservation efforts at important habitat components.

Spruce grouse (Falcipennis canadensis) are a gallinaceous bird distributed across the boreal forest of Canada, Alaska, and the northern regions of the United States. The species is generally associated with short-needled conifer forests, including the spruce-fir forests that occur within landscapes of northern Maine (Robinson 1980). In 2008, the

International Association of Fish and Wildlife Agencies declared that spruce grouse populations in the southeastern region of its range were at risk (Williamson et al. 2008).

The species cannot be hunted in Maine or New Hampshire, and is designated as state 2 endangered (Vermont and New York), threatened (Wisconsin), and uncommon (Nova

Scotia, Minnesota, and Michigan) in many other jurisdictions (Williamson et al. 2008).

Loss of conifer forests, incompatible timber harvesting, and population fragmentation are considered major factors making southeastern populations vulnerable to extirpation

(Williamson et al. 2008).

Regional declines of spruce grouse have been attributed to changes in forest management as well as to demand for spruce (Picea spp.) and balsam fir (Abies balsamea) in timber markets. These factors may diminish habitat for spruce grouse through harvest techniques that fragment residual patches of conifer forest (New

Hampshire Fish and Game Dept. 2006, Anich et al. 2013). Others attribute regional declines to habitat loss associated with urban and agricultural development, as well as to forest maturation in response to decreases in large-scale timber harvesting (Williamson et al. 2008). Since the implementation of the Maine Forest Practices Act in 1991, clearcutting harvest techniques have declined below 5% of harvested volume in Maine, and have been replaced by expansive partial harvests (Jin and Sader 2006). Although clearcuts generally mature into large patches of contiguous forest unless cut or disturbed, the process of removing timber via “partial harvesting” has the capacity to alter forest composition (i.e., increase the deciduous component of the regenerating forest), contribute to fragmentation of mature patches of conifer forest, and reduce the size of conifer-dominated patches.

Regionally, forest practices have changed dramatically in the past 20 years, and have altered the structure of Northeastern forests, with the effects of these changes on wildlife remaining poorly understood (Sader et al. 2003, Hoving et al. 2004, Sader et al. 3

2005, Homyack et al. 2007). Redmond et al. (1982) found differences in nest concealment, nest-site selection, and success between two populations of spruce grouse in Alberta and New Brunswick, suggesting that findings cannot be generalized across the species’ entire range. Thus, effective management of spruce grouse in the Acadian Forest region (Seymour and Hunter 1999) is strengthened by understanding regionally-specific relationships between forest structure and nest-site selection and may contribute to our understanding of regional population declines.

The goal of my research was to understand attributes of forest stands which influence spruce grouse nest-site selection. My specific objectives were: 1) compare vegetation characteristics at nest sites to random points within the stand in which the grouse were captured (i.e. ‘focal stand’) to examine nest-site selection at the patch-scale,

2) compare vegetation characteristics at nest sites to random points thirty meters away to examine nest-site selection at the within-patch scale, and 3) describe the multivariate relationships between habitat characteristics measured at nest sites to make inferences about which forest structural characteristics influence nest-site selection at both the within-patch and patch scales.

METHODS

Sampling Design

We conducted our study in the Telos region to the west of Baxter State Park in north-central Maine during the summer of 2013 (Figure 1). Eight nests were found by locating female spruce grouse which were equipped with radio transmitters during the summers of 2012 and 2013. Females with radios were located during the brooding season using a chick distress call, and were captured using an extendable ‘noose pole’ 4 with a monofilament slip knot loop (Zwickel and Bendell). Locations of nest sites were archived using a hand-held GPS. Four additional nests were opportunistically encountered by observing female spruce grouse on a nest, or by observing nest depressions containing spruce grouse egg shells. Two of these nest sites were encountered in the summer of 2012. I returned to each nest after nesting was completed during August 2013 to ensure that I did not interrupt nesting or alter probability of predation at nests. I measured vegetation characteristics at each nest site and at three

“satellite” points positioned 30 meters away from each nest site at randomly selected azimuths using a random number generator. I assumed that a distance of 30 meters from each nest site would represent fine scale heterogeneity within patches caused by pre- commercial thinning trails and would effectively represent within-patch habitat availability for spruce grouse hens when selecting nest sites.

