Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of manual thinning and glyphosate application on breeding bird communities in southern British… Easton, Wendy Elizabeth 1997

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1997-0405.pdf [ 3.47MB ]
JSON: 831-1.0075273.json
JSON-LD: 831-1.0075273-ld.json
RDF/XML (Pretty): 831-1.0075273-rdf.xml
RDF/JSON: 831-1.0075273-rdf.json
Turtle: 831-1.0075273-turtle.txt
N-Triples: 831-1.0075273-rdf-ntriples.txt
Original Record: 831-1.0075273-source.json
Full Text

Full Text

THE EFFECT OF M A N U A L THINNING A N D G L Y P H O S A T E APPLICATION ON BREEDING BIRD COMMUNITIES IN SOUTHERN BRITISH C O L U M B I A by W E N D Y ELIZABETH EASTON B.Sc , The University of British Columbia, 1989 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES Centre of Applied Conservation Biology Department of Forest Sciences Faculty of Forestry We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A April 1997 © Wendy Elizabeth Easton, 1997 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that, the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or - by his or her representatives! It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of . The University of British Columbia Vancouver, Canada Date DE-6 (2/88) 11 ABSTRACT Removal of deciduous vegetation in managed conifer forests is a major component of most silviculture programs. Despite controversy over non-target effects, research regarding the impact of vegetation management on birds is limited. I examined the effect of removing 90-96% of the deciduous tree volume on breeding bird communities in young conifer plantations during 4 years. Trees were removed by two treatments: manual thinning, and manual thinning plus application of the herbicide, glyphosate. The control and two treatments were each replicated three times. During the 3 post-treatment years, the herbicide-treated sites remained depauperate of deciduous vegetation while the manually thinned sites experienced deciduous regrowth. Variation in richness of bird species across all sites was solely due to the presence or absence of rare species. Richness declined, abundance increased, and common species dominated after herbicide application. Richness, abundance, and evenness increased after manual thinning. Turnover of bird species was highest in the herbicide-treated areas and lowest in the control areas. Residents, short-distance migrants, ground gleaners and conifer nesters increased significantly after herbicide application. Deciduous nesters and foliage gleaners declined in abundance (nonsignificantly) in areas treated with herbicide. The Warbling Vireo (Vireo gilvus), a deciduous specialist, declined in areas treated with herbicide. Although treated areas exhibited similar increases in total abundance, nesting success of open-cup nesting species was significantly lower in the herbicide-treated than manually thinned areas. Warbling Vireos, Swainson's Thrushes, and Dusky Flycatchers preferred to nest in untreated or recovered habitat within larger treated areas. Successful nests were more more likely to be present in areas with willow. Overall, the composition of bird communities became more homogeneous after herbicide application, and showed little change after manual thinning. T A B L E OF CONTENTS iii ABSTRACT i i LIST OF TABLES v LIST OF FIGURES vi ACKNOWLEDGEMENTS vii C H A P T E R O N E : INTRODUCTION 1 C H A P T E R TWO: T H E EFFECT OF V E G E T A T I O N M A N A G E M E N T ON B R E E D I N G BIRD COMMUNITIES INTRODUCTION 5 METHODS Study area and experimental design 6 Bird sampling 8 Data analyses 8 RESULTS Richness, abundance, evenness, and turnover 11 Abundance of species assemblages 17 Abundance of individual species 22 Nesting success 22 DISCUSSION Loss of species 25 Changes in abundance 26 Shifts in community composition 27 Nesting success 27 Population trends 28 C H A P T E R THREE: T H E EFFECT OF L O C A L H A B I T A T ON THE RESPONSE OF BIRDS TO V E G E T A T I O N M A N A G E M E N T INTRODUCTION 30 METHODS Birds 31 Vegetation sampling 32 Habitat variables 33 Data analyses 34 RESULTS Habitat characteristics at the treatment level 37 Relationship between birds and the local habitat 38 DISCUSSION 41 iv C H A P T E R FOUR: T H E R O L E OF DECIDUOUS V E G E T A T I O N IN N E S T H A B I T A T SELECTION A N D N E S T SUCCESS INTRODUCTION 44 METHODS Description of nest site and patch 46 Classification of nests Al Data analyses 47 RESULTS Nest site 48 Nest patch 49 Habitat of successful nests 50 DISCUSSION Nesting habita t 53 Successful nests 54 C H A P T E R F I V E : CONCLUSIONS/MANAGEMENT IMPLICATIONS 56 LITERATURE CITED 59 APPENDIX ONE 67 APPENDIX TWO 68 APPENDIX THREE 69 V LIST OF TABLES TABLE 1. Relative mean abundance of birds/station for each treatment and year. 13 TABLE 2. Nesting success of birds in the control and treated areas, 1994-1995. 24 TABLE 3. Habitat characteristics of the bird station vegetation plots, 1994-1995. 37 TABLE 4. Habitat models for the Dusky Flycatcher, Warbling Vireo, Swainson's Thrush, Chipping Sparrow, and American Robin, 1994-1995. 40 TABLE 5. Characteristics of nest habitat between successful and unsuccessful nests. 49 TABLE 6. Differences in characteristics of nest habitat in the control, manually thinned, and herbicide-treated areas. 49 TABLE 7. Models depicting the most likely nest habitat in the study, control areas, manually thinned areas, and herbicide-treated areas. A model showing the difference in nest habitat between successful and unsuccessful nests. 51 TABLE 8. Nest habitat of the Dusky Flycatcher, Warbling Vireo, Swainson's Thrush, and Chipping Sparrow. 52 VI LIST OF FIGURES FIGURE 1. Differences in total richness of rare birds (a), and mean abundance/station of total (b), common and rare birds (c) between the control and treated areas. 14 FIGURE 2. Evenness of the bird community depicted by relative abundance curves for the control, manually thinned, and herbicide-treated areas. 15 FIGURE 3. Percentage of bird species turnover in the control and treated areas. 16 FIGURE 4. Differences in mean abundance/station of birds between the control and treated areas for each migratory group. 19 FIGURE 5. Differences in mean abundance/station of birds between the control and treated areas within the foraging guild. 20 FIGURE 6. Differences in mean abundance/station of birds between the control and treated areas within the nesting guild. 21 FIGURE 7. Orientation of vegetation plots around each bird station. 35 FIGURE 8. Differences in the growth structure of deciduous vegetation. 36 FIGURE 9. Potential processes influencing habitat selection decisions. 43 Vll ACKNOWLEDGEMENTS This project was the result of input from many people. Fred Bunnell and Tom Sullivan designed the study. Kathy Martin, Holly Hogan, and Brenda Dobson added vegetation plots in the first post-treatment year. Marion Porter conducted the bird census in the first year, while Holly and Brenda did so in the second year. Laura Nagel and Lea Elliott assisted me in collecting data during the last two years. Salmon Arm Forest District provided logistical support. A grant to Fred Bunnell and Kathy Martin from the Fraser River Action Plan (Environment Canada) supported this research. I gratefully acknowledge additional support provided to me from a McPhee Fellowship, and Backman Award. I would like to thank my supervisor, Kathy Martin, who took a chance, and gave me this project to tackle. You always encouraged me to set my goals high, and supported me in pursuing them. I appreciate ALL the time and effort you generously gave (dinners too)! My committee members, Sue Glenn and Jamie Smith, offered words of advice and encouragement throughout my work. This includes Phil Burton in the early moments, who later fled north for better air. I especially thank you, Jamie. Your thoughts, and Animal Behavior course inspired me to look at ecology in a different light (and your weasels, and endless supply of red ink forced me). Tony Kozak taught me the importance of experimental design and degrees of freedom. Iain Taylor improved my writing through his invaluable workshop. Fiona Schmiegelow kindly edited parts of my thesis. Thank you very much! My lab-mates, Rachel Holt, Kari Nelson, and Len Thomas were instrumental during my studies. They showed me the ropes in the beginning and maintained my sanity throughout. Lenny was patient with my almost never-ending questions. Rachel shared her solitude, laughter, and spirit. It would have been a long, and lonely (and VERY quiet) road without your friendship, Rachel. Both of you helped me tremendously with theoretical and technical aspects of ecology ~ thank you. I will miss your passion for conservation and life. Thanks to Pierre Vernier for getting me in this mess in the first place, Tanya Wahbe, my yoga master, and my friends at CACB, who all shared moments of kindness, laughter, and computer help! Holly helped me with logistical difficulties early in this project. Thanks to Laura for tolerating the Canucks in the final, and the budding musicians next door, and Lea for creating an endless number of gourmet, vegetarian dishes. Without both of you my field seasons would not have been nearly as fun, nor successful (being the occasional bear magnet just added to the excitement). Helen Wickett, who kindly shared her home, humour and wisdom, was an inspiration to us all. Finally, I dedicate this thesis to my family. My husband, Owen (it's still strange to say that H word!) is my pillar of strength, ensuring that I do not take myself, or life too seriously. You are the BEST thing about life! My parents, Andrew and Elizabeth, never doubt my ability when I do, nor waver in their support. My brother, Craig, shows me the strength of the spirit. 1 CHAPTER O N E INTRODUCTION Research on forest management has focused primarily on measuring the direct impacts of harvesting and fragmentation on bird communities (Hagan et al. 1996, Schmiegelow et al. 1997). However, habitat alteration continues after logging. Vegetation management is a major component of most silviculture programs (Franklin et al. 1986). In British Columbia alone, harvesting affected more than 53% of the harvested land (118, 466 hectares) in 1992 (Ministry of Forests 1993). By removing vegetation of little commercial value to expedite the growth of coniferous seedlings, vegetation management not only alters habitat structure and composition, but can change the successional trajectory of the forest (MacKinnon and Freedman 1993). Herbicide and manual treatments are two methods of vegetation management. Before the type of treatment is chosen, factors such as the size and tolerance of crop species, proximity of target area to water or residential land (i.e. often avoid herbicide treatment), terrain, and cost are considered (Biring et al. 1996). Glyphosate is the most commonly used herbicide in North America (Campbell 1990, Pimental 1991). It is a non-selective herbicide that damages or kills deciduous vegetation (Sutton 1978, MacKinnon and Freedman 1993, Morrison and Meslow 1984). The toxic effects of glyphosate on plants are related to the dose (amount of absorption), and the inherent susceptibility of the species (Grossbard and Atkinson 1985, Freedman et al. 1993). After glyphosate application, forb and shrub species recover, while hardwood trees show negligible regrowth two to three years after treatment (Sutton 1978, Lund-Hoie and Gronvold 1987, Kimminsetal. 1989). Thinning of deciduous vegetation is one type of manual treatment. As most hardwoods resprout vigorously after cutting, the effectiveness of manual thinning depends on the target species and the moisture status of the site (Biring et al. 1996). For example, manual thinning can 2 provide control of willow species (Salix sp.) for up to 2 years, and of paper birch (Betula papyrifera) for 1 year (Biring et al. 1996). The rapid resprouting of deciduous vegetation after thinning contrasts with the effect of glyphosate application, which generally suppresses deciduous regrowth (Slagsvold 1977, Sutton 1978, Lund-Hoie and Gronvold 1987, Runciman and Sullivan 1996). Conifers may be damaged after manual thinning by solar damage due to the sudden break in canopy or falling slash (Biring et al. 1996). Experimental research on the effects of vegetation management on bird communities to date has been limited (review in Lautenschlager 1993). Much of the research lacks either replication, pre-treatment data, or evaluates only one post-treatment season. After herbicide application the total abundance of songbirds may decrease (Slagsvold 1977, MacKinnon and Freedman 1993) or it may not (Morrison and Meslow 1984). Drops in the densities of individual species can recover two years after glyphosate application (Morrison and Meslow 1984), and five years after manual thinning (Slagsvold 1977). Densities of species that use brushy, deciduous vegetation, or are characteristic of older forests, often decline after treatment, while densities of species that inhabit more open areas increase (Slagsvold 1977, Morrison and Meslow 1984). Untreated areas are colonized by different bird species compared with plots treated with herbicide (Slagsvold 1977, MacKinnon and Freedman 1993). A 4-year study was conducted to investigate the effects of two common methods of vegetation management, manual thinning, and manual thinning plus glyphosate application, on breeding bird communities in south-central British Columbia. After one study season, 90-96% of the deciduous tree volume was reduced on six, 22-47 hectare plots (two treatments replicated three times), while the deciduous tree volume increased 48% on three, 23-36 hectare plots (Runciman and Sullivan 1996). Because herbicide application should suppress the regrowth of deciduous vegetation beyond the one to two year control achieved by manual thinning alone, I 3 predicted that glyphosate application would have a greater negative impact on birds. In Chapter 2,1 evaluate the effect of the two vegetation management treatments on the abundance and diversity of the breeding bird community. I interpret the richness, abundance, and evenness of bird species individually rather than using an index, which can oversimplify an ecologically complex problem (Halpern and Spies 1995). I assess i f alterations in community structure were mediated by birds with similar migratory status or ecological requirements. Generally, birds increased or decreased their abundance in the control and herbicide-treated areas according to their apparent affinity for deciduous trees. Finally, I show that nest depredation was greater in the herbicide-treated areas than the control and manually thinned areas. At the treatment level, bird species appeared to vary in their dependency upon the density of deciduous trees (Chapter 2). For example, in herbicide-treated plots, which remained depauperate of deciduous trees three years after treatment, Warbling Vireos decreased while Chipping Sparrows increased in relative abundance. Three years after manual thinning, the number of deciduous trees was 75% of the number in the control area and both Warbling Vireos, and Chipping Sparrows, increased in number relative to their pre-treatment levels. Although Warbling Vireos responded consistently at the plot level to changes in the volume of deciduous trees, Chipping Sparrows did not. Variation in their response to changes in the density of deciduous trees at the treatment level may be a result of heterogeneity of local habitat. Discrepancies in bird-habitat associations at different spatial scales is common (Wiens et al. 1987). In Chapter 3,1 investigate the relationship between the presence of five bird species and the local habitats they use. I compare these bird-habitat relationships with the broader scale results of Chapter 2, to test if the patterns of bird and deciduous tree presence are consistent. Warbling Vireos showed consistent habitat associations across the two scales, while Chipping 4 Sparrows did not. I also show that, similar to other research (Rice et al. 1983), birds vary in their degree in selectivity of habitat. The presence of suitable nest sites (Martin and Roper 1988, Steele 1993) and the opportunity of nesting success may be major influences during the process of habitat selection (Sieving and Wilson 1997, Martin 1993). Some species prefer to nest in deciduous than coniferous habitats (Lundberg et al. 1981, Willson and Comet 1996). In Chapter 4,1 test if the density of deciduous vegetation is correlated with nesting habitat for the American Robin, Chipping Sparrow, Dusky Flycatcher, Swainson's Thrush, and Warbling Vireo, particularly with the habitat of successful nests. Deciduous vegetation was correlated with nesting habitat in all areas. Successful nests were correlated with willow stems. To evaluate the impacts of vegetation management on birds, both the ecological consequences and the goals of forest management must be considered. In Chapter 5,1 conclude by assessing the management implications of manual thinning, and manual thinning plus herbicide, on birds. It is desirable to produce both specific conclusions for future forest managment in other stands in the study area, and general conclusions that can be extended to other areas. I recommend that to maintain deciduous specialists, manual thinning, or perhaps a less comprehensive herbicide treatment can be used (Chapter 2). Untreated patches within a larger target area of vegetation management, most notably in areas treated with herbicides, should be maintained. Birds occupied and nested in local habitats with more deciduous vegetation in both treated areas (Chapter 3 and 4). Managers should maintain specific habitat features within treated areas for those species that are listed provincially or are declining locally. 