Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Response of carabid species and assemblages to forest practices of British Columbia in Engelmann spruce-subalpine… McDowell, Jocylyn K. 1998

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

Item Metadata

Download

Media
831-ubc_1998-0538.pdf [ 4.79MB ]
Metadata
JSON: 831-1.0088536.json
JSON-LD: 831-1.0088536-ld.json
RDF/XML (Pretty): 831-1.0088536-rdf.xml
RDF/JSON: 831-1.0088536-rdf.json
Turtle: 831-1.0088536-turtle.txt
N-Triples: 831-1.0088536-rdf-ntriples.txt
Original Record: 831-1.0088536-source.json
Full Text
831-1.0088536-fulltext.txt
Citation
831-1.0088536.ris

Full Text

RESPONSE OF CARABID SPECIES AND ASSEMBLAGES TO FOREST PRACTICES OF BRITISH COLUMBIA IN ENGELMANN SPRUCE-SUBALPINE FIR AND INTERIOR CEDAR-HEMLOCK FORESTS by JOCYLYN K. MCDOWELL B. F. A. Queen's University, 1986 B. Sc. University of British Columbia, 1994 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (DEPARTMENT OF ZOOLOGY) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA May, 1998 © Jocylyn K. McDowell, 1998 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. The University of British Columbia Vancouver, Canada Department DE-6 (2/88) A B S T R A C T The global extinction rate is currently accelerating at an alarming pace. Habitat loss through the processes of resource extraction, such as forestry are being blamed for much of the loss of biodiversity. In Canada timber extraction is extremely important economically, but modern silvicultural practices are impinging upon constituant and structural features of forest ecosystems, in particular, arthropods. It is important to characterize both the constituant arthropod species of our forests, and their response to forest practices in British Columbia. In this study the effects of clear and partial cutting and shrub removal, and the role of coarse woody debris were tested in Engelmann Spruce Subalpine Fir (ESSF) and Interior Cedar Hemlock (ICH) forests. Ground beetles (Coleoptera: Carabidae) were collected by pitfall traps, and the diversity of assemblages, and the abundance of individual species were described. Richness was calculated by rarefaction; rank abundance (Whittaker) plots and the log series alpha were calculated to characterize dominance structure; and the Shannon-Weiner and Simpson heterogeneity statistics were calculated. Intraspecific treatment effects were tested with the Kruskal-Wallis single-factor analysis of variance by rank and non-parametric multiple comparison tests. A dendrogram was generated to assess forest type and treatment differences. More than 36,000 carabid beetles consisting of 37 species were collected. A new species, now called Bembidion jocylyn Kavanaugh and Erwin, was collected. ESSF sites had more individuals, but fewer species than ICH sites. Logging had a positive effect on diversity but a negative impact on the total number of individuals. The impact of clear cutting in the ESSF endured longer than that in the ICH. Partial cutting reduced the abundance of individuals, but had little effect on overall diversity. Forest species decreased, but did not disappear in clear cuts; species typical of open habitats increased in clear cuts; habitat generalists were common in most sites. Most carabid species were either unaffected or responded favorably to the removal of shrubs: there was an increase in abundance in the ESSF, and in richness in the ICH. Carabid beetles appeared not to have a relationship to coarse woody debris, but there were problems with this third study. iii T A B L E OF CONTENTS Abstract ii Table of Contents iv List of Tables vi List of Figures vii Acknowledgments x Introduction 1 Biodiversity 1 The Role of Forestry 2 The Impact of Forestry 3 The Diversity of Arthropods 5 The Conservation of Arthropods 5 Insects as Indicator Taxa 6 Carabid Beetles in Interior Forests of British Columbia: this study 7 Experiment 1: Tree harvest 8 Experiment 2: Shrub removal 10 Experiment 3: Coarse woody debris 10 Materials and Methods • 12 Site descriptions : : 12 Sampling methods 15 Sampling design and schedule 1. Tree harvest 15 2. Shrub removal 16 3. Coarse woody debris 16 Sorting and Identification 17 Data analysis 1. Tree harvest 18 2. Shrub removal 22 3. Coarse woody debris 23 iv Results 24 1. Tree harvest 24 : August collections 27 : 1994 seasonal collection 42 2. Shrub removal 50 3. Coarse woody debris 59 Discussion 62 Limitations of pitfall trapping 62 Forest differences 65 Tree harvest : treatment effects 66 : intraspecific comparisons 69 Shrub removal 74 Coarse woody debris 76 Conclusions 78 Literature Cited 80 Appendices 89 v LIST O F T A B L E S Table 1: Total number of specimens caught in all experiments in ESSF sites near East Barriere Lake, British Columbia 25 Table 2: Total number of specimens caught in all experiments in ICH sites near East Barriere Lake, British Columbia 26 Table 3: Mean number of specimens per trap during all August collections (tree harvest experiment) 28 Table 4: Heterogeneity indices for pooled August collections (tree harvest experiment). The Shannon-Weiner is sensitive to changes in abundance of rare species while the Simpson marks changes in the abundance of common species 32 Table 5: Mean number of specimens per trap in the 1994 collection (tree harvest experiment) ESSF 43 Table 6: Mean number of specimens per trap in the 1994 collection (tree harvest experiment) ICH 44 Table 7: Mean number of specimens per trap for shrub removal experiment 51 Table 8: Heterogeneity indices for shrub removal experiment. The Shannon-Weiner is sensitive to changes in abundance of rare species while the Simpson marks changes in the abundance of common species 54 Table 9: P values for differences found (using the Wilcoxon signed rank test) in abundance of species in control and shrub removal sites. Probability values less that 0.05 indicate a significant preference for indicated habitat 58 Table 10: Mean number of specimens per trap by location for coarse woody debris experiment: ESSF 60 Table 11: Mean number of specimens per trap by location for coarse woody debris experiment: ICH 61 vi LIST O F F I G U R E S Figure 1: Engelmann Spruce Subalpine Fir forest ecosystem 13 Figure 2: Interior Cedar Hemlock forest ecosystem 13 Figure 3: Corrected richness for pooled August collections. Rarefaction is used to calculate richness based on a standard sample size of 40 specimens per treatment 29 Figure 4: Alpha values of the logarithmic series: pooled August collections. Alpha is a measure of the dominance structure of the species assemblage 29 Figure 5a: ESSF species rank abundance (Whittaker) plots for all treatments. Percent abundance of each species is plotted on a logarithmic scale against species rank ordered from most abundant to least abundant species 30 Figure 5b: ICH species rank abundance (Whittaker) plots for all treatments. Percent abundance of each species is plotted on a logarithmic scale against species rank ordered from most abundant to least abundant species 30 Figure 6a: Mean number of individuals (with standard error) of Scaphinotus angusticollis trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type 33 Figure 6b: Mean number of individuals (with standard error) of Calathus advena trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type 34 Figure 6c: Mean number of individuals (with standard error) of Nebria crassicornis intermedia trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type 35 Figure 7: Mean number of individuals (with standard error) of Pterostichus adstrictus trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type 37 Figure 8a: Mean number of individuals (with standard error) of Pterostichus riparius trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type 38 vn Figure 8b: Mean number of individuals (with standard error) of Pterostichus herculaneous trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type 39 Figure 9: Dendrogram resulting from an average linkage cluster analysis of arcsine square root transformed percent abundance data of the 20 most abundant species trapped in pooled August collections 41 Figure 10a: Engelmann Spruce Subalpine Fir 1994 seasonal richness calculated by rarefaction with 10 as the standard number of specimens 45 Figure 10b: Interior Cedar Hemlock 1994 seasonal richness calculated by rarefaction with 10 as the standard number of specimens 45 Figure 11a: Alpha values of the logarithmic series with standard error for 1994 ESSF collections. Alpha is a measure of the dominance structure of the species assemblage... 