Vegetation Sampling

At each vegetation sampling point, I measured various forest vegetative characteristics to quantify potentially significant structural attributes. To measure lateral cover, a corrugated plastic cutout of a snowshoe hare was placed at a random orientation directly above the nest site. This cutout was then observed from 0.5 meters above the ground and lateral cover was estimated visually. Canopy closure was estimated by taking four measurements in each cardinal direction using a spherical densiometer. Basal area was measured for live and dead trees (>7.62 cm DBH) and saplings (<7.62 cm DBH) for both coniferous and deciduous species using a two factor prism. Using the point quarter method (Silvy 2012), species, DBH, distance from center point, tree height, low live canopy height, and recess height (height to lowest limb) were recorded. Heights of trees 5 were recorded with a measuring pole and a laser hypsometer. To quantify forest structure at the patch scale, the same vegetation data were measured at 20 systematic locations within each focal stand during summers of 2012 and 2013 (Scott 2009).

Data Analysis

I screened data using a Pearson correlation matrix for each spatial scale, and removed variables with a correlation coefficient |r|> 0.7 to reduce multicollinearity and increase parsimony (Zar 1999, Fuller et al. 2013). In the event of two correlated variables, the variable which related more to spruce grouse ecology or was simpler for natural resource manager to measure, was retained. At the within-patch scale basal area of dead saplings, basal area of dead trees, total basal area, average low canopy, and average recess height were removed, and at the patch scale total basal area and average low canopy were removed from further analyses (Table 1, Table 2). At the within-patch scale, both average height and average DBH yielded a correlation coefficient of 0.791 and basal area of coniferous trees and average canopy closure yielded a correlation coefficient of -0.836; however, both variables were retained because in my opinion they explained unique aspects of forest structure (Table 1). Though often correlated, both of these forest metrics could have unique influences on spruce grouse ecology reflecting stand age, predator access, and other components of grouse habitat.

I conducted 20 univariate logistic regressions with the retained variables using program SYSTAT. I used p-values to evaluate variable significance and set α=0.05. I chose to utilize logistic regression analyses as they provide the ability to quantify a bivariate response between nest-sites and non-nest-sites. Logistic regression analysis produces relationships between a variable and the probability of an event such as nest 6 presence, in this case a response we termed the “probability of selection”. I plotted my regression curves and data points using the statistical program R (R Development Core

Team 2011).

RESULTS

Twelve nest-sites were inventoried with vegetation surveys occurring at each nest.

Vegetation surveys were also conducted at three locations thirty meters away from each nest site (36 locations total) to quantify within-patch availability. Twenty locations distributed systematically throughout each stand (eight stands total) in which ≥ 1 nest site was located or the female spruce grouse was captured were also sampled to quantify availability at the patch scale. At the patch scale, availability data were pooled across all eight stands where nests were located.

Across all nest sites, average canopy closure was 41.3%, average lateral cover was 75.6%, average basal area of live trees was 7.5 m2/ ha, and average total basal area was 26 m2/ ha. Average DBH was 32.26 cm, average tree height was 9.6 m, average low canopy height was 3.35 m, average recess height was 92.73 cm, and average tree density was 0.131 trees/ m2.

Across all sampling locations thirty meters from nest sites, average canopy closure was 36%, average lateral cover was 45.8%, average basal area of live trees was

9.89 m2/ha, and average total basal area was 35 m2/ ha. Average DBH was 34.54 cm, average tree height was 10.0 m, average low canopy height was 4.56 m, average recess height was 112 cm, and average tree density was 0.270 trees/ m2.

Across the eight stands surveyed, average canopy closure was 51%, average lateral cover was 55.69%, average basal area of live trees was 10.36 m2/ ha, and average 7 total basal area was 31.51 m2/ ha. Average DBH was 32.00 cm, average tree height was

10.2 m, average low canopy height was 2.03 m, average recess height was 25.36 cm, and average tree density was 0.271 trees per m2.

At the within-patch scale, average lateral cover was correlated (P = 0.031) positively with presence of a spruce grouse nest (Table 1, Figure 2). Average lateral cover had a regression coefficient of 0.006, and exhibited the strongest relationship with nest-site presence (Figure 2).