5 CHAPTER TWO-: T H E EFFECT OF VEGETATION M A N A G E M E N T ON BREEDING BIRD COMMUNITIES INTRODUCTION The rationale for removing deciduous species by vegetation management is to maximize fiber production by stimulating growth of coniferous seedlings and shortening periods between harvesting (MacKinnon and Freedman 1993). However, vegetation management may have greater consequences for plant species diversity than via the direct effects of logging and burning (Halpern and Spies 1995). In young plantations without vegetation management or young natural forests of the Pacific Northwest, hardwoods may dominate the tree canopy, contributing to the floristic and structural diversity (Halpern and Spies 1995). As the structure (Willson 1974, Rotenberry and Wiens 1980, James and Warner 1982) and species composition of vegetation (Rice et al. 1984) influences the presence of some bird species, vegetation management may have profound effects on bird populations. Herbicide application and/or manual thinning are commonly used to control deciduous vegetation (Biring et al. 1996). Six years after treatment, areas sprayed with herbicide are less complex structurally and floristically than untreated areas, and are colonized by different bird species (Freedman et al. 1993, MacKinnon and Freedman 1993). Within forests, species richness of birds and population density are positively correlated with habitat heterogeneity (Freemark and Merriam 1986). Within clear-cuts, certain bird species are associated with the density of regenerating hardwoods (Titterington et al. 1979, Santillo et al. 1989a). A dense growth of monotypic young conifers may exclude some birds from occupying the area (Schwab 1979). Variation in the structure of the bird community can be measured by changes in the composition, diversity (species richness and evenness), and abundance of bird species (May 1982). Changes in the habitat's horizontal structure can affect the evenness and richness of bird communities differently (Rotenberry 1978). The dominance structure of a community can shift in a polluted or stressed environment (Magurran 1988). After the selective removal of deciduous plants by vegetation management, bird species associated with deciduous cover may either alter their behavior to utilize the remaining vegetation (Morrison and Meslow 1984) or decline in numbers (Slagsvold 1977, MacKinnon and Freedman 1993). However, the quality of habitat cannot be measured by bird presence alone, as birds may occupy marginal or unsuitable habitats due to limited resources elsewhere (Van Home 1983, Pulliam 1988). I investigated the effects of two common methods of vegetation management, manual thinning, and manual thinning plus glyphosate application, on breeding bird communities. Rather than use a diversity index, which can be difficult to interpret (Whittaker 1972), I measured richness, abundance, and evenness individually for each treated area. I predicted that richness and abundance would decline, and evenness would change, after the removal of deciduous vegetation. This decline would be strongest in rare species, as well as in species assemblages with a higher dependency on deciduous vegetation. The shift in evenness would show an increase in the dominance of common species. I hypothesized that species turnover would be higher and nesting success would be lower in areas affected by vegetation management. Because of the longer term suppression of vegetation by herbicide, I predicted that herbicide application would have a greater negative impact on the bird community than manual thinning alone. METHODS Study area and experimental design The experiment was conducted at two sites, Eagle Bay (50° 55'N, 119° 11' W) and Sicamous (50° 52'N, 118° 59'W), in south-central British Columbia, Canada. Both sites are in the Interior Cedar-Hemlock biogeoclimatic zone (Lloyd et al. 1990), where climax forests are dominated by western redcedar (Thuja plicatd) and western hemlock (Tsuga heterophylla). Serai stages are 7 commonly characterized by Douglas-fir (Pseudotsuga menziesii ssp. menziesii), lodgepole pine (Pinus contortd), and paper birch. See Appendix 3 for a complete list of vegetation species surveyed in the study sites. The climate consists of warm, moist summers and cool, wet winters (Ketcheson et al. 1991). The study ran from 1992 to 1995. In the spring of 1992, three replicate plots each ,of a control, manual treatment, and herbicide treatment were established. Two sets of replicates were located at Eagle Bay and one set at Sicamous (Appendix 1). The plots were primarily embedded in a matrix of mature forest. To the north, the Eagle Bay sites were close to a rural, residential area (approximately 1 kilometre). Also in Eagle Bay, a fire in the fall of 1993 destroyed the mature forest border directly south of the control plot "F", the herbicide-treated plot "I", and the manually thinned plots "D" and "E". Pre-treatment data were collected from May to July 1992, prior to the treatment applications from September 25th to October 22nd, 1992. In both the manual and herbicide treatments, most paper birch, black cottonwood {Populus balsamifera ssp. trichocarpd), and trembling aspen {Populus tremuloides) were cut manually with power-saws, except for a few individual stems that were left in openings within the plantations. Alder {Alnus sp.), bitter cherry {Prunus emarginatd), Douglas maple {Acer glabrum), and willow stems within one metre of, or taller than, a planted conifer seedling were also cut. In the herbicide treatment, these stumps were also hand-sprayed with glyphosate. Pre-treatment data was collected from May to July, 1992. Treatments were applied in September and October, 1992, reducing deciduous tree volume by 90-96% (Runciman and Sullivan 1996). Post-treatment data were collected annually from May to July, 1993 through to 1995. Bird sampling Birds were surveyed using point counts (Reynolds et al. 1980) at permanently established stations placed at least 150 metres from each other, and 75 metres from the plantation edge. Due to differences in plot size, the number of stations per plot varied (Appendix 1). Stations were censused four times each year in May and June, from 0530 hours to 0830 hours. Each year, two observers surveyed each station twice at different times in the morning. Six different observers conducted bird surveys during the four study years. A l l birds seen or heard within a 75 metre radius during an 8 minute sampling period were recorded. A complete list of bird species detected is given in Appendix 2. Nests were found by searching thoroughly the plots during the final two post-treatment years, 1994 and 1995.1 standardized nest searching by spending equal search time within each plot. As the treatments mainly affected deciduous trees, I concentrated my efforts on finding nests of species that normally nest in these trees. I monitored nests every 3-4 days for the number of eggs and nestlings. I considered nests to be successful i f at least one host nestling fledged. Data analyses I used Shapiro- Wilks and Bartlett's test for normality of data and homogeneity of variances (SAS Institute Inc. 1989). A l l statistical analyses were conducted using SAS for Windows Version 6.10 (SAS Institute Inc. 1989). The criteria for statistical significance (a-level) was 0.05. Richness. — Weekly variability in bird species richness can be due to sporadic presence of non-resident birds foraging opportunistically, weather, chance, etc. (Kricher 1973). To ensure that I censused regular inhabitants rather than casual visitors I included bird species i f recorded on two or more censuses per station (Willson et al. 1994). As a result, I may have excluded some rarer, resident species. I preferred this possibility to the alternative of inflating the species richness and 9 abundance (Willson et al. 1994). Since grouse, raptors, shorebirds, waterfowl and birds flying over the plot cannot be adequately surveyed by point counts (Schmiegelow et al. 1997), I excluded these species from the analyses. I calculated species richness directly from the total number of species (meeting the above criteria) in each treatment (Whittaker 1972). I compared the presence of common and rare species. Birds present at more than 10 stations/year, and that made up more than 5% of the total abundance/year were considered common, while the remaining species were classed as rare. I graphed the actual effect size of the treatments on the richness of rare birds by removing the annual variation in richness. The richness of rare birds in the control area was subtracted from the richness of rare birds in the treated areas for each year. Since the original richness of rare birds was not equal across the control and treated areas, I was most interested in the change of richness within, rather than between treated areas. To report an index of change in the richness of rare birds from the first to the last study year, the richness in each treated area was divided by the richness in the control area for the pre-treatment year of 1992, and the last post-treatment year of 1995. Each value was multiplied by 100% and the 1992 value was subtracted from the 1995 value. This index does not imply that the changes in richness are statistically significant, rather it shows the relation of the response of birds in the treated to the control areas. The closer the index is to the 0, the more similar the treatment trend is to the control trend. Abundance. — As not all males within a population are mated during the breeding season (Gibbs and Faaborg 1990, Villard et al. 1993), I did not assume that singing males were paired and counted singing, calling, and sighted birds as one individual. I calculated the bird abundance at each station as the mean of the counts at that station in a given year. Since my methods did not allow me to calculate the density of birds per station, the abundance measure is actually the 10 relative abundance of birds within the study. To test the null hypothesis that there was no difference in the response of abundance of birds between treated areas over 4 years, I used a two-factor nested factorial analysis of variance (ANOVA) (Hicks 1993). For the data that did not meet the requirements of a parametric A N O V A , I used a nonparametric Kruskal-Wallis A N O V A (Zar 1996). A significant treatment by year interaction (TxY) indicated that birds responded differently to the treatments over time. Differences in bird responses to the treatments were compared using polynomial contrasts, rather than multiple range tests, because they are more sensitive to mean differences (Day and Quinn 1989, Green 1993). I set up the following a priori contrasts to compare the differences in abundances between years: 1992 versus 1993-1995 (CONTRAST YEAR 1 VS 2-4); 1993 versus 1994-1995 (CONTRAST YEAR 2 VS 3-4); and, 1994 versus 1995 (CONTRAST YEAR 3 VS 4). I graphed the actual effect size of the treatments on relative abundance and reported the percent change of abundance using the same method as described for richness. Evenness. — Although diversity consists of two individual concepts, species richness and evenness (Simpson 1949, Krebs 1989), investigation of the relative abundances of species (evenness) is often ignored (Magurran 1988). The evenness of the bird community in each treatment was evaluated from relative abundance curves and comparisons of the annual variation within treated areas (Whittaker 1965). Turnover. — I used the Jaccard coefficient (J) of similarity to calculate the percentage of species turnover as (1-J) x 100 (Krebs 1989). Assemblages. — To determine if the treatments were affecting birds with similar characteristics, I grouped species by migratory status, and into foraging and nesting guilds, comparing the trends of their relative abundances. Migratory status, predominant forage method, and nest site of each species are given in Appendix 2. Due to their low relative abundance, bark gleaners and 11 nectivores were not examined. I compared the abundance of different assemblages using the same method as for the overall abundance. Species. — Using the same method as for the overall abundance, I compared the abundance of the ten most common species individually across treatments. Nesting success. — Nest survival probability of species with more than one nest in at least two treated areas was calculated using Mayfield's method (Mayfield 1975). I pooled the years, 1994 and 1995, to test whether nesting success differed between the control and treated areas using the log-likelihood ratio test (Zar 1996). RESULTS Richness, abundance, evenness, and turnover I observed 51 species in the study sites (Appendix 2). Seven common species made up 79% of all detections: Dusky Flycatcher (23% of all detections); Dark-Eyed Junco (17%); Orange-Crowned Warbler (10%); Chipping Sparrow (7%); MacGillivray's Warbler (7%); Swainson's Thrush (8%); and Warbling Vireo (6%) (Table 1). Among the remaining 25 rarer species, American Robins, Yellow-rumped Warblers and Nashville Warblers were the most common, and accounted for a total of 13% of all detections. Of the 19 excluded species observed at the sites (Appendix 2), there was evidence that Black-headed Grosbeaks (2 nests) and Ruffed Grouse (drumming) bred on control areas, and Blue Grouse (display) bred on a manually thinned site. Variation in species richness was entirely due to the presence or absence of rare species. The seven common species remained in all areas after treatment. Compared with the control, the number of bird species increased by 8% after manual thinning, and declined by 17% after glyphosate application, three years after treatment (Fig. la). Initial richness was lower in the manually thinned areas (12 species), compared with the control (19), and the herbicide-treated 12 areas (17). Richness of birds in the study area fluctuated annually, resulting in more bird species in 1992 (24) and 1994 (23), than in 1993 (15) and 1995 (19) (Table 1). The abundance of birds increased annually throughout the study due primarily to the significant increase in the numbers of common birds (Fig. lb and c). Although there was no significant difference between the control and treated areas, the abundance of birds in the control remained relatively unchanged (Table 1), while it increased by 40% after manual thinning and 26% after herbicide application (Fig. lb). Relative to the control, common birds increased by 13% and rare birds by 64% after manual thinning, while they increased 15% and 14% respectively after glyphosate application (Fig. lc and d). There were no significant differences between the control and treated areas. Throughout the study, the annual structure of the relative abundance curves was most similar in the control area, where the increase in the abundance of common species was primarily due to the most dominant species (Fig 2a). The evenness of both common, and the most common of the rare species, increased in the manually thinned areas (Fig. 2b). Finally, the herbicide-treated areas had an increase in the dominance of common species (Fig. 2c). The herbicide-treated areas showed a greater turnover of bird species than the control or the manually thinned areas (Fig. 3). Although some of the community trends were similar, they were mediated by different bird species. After glyphosate application, species such as Nashville Warblers, Lazuli Buntings, Cedar Waxwings, Willow Flycatchers, and Lincoln's Sparrows disappeared, while Song Sparrows, Western Tanagers, and Hermit Thrushes arrived. Manually thinned sites lost Golden-crowned Kinglets and gained Song Sparrows, while control areas lost Alder Flycatchers and Willow Flycatchers, and gained Solitary Vireos. 