46 Figure 1 \b\Alpha values of the logarithmic series with standard error for 1994 ICH collections. Alpha is a measure of the dominance structure of the species assemblage... 46 Figure 12a: Mean number of individuals of Scaphinotus angusticollis trapped in old growth and clear cuts of both forests during the summer of 1994 ; 48 Figure 12b: Mean number of individuals of Calathus advena trapped in old growth and clear cuts of both forests during the summer of 1994 48 Figure 12c: Mean number of individuals of Pterostichus adstrictus trapped in old growth and clear cuts of both forests during the summer of 1994 49 Figure 12d: Mean number of individuals of Pterostichus riparius trapped in old growth and clear cuts of both forests during the summer of 1994 49 Figure 13: Corrected richness estimates for shrub removal experiment. Rarefaction is used to calculate richness based on a standard sample size of 580 specimens 52 Figure 14a: Species rank abundance (Whittaker) plots for shrub removal experiment. Percent abundance of each species is plotted on a logarithmic scale against species rank ordered from most abundant to least abundant species 53 Figure 14b: Alpha measures of logarithmic series of shrub removal experiment. Alpha is a measure of the dominance structure of the species assemblage 53 Figure 15a: Mean number of individuals (with standard error) of Scaphinotus angusticollis trapped in control and treatment sites of each forest during shrub removal experiment... 56 viii Figure 15b: Mean number of individuals (with standard error) of Pterostichus ecarinatus trapped in control and treatment sites of each forest during shrub removal experiment... 56 Figure 15c: Mean number of individuals (with standard error) of Pterostichus neobrunnus trapped in control and treatment sites of each forest during shrub removal experiment... 57 Figure 15d: Mean number of individuals (with standard error) of Nebria crassicornis intermedia trapped in control and treatment sites of each forest during shrub removal experiment 57 ix A C K N O W L E D G M E N T S I would like to thank my committee members: Dr. Geoffrey Scudder, my supervisor, for his support and encouragement. He is an inspired naturalist and teacher. Dr. Charley Krebs for his wit, and for his help with every single thing, true or false. Dr. Judy Myers and Dr. Walt Kenner for their advice and confidence in me. I would also like to thank: Dr. George Ball, Mr. Danny Shpeley and Dr. David Kavanaugh for species identifications. Dave Huggard for his help with statistics, and for making me laugh out loud in front of my computer. Kathy Craig, Randy Dailey, Jeff Jarrett, Suzie Lavallee and Karen Needham for their taxonomic excellence and support. Lisa Brumwell and Dan Hay don for their help with statistics. Launi Lucas and Alistair Blachford for their help with figures. Everyone in Kamloops who collected and sorted thousands of pitfall samples. Also, Karen Hodges and Jennifer Ruesink for statistical and editorial help, all kinds of advice and a lot of scones. Marjorie and Garth McDowell, Kirsten and Ken Madsen, and Dick Mahoney for taking me to the edges of the earth, where I found out why this project matters. Euan Mcintosh, Nina, and of course, my family and friends for support and essential distractions. Thanks to the British Columbia Ministry of Forests and Forest Renewal British Columbia for funding this project. x The uncounted products of evolution were gathered there for a purpose having nothing to do with me; their long Cenozoic history was enciphered into a genetic code I could not understand. The effect was strangely calming. Breathing and heartbeat diminished, concentration intensified. It seemed to me that something extraordinary in the forest was very close to where I stood, moving to the surface of discovery (Wilson 1984). I N T R O D U C T I O N Biodiversity Biodiversity is the material basis for all human activities and needs such as agriculture, recreation, ecotourism, and forestry. Our food sources, such as marine stocks and wild germplasms, and many ecological services emanate from earth's diverse biota. Our life-support system and much of our natural heritage and evolutionary innovations spring from the variety of the living world (Myers 1979, Wilson 1988). During the past 600 million years of evolution of the earth's biodiversity, there has been a background extinction rate -of 9% per million years. That rate is accelerating now at an alarming pace and biodiversity is being radically reduced: approximately 17,500 species are going extinct per year world wide (Scudder 1993). An extinction rate of 50% is predicted by the year 2100 if pressures on ecosystems are maintained in the current fashion. With economic development high on the agenda of governing agencies and human populations flourishing globally, resources are being consumed and destroyed. Dam and pipeline construction, and pollution by fertilizers, pesticides, sewage, chemical waste, 1 radioactivity and acid rain have been listed among the causes of species extinction (Shafer 1995, Kim 1993). However, habitat loss and concomitant climate changes are blamed foremost for the loss of biodiversity (Laurance et al. 1997, Desender and Turin 1989, Niemala et al. 1992, Samways 1992, Colwell and Coddington 1994, Butterfield 1996). Habitat loss has come about chiefly through agriculture and urbanization, and by the processes of resource extraction, such as mining and forestry. The Role of Forestry Historically, forests have been valued for timber, game, fodder, medicine, and watershed protection. Today, forests are also recognized for their role in water, carbon and other nutrient cycles, and climatic amelioration (McNeely 1994). Timber extraction in Canada is extremely important economically. Approximately 1 in 10 jobs in this country are related either directly, or indirectly to the forest sector (Anonymous in Smith 1990). Forestry has also been important culturally: logging (in British Columbia) has been a way of life for generations (Guest, pers. comm. 1997). Forests in Canada represent almost 45% of the national land base, and 8% of the world's volume in forest resources (Williams 1991). 