At the patch-scale, lateral cover (P = 0.020) and recess height (P < 0.001) were positively correlated with nest-site presence (Table 1, Figure 3). Basal area of live trees

(P = 0.032) and tree density (P = 0.115) were negatively correlated with nest presence

(Table 3, Figure 4). There was an increase in the probability of presence of a nest-site with increased average lateral cover and increased average recess height (Table 1). At the same scale, there was a decrease in the probability of selection associated with increases in the basal area of live trees and tree density (Table 3). Spruce grouse nest sites were found more frequently in areas with higher lateral cover and recess height, but lower basal area of live trees and tree density (Figure 2, Figure 3, Figure 4).

DISCUSSION

The fundamental decision nesting ground birds face when approached by a nest predator is whether to flush and thereby sacrifice the clutch, or to remain and potentially be killed themselves (Bergerud and Gratson 1988). This decision and the consequent reputation of spruce grouse for exceptional stillness when approached, reflect the life history strategy of this species. Producing small clutch sizes (4-6 eggs) and employing high levels of parental investment in nest protection, spruce grouse retain lower annual 8 fecundity than other similar-sized grouse and make decisions based on long term reproductive opportunities (Sibly and Calow 1985). Because food availability is likely not limiting to spruce grouse, which relies on ubiquitous forage such as blueberries, fungi and spruce needles (Robinson 1980), small clutch sizes suggest a strategy of increased parental survival and investment. Due to the increased relative importance of adult and juvenile survival for spruce grouse relative to other similar-sized grouse, the influence of nest success, and consequently nest-site selection, is increased.

My results suggest that spruce grouse, like other ground nesting birds (Tirpak et al. 2010, Fuller et al. 2013), select different vegetation characteristics at different spatial scales. At the patch scale, we found nest presence to be negatively correlated with basal area of live trees and tree density, suggesting that nests were associated with forest gaps.

Females may select for these open forests to allow for better predator detection as well as easier escape ability compared to densely vegetated stands, as seen in other gallinaceous bird species (Thompson et al. 1987, Tirpak et al. 2006). Increased daily survival rates of ground bird nests have been associated with decreases in stem density, suggesting that there is a demographic benefit associated with female escape cover (Fuller et al. 2013).

Spruce grouse also selected for higher recess height, which may be related to predator detection or escape structure. Because the escape response of this ground bird is to flush to a nearby branch (Robinson 1980), it is expected that this species would prefer areas of higher recess height for predator evasion when nesting. I also detected selection for increased lateral cover, which might play a role in the thermoregulation of incubated eggs. As ground nesting birds that engage in uni-parental care, hens must leave nests to forage, thus decisions about the thermal cover provided by dense vegetation are based on 9 the cooling and desiccation rates of eggs, which are known to be higher in spruce grouse than for sympatric galliformes (Naylor et al. 1987, Bendell and Bendell-Young 2006).

At the within-patch scale, spruce grouse exhibited selection for increased lateral cover, suggesting that on a finer scale, nest concealment is an important factor in nest-site selection. This selection for concealment is consistent with another study on spruce grouse (Anich et al. 2013), which concluded that overall concealment was a good predictor of nest-site selection, and that lateral cover determined nest survival. In the forested landscape of northern Maine, the majority of forest nest predators such as coyotes (Canis latrans “var”), red squirrels (Tamiasciurus hudsonicus), and other terrestrial species, would have nest detection abilities that were most impacted by increases in lateral cover, especially in areas of patchy understory vegetation (Yahner and

Mahan 1996, D’Eon 1997). Increased lateral cover might also provide olfactory concealment as well as visual benefits for nesting grouse by dispersing scent cues employed by nest predators, ultimately making nest detection more difficult for predators

(Conover 2007, Fuller et al. 2013). These observations are consistent with the ecology of many ground-nesting birds, which indicate trade-offs in nest site selection between the need for concealment to reduce detection by predators, but also a need for visibility around the nest to facilitate detection of predators as well as escape ability by the hen

(Thompson et al. 1987, Watters et al. 2002, Tirpak et al. 2006, Anich et al. 2013, Fuller et al. 2013).