13 TABLE 1. Mean abundance of birds/station. Common species are in bold. The total mean abundance/station (A) and species richness (R) are calculated for each treatment and year. See Appendix 2 for scientific and common names of bird abbreviations. CONTROL MANUAL HERBICIDE SPECIES 1992 1993 1994 1995 1992 1993 1994 1995 1992 1993 1994 1995 DUFL 3.60 5.46 4.48 5.18 3.89 4.92 4.68 6.61 4.12 5.49 5.20 6.23 DEJU 3.83 2.23 3.60 2.17 3.58 3.04 4.14 2.70 3.87 4.51 5.35 5.46 OCWA 1.19 1.88 2.12 0.95 1.76 4.13 2.31 1.96 2.30 2.30 2.77 3.40 CHSP 1.83 1.20 1.12 1.29 0.46 1.26 0.98 1.91 0.76 1.34 2.04 4.24 MGWA 0.67 3.06 0.68 1.88 0.65 1.65 2.05 3.02 0.87 2.52 1.15 1.27 SWTH 0.63 0.40 1.20 1.51 3.40 2.44 2.21 2.77 2.00 1.51 1.48 0.96 WAVI 1.02 2.14 2.15 3.68 0.96 1.67 1.23 2.24 0.81 0.57 0.17 0.07 AMRO 1.61 1.37 1.56 1.85 0.18 0.17 0.99 1.20 0.81 0.57 1.27 0.97 YRWA 0.32 0.70 1.00 1.20 0.75 - 0.91 2.01 0.84 0.43 1.26 1.74 NAWA 0.79 1.03 1.94 1.52 0.19 0.73 1.69 1.28 0.11 0.32 0.58 -LABU 1.75 0.72 0.33 0.38 - - • - - 0.35 - 0.30 -SOSP 1.50 0.41 0.50 0.23 - - 0.25 0.20 - - 0.62 0.20 CEWA 1.17 - 0.32 0.45 0.25 - - 0.21 0.40 - 0.25 -BHCO 0.92 - 0.39 0.21 - - 0.17 - 0.40 - - 0.10 humma 0.18 - 0.08 0.23 - - - - 0.52 - 0.20 0.30 ALFL 0.60 0.49 0.13 - - - - - - - - -PISI - - - - - - 0.32 - - - 0.85 -SOVI - - 0.28 0.18 - - 0.07 - - - - -BCCH - - 0.15 - - - - - 0.10 - - 0.25 NOFL - - 0.18 - - - - - - - 0.17 -WIFL 0.13 0.10 - - - - - - 0.10 - - -LISP - - - - - - - - 0.15 0.13 - -RNSA - - - - - - - - 0.11 - 0.07 -WWPE - - - - - - 0.39 - - - - -OSFL - - - - - - 0.08 - - - - -RWBL 0.04 - - - - - - - - - - -GRJA - - - - - - - - 0.11 - - -WETA - - - - - - - - - - - 0.10 GCKI - - - - 0.08 - - - - - - -HETH - - - - - - - - - - - 0.07 EVGR - - - - - - - - 0.07 - - -A 21.8 21.2 22.2 22.9 16.2 20.0 22.5 26.1 18.8 19.7 23.7 25.4 R 18 14 19 16 12 9 16 12 20 11 17 15 a includes Calliope and Rufous Hummingbirds i BO i y i — i * I O I C K I I o c ca T3 C 3 X> < m o H xi * * * p — I — h O O CN ^ 8 II * B » I I - e t a -o o o ^ •\ . \ ". \ •, \ • \ 01 h i cj c x> CQ CD }-< ca o CN o cn Mlt-K — — "2 o 2 '3 5 « IT: i f Q I 9 o s T3 s CD l -ca j w i \ \ \ \ \ HW I ? \ I \ \ I * * CD o c c X) CQ c o S S o O I ( — f t I * i I j j y j 1 ; I • I •' t * , \ i m-o o —5 o \ 1 o CN O co H - - H -•, \ • \ '• \ ' . \ I K> ft) I . i : l ; i -.'• i ii JL II s i / i l! W ft "^1 o o CN O o 1^1 as as o s o s cn o s o s CN os. Os Os Os Os Os cn Os Os CN Os Os ca CD ca c o e s o o ca X> O fc CD T 3 03 -0 C ca -*-» o v\ o 3 CD > CD O CM CO ca C CD CD ca -4-» o -*-» C4H o c o 6 TS O X J2 C/s o CD CJ § - o C 3 X> ca c ca CD c ca CD U-ca i -c« O CD CD T 3 C ^ CD ca o . ^ - £ T" d CM > X ' t3 K ca 3 o I a z o C (X, CD x ? la "3 CD T3 CD •*-> O c CD ca in T 3 C CD CD CD CD CD o ca co C CD CD «C \=i ca CD o _ g CD O c T3 ca C "O ca c on O c x T3 CD C/5 ca CD o .s CD O c ca •a C 3 X) ca ca ca > CD O c ca T3 c 3 • § c ca o CD i= S c ca CD X) w o cn • O V s< •a * o * o ca * o X) c o 6 a o O ?5 - S S « o o CD x : CD CJ a CD ± 3 c/5 * ^ CD CD J2 CD O S © "oo u Q ~ -BJ V u a > » * ^ X ) c fa o * L L CO CO c o II c CD CD X3 Q ^ 1 • —' ca s ^  s P -° 'S g S e fa 2 ca its ° ° H-i O oo xV ^ on — > CO > 1 FIGURE 2. Evenness of the bird community from 1992-1995, depicted by relative abundance curves for the control, manually thinned, and herbicide-treated areas. The more horizontal the curve, the greater the evenness and the lower the dominance of the bird community. Seven common species were identified (vertical line). Species to the right of the vertical line were considered rare. 17 Abundance of species assemblages Neotropical migrants may be more susceptible to habitat alteration than residents and short-distance migrants because they are more specialized in their habitat use (O'Connor 1992). In my study, residents and short-distance migrants responded positively to herbicide application, increasing in numbers by 179% and 86% respectively, from 1992 to 1995 (Fig. 4a and b). Neotropical migrants fluctuated in annual abundance similarly across the control and treated areas (Fig. 4c). Ground gleaners, such as Dark-Eyed Juncos, showed a significant increase of 130% in abundance in the herbicide-treated areas (Fig. 5a). The remaining classes of foragers were not significantly affected by the treatments (Fig. 5). However, foliage gleaners, such as Warbling Vireos, increased significantly in abundance after treatment (Fig. 5b). Although there was no significant difference in their abundance between treatments, foliage gleaners declined 93% in the herbicide-treated areas and 63% in the manually thinned areas compared with the control (Fig. 5b). Although these declines are statistically insignificant, the strong trends imply that some foliage gleaning species are negatively affected by treatment. Hawkers, such as Dusky Flycatchers, were significantly more abundant in all areas after treatment, increasing by 36% in relative abundance after thinning, and by 24% after herbicide application (Fig. 5c). Again, these differences between the control and treatments were not significant. Like ground gleaners, conifer nesters (i.e. Chipping Sparrows), increased significantly in numbers by 250% after herbicide application (Fig. 6a). The remaining classes of nesters were not significantly affected by the treatments (Fig. 6). However, species that nest in deciduous trees, such as Dusky Flycatchers, increased in abundance by 11% after manual thinning, and declined by 21% after herbicide application (Fig. 6b). Mixed tree nesters, such as Swainson's Thrushes, increased in numbers overall during the last two post-treatment years, despite a 39% decline in 18 the herbicide-treated areas (Fig. 6c). Ground nesters, such as Orange-Crowned Warblers, increased in abundance overall after treatment, particularly in 1995 (Fig. 6d). Compared with the control areas, ground nesters increased by 77% increase after herbicide application, and by 38% after manual thinning (Fig. 6d). Shrub nesters, such as MacGillivray's Warblers, increased their numbers non-significantly by 123% after manual thinning and by 25% after herbicide application (Fig. 6e). Herbicide application had the greatest effect on bird assemblages. Short-distance migrants, residents, conifer nesters, and ground gleaners increased their abundance in areas treated with herbicide. Although the response was not statistically significant, deciduous nesters, and more notably foliage gleaners, increased in numbers in the control and manually thinned areas compared with the herbicide-treated areas. a. Residents c o O u xi -*-> B o PH c 0 1 (/) o a 3 X) < c U o c m 2 1.5 1 0.5 0 -0.5 -1 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 2.0 1.0 0.0 -1.0 -2.0 - L 1992 TxY* — x -- o .0-c o n t r o l m a n u a l h e r b i c i d e b. Short-Distance Migrants TxY* ^ ..--I' c. Neotropical Migrants Y** ** ,i K * * t ' -~" — — s i " • < ' >, 1) t 1993 1994 1995 Year FIGURE 4. Differences in mean abundance/station of birds between the control and treated areas for each migratory group. See FIG. 1 for abbreviations. Residents (FT x Y i6,i8=3.17*) and short-distance migrants (FX x Y )6,i8=3-70*) increased significantly in the herbicide-treated areas only. Neotropical migrants fluctuated in abundance (FC O N TRASTYEAR2 vs3-4,i,i8=57.80**; F 3 V S 4 U G =101 .67**) . a. Ground Gleaners 4.0 ~ r TxY** c o V B o G O on o c 0 3 -a a D <: c c3 D s o a ID s 2.0 o.o 4 -2.0 -4.0 - x — - o -.0 . -3.0 2.0 b. Foliage Gleaners c o n t r o l m a n u a l h e r b i c i d e Y* 1.0 $ O.o 4 -1.0 1.0 0.5 0.0 -0.5 ** ** — I - - - - : : c. Hawkers -1.0 ^ r i ..•> 1 • ' \ ** <>- - . J ^ ~~ ~ " ~* f 1 1992 1993 1994 A Year 1995 FIGURE 5. Differences in mean abundance/station of birds within the forag guild between the control and treated areas. See FIG. 1 for abbreviations. Abundance of foliage gleaners and hawkers were different after treatment (FCONTRAST YEAR ivs 2-4,i,i8=24.07** and F , vs2-4,i,i8=23.06** respectively), and in 1995 compared with 1994 (F C O N T R A S T Y EAR3VS4,U8 = 11-16** and F1VS2.4,148=23.06** respectively). I H»-H—h-3 O 3 T3 * H M » in o d d \ • \ '. \ ', \ ', \ • \ '• I I CplOII / .' / . ' /.' J * >'d»i n h z T 3 c 3 O o KH O — i p CN II Q I ¥ \ \ \ \*fk I .7 •I I r. / ; / • i a a h-v. 11 AJWI o s I «-i o >n - i —^  d p d os os OS ON CN OS OS p d -H c o U I 1 OK II 0 <> 1 i A "A __i^Wi_t_ iv u i i— o d o - H Q I X \.1 >' 1 a -* * i o d p d C/3 —I u h -p o o — i cap; ptc \ . \ • \ t> I s I i Os os OS OS OS OS CN OS OS d o d d 10JJU03 UIOJJ uoijBJS/souepunqv ui33j/\j in aouajsmrj 22 Abundance of individual species Only chipping sparrows showed a significant response to vegetation management ( F X x Y ; 6 , i 8 = 4 - 9 1 , p=0.0044), increasing their abundance by 287% in the herbicide-treated areas, and by 123% in the thinned areas (Table 1). Dusky Flycatchers and Dark-Eyed Juncos were the most abundant species, on all plots in all years. I expected dusky flycatchers, as species that nests in deciduous trees, to be affected by the removal of deciduous vegetation. Yet, their abundance increased similarily in all areas over the study (F Y j 3 ; i8 = 10-09, p=0.0005) (Table 1). Conversely, Yellow-rumped Warblers showed an overall increase in numbers over the study in all areas (FY,3,I8=9-410, p=0.0007), compared with the control, they dropped by 118% after glyphosate application and by 66% after thinning. The trends for the remaining 7 species were not significant. However, Warbling Vireos almost disappeared from the herbicide-treated areas, dropping 77%, while increasing 261% in the control and 133% in the manually thinned areas (Table 1). Also, Nashville Warblers disappeared after herbicide application (Table 1). Nesting success Including all active nests with known outcomes (n=99 nests), nesting success was not independent of treatment (G=8.01, df=2, p<0.05). Predation was the main cause of clutch loss across all sites. Nesting success was highest in manually thinned areas with 46% of nests being successful (12/26 nests), intermediate in the control areas with 28% (13/47), and lowest in the herbicide-treated areas with 12% (3/26). The nesting success was very low for Chipping Sparrows in the control areas (2 nests were parasitized by Brown-headed Cowbirds, 1 abandoned, 1 depredated), and Cedar Waxwings in the herbicide-treated areas (Table 2). Overall, assuming an average of 26 days for the nest cycle and using the total daily probability of survival, an average nest would have a 48% chance of survival in the manually thinned areas, 30% in the control areas, and 17% in the herbicide-treated areas (Table 2). a ° 73 3 £ a co o Is i 2 | >1 It-! T3 O & J X W ^ Os C O ° ^ § a I CO CO +-» to o 3 O c U o CU ft &0 <o oo ^ j - r.-) H O (V) ,f | ,4 >o M * M 1 1 ^ " TJ- IT) oo os _ os <—i i n © •—i co o ir, io CN ro >n o co o —c r - -3-CN os i n TT —i , ro ON r*- o s os o s o s m oo © so oo CN so oo so r~ Os Os Os Os Os Os O S O s r ^ O s O S O s O S O S CN CO S ^ in rt —' a s oo ro CN — O O oo CO O CN — C O C O s o CO in "1 co as g O O - H - H O CN CN «n O 0 CN CN <—i ro ' w ' ^ - ^ * w ' co oo r -2 ( N M 2 °° CN CN «n co 25 DISCUSSION As anticipated, areas treated with herbicide had a higher loss of bird species (3 species), and a greater turnover of species compared with the control areas. Contrary to my predictions, the abundance of birds increased after both treatments. This was mediated by shifts in the dominance structure. Generally, the abundance of both common and rare birds increased in the manually thinned areas, while only common birds did so in the herbicide-treated areas. As expected, the dominance structure of the control areas was more constant than in the treated areas. Nesting success was particularly low in the herbicide-treated areas. Loss of species The decline in richness in the herbicide-treated areas compared with the thinned and control areas may be due to: food availability; delayed deciduous regrowth; and species vulnerability. Other researchers have noted lower densities of invertebrate herbivores (e.g. caterpillars), key foods for nesting birds, in herbicide-treated areas (Slagsvold 1977, Santillo et al. 1989b). There are higher densities of insects in the understory of deciduous forests than in coniferous forests (Werner 1983, Willson and Comet 1996). Second, and similar to other studies (Lund-Hoie and Gronvold 1987, Slagsvold 1977), manual thinning promoted deciduous sprouting while glyphosate application depressed deciduous tree regrowth (Runciman and Sullivan 1996, personal observation). Thus, glyphosate application encourages the succession of conifer forest stands that are homogeneous in structure and species composition (Franklin and Forman 1987, MacKinnon and Freedman 1993). The number of bird species positively correlates with the richness of tree species and canopy height (Willson 1974, James and Warner 1982), as well as habitat complexity (Recher 1969, May 1986). Loss of deciduous vegetation may result in loss or reduced abundance of associated deciduous specialists, plus other bird species that occasionally use this habitat during periods of 26 environmental stress (Karr and Freemark 1983). The Warbling Vireo, a deciduous specialist, declined strongly in abundance in herbicide-treated areas while generally increasing in the control and herbicide-treated areas. By depressing deciduous tree regrowth, glyphosate application may have reduced tree richness and habitat variability, preventing the recovery of the original bird fauna. Finally, the original bird community in the manually thinned area, dominated by common species, may have been less vulnerable to disturbance than the bird communities in the control and herbicide-treated areas. None of the seven common species were lost in either the control or treated areas, showing that they were generally more resilient to habitat manipulation. Specialists, such as Warbling Vireos, may be more susceptible to habitat alterations than habitat generalists, like Dark-eyed Juncos, that could use alternative habitats (Hagan et al. 1996). I do not know whether the response to treatment would have remained similar to the control areas i f the manually thinned areas had supported more rare species originally. Changes in abundance Unlike previous research that documented declines in bird abundance immediately following glyphosate application (Santillo et al. 1989a, Morrison and Meslow 1984, MacKinnon and Freedman 1993), and higher bird abundance in manually thinned areas compared with herbicide-treated areas (Slagsvold 1977), I observed similar increases in the abundance of birds after both treatments. This increase was due to the influx of more common individuals in the herbicide-treated areas, and common and moderately common (American Robin, Nashville Warbler, Yellow-rumped Warbler) in the manually thinned areas. As a result, evenness of the bird community increased after manual thinning. 