200 years ago, when non-native logging was new in Canada, timber licenses and leases were easily obtained, so the best and most easily accessible timber was selected and logged (Gold 1985, Gould 1975). Today, technology allows foresters access to many areas that had previously been viewed as too rugged or distant. Silviculture practices, developed in Europe, have been in use in Canada for about 100 years. Clear cutting is 2 probably the most visible of these: approximately 90% of the trees cut in British Columbia's public forests are harvested by clear cutting (Hammond 1993). It is often followed by slashburning or mechanical scarification, and herbicide and fertilizer spraying, which encourage seedling development and reduce shrub and herb growth. Managed stands are often thinned at intermediate stages, which results in very little natural mortality or wasted timber. The Impact of Forestry Some believe that nature can take care of itself and that forest practices are not disruptive, since they mimic natural disturbances (McNeely 1994, Pojar 1993, Hammond 1993). Disturbances are defined as discrete events disrupting preexisting biological and physical organization (Pickett and White 1985). Small and even large scale disturbances, such as fire and insect outbreaks, are now understood to be integral to natural forest ecosystem dynamics (McCleod 1980). However, unlike modern forest practices, natural disturbances tend to be localized, and generally individual trees or fragments of the forest are left intact. Furthermore, natural disturbance leaves the majority of the biomass to decay, thereby returning nutrients to the soil and acting as physical barriers to erosion: clear cutting and slash burning remove biomass. Natural disturbances tend to be random in space and time, unlike timber harvesting, for which the time and place are calculated according to economic and logistic variables. The result is that large areas are being systematically stripped of vegetation and left to erode (Hammond 1993). Finally, it is 3 impossible to recreate the large and fierce forest fires of the past which, historically, have played a significant role in forest succession (Haila et al. 1994). Clear cutting in the tropics has been reported to alter the diversity and composition of fragment biotas, and to change ecological processes like nutrient cycling and pollination (Laurance et al. 1997). Temperate systems may be more resilient than tropical ones, since they are composed of fewer, more robust species than tropical systems (Pianka 1966). However, some believe that temperate forests harbor a diversity of species that approaches the much touted biological diversity of tropical rainforests (Luoma 1991). Clear cutting in temperate forests, undoubtedly causes many of the same problems. Mixed shrub and herb complexes often occupy moist and rich sites very soon after harvest and site preparation, and early survival and growth of conifer plantations can be severely reduced by physical damage, or by competition for light and space (Coates et al. 1994, Whitehead and Harper 1998). Silviculture practices may include the systematic reduction of competing vegetation in the early stages of seedling development. The reduction of shrub and herb layers during succession, however, inevitably has an impact on organisms that rely upon them. The clearing of sapling stands, subsequent thinning and intermediate cutting have drastically altered forest habitat by decreasing the natural mortality of trees so that dead, decaying wood is rarely found in a managed forest stand. This has caused species 4 dependent on coarse woody debris for nesting and food to become endangered (Kuusipalo and Kangas 1994, Lattin 1993). The Diversity of Arthropods Arthropods, in particular, are suffering from forest practices (Samways 1992, Lattin 1993, Lenski 1982). Having existed for more than 400 million years, and survived the Permian and Cretaceous mass extinctions, arthropods, until now, have been the most successful of all living things. Along with other invertebrates, they constitute more than three quarters of global biodiversity (Kim 1993). There are over a million species described (Stork 1988) and in Canada alone, there are over 66,000 insect species (Lehmkuhl et al. 1984). Arthropods have a large role in nutrient cycling as herbivores, as natural predators, and as parasites of other arthropod species (Kim 1993, Lattin 1993 ). They are significant in terms of energetics and biocycles: some grasshopper species transfer 5 to 10 times the amount of energy by feeding on plants as do the bird or mammal species in the area (Samways 1993). They are major components of ecosystems and their demise is significant. The Conservation of Arthropods Many have argued that the conservation of arthropods should be a goal unto itself: they have an inherent biological right to exist in an evolutionary context with ecological and instrumental values (Callicott 1986, Wilson 1992, Anderson et al. 1990, Lattin 1993). It 5 is generally agreed that effective arthropod conservation requires a better knowledge of species distribution throughout the region, habitat requirements and ecological function of at least the key taxa, and the impacts of different disturbance regimes upon special habitats and the associated fauna (Wilson 1985, 1989, Janzen 1987). Insects as Indicator Taxa The presence or absence of arthropods is also important to the distribution, abundance and diversity of plants and vertebrates, which typically are the premier species in conservation efforts (Miller 1993). To this end, insects have been used as indicator taxa, which are organisms so intimately associated with particular environmental conditions that their presence indicates the existence of those conditions (Burton et al, 1992). Insects make •good indicator taxa because severe disturbances are known to lead to substantial mortality, and reductions in diversity (Weaver 1995, Niemela et al. 1992, Noss 1990, Pearson 1994). Niemela et al. (1993) suggested that invertebrates are more sensitive to habitat changes, perhaps because they operate at smaller spatial and temporal scales than vertebrates. Ground dwelling insects are generally favoured over canopy fauna as bioindicators (Koen and Crowe 1987, Adis and Shubert 1985), and ground beetles (Coleoptera: Carabidae) in particular, are becoming important bioindicators for comparative ecological studies. These insects are generalist predators and are ubiquitous. They have differentiated habitat requirements and diverse morphology. They are relatively well known taxonomically and 6 like most insects, they respond rapidly to environmental changes (Thiele 1977, den Boer 1986, Duschene et al. 1994). Carabid Beetles in Interior forests of British Columbia: this study Whether to conserve arthropods or to save other taxa, it is important to characterize both the constituant species of our forest ecosystems, and the response of assemblages and individual species to disturbances which are integral to forest practices in British Columbia. In this study, ground beetles were surveyed in order to characterize the effects of forest practices on that group of arthropods in two interior British Columbia forest ecosystems. In Engelmann Spruce Subalpine Fir (ESSF) and Interior Cedar Hemlock (ICH) forests in British Columbia, three habitat manipulations which mimic current forest practices were imposed. Between 1991 and 1996, carabids were collected by pitfall traps, and the diversity of assemblages and the abundance of individual species were subsequently described. Thiele (1977) stated that carabid diversity is directly linked to habitat productivity. The ESSF, among other things, has a much shorter growing season than the ICH, therefore is quite different in terms of productivity. In this study I will test these hypotheses: carabid assemblages collected in ESSF and ICH forests may have species in common since they are adjacent. However, ESSF assemblages will be less diverse than those in ICH sites, owing to the differing productivity of the two forests. 