This understanding of spruce grouse ecology also sheds light on how spruce grouse interact with the complex industrial forestry landscape of northern Maine. Harvest practices and resource management decisions that impact these areas have the capability 10 to alter current spruce grouse habitat and might potentially influence nest success. The current forest structures that spruce grouse select in Maine are relicts of past harvests, which are no longer employed. Many of the stands where we found nest-sites had a silvicultural history of clear cutting, herbicide treatment, and pre-commercial thinning.

These practices, while common in the 1970’s, have decreased to less than 5% of harvests on the Maine landscape (Jin and Sader 2006), and have been replaced with partial harvests that influence more land area and result in unknown forest structure and habitat quality for spruce grouse (Sader et al. 2003, Hoving et al. 2004, Fuller and Harrison

2005). The regional decline of this species is speculated to be connected with changes in forest structure that provide inadequate habitat for spruce grouse, which may potentially result in larger demographic issues for the regional population (Pietz and Tester 1982,

Williamson 2008, Seibold et al. 2013).

Suggestions for future research

There is a need for further research into nest survival rates of spruce grouse and the status of Maine’s population to understand what factors influence regional declines and what components of the landscape this species utilizes for other life stages. Future research should also focus on quantifying changes in forest structure that result from shifts in harvest practices to more accurately relate changes in forest structure from harvesting to population dynamics.

Future research should employ more robust study designs and statistical modeling approaches. For example, my analyses examined selection using combined use and combined availability across all individuals. This method contains many assumptions as all data collected on availability was assumed to be truly available to every individual. A 11 more robust sampling design would compare individual habitat use to individual habitat availability by comparing the vegetation at the nest site to vegetation within that individual’s home range. Another assumption of this study was that the focal stand in which a grouse was captured was considered as its home range, but more likely, a more diverse heterogeneous mixture of the focal stand with occasional sallies into berry producing harvest areas would likely represent the home range of an individual. For statistical analysis, I conducted many uni-variate logistic regressions that could have inflated my probability of type I statistical errors (Zar 1999). With a greater number of nests, I could have conducted more robust and complex modeling. Given a greater number of nest-sites, I would have conducted a principal component analysis (PCA), which would have examined the spread of the data and returned ranked principal components (PCs) composed of different suites of measured variables. From this I could have reduced the number of variables we considered in subsequent models. Using a reduced suite of variables, I could have used multi-variate model selection within an information-theoretic framework to explore multiple model structures explaining nest-site selection across my 2 spatial scales.

Management Implications

Natural resource managers interested in promoting viable spruce grouse populations should attempt to provide forest structure (Hewitt et al. 2001) required by different life stages of spruce grouse. Timber extraction or conservation practices which promote areas of dense, low understory vegetation as well as open forest structure provide simultaneous nest concealment and predator detection abilities during the nesting season. These conditions have been created in northern Maine’s industrial forests through 12 clearcutting, herbicide treatment, and pre-commercial thinning. The dense conifer stands resulting from past management provide ample canopy cover to protect spruce grouse from avian predators. Further, gaps in the canopy from thinning or harvest trails allow increased light ingress and contribute important patches of high lateral cover selected for by females for nesting. This study contributed to our understanding of how spruce grouse choose nesting sites within the commercially managed forests of north-central Maine and provides insight into the ecological interactions among life history strategies and resource selection.