27 Shifts in community composition A review of eight studies in northern coniferous ecosystems showed that bird species responded differently to vegetation management according to their foraging and nesting preferences (Lautenschlager 1993). I observed similar patterns. Species turnover was greatest in the herbicide-treated areas, and lowest in the control areas. The herbicide-treated areas became dominated by conifer nesting and ground gleaning species, species less dependent upon deciduous vegetation. Deciduous tree nesters and foliage gleaners showed a decline (non-significant) in abundance, relative to the control, in the herbicide-treated areas despite the increase of Dusky Flycatchers. Three years after treatment, the manually thinned areas had both a higher richness, and high relative abundance of birds. Lost foliage gleaning and deciduous nesting individuals returned and new individuals from the remaining foraging and nesting guilds appeared. Manual thinning may have added structural diversity, providing more foraging and nesting sites than were present prior to treatment. Despite concern that neotropical migrants are declining in numbers due to forestry practices (see papers in Hagan and Johnston 1992), as a group, neotropical migrants were not negatively affected by vegetation management. However, they may take two to three years to respond to habitat manipulations in their temperate breeding grounds (Hagan et al. 1996). Residents and short-distance migrants may occupy a broader range of habitats (O'Connor 1992). In herbicide-treated areas, residents increased in abundance due to Dark-eyed Juncos, and short-distance migrants due to Chipping Sparrows. Nesting success Nesting success was particularly low in the herbicide-treated areas. Although songbird nesting success estimates range from 11-77% («=32 species), the mean nesting success was 44% (review 28 in Martin 1992). The difference in productivity between the treatments may be related to vegetation complexity. Greater nest predation is found in open, rather than dense, riparian habitat (Blancher and Robertson 1984) and greater nest success with lower foliage density (Conner et al. 1986). Productivity may be lower in young plantation forests (Holt and Martin 1997). These results suggest that my study areas, particularly the herbicide-treated areas, are population sinks (sensu Pulliam 1988). I cannot determine if the treatments caused the disparity in nesting success. Population trends In the control areas, fluctuations in overall bird abundance were low and the variation in community evenness depended on the addition or subtraction of species rather than an increase in the relative abundance of species already present (Tramer 1969, Rotenberry 1978). This may be characteristic of a more predictable, stable environment. However, I found that bird abundance increased in the herbicide-treated areas, increasing the dominance of the more common species, despite poor local productivity. This implies that trends in abundance may be decoupled from trends in productivity, a characteristic of "source-sink" population regulation (Brawn and Robinson 1996). The population density and habitat quality relationship may be decoupled in a habitat, when birds opportunistically increase their density in low-quality habitats during periods of high production and high overall density (Van Home 1983). Despite lower nest survival in the herbicide-treated areas, Dusky Flycatchers increased in abundance. This increase in abundance may have been due to an increase in the numbers of unmated males rather than pairs. As their overall abundance was high, they may have opportunistically inhabited the poorer quality habitat of the herbicide-sprayed areas. The effects of disturbances should not be measured by only changes in total abundance and diversity as the quality of habitat may not be evident by the presence of birds alone (Van Home 1983). The herbicide-treated areas had higher bird abundance than the control areas, and more species than the manually thinned areas despite lower nesting success. Birds may be selecting different nesting habitat in the herbicide-treated areas, compromising their nesting success (Chapter 4). Vegetation management also altered the species composition of birds. However, birds appeared to vary in their response to the density of deciduous trees. As guilds, foliage gleaners and deciduous tree nesters appeared to have the flexibility to respond to the subsequent increase in deciduous trees two and three years after manual thinning. However, these patterns were not always evident at level of individual species (i.e. Swainson's Thrushes increased only in the control areas and declined in the treated areas). Variability in the local habitat may further explain differences in bird presence across treated areas (Chapter 3). 30 C H A P T E R THREE: T H E EFFECT OF L O C A L H A B I T A T ON THE RESPONSE OF BIRDS TO V E G E T A T I O N M A N A G E M E N T INTRODUCTION Patterns of bird-habitat relationships vary at different spatial scales (Wiens et al. 1987). At the treatment level in my study, some birds appeared to vary in their response to the removal of deciduous trees (Chapter 2). For example, the relative abundance of Chipping Sparrows increased after both treatments despite the difference in the number of deciduous trees; 4 trees per bird station in the herbicide-treated areas versus 32 trees in the manually thinned areas in 1995 (Table 3, page ). As the scale of the study decreases, habitat tends to be more heterogeneous (Collins 1983). Therefore at the local habitat, Chipping Sparrows may be responding consistently to the presence or absence of deciduous trees. Problems of scale can be circumvented by studying bird-habitat relationships at several hierarchical levels (Wiens et al. 1987). For several reasons, the removal of deciduous trees by vegetation management may not be consistent within treated areas. Often, there will be untreated sections (skips) within the application area that may leave pockets of deciduous trees (Santillo et al. 1989a). Variation in nutrient and moisture regimes of the soil, as well as differences in planting regimes, will also contribute to inconsistent habitat characteristics within treated areas (Appendix 1). Therefore, birds may be choosing local habitat that is atypical of the entire treated area. The scale of the study will influence the patterns in habitat occupancy that are detected (Wiens et al. 1987). Many factors and processes can influence where birds reside (Fig. 9). The species-specific responses of birds to environmental characterisitics are often thought to be the infrastructure on which the other processes interact (Wiens 1989b). Bird species vary in the breadth of habitats they will occupy (i.e. specialist versus generalist), and consequently, should differ in their 31 selectivity of habitats. To maintain selected bird species in managed forests, it would be beneficial to know not only the particular vegetation features that birds are associated with, but their degree of habitat selectivity. Variations in patterns of habitat occupancy by a species may also be affected by population density (Brown 1969, Fretwell and Lucas 1969). Plots may vary in the degree to which they are saturated by breeding individuals, influencing the response of individuals to habitat variability (Wiens et al. 1987). For instance, Dusky Flycatchers appeared to saturate entire study area despite vegetation management and their apparent preference for nesting in deciduous trees (Chapter 2 and 3). As a result, I would expect their habitat selectivity to be low compared with other species. Correlations of habitat characteristics with bird density are often based on the doubtful assumption that density means the same thing in different habitats (Wiens 1989b). Bird-habitat relationships constructed with presence-absence data will not be affected i f the density of birds at some local habitat varies independently of environmental factors (Rotenberry 1986). I chose to look at the relationship between the presence and absence of five bird species with local habitat features (bird-habitat associations). I compared these bird-habitat associations with patterns of bird abundance measured at the treatment level to see i f consistent patterns revealed habitat preferences. I also evaluated the differences in the degree of habitat selectivity of the five bird species. As the treatments altered the density and species composition of trees, I chose tree nesting species: Dusky Flycatcher, Warbling Vireo, Swainson's Thrush, Chipping Sparrow, and American Robin. METHODS Birds Chapter 2 contains methods for the census and calculation of abundance of birds. The trends in abundance of the five birds are found in Table 1. Due to the widespread abundance of Dusky 32 Flycatchers, the logistic regression model was used to distinguish between areas of low (< 2 individuals/bird station) and high (> or = 2 individuals/bird station) abundance. The remaining four models predicted the habitat in relation to the probability of occupancy of birds. I used the percentage of correctly classified events (presence or absence) as a measure of the degree of habitat selectivity (Rice et al. 1983). The sensitivity of a habitat model refers to the correct prediction of the presence of a bird, while the specificity refers to the correct prediction of the absence of a bird. If a bird was highly selective in its habitat choice, one would expect the habitat model to correctly predict both the presence and absence of the bird (model has high sensitivity and specificity). Vegetation sampling In 1993, four permanent vegetation plots were established at each bird station. Two configurations of the four vegetation plots were made: one for all of the even numbered bird stations, and one for all of the odd numbered bird stations (Fig. 7). One vegetation plot was placed along each north, south, east, and west bearing, at a randomly chosen distance from the bird station (Fig. 7). Each plot contained three nested subplots: 100 m (7.07 x 14.14 m) subplot 2 ' 2 for sampling trees; 10 m (2.24 x 4.47 m) subplot for sampling shrubs; and 1 m (.71 x 1.41 m) subplot for sampling ground cover. Tree and shrub species were considered 'trees' i f greater than 2 m in height. Numbers of all 'trees' in the large subplot were counted and classified into species and two size classes: less than, and greater than, 5 cm in diameter at breast height (dbh). In the 5m subplot, all shrub and tree species 0.5-2 m in height were considered 'shrubs'. Within this medium subplot, each shrub species was identified and its percent cover of the subplot estimated. The entire area of the 1 m subplot was categorized into type of cover (i.e. rock, herb species, moss, etc.) and the area they covered (%). 33 To distinguish the difference in growth structure in trees, I classified three types of vertical growth: stems, trees and sprouts (Fig. 8). Athough a sprouting tree had a group of sprouts emerging from the same cut stump, I classed it as one tree. The number of deciduous stems included the total number of individual sprouts and trees. Habitat variables For each tree species (Appendix 3), I counted the number of trees, sprouts, and stems. In addition, I grouped several tree species and size groups together: the two Populus sp. (P. balsamifera ssp. trichocarpa and P. tremuloides); deciduous species that prefer drier sites (bitter cherry, Sheperdia canadensis, Amelanchier alnifolia, Ceanothus' sp.) (Klinka et al. 1989); deciduous species that prefer wetter sites (Douglas maple, Alnus sp., Lonicera sp., Cornus stolonifera, Sambucus racemosd) (Klinka et al. 1989); lodgepole pine stems greater, and less than 5 cm in dbh; all deciduous species, both greater, and less than 5 cm in dbh; and, all conifer species, both greater and less than 5 cm in dbh. I combined the percent cover of shrubs for: each shrub genus; deciduous species that prefer drier sites (same as the tree species); deciduous species that prefer wetter sites (same as the tree species); all deciduous species; all conifer species; all shrub species. For the data surveyed in the 1 m subplot, I grouped the herbaceous species as: forbs; horsetails; moss, lichen, and fungi; grass; and, shrub and tree species. The habitat characteristics for each bird station were considered the local habitat. I counted the number of species of shrubs and trees in each N , S, E, or W vegetation plot to describe the total number of vegetation species for that bird station. I combined the values of each of the remaining habitat variables for each vegetation plot (N,S,E, and W) and calculated the mean of the individual variables. At the treatment level, the habitat was described by summing the 34 measures for each local habitat variable, and dividing by the number of bird stations within that particular treatment. These data were used for descriptive purposes and not analyzed statistically. Data analyses Using data collected in 1994 and 1995,1 reduced the number of vegetation variables in two steps. By examining a correlation matrix, I ensured that no variables were highly correlated (>75% correlation), dropping the number of variables from 105 to 68. Then I conducted a principal components analysis (PCA) to reduce 68 vegetation variables to 26 (Nichols 1977). The broken-stick model was used to determine the number of significant principal components (Jackson 1993). To build the bird-habitat association models, I pooled the data from 1994 and 1995.1 then used both stepwise and elimination logistic regression procedures (Hosmer and Lemeshow 1989), and chose the model that best predicted the presence or absence of a bird species (SAS Institute Inc. 1989). I chose logistic over linear regression, because the former does not require the data to have multivariate normality (Hosmer and Lemeshow 1989). Habitat variables had to have a significance level of 0.05 to enter, and to remain, in the models. I used the -2 times the log of the likelihood to test the significance of the model. The Hosmer-Lemeshow goodness of fit test, and the percentage of correct predictions compared with the observed outcomes, were used to evaluate how well the model explained the data (Hosmer and Lemeshow 1989). If the model closely fitted the observed data, the null hypothesis for the goodness of fit test (no difference between the observed and predicted values) would not be rejected, and the percentage of correct predictions would be closer to 100. The Wald chi-square test evaluated the significance of each variable (Hosmer and Lemeshow 1989). The parameter estimates can be used to calculate the estimated logit of the probability of an event (i.e. the presence of a bird). The standardized estimates are not influenced by the magnitude of each variable. Odd numbered bird station Even numbered bird station north west 5 0 m 10m south north 3 0 m east west 4 0 m 3 0 m south 1 0 m e Example of a north vegetation plot with the 3 subplots (tree, shrub, and herb). 7.07m 2.24m 14.14m .71m 4.47m 1.41m F I G U R E 7. Orientation of vegetation plots around each bird station. • • 1 Tree 0 Sprouts 1 Stem A 1 Tree 4 Sprouts 4 Stems B OTree 0 Sprouts 0 Stems C FIGURE 8. Idealized drawing of the growth structure of deciduous vegetation. Typically, (A) depicts the structure of deciduous vegetation in the control areas, (B) in the manually thinned areas, and (C) in the herbicide-treated areas (the cut-stump only). 