7 Experiment 1: Tree harvest In the first experiment of this study, carabid assemblages from old growth stands and from clear cut blocks are compared. Several authors have documented a response by carabid beetles to tree harvest (Lenski 1982, Jennings et al. 1986, Morris and Rispin 1988, Niemala et al. 1993, Craig 1995, Neave 1996, Lemieux 1998). These studies suggest that old growth forests assemblages are less diverse than open grassland habitat assemblages, since the latter contain more rare species. The invasion of logged sites by species of open habitat, together with the short term persistence of several mature forest generalists, initially increased carabid diversity in regenerating stands, in comparison to the mature stands. There has also been a general decline in the abundance of carabids with logging (Niemela et al. 1992, 1993). Assemblages are also compared from old growth and partial cut blocks. In a comparison of soil arthropods 14 years after harvest, in litter layers of selectively harvested and nonharvested coastal redwood forests, predator guilds showed significantly reduced abundance in selectively harvested forest (Hoekstra 1995). In a study looking at ground beetle assemblages in boreal forests, Niemala et al. (1992) found that some species disappeared from cuts because they died or dispersed. Others survived the cut, but declined and disappeared later due to pressures on reproduction or other populations level processes. Fazekas et al. (1992) found that species living in forests were large in size, brachypterous and often autumn breeders. These characteristics reflect 8 a stable ground beetle community. Parmenter and MacMahon (1984) found that forest species were adapted to cool temperatures, low vapor pressure deficits and reduced light intensities. Forest species are ecologically highly specialized and evolutionarily conservative, and small populations of mature forest specialists are at great risk of local extinction (Niemala et al. 1992). Open habitat species, being day active and having adapted to bright, warm growing seasons would have an early summer reproductive period and would thus overwinter as adults. They would tend to peak earlier in the season, although these parameters would vary with altitude and site specific climatic conditions (Thiele 1977, Holliday 1991). According to Baguette et al. (1993), carabid species living in clear cuts are specialists. Lindroth (1961-1969) said that forest generalists form the basic forest carabid assemblage. Habitat generalists are able to maintain breeding populations in a variety of environments: some species have broad habitat requirements and for some, wide habitat distribution is due to directed movement between reproduction habitat and hibernation habitat. Associations with environmental types are strictest amongst field and forest species, but most species are found in low numbers outside each preferred environmental type (Niemala and Halme 1992). In this study, I will test the following hypotheses: clear cutting will reduce the number of individuals, but will increase the diversity, relative to that of uncut forest assemblages. The 9 abundance of carabids in partial cut blocks will be lower than that in old growth stands. Some species will be negatively effected by the opening of habitat; others, particularly the uncommon species, will thrive in open habitat; and still others will be unaffected by the removal of trees and will be found in all habitats. Experiment 2: S h r u b removal In the second experiment, the impact of shrub removal was tested by comparing carabid assemblages from sites (in cutblocks) where shrubs had been physically removed, to sites where shrubs were allowed to grow. Habitat in which vegetation is simple usually has a minimal litter layer, and a seasonally variable herbaceous ground cover. Arthropods living in these habitats, particularly ground-dwelling species, may be subjected to severe environmental conditions (Parmenter and MacMahon 1984). In this study I will test the following hypotheses: the removal of shrubs will have an effect on carabid diversity, and will reduce the abundance of species dependent on shrubs. Experiment 3: C o a r s e woody debris The third experiment was designed to characterize the use of coarse woody debris by carabids. In cut blocks, assemblages and individual species abundances at various locations relative to logs, were compared. Certain carabid species are known to be dependent upon large old dead wood (Lattin, 1993). In this study I will test the following hypotheses: carabid assemblages in close proximity to logs will differ from those further away. Some 10 species will be found consistantly near logs, while others will be found in random locations. 11 MATERIALS AND METHODS Site Descriptions Two biogeoclimatic zones were sampled in three experiments; Engelmann Spruce-Subalpine Fir (ESSF) and Interior Cedar-Hemlock (ICH). Both forests had been harvested extensively during the previous 30 years using both clear-cutting and partial cutting systems. Many clear cut sites had been burned and some of the harvested areas had been mechanically prepared and planted at various stages. The general characteristics of these zones were described by Meidinger and Pojar (1991). The ESSF is the uppermost forested zone in the southern three-quarters of British Columbia (Figure 1), and typically occurs above the ICH: the sites for this study were between 1500 and 2300 meters. ESSF occurs predominantly in mountainous terrain which is often steep and rugged. It has relatively a cold, moist, and snowy continental climate with cool, short (less than 4 months) growing seasons and long, cold winters during which soils are usually frozen. Fifteen forested subzones are recognized in the ESSF which can be broadly grouped into three climatic types: dry, moist and wet. The sites for this study were "wet cold 2" (wc2). Engelmann spruce (Picea engelmannii Parry) typically dominates the semi-open canopy of mature stands while subalpine fir (Abies lasiocarpa (Hook.) Nutt.) is most abundant in the understory. There is a moderately dense shrub layer, and a very productive and luxuriant herbaceous layer. Aside from timber harvest, the ESSF is valued for the grazing of domestic livestock, fur trapping and recreational activities. 12 Figure 1. Engelmann Spruce-Subalpine Fir (ESSF) biogeoclimatic zone in British Columbia. Figure 2. Interior Cedar-Hemlock (ICH) biogeoclimatic zone in British Columbia. 13 The ICH occurs at lower to middle elevations of southeastern British Columbia (Figure 2), usually below the ESSF. The sites for this study were between 400 and 1500 meters. This zone has cool, wet winters and warm, dry summers although it does experience a few very cold winter days and a few very hot summer days. Eleven subzones are found in the ICH with most of them being wet or moist. The sites for this study were "moist warm 3" (mw3). There are moderately developed shrub and herb layers. Western red cedar (Thuja plicata Donn.) and western hemlock (Tsuga heterophylla (Raf.) Sarg.) usually dominate the closed canopy of mature climax forests, although the ICH has the highest diversity of tree species in the province. This diversity, coupled with high productivity, have made the ICH very valuable for timber harvest. Livestock grazing is restricted to cutblocks and roadsides, but fur is harvested here and recreational activities are many. The two forest types were sampled in uncut stands, and partial-cut and clear-cut blocks. In spite of the fact that clear cuts in both forests had been logged primarily in the 1970s, the more productive ICH clear cuts had dense shrub and deciduous tree cover, while the ESSF clear cuts had little regrowth: newer clear cuts in the ICH, which had been logged in the late 1980s, were also sampled since they were at the same successional stage as the ESSF cuts. One of the ESSF partial cuts was new enough to be bare of shrubs, but the other two were older and had dense shrub cover. A l l of the ICH partial cuts were older with very dense shrub cover. 