13

FIGURES AND TABLES Table 1. Pearson Correlation Matrix for within-patch scale vegetation data. Variables with r>0.70 were removed from subsequent analyses and are highlighted. AVG$ BA$ BA$ BA$ BA$ BA$ Rel.$ AVG$ AVG$ Total$ AVG$ AVG$ AVG$ Tree$ Variable Lat.$ Decid$ conif$ dead$ decid$ conif$ conif$ Low$ CC BA DBH Ht Recess Density Cov. Sap sap sap trees trees BA Cano AVG*CC 1.000 AVG*Lat.*Cov. 0.631 1.000 BA*Decid*Sap ;0.358 ;0.187 1.000 BA*conif*sap ;0.539 ;0.430 0.355 1.000 BA*dead*sap ;0.395 ;0.447 0.160 0.708 1.000 BA*decid*trees ;0.299 ;0.377 0.313 0.076 0.239 1.000 BA*conif*trees %0.836 ;0.621 0.334 0.318 0.419 0.249 1.000 Total*BA %0.800 ;0.632 0.412 0.717 0.745 0.250 0.859 1.000 Rel.*conif*BA ;0.615 ;0.182 0.061 0.251 ;0.067 0.025 0.330 0.234 1.000 AVG*DBH 0.066 ;0.227 ;0.116 ;0.219 0.018 0.117 0.179 0.076 ;0.430 1.000 AVG*Ht ;0.244 ;0.278 0.036 ;0.145 0.116 0.098 0.556 0.409 ;0.221 0.791 1.000 AVG*Low*Canopy ;0.146 ;0.276 0.136 ;0.016 0.283 0.067 0.498 0.466 ;0.454 0.726 0.919 1.000 AVG*Recess 0.359 0.085 ;0.163 ;0.442 ;0.176 ;0.062 ;0.026 ;0.138 %0.717 0.803 0.723 0.764 1.000 Tree*Density ;0.602 ;0.433 0.367 0.366 0.552 0.429 0.678 0.660 0.221 ;0.165 0.056 0.114 ;0.266 1.000

Table 2. Pearson Correlation Matrix for patch-scale vegetation data. Variables with r>0.70 were removed from subsequent analyses and are highlighted. BA BA BA AVG AVG BA live AVG TOTAL AVG AVG AVG Tree Variable dead dead live Lat Low splings CC BA DBH Ht Recess Density splngs trees trees Cover Canop BA live splings 1.000 BA dead splngs 0.130 1.000 BA dead trees -0.109 0.043 1.000 BA live trees 0.166 -0.051 -0.032 1.000 AVG CC 0.232 -0.153 0.051 0.274 1.000 AVG Lat Cover -0.012 -0.065 0.046 -0.289 0.055 1.000 TOTAL BA 0.668 0.289 0.107 0.778 0.290 -0.219 1.000 AVG DBH -0.214 0.146 0.100 -0.020 -0.409 -0.117 -0.075 1.000 AVG Ht -0.066 0.313 0.098 0.229 -0.263 -0.271 0.221 0.692 1.000 AVG Low Canopy 0.290 -0.075 0.062 0.490 0.866 -0.093 0.495 -0.338 -0.064 1.000 AVG Recess 0.032 -0.078 0.182 0.051 0.456 0.068 0.066 0.002 0.123 0.334 1.000 Tree Density 0.304 -0.037 0.051 0.444 0.320 -0.092 0.478 -0.261 -0.057 0.445 0.064 1.000

Table 3. Results of univariate logistic regressions for structural variables hypothesized to influence nest-site selection by spruce grouse at within-patch and between-patch scales. Intercept and coefficient values, as well as corresponding P-values are presented and significant variables (α=0.05) are highlighted. Variable Coefficient (Bi) Intercept (Bo) p-value # BA Decid Splings -0.065 0.513 0.780 # BA conif splings -0.012 0.552 0.647 #BA decid trees -0.720 0.560 0.097 Within AVG CC 0.002 0.432 0.654 Patch AVG LatCov 0.006 0.106 0.031 Scale RelConif BA -0.032 0.526 0.945 AVG Ht -0.019 0.685 0.572 Tree Density -0.835 0.667 0.103 #BA live splings 0.000 0.068 0.944 #BA dead splings -0.004 0.073 0.73 #BA dead trees 0.01 0.063 0.565 #BA live trees -0.009 0.164 0.032 Patch AVG CC -0.001 0.096 0.344 Scale AVG Lat. Cov 0.002 -0.021 0.020 AVG DBH 0.001 0.058 0.906 AVG Ht -0.011 0.179 0.264 AVG Recess 0.001 0.026 0.000 Tree Density -0.173 0.115 0.041 14

Maine

Figure 1. My study area with the Telos region of north-central Maine highlighted. Map generated by Stephen Dunham. 15

Figure 2. The positive relationship between lateral cover and the probability of nest site selection of spruce grouse at the within-patch scale as described using logistic regression.

Figure 3. The positive relationship between lateral cover and recess height in stands and the probability of nest site selection of spruce grouse at the patch scale as described using logistic regression.