37 RESULTS Habitat characteristics at the treatment level Compared with the control and manually thinned areas, the herbicide-treated areas had a very low density of deciduous trees or stems, and a higher density of coniferous stems (Table 3). During the second and third post-treatment year, the control and manually thinned areas had similar densities of deciduous stems, although there was generally more examples of regular growth structure (i.e. trees versus sprouts), and larger deciduous trees, in the control area. Typically, the herbicide-treated areas had more Douglas-fir stems, while the control and manually thinned areas had more stems of willow species. Finally, the control and herbicide-treated areas had larger lodgepole pine stems than manually thinned areas. Except for Taxus brevifolia, conifers rarely had sprouts. There was much heterogeneity between local habitats within the control and treated areas; delineated by generally high standard deviations (approximately 50-200% of the mean) for the mean habitat variables (Table 3). Table 3. Mean values of the habitat variables/bird station for the control (control), manually thinned (manual), and herbicide-treated (herbicide) areas in the second (1994) and third (1995) post-treatment year. The standard deviations are in parentheses. Habitat variables maintained in the regression analyses (Tables 4, 7-8) are presented. Control Manual Herbicide Habitat variable 1994 1995 1994 1995 1994 1995 no. deciduous trees 29(17) 42(18) 20 (20) 32(15) 1(2) 4(4) no. deciduous stems 41 (19) 58(21) 33 (29) 60 (30) 2(3) 7(6) no. deciduous sprouts 15(11) 22 (17) 17(16) 42 (26) 1(2) 4(4) no. deciduous trees (>5cm dbh) 2(2) 3(2) 1(2) 1(2) 0(0) 0(0) no. deciduous stems (>5cm dbh) 3(3) 5(4) 1(2) 1(2) 0(0) 0(0) no. Populus sp. stems 12(12) 17(15) 10(17) 15(21) 0(1) 1(1) no. of stems of water-shedding species 1(3) 4(6) 0(0) 1(3) 0(0) 0(1) no. of stems of Salix species 4(6) 5(6) 4(4) 5(5) 1(1) 2(3) no. coniferous stems 8(8) 10(7) 12(5) 14(7) 15 (15) 18(13) no. coniferous stems (>5cm dbh) 2(3) 2(3) 2(3) 3(3) 3(4) 3(4) no. lodgepole pine stems 4(4) 5(4) 4(5) 6(6) 3(6) 3(5) no. lodgepole pine stems (>5 cm dbh) 2(2) 2(3) 0(1) 1(2) 2(4) 2(4) no. Douglas-fir stems 0(1) 1(1) 2(2) 3(2) 3(4) 5(4) shrub cover 29(15) 26(14) 46(16) 31 (13) 35(11) 25 (10) richness of tree species 7(2) 9(2) 6(2) 8(1) 6(2) 7(2) richness of shrub species 8(3) 8(3) 7(2) 8(3) 9(2) 10(2) 38 Relationship between birds and the local habitat Habitat features associated with birds were also examined. Birds may not be dependent upon the associated habitat feature itself. Rather, the selected habitat variable may be correlated with environmental factors, resources, or processes that were not measured. Areas with more stems have smaller trees and the habitat is more dense. Higher density and species richness of vegetation are indicators of higher vegetative productivity. Areas with many conifers were usually naturally regenerated (i.e. western redcedar and western hemlock). Areas with more lodgepole pines and Douglas-firs were often more open because trees preferred drier sites and were planted at regular intervals (Lavender et al. 1990). Drier sites are usually less productive, supporting fewer trees than wetter sites, which can be densely populated with Populus species (Klinkaetal. 1989). Warbling Vireos were highly selective, choosing areas with deciduous trees and avoiding areas with more Douglas-fir stems (model had sensitivity and specificity of -80%) (Table 4). These trends in habitat choice were also evident at the treatment level, where the relative abundance of Warbling Vireos declined in the herbicide-treated areas, and increased in the control and manually thinned areas (Table 1). Compared with the control and manually thinned areas, the herbicide-treated areas had a lower number of deciduous trees, and more Douglas-fir stems (Table 3). Swainson's Thrushes, foliage gleaners and mixed tree nesters, were related to habitat features often associated with greater vegetative productivity, high richness of tree species and fewer lodgepole pine stems (Table 4). The reason for the decline (non-significant) in abundance of Swainson's Thrushes after treatment (Chapter 2) is not readily apparent by examining their habitat associations. Their habitat model could correctly predict their presence about 65% of the time (Table 4). 39 In contrast, the other mixed tree nester, the American Robin, was positively related to the presence of large lodgepole stems (Table 4). American Robins were also positively correlated to those deciduous species that prefer drier sites, and areas with low species richness of shrubs and trees. These habitat characteristics are often indicative of drier areas with lower vegetative productivity and were not typical of manually thinned areas (Table 3), where robins increased non-significantly in relative abundance (Chapter 2). Their habitat model had similar predictive abilities as the Swainson's Thrush model (Table 4). Chipping Sparrows, ground gleaners and conifer nesters, also avoided wetter sites. They preferred sites without Populus species and with more Douglas-firs (Table 4). Chipping Sparrows were associated with habitat characteristics typical of the herbicide-treated areas, where they increased significantly in abundance (Chapter 2). However, their habitat associations were inconsistent with those typical of the manually thinned areas (Table 3), where they showed a general increase in abundance (Chapter 2). Compared with the thrushes, the habitat model for Chipping Sparrows had a sensitivity of almost 80% (Table 4). Higher relative abundances of Dusky Flycatchers, deciduous tree nesters, were negatively correlated to large deciduous trees, and deciduous shrub cover (Table 4). Differences in their habitat selectivity between areas of low and high relative abundance were -70% (Table 4). At the treatment level, Dusky Flycatchers did not display such habitat preferences evident in the habitat model. cj is 3 H Q to a> 73 • £ I ° « S CJ o 3 OJ CO OJ ° - o a "8 9 aj V CJ . „ ' * s V 2 f-„ CJ * J> 3 . . > Q >s C > cj ? O s •8 CJ 43 V X cj +-» o CJ " 5-3 CJ C eo .—, cj OH 3 .B<7s CJ xi > to 1 3 CS CJ CS co .ta CS e CJ CJ CJ X co s SP to cj o T 3 3 3 • s cd •a u „ CJ -j= 3 y > CJ 43 60 cj C X! '3 u .a ts & p CJ T3 CO 73 * o O CJ > o CO CS CJ 3 "S Q ts 2 S § •S CJ «g 43 o , « —< <o 1 3 3. _ o S —1 2 " ts bO J-cj •9 XI CJ o 43 I S S 2 § S I co !a CJ 52 o T 3 1 o 00 o O 4= J 73 - a cj N CJ -3 ts 3 CS 3 -a -3 CS 4> GO 73 00 CJ CS o 3 cs > CJ 'o a. &o in o r--o 00 OS i n 00 SO 00 o 00 OS p in r -00 OS OS co SO i n CN i n o ro ro CN O 0 cn CN o s £ J s o * ° o" P <^  ro © 0 9 o * * * * * * OS * m 0 so »—' i n 06 O SO —c 00 OS # * * * * * SO i n CN cn 00 CN cn cn O OS o CN d o so o s 0 0 o CN C O O o o 00 a s v o 00 S CN CN so CN U-> O CO d o o ^ "~> OS o s >o 2 OS C O ^ ^ C O -r 9 d _ >o ^ - 1 s ^ _ H r - so r~- 9 d ^ d 00 -rr CO OS CO CO CO CN d X T3 w Q o w Q o W Q O w Q O i n A cj is CO 3 O l-i T3 CJ 3 O X 3 O 3 CJ -a to 6 CJ CJ CJ -*-» CO -«-» CO UOU glas-12 glas-cx 3 CJ CJ CJ T3 0 0 i-i Q CJ •+-» a 0 d 3 CJ 3 CO a, M 3 a, u »<^p -CJ T3 O O CJ CJ • S c o CJ a. co to S I • K A SP O. £5. 3 1) Cj O CJ e O O • S e c o CJ D. 3 O 3 12 u CJ T 3 m so o CN m CN CN SO CN OS v o 4 CO OS m "* d o o * * * O CO — 1 CN CO CO v o 00 O O 0O 0 CO OS 00 v o SO i n CN CO m CN SO 00 00 0 0 OS <n 1—1 m i n ro H t-H . 1 CN 0 VO CN SO >n 0 CO ro o s 0 O 00 p *—< CO O < CO *—1 »—< 0 d 0 O O 0 d d d d d d d d O O d so <=> r~ £ £ s: a " i f~ r o co CN 0 9 9 43 X -a o i n CO s CJ 83 " « CJ _1 W Q O cj •r- CJ O OH CJ CJ e 0 .3 3 xi . O to ts > o >^ so E if Q 5 v o i n II 3 3 CS -a 3 3 • s O cj . 3 > 3 00 s > o 00 i n 2 cj C lo cj CJ 23 M X & cs 3 l 1 3 CS to CJ > CJ co P Si X GO O , CS OH 00 ao .3 OH OH 4= u 5o II CO CJ 3 to CJ CJ CO 6 - ° OH CS .3 X o 00 j? i " .3 bp fc -§ c^  -£ 6 OH 8 CO CO CO CJ CO Jin 3 ^ CO CO . ^ H QJ a "§ ^ ft " to X CJ P cj _2 -3 to § 3 « CJ CJ to C X OH CS 41 DISCUSSION Greater consistency between treatment effects and local habitat associations of birds in my study implied habitat preferences. Chipping Sparrows, and more notably, Warbling Vireos, were more selective than American Robins, Swainson's Thrushes, and Dusky Flycatchers. Local habitat associations of Warbling Vireos confirmed their preference for areas with small deciduous trees. Chipping Sparrows preferred drier, conifer-dominated habitats. A species' flexibility in habitat choices will determine its vulnerability to habitat manipulation. The probability of Warbling Vireos disappearing in areas where hardwood regeneration is removed permanently (eg. glyphosate treatment) is high. Variation of the local habitat was higher within manually thinned than the control areas despite both areas having similar mean values of deciduous density (Table 3). The moisture status of the local habitat, which influences productivity levels, affects the response of vegetation to manual thinning (Biring et al. 1996). The more productive a site is, the more likely it will quickly promote vigorous sprouting of thinned deciduous vegetation. Thinning in less productive sites (little sprouting) will negatively affect the presence of those bird species using dense deciduous vegetation. Both American Robins and Chipping Sparrows, which increased in numbers in the manually thinned areas, were associated with vegetation variables typical of less productive sites. Swainson's Thrushes, which declined in thinned areas, occupied more productive local habitats. This was similar to other studies (Hansen et al. 1995), where American Robins and Swainson's Thrushes occupied different habitat types. This may be a result of historical or present level of interspecific competition. Varying effects of vegetation management on the presence of birds have also been documented elsewhere (Slagsvold 1977, Lautenschlager 1993). 42 More Dusky Flycatchers were present in areas without large deciduous trees and shrub cover. They appeared to be habitat generalists, not strongly selecting for particular habitat features in the study area. Theoretical models of habitat occupancy predict that the range of habitats occupied by a species will increase with increasing density (Brown 1969, Fretwell and Lucas 1969). Less competitive flycatching individuals may have been forced into poorer habitats by intraspecific competition. In other areas, where their densities are lower, Dusky Flycatchers may not act as habitat generalists. Habitat models were unable to always predict the presence or absence of birds because: 1) suitable habitat was not occupied (i.e. the species was living below capacity in the study area-unsaturated); 2) relevant vegetation features that could further distinguish between occupied and unoccupied sites were not measured; or 3) additional factors/processes (other than vegetation characteristics) were influencing their habitat decisions (Fig. 9). Habitat selection involves a series of decisions. Theoretical models of habitat selection assume that species have a good knowledge of their environment and can assess risk (Fretwell and Lucas 1970, Pulliam and Danielson 1991). In this chapter, I only assessed the potential of local habitat characteristics to explain the variation in bird presence noted in Chapter 2 and the variation in the habitat selectivity of bird species. However, predation can also influence habitat selection. Chipping Sparrows and Warbling Vireos, species which showed high habitat selectivity, both decreased in habitats, the control and herbicide-treated areas respectively, where there probability of nest survival was lower (Chapter 2). In Chapter 4,1 investigate the characteristics of nesting habitats including those of successful nests. CD O tu o P « Disl U CO <u tu ^ 1 JS > ristic & , <«_, CD to 2 ° O O 6 00 OJ e3 LH f— ° y BJ O > i o s u u_t CO 1 CD »-< ft 1 w -soi crj O S 0 0 OS i—i co tu o tu +-» O o tu d o H-» o SO 03 X 3 H-» o c3 cd -+-» o O H CCS H—» c/> CU to CO <U o o l-l O H tu H O S en 44 C H A P T E R FOUR: T H E R O L E OF DECIDUOUS V E G E T A T I O N IN N E S T H A B I T A T SELECTION A N D N E S T SUCCESS OF BIRDS INTRODUCTION Songbirds in fragmented forest landscapes may have lower reproductive success than birds in contiguous forest due to lower pairing success (Gibbs and Faaborg 1990, Villard et al. 1993), cowbird parasitism, and nest depredation (Robinson et al. 1995). In many areas, the rate of timber removal is faster than forest regeneration (Bunnell and Kremsater 1990), creating a landscape with a higher frequency of young to mid-seral forest (O'Hara et al. 1994). Although studies have measured patterns of passerine presence in landscapes altered by forest management (Schmiegelow et al. 1997, McGarigal and McComb 1995), there are few studies on natural, open-cup nests in young, regenerating forests (Hanski et al. 1996). Habitat manipulation continues in these young, regenerating forests where, through vegetation management, deciduous vegetation is selectively removed (MacKinnon and Freedman 1993). Nesting success was 12-46% in my study area and may be further exacerbated by vegetation management (Chapter 2). The availability of suitable nest sites can influence the distribution of open-cup nesting species (Martin and Roper 1988, Steele 1993). Although they will nest in both habitats, Pied Flycatchers (Ficedula hypoleucd) preferred to nest in deciduous than coniferous habitats (Lundberg et al. 1981). In southern Alaska and northern British Columbia, there is a higher concentration of nests in the deciduous than coniferous understories (Willson and Comet 1996). With equal search time spent in each area, I found more nests in the control compared with the treated areas (Table 2). As patterns of habitat use often reflect nest site requirements (MacKenzie et al. 1982, Steele 1993, Willson and Comet 1996), I predicted that those bird species preferring to nest in deciduous vegetation will select nesting habitat with more deciduous trees regardless of the treatment. 45 Predation is the most frequent cause of nest failure of open-nesting passerines breeding in temperate regions (Chapter 2, Ricklefs 1969, Martin 1992, Hanski et al. 1996). Birds choose nest site characteristics that reduce the risk of predation (Martin and Roper 1988, Marzluff 1988). Artificial nests are depredated more frequently in coniferous than deciduous forest (Sieving and Willson 1997). I proposed that nesting success is greater in sites with a higher density of deciduous trees. Nesting habitat may be measured at two spatial scales: the nest site (nesting substrate, position, and concealment of nest); and nest patch (habitat surrounding the nest site) (Martin and Roper 1988). To minimize depredation rates, I hypothesized that birds should select nest sites with greater concealment (see review Martin 1992), and closer to the next clump of vegetation (Kelly 1993). I also expect that birds will select nest patches: with greater plant density (Conner et al. 1986, Norment 1993); with a higher number of potential nest sites (Kelly 1993, Martin 1993); and away from edges (although some species show no avoidance, see review Paton 1994). Higher densities of deciduous vegetation within conifer plantations should increase many of these nesting habitat characteristics. This relationship should be stronger for Warbling Vireos, a deciduous specialist (Chapter 2 and 3). In 1994 and 1995,1 studied the nesting habitat of open-cup nesting species in areas treated by vegetation management, and control sites. I predicted that the following vegetation measures would be greater in habitat with nests than without nests, and in habitat with successful nests compared with unsuccessful nests: species richness of trees; clumping of vegetation; density of deciduous vegetation; and the amount of foliage cover above, and around the nest. To avoid detection from predators, nests should be placed at the side of the shrub/tree facing the nearest clump of vegetation, and at a height that is similar to the height of surrounding foliage with the highest density. I compared the nesting habitats of five species, Dusky Flycatcher, Warbling 46 Vireo, Swainson's Thrush, Chipping Sparrow, and American Robin. They are all open-cup, tree nesting species with nests that were most commonly found. METHODS Description of nest site and patch At each nest site, I recorded the species and height of the tree or shrub in which the nest was placed. The height of the nest was also noted. I classed the amount of vegetation cover directly above the nest as: 1 (0-33% cover); 2 (34-66%); or 3 (67-100%). In 1995,1 used the same classifications as above to estimate the amount of vegetation cover around the nest, from a distance of 1 metre, in the four cardinal directions. I described the spatial characteristics of the nest patch in 1995 by recording the following categorical variables (yes/no) within a 5m radius of the nest: i f the nest was located in a clump of trees (tree clump); if the nest was at an edge of a main or old logging road, or opening (edge); i f the nest was placed at the side of shrub/tree facing the nearest clump of vegetation (aspect); and i f the nest was at the same height as the surrounding foliage with the highest density (height foliage). To characterize the species composition and density of the vegetation of the nest patch, I recorded vegetation measures within a 5m radius of the nest. I counted the number of trees by species and size class (<5cm dbh, >5cm dbh), and estimated the percentage of shrub cover. In 1995,1 recorded the species composition of shrubs, and their proportion of cover. The nearest permanent vegetation plot (either the north, south, east, or west plot, located at each bird station) to the nest was used to compare nest habitat with non-nest habitat (see Chapter 3 for description). The permanent vegetation plots were larger (100.0 m ) than the plots 2 describing the nest patch (78.5 m ). Therefore, vegetation measures in the non-nest habitat were scaled (i.e. number of stems/78.5 m ) so that they were comparable with the nest habitat. 