14 Sampling methods Collecting procedures (Hicks 1993) have been described by Huggard and Klenner (1996) as follows. In the summer months, the beetles were trapped with pitfall traps (400 ml. plastic cups, opening diameter 9.5 cm). One cup was set into the ground with its top flush with the ground surface, and a second, close fitting cup put inside the first. 100 ml of 65% propylene glycol, a non-toxic and non-volatile liquid, was poured into the inner cup. A 30x30 cm board was held 15 cm above the pitfall trap on 3 pegs to keep rain and debris out of the trap. Contents of the traps were collected after 14 days by straining the glycol through a fine mesh, and rinsing the contents of the strainer into a sample cup with ethanol. Each sample was given a unique number to identify the trap and trapping session. In the winter months an insulated 125 cm-tall plywood chimney was mounted on a 55 cm x 55 cm platform, held 10 cm above the ground. A single pitfall trap was placed at the center of the platform, and was accessed through the top of the chimney. The chimneys were placed in position before snowfall. After snow had accumulated in the middle of the winter, the pitfalls were set with propylene glycol through the chimney, without disturbing the snow cover. The contents of the traps were collected 2 months later. Sampling Design and Schedule 1. Tree harvest Three sites were sampled in each uncut stand, and partial cut, and clear cut block in both forest types. A doubly-nested sampling design was used. Five individual pitfall cups were 15 set in a circle of radius 4 meters to form a trap circle. Within each of three replicate cutblocks or stands three trap circles were placed 75 meters apart, making a total of 45 traps, or 9 rings per treatment. Trapping was done in late summer from 1991 to 1995 in two successive 14 day sessions. In 1994 one third of the traps were set in June, two thirds in July, and all in August and September. The final collecting sessions were in July 1996. A single winter trap was set up in the center of each summer circle. Winter trapping was done between 1992 and 1995 in two month mid-winter intervals. (Collection dates for all three experiments are listed in Appendix 1.) 2. Shrub removal In the summer of 1994, at each of three sites in ESSF partial cuts and older ICH clear cuts, shrub cover was removed from three 30X30 meter plots. The treatment plots were each paired with controls, where shrubs were not removed. Each of the shrub removal and . paired control plots had one circle of six traps in its center. Trapping was conducted in late summer 1994, early and late summer 1995 and winter 1995/1996. 3. Coarse woody debris Six sites were set with traps placed at 0 cm, 50 cm, 100 cm, 150 cm, 200 cm and 350 cm from a log. There were two sites in ICH new clear cuts, one in an older ICH clear cut, two in ESSF clear cuts, and one in an ESSF partial cut. The five nearest traps were placed in a ring so that each was sampling a similar area. Three logs were sampled in this way at 16 each of the six sites. Collections were made in August 1994 and 1995, and June and July 1996. Sorting and Identification The contents of each pitfall sample were sorted to class, order or family depending upon available taxonomic keys, by the B.C. Ministry of Forests. A l l adult carabids were sent to Dr. G.G.E. Scudder's laboratory at the University of British Columbia in Vancouver. Between January 1995 and October 1997 all specimens were identified to species using the Spencer Museum reference collection and Lindroth's keys (1961-1969). The identity and number of each species was recorded for each sample. Most specimens were replaced in vials of 70% ethyl alcohol. For species that were very abundant, at least 30 specimens of each were pinned. Pinned beetles were individually labeled with the location, treatment, trap number, cardinal points, the range of dates.the trap was in the field, and the collector's, species' and determiner's names as recommended by Winchester and Scudder (1993). For species of low abundance, all specimens were pinned and labeled. Pinned and identified representatives were sent to two carabid experts, Dr. George Ball and Mr. Danny Schpeley at the University of Alberta in Edmonton for verification. In addition, specimens of one species were sent to Dr. David Kavanaugh at the California Academy of Sciences in San Francisco, for identification. Voucher specimens have been placed in the Spencer Entomological Museum, and will be sent to the Ministry of Forestry in Kamloops, British Columbia, the Royal British Columbia Museum (Victoria), the Pacific Forestry Center (Victoria) and the Canadian Collection (Ottawa). 17 Data analysis Once the identity of all species was verified and missing samples noted, data were entered in Microsoft Excel and combined using both Microsoft Excel and Access. A l l heterogeneity indices were calculated with Pisces Species Diversity and Richness (Richard Seaby at IRC House, UK) . A l l other statistical calculations were done with Systat Version 5.0, and all figures made with Sigmaplot Version 4.0, unless otherwise specified. A species list was assembled for the tree harvest, shrub removal and coarse woody debris collections with total number of individuals collected. Individual traps were too close together to be considered independent samples. However, rings were situated further apart than the typical home range of a large carabid, so each ring was considered to be an independent sample within the block. This alleviated statistical problems posed by the numerous "0" values found in many individual cups, as well as avoiding the problem of missing cups, since some of the samples were lost or destroyed in the field. In some collection sessions, an entire ring of traps was missing. Thus when appropriate, data were pooled over time to maintain a balanced sample design. 1. Tree Harvest Since sampling effort was not consistant from year to year, analysis was limited to two portions of the data. The five August collections were selected and analyzed as representatives of the entire collection, and samples from 1994, which spanned the entire season, were analyzed in the hope that seasonal differences would embellish conclusions 18 about the results of the August collections. Species lists were assembled for both portions of data, with number of individuals per trap recorded. a. Tree Harvest: August collections Richness Rarefaction calculates species number based on an hypothetical smaller sample size than what is available. This is important since it allows the comparison of collections with unequal numbers of specimens. The method relies on robust sample sizes, hence errors may occur with small sample sizes, but generally rarefaction curves are reliable (Simberloff 1978). Richness estimates for each of the five August collections were obtained by rarefaction using 40 as the standard number of specimens (Ecological Methods computer package by Krebs, 1989). These values were used to test for a year effect by A N O V A . Richness estimates, based on pooled August samples from 1991 to 1995, were then obtained for all treatments. Evenness Evenness is a way of quantifying the dominance structure of an assemblage. An unequal representation is compared to an hypothetical community in which all species are equally present (Krebs 1989). Evenness was presented graphically with rank abundance plots (Whittaker plots, Krebs 1989). In these graphs, percent species abundance is plotted on a logarithmic y-axis against species rank on a linear x-axis. Alpha values were calculated using Logserie (Ecological Methods computer package, Krebs 1989) for each of the 19 resultant plots. A series of studies (Kempton and Taylor in Magurran 1988) investigating the properties of the log series alpha have come out strongly