Figure 4. The negative relationship between basal area (m2/ha) of live trees and tree density (trees/m2) in relation to the probability of nest site selection of spruce grouse at the patch scale as described using logistic regression.

16

LITERATURE CITED

Anich, N., M. Worland, and K. Martin. 2013. Nest-site selection, nest survival, productivity, and survival of spruce grouse in Wisconsin. The Wilson Journal of Ornithology 125:570-582.

Bendell, J., and L. Bendell-Young. 2006. Eggs of spruce grouse dry at a faster rate than those of ruffed grouse. Canadian Journal of Zoology 84:1688-1692.

Bergerud, A., and M. Gratson. 1988. Adaptive Strategies and Population Ecology of Northern Grouse. University of Minnesota Press, Vol. 2. Minneapolis, USA.

Conover, M. 2007. Predator-prey dynamics: the role of olfaction. CRC Press, Boca Raton, Florida, USA.

D’Eon, R. 1997. Vegetative concealment, proximity to trails, and predator activity as relative factors affecting nest success and egg loss in spruce grouse, Dendragapus canadensis. Canadian Field-Naturalist 111:399-402.

Fuller, A., and D. Harrison. Influence of partial timber harvesting on American martens in north-central Maine. 2005. Journal of Wildlife Management 69: 710-722.

Fuller, A., and D. Harrison. Modeling the influence of forest structure on microsite habitat use by snowshoe hares. 2013. International Journal of Forestry Research 2013:1-7

Fuller, A., S. Spohr, D. Harrison, and F. Servello. 2013. Nest survival of wild turkeys Meleagris gallopavo silvestris in a mixed-use landscape: influences at nest site and patch scales. Wildlife Biology 19:1-9.

Hewitt, D., D. Keppie, and D. Stauffer. 2001. Predation effects on forest grouse recruitment. Wildlife Society Bulletin 29:16-23.

Homyack, J., D. Harrison, and W. Krohn. 2007. Effects of precommercial thinning on snowshoe hares in Maine. Journal of Wildlife Management 71:4-13.

Hoving, C., D. Harrison, W. Krohn, W. Jakubas, and M. McCollugh. 2004. Canada lynx habitat and forest succession in northern Maine, USA. Wildlife Biology 10:285- 294.

Li, P., and T. Martin. 1991. Nest-site selection and nesting success of cavity-nesting birds in high elevation forest drainages. The Auk 108: 405-418.

Lovell, G., G. Ruxton, K. Langridge, and K. Spencer. 2013. Egg-laying substrate selection for optimalcamouflage by quail. Current Biology 23:260-264.

17

Martin, T., and J. Roper. 1988. Nest predation and nest-site selection of a western population of the Hermit Thrush. Condor 90:51-57.

Martin, T. E. 1992. Breeding productivity considerations: what are the appropriate habitat features for management? Pages 455-473 in J.M.Hagan III, and D.W.Johnston, editors. Ecology and Conservation of Neotropical Migrant Landbirds. Smithsonian Institution Press, Washington, D.C.,USA.

Naylor, B., K. Szuba, and J. Bendell. 1987. Nest cooling and recess length of incubating spruce grouse. The Condor 90:489-492.

New Hampshire Fish and Game Department. 2006. Species Profile: Spruce Grouse. New Hampshire Wildlife Action Plan, Appendix A: Species Profiles – Birds. A-580 – A586.

Newton, I. 1998 Population limitation in birds. Academic Press INC. London, UK.

Pietz, P. and J. Tester. 1982. Habitat selection by sympatric spruce and ruffed grouse in north central Minnesota. Journal of Wildlife Management 46:391-403.

Jin, S., and S. Sader. 2006. Effects of forest ownership and change on forest harvest rates, types and trends in northern Maine. Forest Ecology and Management 228:177- 186.

Redmond, G., P. Herzog, and D. Keppie. 1982. Vegetative structure, concealment, and success at nests of two races of spruce grouse. Canadian Journal of Zoology 60:670-675.

Robinson, W. 1980. Fool Hen: The spruce grouse on the yellow dog plains. The University of Wisconsin Press, Madison, Wisconsin, USA.