47 Classification of nests To test hypotheses on the selection of nest habitat, I used only those open-cup nests that were thought to have been constructed that year, whether they were found active or inactive. I considered inactive nests to be of the present year only if the nest materials appeared fresh, and the nest did not contain cobwebs, dead leaves, or other old vegetative material. Inactive nests were classed as (a) successful if nests were very worn and had feather shaft debris, (b) unsuccessful i f nests had signs of depredation or abandonment (i.e. egg shell or holes in nest, eggs or dead nestlings), or (c) uncertain for nests without such signs and wear. For the comparison of successful nesting habitats, I did two comparisons. First, I compared unsuccessful with successful nests, second, I added the uncertain nests to unsuccessful nests and compared them with successful nests. Nests known to be successful were worn and usually contained feather shaft debris {personal observation), so nests classed as uncertain were most likely unsuccessful. I may have overestimated failed nesting attempts, as some birds construct nests that they never use (Ehrlich et al. 1988), and successful nests may be falling apart and were considered old. However, this bias should have been equal across the study. Chapter 2 describes methods for nest searching and monitoring of active nests. Data analyses I compared the discrete measures of the nest site and patch using the log-likelihood ratio test. I used logistic regression (Hosmer and Lemeshow 1989) to evaluate which habitat variables (of the reduced variable set used in Chapter 3) were best correlated with nest patch selection for all nests. See Chapter 3 for details on logistic regression models. The models were produced by backward and stepwise methods (SAS Institute Inc. 1989). Habitat models, comparing the difference between nest patches and non-nest habitat, were calculated using all nests in the: 1) study area; 2) in control areas; 3) in manually thinned areas; and 4) in herbicide-treated areas. I 48 compared the habitat differences between all successful and unsuccessful nests (the sample size was too low to compare the control and treatments individually). Finally, I compared the nest patches of the five bird species individually with non-nest habitat. RESULTS I monitored 184 nests (102 active, 82 inactive) of 13 open-nesting species. Nests of Dusky Flycatchers, Warbling Vireos, Swainson's Thrushes, Chipping Sparrows, and American Robins made up 82% of all nests found. Of the 82 inactive nests, 31 were classed as unsuccessful, and three as successful. The outcome of the remaining 47 nests was uncertain. Nest site Dusky Flycatchers nested predominantly (39/43 nests) in deciduous trees, while Warbling Vireos did so exclusively (10/10). Swainson's Thrushes nested in deciduous trees about half of the time (14/30), American Robins less then half (11/45). Chipping Sparrows were predominantly conifer nesters (23/24). Nests were often placed on the side of the shrub/tree closest to the nearest clump of vegetation, sheltered from above by high vegetation cover, and were nearly always at a height similar to the surrounding foliage of the nest patch (Table 5). These three trends were similar across the control and treated areas (Table 6). Although birds were more likely to avoid placing their nests near an edge (Table 5), nests were more likely to be placed at an edge in the control area than in the treated areas (Table 6). Birds did not prefer to place their nests in a clump of trees (Table 5). 49 TABLE 5. The number of all nests, and successful nests with certain nest site characteristics. See methods section for explanation of the habitat variables. Relationships that are not tested (NT) are noted. Significant values are noted by: *=p<0.05; **=p<0.01. Habitat variable n All nests no. nests (%) G statistic (df) Successful nests no. nests (%) G statistic (df) tree clump 112 56 (50) NT 9(16) 1.38 edge 112 41 (37) 8.13** (1) 11(27) 1.53 aspect 93 61 (66) 9.18** (1) 10(16) 0.97 height foliage 111 105 (95) 1 NT 22 (21) 1 0.07 above cover <33% 124 15(12) 1(7) above cover 33-66% 115 27 (23) r 43.36** (2) 8(30) r 3.51 (2) above cover >66% 113 70 (62) J 14 (20) J TABLE 6. Difference in the number of nests with certain nest site characteristics between the control (control), manually thinned (manual), and herbicide-treated (herbicide) areas. See methods section for explanation of the habitat variables. Relationships not tested (NT) are noted. Significant values are noted by: *=p<0.05; **=p<0.01. control manual herbicide G statistic Habitat Variable n no. nests (%) n no. nests (%) n no. nests (%) (df) tree clump 44 23 (52) 29 13 (45) 41 20 (49) 0.39 (2) edge 44 23 (52) 29 9(31) 40 10(25) 7.31** (2) aspect 39 28 (72) 23 14(61) 32 20 (63) 1.04 (2) height foliage 44 42 (95) 27 26 (96) 40 40(100) NT above cover <33% 44 7(16) 29 6(21) 41 3(7) 1 above cover 33-66% 44 12 (27) 29 7(24) 41 8(20) (•4.08 (4) above cover >66% 44 26 (59) 29 16(55) 41 30 (73) J Nest patch Nests were more likely to be found in areas with more deciduous trees and coniferous stems, greater shrub cover, and fewer large (>5 cm dbh) coniferous stems (Table 7). Despite treatment-level differences (Chapter 2-3), the presence of deciduous trees/stems and shrub cover was a good predictor of the presence of nests in the control and treated areas (Table 7). The presence of nests was also positively correlated to the number of Douglas-fir stems in the control areas, and to the number of coniferous stems in the manually-thinned areas. Nests of deciduous nesters, Dusky Flycatchers and Warbling Vireos, were positively related to the number of deciduous trees (Table 8). However, Dusky Flycatchers preferred sites where the tree strata was more open (i.e. drier) and the shrub cover greater. 50 Swainson's Thrushes preferred to nest in areas with a high density of deciduous and coniferous stems (Table 8). This type of habitat is not typically maintained by glyphosate application. In contrast, Chipping Sparrows nested in patches with many lodgepole pine and Douglas-fir stems. Finally, I could not produce a nest patch selection model for American Robins because either the sample size (n=45) was too small to detect a relationship, or there was no clear relationship (i.e. characteristics of nest patch for robins had a high variance). Habitat of successful nests The outcome of a nest was independent of its placement in a clump of trees, edge, or at a height similar with surrounding foliage in the patch (Table 6). The outcome was also independent of whether the nest was placed at the side of shrub/tree facing the nearest clump of shrub/trees, and the amount of vegetation cover above the nest (Table 6). For the individual five bird species, there were no differences in the nest site characteristics of successful compared with unsuccessful nests. Overall, successful nests were positively related to the presence of willow stems (Table 7). The relationship between successful nests and the habitat was similar for both comparisons (successful versus unsuccessful nests, successful versus unsuccessful plus uncertain nests). Including all of the nests provided more power for the test. 45 o ft 3 a u cj s OJ 45 T 3 3 co CO CJ -o 3 CO CO JJ o . a "5 CJ cj 43 cj o -a © V CJ 45 cj to ex CJ 0 6 1 s CO • CO O £ v CJ II a o c T 3 3 CO co CJ 45 CJ CS / — N & J2 .2 £ 60 , O ' T 3 CJ CO _ra OJ o 3 u OJ C 3 CD CD CJ 43 ca > c ca CJ eg ' 3 60 T 3 CJ • J - ^ J CJ 43 ,g CJ o 3 CJ CJ 43, 60 .3 ca cj i s • CJ '5 CJ 43 T3 3 ft CJ • T 3 CO T J T3 O ca CJ ca •O cj C .3 43 CJ o o 3 CO 3 3 .2 3 c S § p 2 B 3 00 — CJ cr-»- o « § '60 O !—1 ca . CJ W —1 P Q cj CJ CJ 3 CO 3 OJ CJ £ CJ 4 3 43 ca 43 3 cj 15 S -3 i l co • -OJ S o C<-I O C*H T 3 T 3 1 O 60 O O 43 J 13 CN M N CJ •3 ca 3 CO 5 &> -*-» t o H "ca 00 CJ cj CJ ca l | is <f> > SO cn vo o i n i n o t-OS so CN cn 00 O 0 O SO CN r-o CN SO o m CN >n OS SO CN 0 00 OS i n SO _ _ Os i n in so 0 0 O SO O 00 in 00 00 cn •a- CN 00 in in i n OS 0 SO i—t CN CN CN CN CN m m so cn 0 0 © © 0 0 O O d d d 0 0 * * * * * * # * * # * O OS i n r~ CN OS •a-00 cn 00 i n r-; Os d i n 00 i—1 o cn 00 ^ V S O 1^ * * so t-; * * OS * i n 06** * * cn p i n CN o r-i n C N 00 OS * * ^ 00 °° os g r- m o so 1 = 5 r— *a-c*l p ""I p p T 1 d P 0 d OS 0 O —1 CN J5 so O cn CN i n CN s O Csl O "T* d d d £ : c- so os ?? cn CN 1—1 ° > —1 so m n o o o d d d 00 i n so 00 CN CN SO CN m — o d d OS 1-5 "1 o 43 43 O m A W Q O w Q O w Q O W Q O w Q O CJ CJ is CO 3 O 3 3 1 IH T 3 cj 3 O .3 3 6 cj CO CO O S 3 O t. LH CJ c 2 > • 3 O 3 o 8 x> CO cj is ^ C? CO IH 0 \ 3 I s _ -B B CJ & S i E ft^ 3 © r r o K cj O -3 o !T " o -2 -S Q •§ cj . . P O 3 co S2 ? 3 2 v ? o s c 2 2 & . 3 3 3 co CJ CJ is CO 3 o 3 CJ .3 3 s ? o s 00 m SO CJ CJ « 3 CO CO CJ .3 i n * J 00 S i 3 I  H CJ 3 er1 CO W ft 2 o ^ • 3 3 co 8 cj o co 3 o > i f i n >S -a-a ca 0 - 3 co 3 CJ S co 3 A ^ a — CO OJ CJ 3 ft I S • • -P 3  3 > CO H CJ CJ 3 45 ft -a CJ .3 c CO ^—^ cj cn o\ a 11 os — s 11 3 3 2 o 3 C CJ co 3 r-* CO Lo .3 > o 00 SO 00 >n 06 so 00 so Os CN O * * * * ™ 00 55 ON cn m CN i n SO 0 0 OS i n ! O m i n i n cn i n so i n 00 SO i n CN 00 SO m OS so 0 0 7—< 0 t - H 0 OS 0 1—1 0 CN m r - H so CN m CN cn p p p p •a; p 0 0 00 0 0 p i n p O CN p CN p d d d d d d d d d 0 d 0 d d 0 d d O 0 d ••a" so OS £4 os 0 W Q O cj pi CJ •3 c cj rn 3 "• — 3 3 •3 ^ ft.3 cj u o 3 co ra so t -a-u o 3 3 a tU to a .o < 11 '£ p 2 "5 to co g ° a T 3 •a IcN Jg I'E S 135 w c/3 tu a> l | •8 > o 1/3 SO SO so o oo o so in CN •n o oo OS OS oo CO CO o CN r-in co CN CN OS CN CN "3- in r- OS SO m r-- o o 00 r~ oo oo CO m •*r OS OS o oo OS ro Os SO CN co CN >n so *o CN CN CN (N d o o d d © © © d d d * * » * * * * # * * # # * * * * * * CN OS CN \0 OS oo CN o SO oo ro OS CN CO SO OS SO o <r~-00 d od OS ro d SO so in CN ro CN in oo oo oo O CN so o d d O \c o) © T d o tu D. SO OS —< OS TJ- OS o o o o o o d d SO 00 CN O -H CN m en © —< © o d d >n in co co O r- o d d CN vo CN Co 3 ^ o T d oo as oo o OS OS OS o T J - o m —• sq © © © d o d o ^ CN OS T f ~ *f m ? ° TI- in C N CN o — O *•? © © © OS o SO 00 CN ro p d © * * o £ 1< ^ oo so CN m o p © d oo p-r- ro o —' W Q O w Q O w Q O o CU w Q O 3 o <u T 3 a, 3 o a to U to .15 <o to tu -O o °o .3 C cu i2 s (L> CO a s S3 «J b s P CO O -> o o * i o ^ <L> c tu io u Q to _ o B 43 to 3 O t i l tu to to 3 o to os 3 ' O l H O 3 tu C o 3 to 3 O 3 12 ' u tu T 3 fc- to tu .3 a to 6 £ to S " s *5 tU tu w c w •a 2 c3 1 -2 Q 2 d d G. a a to ^O CO o" si to O ro o II ^—^ 2 II ^—^ <u a CO a ro .a ^ — - SO so > II CO •*-» a II 60 ese a -fl ese a .3 ese O CO ese ~ — ' -i-» !-• 1 O .3 OH o a Ctj H-» a to to tu to tu to a > C/3 a > ro so ft H-> oo a || oo 8 a .3 — "B. a. o & *- a '3 to S3 tu to U a > 53 DISCUSSION Across the study, nesting habitat was located in areas with more deciduous vegetation. In contrast to manually thinned areas, nesting habitat in the herbicide-treated areas was typical of vegetation that escaped treatment. Most of the species selected specific nest sites, typical of the local habitat they occupied (Chapter 3). Successful nests were associated with more willows. Nesting habitat Despite variation in associations of bird presence with deciduous trees (Chapters 2-3), nest patches were always positively correlated with deciduous vegetation. Nests in herbicide-treated areas were associated with habitat that escaped herbicide treatment. Nests in the manually thinned areas were positively related to deciduous stems, implying that thinning in wetter areas (higher number of sprouts and therefore stems), was not inhibitory to nesting. Overall, nests were positively related to stem density, implying that the clumping of trees at a scale larger than the nest patch influenced nest patch selection. Many silviculture practices discourage the clumping of trees (Hansen et al. 1991). The nest patches of Chipping Sparrows, Swainson's Thrushes, and Warbling Vireos were typical of the local habitats they occupied (Chapter 3). Birds select patches that have a large number of potential nest sites (Martin and Roper 1988). Chipping Sparrows preferred nesting in habitat created by herbicide spraying, and drier areas in the manually thinned areas. Swainson's Thrushes nested in dense habitat, typical of productive and/or naturally regenerated sites. Warbling Vireos preferred nesting in habitat with more deciduous trees (tree implies deciduous vegetation had escaped treatment). In contrast, Dusky Flycatchers were more selective in the habitat they nested in than in the habitat they occupied (Chapter 3). This may indicate that the population had non-breeding, "floating" individuals (Gibbs and Faaborg 1990, Villard et al. 54 1993). Dusky Flycatchers avoided nesting in habitat that was densely treed, choosing drier areas with deciduous trees. American Robins did not appear to be selective in their nesting habitat. Successful nests Although nests generally had greater foliage cover, I found no difference in these characteristics between successful and unsuccessful nests. This contrasts with other findings where higher nest concealment (see review Martin 1992), and nest presence in a clump of trees (Kelly 1993), were characteristics correlated with successful nests. Although I may have not had enough power to detect a difference, other researchers have also failed to find a difference in nest concealment between successful and unsuccessful nests (Hanski et al. 1996). This may indicate that olfactory, rather than visual, clues are more important to predators at my site, which supported many mammals (Runciman and Sullivan 1996) and few corvids (Table 1). As in other studies (Hanski et al. 1996, Rudnicky and Hunter 1993), I found no influence of edge on nesting success. However, most of my edge habitat was between the plantation forest and active or non-active logging roads. These edges had low contrast between habitats, and probably supported similar predator assemblages, than edges between forest-agriculture lands or mature forest-young plantations. Successful nests may have more likely to be in willow for several reasons. First, willows may have inherent properties, that make it more difficult for predators to detect nests. Despite similarities in nest predation, Prothonotary Warblers (Protonotaria citred) prefer to nest in flooded habitat, where willows are a dominant species, than dry habitat (Petit and Petit 1996). However, these authors did observe a change in predator species from dry to flooded areas. Second, willows are commonly found in wetter sites where nesting success may be greater. Higher nesting success of Prothonotary Warblers in flooded, compared with dry areas, was 55 probably due to greater abundances of food, particularly mayflies, dipterans, and spiders (Petit and Petit 1996). Males in flooded areas made more feeding trips but spent less time foraging. In riparian areas, birds also prefer to nest in willows. Western and Eastern Kingbirds (Tyrannus verticalis and T. tyrannus) nest most frequently in peach-leaved willow (Salix amygdaloides) and green ash (Fraxinus pennsylvanicd) in southern Manitoba (MacKenzie and Sealy 1981). Nest sites of Willow Flycatchers (Empidonax traillii) were most most frequently associated with high willow density (Sedgwick and Knopf 1994). It is possible to maintain all of described nesting habitats after vegetation management. However, herbicide application plus planting practices tend to homogenize the habitat, maintaining fewer nesting sites. Patches of willow within a target area of vegetation management should be maintained i f possible. 