in favour of its use, even when the log series distribution is not the best descriptor of the underlying species abundance pattern. Heterogeneity Heterogeneity indices are based on both the evenness and the richness of species assemblages. Two heterogeneity statistics, the Shannon-Weiner (H'), and the inverse of the Simpson (1/H) were calculated. The higher the value of the statistic, the more diverse is the community. Jarosik (1991) found that both the Shannon-Weiner and the Simpson indices could be used in long term observations (on natural sites), as sensitive and reliable indicators of differences among plots: short term catches may yield an unreliable result. The Shannon-Weiner is known to detect differences in diversity based on the presence of rare species and the Simpson, those based on the abundance of common species (Magurran, 1988). Intraspecific Comparisons Three rings per block and three blocks resulted in nine values representing "number per trap" for each species' abundance. For species represented by more than 100 specimens the mean and standard error of these nine numbers were plotted as histograms for both forest types and all treatments. 20 The Kruskal-Wallis single-factor analysis of variance by rank was used to test treatment differences for each of these species. This test is applicable when populations are not normal and variances are heterogeneous, as is the case with these data (Zar 1996). If a significant difference was detected between treatments, a non-parametric multiple comparison test, similar to the Tukey test, was performed on the mean numbers of species. This calculation was done by hand following the method outlined by Zar (1996). Significantly different treatment abundances were recorded. Cluster Analysis A dendrogram was produced to assess forest type and treatment differences based on all species (20) present in more than one of the 7 possible variants of treatment and forest type. The abundance of each species was calculated as a percent of the total collection, and these data were arcsine square root transformed for the generation of the dendrogram. The cluster analysis was based on a hierarchical similarity measure (Euclidian distance). b. Tree Harvest: 1994 Collection Richness, Evenness and Heterogeneity measures Richness estimates for each treatment and month were obtained by rarefaction. Only one of three rings per block was sampled in June, and only two in July so initially calculations were made using data from the ring that had been used continuously (Ring 2) (using 20 as the standard number of specimens). Estimates were then derived for all rings (using 30 as 21 the standard number of specimens). The two calculations were graphed together and compared. Subsequent analyses were based on all rings. The richness estimates for each ring (using 10 as the standard number of specimens) were graphed to compare differences between months and treatments. Whittaker plots were made for each forest type and treatment by month. Alpha values of the logarithmic series were calculated and graphed with standard error. The Shannon-Weiner and the inverse of the Simpson were calculated as above. 2. Shrub removal The data from the shrub removal experiment were pooled by ring, and over time to obtain average numbers per trap as in the tree harvest experiment above. A species list was assembled for the shrub collections comparing numbers per trap for treatment and control of both forest types. Richness, evenness and heterogeneity statistics were calculated as above. Intraspecific comparisons Again, there were nine values representing "number per trap" for each species' abundance. (See August collections above) The mean and standard error of these nine numbers for four abundant species were plotted as histograms, and a non-parametric test, the Wilcoxon signed rank test, was done to test for differences between control and treatment within 22 each forest type. This test is useful when populations are not normal and variances are heterogeneous, as is the case with these data (Zar 1996). 3. Coarse Woody Debris Pooling over time was the only option for the coarse woody debris experiment, in spite of the imbalance of the collections (2 Augusts versus 1 early summer), since sample sizes were small, and there were missing samples. A species list for both forest types was assembled with number of individuals per trap listed comparing location with respect to the log. Richness and evenness measures were calculated as above. 23 R E S U L T S A total of 36,922 carabid beetles were collected in the three experiments. These consisted of 37 species in 18 genera (Tables 1 and 2). Nine species were present in numbers greater than 1000, 24 were represented by less than 100 specimens and of these, 4 species were represented by only 1 specimen. There were over 12,000 specimens of Calathus advena Leconte, almost 8000 of Scaphinotus angusticollis Mannerheim and more than 4500 specimens of Pterostichus riparius Dejean. One species, of which 13 specimens were collected, is now called Bembidion jocylyn Kavanaugh and Erwin, and is a previously unrecorded species (Kavanaugh et al. in press). One specimen of Dyschirius planatus Lindroth was found in an old clear cut of the ICH: this is the first record of this species for British Columbia. No adult carabids were trapped during the winter sessions. 1. Tree Harvest There were 34 species in the ICH and 24 in the ESSF with 22 species in common. The mean number of carabids per trap was 7.9 in the ESSF and 4.3 in the ICH making the abundance almost twice as high in the former as in the latter. Without adjusting for effort, some general trends in abundance are apparent (Tables 1 and 2). In both forest types the number of individuals in the old growth stands was greater than 50 % of the forest total, while the number in the partial cuts accounted for about 25%, and the number in the clear-cuts (combined for ICH), contributed slightly less than that. The number of species was highest in the ESSF clear cut and the ICH new clear cut (in which the vegetation was most similar successionally) and lowest in old growth stands. The number of genera was highest in the ESSF clear cut and the ICH old clear cut and in both forests, lowest in the uncut 24 ro o n i - i n < D W i - s c \ i i - c s r - a ) < t o i D i f l ' * T- O l T - r N S O 5 T - C D I - C 0 I - C O i - CM CO rj- LO CM N O) CO N W i -C\l O S CO i -LO r-» LO CO •"tf CM CM CM CM CM J 3 ii ns >> T3 O O es o U J3 "3 U J S j * rt i—I fc sa-cs s V L-J 2 s u J = c p — I/-. S.ONN vo O ~ : — S.Tt-h — t ~ CM i: CN iiOOii t O : !OV: ii sCM ^ ^ CO K £ = CD CM CM rt W i— rt D c 3 - - cn 2> 3 IT. r~ c - 0 > C N O \ V D _ _ i s l V O O O t ^ _ _ - v o x CM* ~ ig;1 m u~. t— 00 00 W CD CU > CS J = cu V t-H 3 3 O c 3 ; in .':: -i — 0 0 , ON „ _ _ o \ . C M : W-l O V i C M , CT\,0 — r— ^ $ 5 « ^ s ^ 5 CM — _ C?> £ 05 S CM _ OJ 00 J S 3 rt o s E CS 01 Ui o a. ^ =tf  o S iiiaii <::4>: o E3; - C cs Cj 60 , 6 0 u .2 _ w J 5 oo ft. 2 g a a £ 5 3 : a s c :§ !§ -S -S £ £ oa CQ 5 f ; o ; •fe § § 1 c a •S 1 I s i CQ O J3 a ca; E > a. a. -2 -2 U 5 a -3s ^ ;;sj:' c o •Si •<=. <J .2 3 ^ •« "Si •I' ~ 2 = 3 s ii - „ S 'C 5 'C -S ^  c S s : j o w.-Hv ca • ^ •• « e <u -s: S ^5 ' i s i ^ : :;j=:i 1) :i •JT-2 Q ^ .3 C S 60 •&• c >» a •a .S . 3 t—t •a § p p o o o £ £ £ £ )^ :C3' *^ cj 3 -a a e >u CO C o L , 25 u .A u o o CO i - T- c\i co i - r ^ . c o o r - . i - 0 5 i - i - O N C I i - I O i - ' t O f l D t <D00(DC0KCOWT-NO)tD CM -i- 7 - O) CO t-r-i - CM ^ CO c^  tN'.rf S - t- -:;C<V r-,rr-A O N OO — r:i:ii!;ii O N ; | T * : ! i - If) CO CO N O N O 2 CO I ^ 2 ^\ O N : C J _ — ; :! O N „ : ^ c". OO O N — O N O O N — CN — r-( v~i L- c^  H — OO e > ao'S a' £ 3 P §'•§ ^ 3 I 1 5 11 s y u « S 3 .3 w 5 t 3 =c:i;oe a -c •< 5 2 2 5 5 3 3 3 £ £ £ ire; 5 % i ^ -5 =3 1 I ;.