Roper, J., K. Sullivan, and R. Ricklefs. 2010. Avoid nest predation when predation rates are low, and other lessons: testing the tropical-temperate nest predation paradigm. Oikos 119:719-729.

Sader, S., M. Bertrand, and E. Wilson. 2003. Satellite change detection of forest harvest patterns on an industrial forest landscape. Forest Science 49:341-353.

Sader, S., M. Hoppus, J. Metzler, and S. Jin. 2005. Perspectives of Maine forest cover change from landsat imagery and forestry inventory analysis (FIA). Journal of Forestry 103:299-303.

Scott, S. 2009. Spatio-temporal dynamics of snowshoe hare density and relationships to Canada lynx occurrence in northern Maine. M.S. Thesis, University of Maine.

18

Seymour, R. S. and M. L. Hunter, Jr. 1999. Principles of Ecological Forestry. Cambridge Univ. Press. London, UK.

Seibold, S., A. Hempel, S. Piehl, C. Bassler, R. Brandl, S. Rosner, and J. Muller. 2013. Forest vegetation structure has more influences on predation risk of artificial ground nests than human activities. Basic and Applied Ecology 14:687-693.

Sibly, R., and P. Calow. 1985. Classification of habitats by selection pressures: A synthesis of life-cycle and r/K theory. Proceedings of the Symposium of the British Ecological Society.

Silvy, N. J., 2012. The Wildlife Techniques Manual. 7thed. Johns Hopkins University Press. Baltimore, Maryland, USA.

Thompson, F., D. Freiling, and E. Fritzell. 1987. Drumming, nesting, and brood habitats of ruffed grouse in an oak-hickory forest. Journal of Wildlife Management 51:568-575.

Tirpak, J., W. Giuliano, C. Miller, T. Allen, S. Bittner, J. Edwards, S. Friedhof, W. Igo, D. Stauffer, and G. Norman. 2006. Ruffed grouse nest success and habitat selection in the central and southern Appalachians. Journal of Wildlife Management 70:138-144.

Tirpak, J., W. Giuliano, T. Allen, S. Bittner, J. Edwards, S. Friedhof, C. Harper, W. Igo, D. Stauffer, and G. Norman. 2010. Ruffed grouse-habitat preference in the central and southern Appalachians. Forest Ecology and Management 260:1525-1538.

Watters, M., T. McLash, C. Aldridge, and R. Brigham. 2002. The effect of vegetation structure on predation of artificial greater sage-grouse nests. Ecoscience 9:314- 319.

Williamson, S. J., D. Keppie, R. Davison, D. Budeau, S. Carriere, D. Rabe, and M. Schroeder. 2008. Spruce Grouse Continental Conservation Plan. International Association of Fish and Wildlife Agencies, Washington, DC. USA.

Yahner, R., and C. Mahan. 1996. Depredation of artificial ground nests in a managed, forested landscape. Society of Conservation Biology 10:285-288.

Zar, J., Biostatistical Analysis. 1999. Prentice-Hall, 4th ed. New Jersey, USA. Zwickel, F. C., and J. F. Bendell. 1967. A snare for capturing Blue Grouse. Journal of Wildlife Management 31:202-204.

19

AUTHOR’S BIOGRAPHY

N. Scott Parkhill was born in the Commonwealth of Massachusetts in the town of

Winchester on June 4th, 1992. He was raised in the Greater Boston Area until graduation from Winchester High School in 2010, when he also earned the rank of Eagle Scout. He pursued his B.S. at the University of Maine, majoring in Wildlife Ecology with a concentration in Wildlife Science and Management. He served as the Vice-President

(2011-2012) and President (2012-2013) of the University of Maine Student Chapter of the Wildlife Society, Secretary of the University of Maine Subunit of the American

Fisheries Society (2013-2014), and President of Xi Sigma Pi Forestry Honors Society

(2013-2014). He received the 2014 New England Outdoors Writers Scholarship, the 2014

Dwight B. Demeritt Award, and the 2014 P.F. English Memorial Award, as well as placing on the Dean’s list over four consecutive years. Upon graduation, he hopes to secure regional experience in the field, as well as a Master’s degree in wildlife ecology.

An avid outdoorsman, Parkhill also intends to become a much better hunter and fisher.