56 C H A P T E R FIVE: CONCLUSIONS MANAGEMENT IMPLICATIONS The prevailing idea about how best to maintain biodiversity in temperate forest regions, is to promote the diversity of vegetation structure and composition in intensively managed landscapes (Franklin 1988). Although methods to maintain and develop stands diverse in structure and species composition are described in silviculture texts (Smith 1986), they are not widely practiced (Hansen et al. 1991). As early successional habitat becomes increasingly prevalent in our forested landscapes, it is still largely ignored as an area for conserving biological diversity (Franklin 1993). I found that promoting a more homogeneous forested habitat by manual thinning plus herbicide application encouraged the proliferation of common species, particularly those residents and short-distance migrants that were also ground gleaners and conifer nesters. Some deciduous nesting and foliage gleaning species declined in abundance in areas treated with herbicide. As a deciduous specialist, Warbling Vireos were particularly vulnerable to the loss of deciduous trees after glyphosate application. In addition, their nests (3 nests/7 nests found), along with those of Chipping Sparrows (2/10) and Dusky Flycatchers (2/26), were sometimes parasitized by Brown-headed Cowbirds. * Warbling Vireos, Swainson's Thrushes, and Dusky Flycatchers (three of four common birds tested) preferred to nest in untreated or recovered habitat within the larger treated area. Successful nests were generally absent in areas without willow. I recommend that patches of untreated areas are maintained within a larger target area for vegetation management, particularly for herbicide treatments. These untreated sections should include some productive sites and more willow stems. The flexibility of species in their habitat choices will determine their vulnerability to habitat manipulations. Foliage gleaners and deciduous tree nesters appeared to have the 57 flexibility to respond to the subsequent recovery of deciduous trees in the manually thinned areas during the second and third post-treatment year. The general impact of vegetation management on bird communities will clearly depend on the species and the population dynamics of birds present, the intensity and extent of treatment, and the landscape context of the treated area. Manual thinning increased the abundance of common and moderately common species already present on the sites. Herbicide-treated areas experience a net loss of species, a higher turnover of species, and an increase in species dominance. Glyphosate application may encourage a greater loss of community structure in the future as these sites progress in different successional trajectories (MacKinnon and Freedman 1993). I recommend that on a landscape scale, methods of vegetation management be varied. Presently, data on the long-term effects of vegetation management on birds on a large spatial scale are negligible (Enns et al. 1993). Very low nesting success coupled with the increase in abundance of birds suggest that herbicide-treated areas may be acting as "sink" habitats for some species of birds (eg. Dusky Flycatchers). The viability of such populations would be susceptible to fluctuations in the source populations and be linked to events in the surrounding landscape. I recommend that particular attention is paid to some of the more vulnerable or rare bird species expected to be present in early successional forests. This may include those wetland/marsh species that inhabit such patches within forested areas. In my study area, forest management often removed trees from such sites (personal observation). In the Interior Cedar-Hemlock biogeoclimatic zone, I would maintain habitat for Alder and Willow Flycatchers, Nashville Warblers, and Warbling Vireos. I also would also test the impact of vegetation on nesting success with controlled, replicated experiments. 58 The effect of vegetation management on the growth of conifers is unclear. Maintaining black cottonwoods has positive effects on the growth of western redcedars in southwestern British Columbia (Burton 1993). Three years after treatment, Douglas-fir saplings do not grow more on sites where deciduous vegetation is removed by manual thinning or glyphosate application than control sites (Karakatsoulis et al. 1989). Still other studies in the Pacific Northwest show that conifers grow best where deciduous vegetation is minimized (Walstad and Kuch 1987). Results on the effects of vegetation management on conifer growth are inconclusive and prohibit a single approach (eg. removal of deciduous vegetation) to increase timber yields. Hardwood competition may not be the most influential factor affecting the growth of conifers (Weldon and Slauson 1986). There are many beneficial effects of deciduous vegetation: captures, stores, and recycles considerable quantities of micro- and macronutrients from the ecosystem that may otherwise be lost; adds organic matter and nitrogen to the soil; prevents soil erosion; interferes with the movement of insects; and protects conifer seedling from browsing animals (Lavender et al. 1990). I recommend exploration of non-traditional methods of vegetation management. For example, hardwoods can act as nurse trees, inhibiting the growth of deciduous shrubs and encouraging the regeneration of shade-tolerant conifers (McLennan and Klinka 1990). The presence of bird species in forests exposed to different harvest regimes varies (Hansen et al. 1995). It is unrealistic that a single management goal plan will achieve the necessary habitat characteristics for all species. Unless specific goals are desired, a range of habitat composition and structure should be maintained across the landscape. Finally, the changes I observed were not apparent immediately following treatment. Rather, some shifts in the community composition (i.e. the increase in conifer nesters) were only evident three years after treatment. Research on the effects habitat disturbances need to extend for 59 periods longer than one or two years. The management of hardwoods in western North America is a fairly recent event and research on its effects on wildlife in British Columbia is restricted (Enns et al. 1993). There is a major opportunity for ecologists to use adaptive management ideals to measure large-scale responses of vegetation and wildlife to vegetation management (Walters and Holling 1990). LITERATURE CITED Biring, B.S., P.G. Comeau, J.O. Boateng. 1996. Effectiveness of forest vegetation control methods in British Columbia. Canada-British Columbia Partnership Agreement on Forest Resource Development: FRDA Handbook 011. Blancher, P.J., R.J. Robertson. 1984. Resource use by sympatic kingbirds. Condor 86: 305-313. Brawn, J.D., S.K. Robinson. 1996. Source-sink population dynamics may complicate the interpretation of long-term census data. Ecology 77: 3-12. Brown, J.L. 1969. Territorial behaviour and population regulation in birds. Wilson Bulletin 81: 293-329. Bunnell, F., L . L . Kremsater. 1990. Sustaining wildlife in managed forests. The Northwest Environmental Journal 6: 243-269. Burton, P.J. 1993. Differential conifer establishment in pure and mixed neighbourhoods of early successional post-logging vegetation. Proceedings of the Integrated Forest Vegetation Management: Options and Applications Workshop. F R D A . Campbell, R . A . 1990. Herbicide use for forest management in Canada: Where we are and where we are going? Forestry Chronicles 8: 355-360. Cannings, R.A., R.J. Cannings, S.G. Cannings. 1987. Birds of the Okanagan Valley. British Columbia Royal British Museum, Victoria. Collins, S.L. 1983. Geographic variation in habitat structure for the wood warblers in Maine and Minnesota. Oecologia 59: 246-252. Conner, R.N. , M.E . Anderson, J.G. Dickson. 1986. Relationships among territory size, habitat, song, and nesting success of Northern Cardinals. Auk 103: 23-31. Day, R.W., G.P. Quinn. 1989. Comparisons of treatments after an analysis of variance in ecology. Ecological Monographs 59: 433-463. Ehrlich, P.R., D.S. Dobkin, D. Wheye. 1988. The Birder's Handbook: A Field Guide to the Natural History of North American Birds. Simon & Schuster Inc., Toronto. 60 Enns, K . A . , E.B. Peterson, D. McLennan. 1993. Impacts of hardwood managment on British Columbia wildlife: problem analysis. B.C. Ministry of Forest and Forest Canada, Victoria, B.C. F R D A Report No. 208. Franklin, J.F., R.T. Forman 1987. Creating landscape patterns by forest cutting: ecological consequences and principles. Landscape Ecology 1:5-18. Franklin J.F., T. Spies, D. Perry, M . Harmon, A. McKee. 1986. Modifying Douglas-fir management regimes for nontimber objectives. Pages 373-379 in C D . Oliver, D.P. Hanley, and J. A . Johnson, editors. Modifying Douglas-fir Management Regimes for Non-timber Objectives. U S D A Forest Service, Seattle, Washington. Freedman, B., R. Morash, D. MacKinnon. 1993. Short-term changes in vegetation after the silvicultural spraying of glyphosate herbicide onto regenerating clearcuts in Nova Scotia, Canada. Canadian of Journal Forest Research 12: 2300-2311. Freemark, K .E . , H.G. Merriam. 1986. Importance of area and habitat heterogeneity to bird assemblages in temperated forest fragments. Biological Conservation 36: 115-141. Fretwell, S.D., H.L. Lucas. 1970. On territorial behaviour and other factors influencing habitat distribution in birds. Acta Biotheoretica 19: 16-36. Gibbs, J.P., J. Faaborg. 1990. Estimating the viability of Ovenbird and Kentucky Warbler populations in forest fragments. Conservation Biology 4: 193-196. Green, R.H. 1993. Application of repeated measures designs in environmental impact and monitoring studies. Australian Journal of Ecology 18: 81-98. Grossbard, E., D. Atkinson. 1985. The Herbicide Glyphosate. Butterworths. London. U .K . Hagan, J .M. I l l , D.W. Johnston, editors. 1992. Ecological and Conservation of Neotropical Migrant Landbirds. Smithsonian Institution Press, Washington. Hagan, J.M., W . M . Vander Haegen, P.S. McKinley. 1996. The early development of forest fragmentation effects on birds. Conservation Biology 10: 188-202. Halpern, C.B., T.A. Spies. 1995. Plant species diversity in natural and managed forests of the pacific northwest. Ecological Applications 5: 913-934. Hanski, I.K., T.J. Fenske, G.J. Niemi. 1996. Lack of edge effect in nesting success of breeding birds in managed forest landscapes. Auk 113: 578-585. Hansen, A.J . , T.A. Spies, E.J. Swanson, J.L. Ohmann. 1991. Lessons from natural forests: implications for conserving biodiversity in natural forests. Bioscience 41: 382-392. Hansen, A.J . , W.C. McComb, R. Vega, M.G. Raphael, M . Hunter. 1995. Bird habitat relationships in natural and managed forests in the west cascades of Oregon. Ecological Applications 5: 555-569. 61 Hicks, C.R. 1993. Fundamental Concepts in the Design of Experiments. Saunders College Publishing, New York, N Y . Holt, R.F., K . Martin. 1997. Landscape modification and patch selection: The demography of two secondary cavity-nesting species colonizing clear-cuts. Auk. In press. Hosmer, D.W., S. Lemeshow. 1989. Applied Logistic Regression. John Wiley and Sons, New York, New York. Jackson, D.A. 1993. Stopping rules in principal components analysis: A comparison of heuristical and statistical approaches. Ecology 74: 2204-2214. James, F.C., N.O. Warner. 1982. Relationships between temperate forest bird communities and vegetation structure. Ecology 63: 159-171. Karakatsoulis, J., J.P. Kimmins, R.E. Bigley. 1989. Comparison of the effects of chemical (Glyphosate) and manual conifer release on conifer seedlings physiology and growth on Vedder Mountain, British Columbia. Proceedings of the Carnation Creek Herbicide Workshop. F R D A Report No. 063. Karr, J.R., K . E . Freemark. 1983. Habitat selection and environmental gradients: dynamics in the "stable" tropics. Ecology 64: 1481-1494. Kelly, J.P. 1993. The effect of nest predation on habitat selection by Dusky Flycatchers in limber pine-juniper woodland. Condor 95: 83-93. Ketcheson, M . V . , T.F. Braumandl, D. Meidinger, G. Utzig, D.A. Demarchi, B . M . Wikeem. 1991. Pages in 167-181 D. Meidinger, J. Pojar, editors. Interior Cedar-hemlock Zone. Ecosystems of British Columbia. British Columbia Ministry of Forests, Victoria. Kimmins, J.P., K . M . Tsze, R.E. Bigley. 1989. Development of non-commercial vegetation following clearcutting and clearcutting-plus-slashburning, and some biogeochemical effects of herbicidal control of this vegetation. Proceedings of the Carnation Creek Herbicide Workshop. Forest Resource Development Agreement Report 063. Forestry Canada. Klinka, K. , V.J . Krajina, A . Ceska, A . M . Scagel. 1989. Indicator Plants of Coastal British Columbia. University of British Columbia Press, Vancouver. Krebs, C.J. 1989. Ecological Methodology. Harper Collins Publishers, New York, New York. Kricher, J.C. 1973. Summer bird species diversity in relation to secondary succession on the New Jersey piedmont. American Midland Naturalist 89: 121-137. Lautenschlager, R.A. 1993. Response of wildlife to forest herbicide applications in northern coniferous ecosystems. Canadian Journal of Forest Research 23: 2286-2299. 62 Lavender, D.P., R. Parish, C M . Johnson, G. Montgomery, A . Vyse, R.A. Willis, D. Winston. 1990. Regenerating British Columbia's Forests. University of British Columbia, Vancouver. Lloyd, D.A., K . Angove, G. Hope, C. Thompson. 1990. A Guide for Site Identification and Interpretation of the Kamloops Forest Region. Two volumes. Land Management Handbook Number 23. British Columbia Ministry of Forests, Victoria. Lundberg, A. , R.V. Alatalo, A . Carlson, S. Ulfstrand. 1981. Biometry, habitat distribution, and breeding success in the Pied Flycatcher Ficedula hypoleuca. Ornis Scandinavica 12: 68-79. Lund-Hoie, K. , S. Gronvold. 1987. Glyphosate application in forest-ecological aspects: successional vegetation changes. Scandinavian Journal of Forest Research 2: 455-468. MacKenzie, D.I., S.G. Sealy. 1981. Nest site selection in Eastern and Western Kingbirds: a multivariate approach. Condor 83: 310-321. MacKenzie, D.I., S.G. Sealy, G.D. Sutherland. 1982. Nest-site characteristics of the avian community in the dune-ridge forest, Delta Marsh, Manitoba: a multivariate study. Can. J. Zool. 60: 2212-2223. MacKinnon, D.S., B. Freedman. 1993. Effects of silvicultural use of the herbicide glyphosate on breeding birds of regenerating clearcuts in Nova Scotia, Canada. Journal of Applied Ecology 30: 395-406. Magurran, A . E . 1988. Ecological Diversity and its Measurement. Princeton University Press, Princeton, New Jersey. 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. Ecological and Conservation of Neotropical Migrant Landbirds. Smithsonian Institution Press, Washington. Martin, T.E. 1993. Nest predation and nest sites. Bioscience 43: 523-532. Martin, T .E . , J.J. Roper. 