-o CQ oa oa i j ;<£. : .:NJ:; ^ ^ ^ 3 "S s -a -3,5 H O O Q 3 .3 3 £ § 3 S S 5i 2 5s 5 5 i '£ Q == -3 a a 'C Q a _ oT c/3, 5 Q , a " s s | a 2 • sj 3 ' - ^ NJ ^ 1~ i a a a § .3 "° P- P- P- s ^ -2 .2 .2 .2 S 3 -3 ; s e * 3 3 3 3 , 3 ; . •S "a 0, Q. 0, Co to Co Co h~ — !^ "g 8 Si 3 o 26 stands, although the differences were not dramatic, la. Tree Harvest: August collections Trends for numbers of species and percents of forest total by treatment were roughly the same within the August collections (Table 3) as for the entire main collection. The ESSF and the ICH had similar numbers per trap with 4.06 and 3.70 respectively. Richness, Evenness and Heterogeneity measures Differences between richness estimates calculated by year were not significant (Appendix 2), so all August collections were pooled to calculate richness by treatment (Figure 3). Richness in the uncut sites for both forest types and partial cut sites for the ICH was the same. ESSF partial cut site and ICH old clear cut site values were similar and the ESSF clear cut site value was equal to the ICH new clear cut site value. Alpha values (Figure 4) were calculated in tandem with Whittaker plots (Figures 5A and 5b) which were consistant with the log series distribution model. Overlapping alpha values for uncut and partial cut sites of both forest types suggest that the dominance structure of these are similar. By the same token, the structure of these was different from that of the respective clear-cuts. The pattern of alpha values mimicked that of corrected richness above, which suggests that richness and evenness changed concurrently in response to tree harvest. 27 u o IK I I . v u .2 la , a-B S cs a em , c i:;cn:i cn ;!rf:i cn Pos o NS*;? O SO;: O h.Oj; O il iipiidh-diidii : cn ii\oij ii O N;<Ni ii O iiOii : CN i rf CN i ; o : - : o i :! O (Of O I oo in ii o io ! o o : O j O'1 |:r>.i in :• mi; oo CN :J <^;> i:S-;i rf iiOii Coli d iidj innii cn iicvii — oo: cn Oil i;VOi O HOVi O ^ — i VD i '.Oi: p . O:; — ;;,Oi! O ii.Oi! lid i d i.o:i d Ho] d HoJ CN CN * O CN : ;.oi o i.^.: rr io; i:p:i o ?cn • 0 : 0 : i:d.i d r:d i d [' d :; -O in loo.: O :' 'CN;; o ['Sf: d! : : rf .= — i:,rf.:: ;;00.: — '.—' ' • ~ : : O ii'Oi! '••—:i in ilOO!: cn !! .rr-: O i: Oi: cn ii : -e-li O i:r?i; CN !! h'p'i! d iio:i — ii ':CN:; r-~ 00 VO : Oii o ::—; — : CN o : cni cn ii, i p i d -cNidl nrf ii O iiiCN;: Ov ? ;;.p.ii d ir§ij CN ji i! r~- CN in 1: ii O Lrf J O I: 'I O NO!! O i! : Oi! CN ii iOi:Oi to'1! O ii £ o <o S rj CN vo _ ; h:s*ii VD ; :iCvi 00 :i :..OI| O f i;'0| O ! rf vo ° . — r-' , m m oo ; m ;:0:; vo CN ; cn ~-O 00 O CN — CN ; O ;!—•: O -O-; O :.cn:: ' iOEOidid id i 'd: : "O-: o: d rf rf cn r-~ o o d —' o rf CN : cn ON"< CV ' cn:; . — o :•:—'! O ' o o o , i d i o i i o i . d : — sC cn .: — ' O ; o • •o o — d d ' d :;^ H. rf ;:lni" cn IO O ; cn : :VO:: CN : rf o cn o cn ON o- cn i O \ — i.O'i O f'O : O0 iiO : — :: . d ! d i d i ; d h d : i d i d d i VD : IH ; ' CN CV o o d d -o o d 3* O i ON — O rf . oo cn cn in i, : O ! 00 vO O : d d d d o c r>:; cn i oc rr cn-. rf;: :0:CN 00 cn : d d oi d o d o d o d rr,2 .n P VO _ cN 2 * — a. 'i <3: 3 5 C <3 5 -S § 5 ^ -2? a S 3 - i s 'i'-C ^ '•'.^ * a a ° C :'• C:: c o a a to fc i'^ S: :: S ii 5-S 5 c a a a a g 6- &• 6-1 i u a; a; a; cj.fe: rc rg CQ CQ CQ ^ ^ -« ' - § .a ?j ? c c S:.ci. a. a. a, a. a, a. ^ o, o, CJ Shi i3i; 5;:;: • 5 is1 3 ^ < 3 < v. C U 3 C C <U tO S 28 12 10 I . 4 clear cut { new clear cut partial cut old clear cut i uncut 1 partial cut uncut ESSF ICH Figure 3: Corrected richness for pooled August collections. Rarefaction is used to calculate richness based on a standard sample size of 40 specimens per treatment. 6 i I ^ 3 a <3 clear cut • partial cut uncut new clear cut old clear cut partial cut uncut ESSF ICH Figure 4: Alpha values of the logarithmic series: pooled August collections. Alpha is a measure of the dominance structure of the species assemblage. 29 100 10 <U u c a -a c 3 0.1 0.01 —•— uncut • O- • partial cut — T — clear cut \ 6 e C3 -a c o.i 0 10 20 25 30 s p e c i e ' l W Figure 5a: ESSF species rank abundance (Whittaker) plots for all treatments Percent abundance of each species is plotted on a logarithmic scale against species rank ordered from most abundant to least abundant species. 100 n 10 H •o uncut partial cut old clear cut new clear cut 0.01 0 10 20 25 30 15 species rank Figure 5b: ICH species rank abundance (Whittaker) plots for all treatments Percent abundance of each species is plotted on a logarithmic scale against species rank ordered from most abundant to least abundant species. 30 In both forest types, the heterogeneity indices (Table 4) indicated highest diversity for the successionally similar ESSF and ICH new clear-cuts. In the ICH the values of the indices were lowest for partial cut while in the ESSF the lowest value was in the uncut. The values for the indices of partial cut and clear-cut in the ESSF were closest. In the ICH, partial cut and old clear cut values were very close, although the Shannon-Weiner statistic barely distinguished uncut from old clear cut. The Simpson values differed across treatments more than the Shannon-Weiner values. Intraspecific comparisons The intraspecific comparisons for the pooled August collections revealed various trends for the different species. Most species were found in both ESSF and ICH, but many of the more abundant species were found predominantly in one or the other. Most of these species were found in all of old growth, partial cut and clear cut sites, but their abundance varied. Forest species: No species were found exclusively in the old growth. However, some of the more abundant species obviously preferred the old growth, based on the Kruskal-Wallis and multiple comparison tests. Scaphinotus angusticollis was most abundant in uncut stands in both forests, but in the ICH, no difference was detected between the uncut and partial cut stands (Figure 6a). S. marginatus was most abundant in uncut stands, but evidently 31 Table 4: Heterogeneity indices for pooled August collections of the tree harvest experiment. The Shannon-Weiner is sensitive to changes in abundance of rare species while the Simpson marks changes in the abundance of common species. Engelmann Spruce Subalpine Fir Interior Cedar Hemlock uncut partial cut clear cut uncut partial cut old clear cut new clear cut Shannon-Weiner H' 1.268 1.673 1.963 1.312 1.108 1.334 1.614 1/Simpson 1/H 2.758 4.073 4.310 2.937 1.994 2.309 3.187 32 8 a* 6 5 H 34 1 H uc ESSF pc CC uc ICH pc occ ncc Figure 6a: Mean number of individuals (with standard error) of Scaphinotus angusticollis trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type. uc = uncut, pc = partial cut, cc = clear cut, occ = old clear cut, ncc = new clear cut ESSF H = 23.14, p < 0.001, uc * pc, pc P cc ICH H = 27.31, p < 0.001, uc £ occ, pc ^ occ, uc £ ncc, pc £ ncc 33 6 I uc pc ESSF-uc ICH pc occ Figure 6b: Mean number of individuals (with standard error) of Calathus advena trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type. uc = uncut, pc = partial cut, cc = clear cut, occ = old clear cut, ncc = new clear cut ESSF H = 11.185, p< 0.004, uc * cc ICH H = 16.87, p< 0.001, pc * occ 34 Figure 6c: Mean number of individuals (with standard error) of Nebria crassicornis intermedia trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type, uc = uncut, pc = partial cut, cc = clear cut, occ = old clear cut, ncc = new clear cut ESSF H = 12.81, p< 0.002, ucjtcc 35 recovered over time, since its abundance in the old clear cut was significantly higher than that in the new clear cut (Appendix 3a). Calathus advena was the most abundant species overall and definitely preferred