1988. Nest predation and nest site selection of a western population of the Hermit Thrush. Condor 90: 51-57. Marzluff, J .M. 1988. Do Pinyon Jays alter nest placement based on prior experience? Animal Behaviour 36: 1-10. May, P.G. 1982. Secondary succession and breeding bird community structure: Patterns of resource utilization. Oecologia 55: 208-216. May, R . M . 1986. The search for patterns in the balance of nature: advances and retreats. Ecology 67: 1115-1126. Mayfield, H . 1975. Suggestions for calculating nest success. Wilson Bulletin 87: 456-466. 63 McGarigal, K. , W.C. McComb. 1995. Relationships between landscape structure and breeding birds in the Oregon Coast Range. Ecological Monographs 65: 235-260. McLennan, D.S., K . Klinka. 1990. Black cottonwood-a nurse species for regenerating western redcedar on brushy sites. B.C. Ministry of Forest and Forest Canada, Victoria, B.C. F R D A Report No. 114. Ministry of Forests. 1993. Annual Report of the Ministry of Forests for the Fiscal Year Ended March 1993. British Columbia Ministry of Forests, Victoria. Ministry of Forests. 1995. The Forest Practices Code of British Columbia Act. British Columbia Ministry of Forests, Victoria. Morrison, M . L . , E.C. Meslow. 1984. Effects of the herbicide glyphosate on bird community structure, western Oregon. Forest Sciences 30: 95-106. Nichols, S. 1977. On the interpretation of principal components analysis in ecological contexts. Vegetatio 34: 191-197. Norment, C.J. 1993. Nest-site characteristics and nest predation in Harris' Sparrows and White-crowned Sparrows in the Northwest Territories, Canada. Auk 110: 769-777. O'Connor, R.J. 1992. Population variation in relation to migrancy status in some North American birds. Pages 64-74 in in J .M. Hagan III and D.W. Johnston, editors. Ecological and Conservation of Neotropical Migrant Landbirds. Smithsonian Institution Press, Washington. O'Hara, K . L . , R.S. Seymour, S.D. Tesch, J .M. Guldin. 1994. Silviculture and our changing profession. Journal of Forestry 8-13. Paton, P.W. 1994. The effect of edge on avian nest success: How strong is the evidence? Conservation Biology 8: 17-26. Petit, L.J. , D.R. Petit. 1996. Factors governing habitat selection by Prothonotary Warblers: field tests of the Fretwell-Lucas models. Ecological Monographs 66: 367-387. Pimentel, D., L . McLaughlin, A . Zepp, B. Lakitan, T. Krause, P. Kleinman, F l Vancini, W.J. Roach, E. Graap, W.S. Keeton, G. Selig. 1991. Environmental and economic effects of reducing pesticide use. Bioscience 41: 402-409. Pulliam, H.R. 1988. Sources, sinks, and population regulation. American Naturalist. 132:652-661. Recher, H.F. 1969. Bird species diversity and habitat diversity in Australia and North America. American Naturalist 103: 75-80. Reynolds, R.T., J .M. Scott, R.A. Nussbaum. 1980. A variable circular-plot method for estimating bird numbers. Condor 82: 309-313. 64 Rice, J., B.W. Anderson, R.D. Ohmart. 1984. Comparison of the importance of different habitat attributes to avian community organization. Journal of Wildlife Management 48: 895-911. Rice, J., R.D. Ohmart, B.W. Anderson. 1983. Habitat selection attributes of an avian community: A discriminant analysis investigation. Ecological Monographs 53: 261-290. Ricklefs, R.E. 1969. An analysis of nesting mortality in birds. Smithsonian Contributions to Zoology 9:1-48. Robinson, S.K., F.R. Thompson III, T .M. Donovan, D.R. Whitehead, J. Faaborg. 1995. Regional forest fragmentation and the nesting success of migratory birds. Science 267:1987-1990. Rotenberry, J.T. 1978. Components of avian diversity along a multifactorial climatic gradient. Ecology 59: 693-699. Rotenberry, J.T., J.A. Wiens. 1980. Habitat structure, patchiness, and avian communities in North American steppe vegetation: a multivariate analysis. Ecology 61: 1228-1250. Rudnicky, T.C., M X . Hunter. 1993. Avian nest predation in clearcuts, forests, and edges in a forest-dominated landscape. Journal of Wildlife Management 57: 358-364. Runciman, J.B., T.P. Sullivan. 1996. Influence of alternative conifer release treatments on habitat structure and small mammal population in south central British Columbia. Canadian Journal of Forest Research 26: 2023-2034. Santillo, D.J., P.W. Brown, D . M . Leslie, Jr. 1989a. Response of songbirds to glyphosate-induced habitat changes in clearcuts. Journal of Wildlife Management 53: 64-71. SAS Institute Inc. 1989. SAS/STAT® User's Guide. Version 6.0, Fourth Edition, Volume 2, Cary, N C : SAS Institute Inc. Schmiegelow, F .K.A. , C.S. Machtans, S.J. Harmon. 1997. Are boreal birds resilient to forest fragmentation? A n experimental study of short-term community responses. Ecology. In press. Schwab, F.E. 1979. Effect of vegetation structure on breeding bird communities in the dry zone Douglas-fir forests of southeastern British Columbia. M.Sc. thesis. University of British Columbia, Vancouver. Sedgewick, J.A., F.L. Knopf. 1994. Describing Willow Flycatcher habitats: scale perspectives and gender differences. Condor 94: 720-733. Sieving, K .E . , M.F. Willson. 1997. Experimental assessment of avian nest predation in northwestern coniferous and deciduous forest understory. In Review. Ecology. Simpson, E.H. 1949. Measurement of diversity. Nature 163: 688. Slagsvold, T. 1977. Bird population changes after clearance of deciduous scrub. Biological Conservation 12: 229-244. 65 Smith, D . M . 1986. The Practice of Silviculture. John Wiley & Sons, New York. Steele, B .B. 1993. Selection of foraging and nesting sites by Black-throated Blue Warblers: their relative influence on habitat choice. Condor 95: 568-579. Sutton, R.F. 1978. Glyphosate herbicide: an assessment of forestry potential. Forestry Chronicle 54: 24-28. Titterington, R.W., H.S. Crawford, B .N. Burgason. 1979. Songbird responses to commercial clear-cutting in Maine spruce-fir forests. Journal of Wildlife Management 43: 602-609. Tramer, E.J. 1969. Bird species diversity: components of Shannon's formula. Ecology 50: 927-929. Van Home, B. 1983. Density as a misleading indicator of habitat quality. Journal of Wildlife Management 47: 893-901. Villard, M. -A . , P.R. Martin, C.G. Drummond. 1993. Habitat fragmentation and pairing success in the Ovenbird (Seiums aurocapillus). Auk 104: 759-768. Walstad, J.D., P.J. Kuch, editors. 1987. Forest Vegetation Management for Conifer Production. John Wiley & Sons, New York, N . Y . Walters, C.J., C S . Holling. 1990. Large-scale management experiments and learning by doing. Ecology 71: 2060-2068. Weldon, C.W., W.L. Slauson. 1986. The intensity of competition versus its importance: an overlooked distinction and some implications. Quarterly Review of Biology 61: 23-44. Werner, R.A. 1983. Biomass, density, and nutrient contrent of plant arthropods in the taiga of Alaska. Canadian Journal of Forest Research 13: 729-739. Whittaker, R.H. 1965. Dominance and diversity in land plant communities. Science 147: 250-260. Whittaker, R.H. 1972. Evolution and measurement of species diversity. Taxon 21: 213-251. Wiens, J.A. 1989a. The Ecology of Bird Communities. Volume 1: Processes and Variations. Cambridge University Press, Cambridge. Wiens, J.A. 1989b. The Ecology of Bird Communities. Volume 2: Processes and Variations. Cambridge University Press, Cambridge. Wiens, JA., J.T. Rotenberry, B. Van Home. 1987. Habitat occupancy patterns of North American shrubsteppe birds: the effects of spatial scale. Oikos 48: 132-147. Willson, M.F. 1974. Avian community organization and habitat structure. Ecology 55: 1017-1029. 66 Willson, M.F. , T.A. Comet. 1996. Bird communities of northern forests: ecological correlates of diversity and abundance in the understory. Condor 98: 350-362. Willson, M.F. , T.I. De Santo, C. Sabag, J.J. Armesto. 1994. Avian communities of fragmented south-temperate rainforests in Chile. Conservation Biology 8: 508-520. Zar, J.H. 1996. Biostatistical Analysis. New Jersey: Prentice-Hall Inc. APPENDIX 1 Physical Characteristics, Treatments and Census Intensity of Study Plots P l o t A B C D E F G H I S i t e S i c a m o u s S i c a m o u s S i c a m o u s E a g l e E a g l e E a g l e E a g l e E a g l e E a g l e B a y B a y B a y B a y B a y B a y A s p e c t E a s t S o u t h E a s t N o r t h N o r t h N o r t h N o r t h N o r t h N o r t h E l e v a t i o n 7 0 5 m 1 0 3 0 m 1 1 4 3 m 7 8 5 m 7 6 0 m 6 9 0 m 6 1 0 m 5 8 0 m 8 0 0 m A r e a 2 6 h a 3 8 h a 4 7 h a 2 2 h a 2 3 h a 2 3 h a 3 6 h a 2 2 h a 2 5 h a L o g g e d 1 9 8 5 - 8 8 1 9 8 3 1 9 7 7 - 8 1 1 9 8 3 1 9 7 9 - 8 2 1 9 7 8 1 9 8 4 - 8 5 1 9 8 6 - 8 8 1 9 8 3 P r e p a r a t i o n 3 2 1 1-3 1-2 1-2 2 , 4 2 2 1-2 P l a n t e d 1 9 8 9 1 9 8 7 1 9 8 4 - 8 6 1 9 8 5 1 9 8 6 1 9 8 5 - 8 7 1 9 8 7 1 9 8 7 - 9 0 1 9 8 7 - 8 8 S p e c i e s F d , P I P I , F d , S F d , S L w P I P I , S P I , L w , S , P I , F d , S P 1 , S P l a n t e d b F d T r e a t m e n t C o n t r o l M a n u a l H e r b i c i d e M a n u a l M a n u a l C o n t r o l C o n t r o l H e r b i c i d e H e r b i c i d e # S t a t i o n s 6 7 7 6 6 5 6 5 3 Plots were preparedfor planting by l=broadcast burn, 2=rough bunch and burn,3=windrow and burn, and 4=rough bunch only. bSpecies planted were Fd=Douglas-fir, Pl=lodgepole pine, S=hybrid spruce (Picea engelmannii x P. glauca), Lw=western larch (Larix occidentalis). 68 APPENDIX 2 Bird species recorded at permanent stations, 1992-1995 (n=51). The seven common species are in bold. Birds are classified by migratory status, nest and forage guilds (Ehrlich et al. 1988, Cannings et al. 1987). G U I L D S S c i e n t i f i c N a m e C o m m o n N a m e A b b r e v i a t i o n M i g r a n t N e s t F o r a g e Anas platyrhynchos M a l l a r d M A L L * " Gallinago gallinago C o m m o n S n i p e C O S N * Accipiter gentilis N o r t h e r n G o s h a w k N O G O * 3 Buteo jamaicensis R e d - t a i l e d H a w k R T H A * Glaucidium gnoma N o r t h e r n P y g m y - o w l N P O W * 3 Dendragapus obscurus B l u e G r o u s e B L G R * 2 ' 4 Bonasa umbellus R u f f e d G r o u s e R U G R * 3 ' 4 Selasphorus rufus R u f o u s H u m m i n g b i r d R U H U N T M M N Stellula calliope C a l l i o p e H u m m i n g b i r d C A H U 1 N T M M N h u m m i n g b i r d s p . h u m m N T M M N Sphyrapicus nuchalis R e d - n a p e d S a p s u c k e r R N S A N T M D B Colaptes auralus N o r t h e r n F l i c k e r N O F L R M G Picoides villosus H a i r y W o o d p e c k e r H A W O * R D B Contopus borealis O l i v e - s i d e d F l y c a t c h e r 0 S F L 2 , 3 ' 4 N T M C H C. sordidulus W e s t e r n W o o d - p e e w e e W W P E N T M D H Empidonax alnorum A l d e r F l y c a t c h e r A L F L N T M D H E. Iraillii W i l l o w F l y c a t c h e r W I F L N T M D H E. oberholseri Dusky Flycatcher D U F L N T M D H Tachycineta thalassina V i o l e t - g r e e n S w a l l o w V G S W * 1 2 N T M D H Perisorius canadensis G r a y J a y G R J A 1 ' 3 R C G Cyanocitta stelleri S t e l l a r ' s J a y S T J A * 1 R C G Corvus corax C o m m o n R a v e n C O R A * 3 ' 4 R c G Parus atricapillus B l a c k - c a p p e d C h i c k a d e e B C C H 1 ' 3 ' 4 R D F P. rufescens C h e s t n u t - b a c k e d C h i c k a d e e C B C H * 1 R M F Sitta canadensis R e d - b r e a s t e d N u t h a t c h R B N U * 3 R M B Regulus satrapa G o l d e n - c r o w n e d K i n g l e t G C K I 1 R C F Catharus swainsoni Swainson's Thrush S W T H N T M M F C. guttatus H e r m i t T h r u s h H E T H 1 ' 3 ' 4 N T M C G Turdus migratorius A m e r i c a n R o b i n A M R O S D M M G Ixoreus naevius V a r i e d T h r u s h V A T H * 3 S D M C G Bombycilla cedrorum C e d a r W a x w i n g C E W A S D M M F Vireo solitarius S o l i t a r y V i r e o S O V I N T M C F V. gilvus Warbl ing Vireo W A V I N T M D F V. olivaceus R e d - e y e d V i r e o R E V I * 4 N T M D F Vermivora celata Orange-crowned Warbler O C W A N T M G F V. ruficapilla N a s h v i l l e W a r b l e r N A W A N T M G F Dendroica coronata Y e l l o w - r u m p e d W a r b l e r Y R W A S D M M F D. townsendi T o w n s e n d ' s W a r b l e r T 0 W A * 3 4 N T M C F Oporornis tolmiei MacGil l ivray's Warbler M G W A N T M S F Pheucticus melanocephalus B l a c k - h e a d e d G r o s b e a k B H G R * 4 N T M D F Passerina amoena L a z u l i B u n t i n g L A B U N T M S G Spizella passerina Chipping Sparrow C H S P S D M c G Melospiza melodia S o n g S p a r r o w S O S P R S G M. lincolnii L i n c o l n ' s S p a r r o w L I S P N T M S G Junco hyemalis Dark-eyed Junco D E J U R G G Agelaius phoeniceus R e d - w i n g e d B l a c k b i r d R W B L 1 ' 3 S D M D G Molothrus ater B r o w n - h e a d e d C o w b i r d B H C O S D M G Piranga ludoviciana W e s t e r n T a n a g e r W E T A 1 ' 3 ' 4 N T M C F Carduelis pinus P i n e S i s k i n P I S 1 S D M M F Loxia curvirostra R e d C r o s s b i l l R E C R * 3 S D M C F Pinicola enucleator P i n e G r o s b e a k P I G R * 3 ' 4 R c F Coccolhraustes vespertinus E v e n i n g G r o s b e a k E V G R 1 R c G humm includes both Rufous and Calliope Hummingbirds *Species excludedfrom the analysis (n=19). Species censused in: 1 1992 only (n=4); 2 1993 only (n=0); 3 1994 only (n=4); 4 1995 only (n=3). Migrant is migratory guild where NTM=neotropical migrant, SDM=short-distance migrant, and R= resident species. NEST is nesting guild where M=mix tree, C=conifer, D=deciduous tree, S=shrub, and G=ground nesting species. FORAGE is foraging guild where N=nectivore, B=bark gleaner, F=foliage gleaner, G=ground gleaner, and H=hawker species. APPENDIX 3 Tree and shrub species recorded in bird station vegetation plots. S c i e n t i f i c N a m e C o m m o n N a m e A b b r e v i a t i o n Abies t r u e f i r A B Acer glabrum ssp. douglassii D o u g l a s m a p l e A C G L Alnus incana ssp. tenuifolia m o u n t a i n a l d e r A L I N Amelanchier alnifolia S a s k a t o o n A M A L Apocymum androsaemifolium s p r e a d i n g d o g b a n e A P A N Arctostaphylos uva-ursi k i n n i k i n n i c k A R U V Betula papyrifera p a p e r b i r c h B E P A Ceanothus velutinus s n o w b r u s h C E V E Cornus stolonifera r e d - o s i e r d o g w o o d C O S T Gaultheria ovatifolia w e s t e r n t e a - b e r r y G A O V Holodiscus discolor o c e a n s p r a y H O D I Larix occidentalis w e s t e r n l a r c h L A O C Lonicera ciliosa o r a n g e h o n e y s u c k l e L O C I Lonicera involucrata b l a c k t w i n b e r r y L O I N Lonicera utahensis r e d t w i n b e r r y L O U T Mahonia aquifolium t a l l Oregon g r a p e M A A Q Pachistima myrsinites f a l s e b o x P A M Y Picea sp. s p r u c e ( s p . ) P I S P Pinus contorta l o d g e p o l e p i n e P I C O Pinus monticola w e s t e r n w h i t e p i n e P I M O Populus tremuloides t r e m b l i n g a s p e n P O T R E Populus balsamifera s s p . trichocarpa b l a c k c o t t o n w o o d P O T R I Prunus emarginata b i t t e r c h e r r y P R E M Pseudotsuga menziesi s s p . menziesii D o u g l a s - f i r P S M E Ribes lacustre b l a c k g o o s e b e r r y R I L A Ribes viscosissimum s t i c k y c u r r a n t R I V I Ribes sp. c u r r a n t ( s p . ) R I S P Rosa sp. r o s e ( s p . ) R O S P Rubusidaeus r e d r a s p b e r r y R U I D Rubus leucodermis b l a c k r a s p b e r r y R U L E Rubus parviflorus t h i m b l e b e r r y R U P A Salix sp. w i l l o w ( s p . ) S A S P Sambucus racemosa r e d e l d e r b e r r y S A R A Sherperdia canadensis s o o p a l a l l i e / s o a p b e r r y S H C A Sorbus scopulina w e s t e r n m o u n t a i n - a s h S O S C Spirea betulifolia b i r c h - l e a v e d s p i r e a S P B E Symphoricarpos albus c o m m o n s n o w b e r r y S Y A L Taxus brevifolia P a c i f i c / w e s t e r n y e w T A B R Thuja plicata w e s t e r n r e d c e d a r T H P L Tsuga heterophylla w e s t e r n h e m l o c k T S H E Vaccinium alaskaense A l a s k a n b l u e b e r r y V A A L Vaccinium membranaceum b l a c k h u c k l e b e r r y V A M E Vaccinium ovalifolium o v a l - l e a v e d b l u e b e r r y V A O V Vaccinium parvifolium r e d h u c k l e b e r r y V A P A Vaccinium sp. b l u e b e r r y ( s p . ) V A S P 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items