uncut areas (Figure 6b). Nebria crassicornis intermedia was most abundant in the ESSF old growth although no difference was detected between its abundance there and in partial cut blocks (Figure 6c). Clear cut species: Pterostichus adstrictus was most abundant in clear cuts of both forest types (the successionally similar ICH new clear cut and ESSF older clear cut) (Figure 7). Analysis did not reveal a significant preference for clear cuts for any other species. Most of the less common species (of which there were less than 100 specimens) (17 in the ESSF, 25 in the ICH) were found exclusively or mainly in clear cuts (Tables 1 and 2: Tree harvest): the presence of these species increased the species diversity in those sites. Exceptions to this trend in the ESSF were Bembidion quadrifoveolatum and Harpalus nigratarsus, a large proportion of which were found in partial cuts. Sericoda quadripunctata was found exclusively in clear cut sites that had been burned in 1988. Habitat Generalist Species Pterostichus riparius (Figure 8a) and P. herculaneous (Figure 8b) were present in all treatments within a forest type. P.riparius, one of the three most abundant species, showed no preference for any habitat type. P. herculaneous however, was less abundant in new clear cut sites, but no significant differences were found between its abundances in 36 03 ft 03 3 12 -3 c 03 <U ESSF ICH-Figure 7: Mean number of individuals (with standard error) of Pterostichus adstrictus trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type. uc = uncut, pc = partial cut, cc = clear cut, occ = old clear cut, ncc = new clear cut E S S F H = 11.11, p<0.004, u c ? c c ICH H = 15.79, p < 0.001, pc t ncc, uc ?ncc 37 03 — 03 -o c c 03 CD Figure 8a: Mean number of individuals (with standard error) of Pterostichus riparius trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type, uc = uncut, pc = partial cut, cc = clear cut, occ = old clear cut, ncc = new clear cut ESSF H = 2.08, p< 0.354 38 0.4 0.3 •S 0.2 0.1 X 0.0 uc ESSF pc cc uc ICH-pc occ ncc Figure 8b: Mean number of individuals (with standard error) of Pterostichus herculaneous trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type, uc = uncut, pc = partial cut, cc = clear cut, occ = old clear cut, ncc = new clear cut ICH H = 11.87, p < 0.008, pc ncc, occ ncc 39 the uncut stands and the cut blocks. P. neobrunnus appeared to be most abundant in the uncut ICH forest, but differences were not significant between that and the older clear cut or the partial cut (Appendix 3b). Leistus ferruginosus was present in all treatments and differences in abundance were not significant (Appendix 3d). Unresolved species: For some of the abundant species, the Kruskal-Wallis test was able to detect differences amongst treatments and yet patterns were not resolved with the non-parametric multiple comparison test. Treatment effects on Synunchus impunctatus (Appendix 3e), P. ecarinatus (Appendix 3c) and Trechus chalybeus (Appendix 3f) were unresolved. Cluster Analysis The ESSF clear cut and ICH new clear cut assemblages closely resembled each other according to the cluster diagram (Figure 9), but the ICH old clear cut assemblage was unlike either of them. The partial cut and uncut assemblages, within forest types, were most similar to one another. 40 o o CN cn 00 d + 10. + o CN o c n oo o + + CN m o o + o o o u CJ o o o o c X U o cj o o ft 3 U PL, PH 00 00 00 00 00 00 W w W cj 3 K U ft u 4—» ca T3 (U O c -a C 3 X> 03 w C (U o <u tU c ft 03 *T3 OO c e O =3 T3 C X ! 03 O ~ w o .5 T3 tu 3 3 cr 03 on to tu E .£ ^ 'Lo o >. a -c M— 03 O S 03 oo C C 03 O <u o ts » = o o o tu * i 3 O i -03 O "o . 3 £ PH U CU C c 1 'ft 8 03 C X i 3 00 3 3 * 5 tu ft oo x < = 2 <U T3 OJO tu 03 "T; tU O c - S 03 -a 5 ^ O ft * 2 O0 ^ 3 8 IS ^ CJ "  ocu 5 l CJ P-i is 00 (U oo 77 E S ffi -a 03 03£ T3 00 C O 3 T3 o3 (U to 2 o 3 X I 00 ' CJ CJ 3 CJ 1— 03 T3 (U u tu c "5 K _e3 03 ft cj ft PH O a y g 41 lb. Tree Harvest: 1994 Collection Numbers per trap (Tables 5 and 6) were usually highest in June or July of the 1994 collections. The exception to this trend was in the ICH uncut sites where the most carabids were caught in August. In the uncut stands, numbers of specimens per trap were 2 to 3 times higher in the ESSF than in the ICH in June, July and September. In clear cuts, numbers per trap were up to 6 times higher in the ESSF throughout the summer. (More than 80 % of the July samples from the ESSF clear cut were missing.) Richness, evenness and heterogeneity measures A comparison of the corrected richness of uncut stands based on the data from the one ring that had been used consistantly, versus all available data, revealed no differences in treatment values relative to one another (Appendix 4 a, b). Subsequent analyses were based on all available data. In June, richness values overlapped for all treatments within a forest type (Figures 10a and 10b). As the summer progressed, differences evolved. In August, the month in which most collections were made for the tree harvest experiment, differences in richness were most pronounced. Differences in the dominance structures of the 1994 ESSF collections were most pronounced in August and September, and seasonal patterns in dominance structures in uncut and partial cut sites were very similar. Whittaker plots (Appendix 5 a-g) and log series alpha values (Figure 1 la) illustrate these trends. In the ICH (Figure 1 lb), alpha 42 i in M in cn — o- cs in rt vo cs — W V O — V D CS rf cn —' CN —' O O vo O O O O O rf o O d o d d d d d d d d d 3 h s . K — o d d -vo : •;—<-; cn r~ — ;;0-: — :cn; oo cs in — o-inONin—<cn o o o o o > n o o o d d d d d d d d d "3 : cs VD rr cn vo cs ; <n- rr sen - in. CN :;in; rr ;cn-; in o ei o o o 3 — m. O N rf os :o d d i 00 : o\ — rt CS : — • V D 00-: O N . cn . — O d d d . O N in O N O rt O rr p rt d d d in ;vS;: O S . d ' d : d :• o .d-: —, in o — O O rr 'Sr.:. -; oo O J cn — • O os O O d d d d 3 : 3 in O cs o o vo d — in •••••••• • •-• oo: cn:- in cn;: in • O - :: : ; cn — CS V O i CS o o cn V D o o d d — d d d 2 2 •3 in in O 00 . cn V D in in 0 in, in : 0: : V O in e cs O O' — d 0: cs O d 01 :d d ; — — in ;r> in •. O N :-ov in cs, in S O CS rr - —: O :i O irr'.: oi >d- d u [«l Ul OJ OO, O ; cn O N ; i . , V O : O N .ON'-! • V O '-VO-; :• O .O.' d d ,o-: — . d d ;-• 01 in — cn O N ;;o- :,rt:- cs O" V D f — ;;cnv vo —r-, o d oi o d d ! o o tH' rr1, — ^  O "VO- VO ci ; O © — CM: to O • O . o.io-did o d d 1 d • — . 00 "i rn vo • '• m : - — Os 10 00 —1 O N : . C S 0 vo 0 • O • O • O • O O d cn d d d d cn O O V O o o- o| C o CJ u .§ =3 o o <D ^ o 51 3 & 5 o-CO ? I S 1 ' vo 81 — cn: Z o — - O N rt O N ; in — vos o- rt O t< O —' cn g o o 00 O 'n O O — rt — :: O rT . O N rt cn . O N p o o-* H : vo ;cni: o sin;1 o —> 1 v-, o o — o- o cn; in : . .NO - d © n o d- oi d d •a 5 •a § -a u a 3^ 3 .3 t: s 3 O  iu v. 3 . a ^ ^ ^ 3 3^ -w;- k. -Cs ^ £ S 05 CQ 3 3 J J i u S •- 1 '::2:; t^ ,- t5 •i l l s s J % a, a. j a a S s 5 -c .5 C c i a. a. -o )^ 3 <3 ^ a; a; a; si N D O N _ | 43 a. co cn — —• o d o d o d o O N d CN O O O O d d d S o u B S : 3 "3 J CN r~ r CN r-o CN o CN CN cn o CN o vo o o VO o CN = CN o •* CN O d d d ° d d d d d °' cs o CN o CN CN O '. cn CN CN iO CN CN •* o cn cn •tf o o cn cn o CN CN O Ov o cn o Ov O cn 0 0 © o d © . ' © d d d d d d d d °" d © ' 0 0 :-m : iO.: 100 :cn :o oo t— cn r-O © . cn CN •* .oo cn o in r~ 0 0 cn o CN VO ^-oo o cn © d - d,o CN d d d d d — d oo _ | o q vo .©•; o !©•; o ;: Oi d ;©;•; d ' cn «osi: iO-VOf •; — 'O i ; s • BO .. I* 3' s; © o : in;: — • co • T J - o : o^ i \o ;o; \o ^ col lie*,;; m V"; O "0\? co — ;j ;o.; — !l ^ — o o" P o I a *t — — ! vo cn : — : in -.Ot O .O • O sd:. d d d .. vo 'in ^ H -3 ' 13 ; (N tn C I :CO!i ON ^ • O O —; © —; p • O 0 V i O ,:0!( <n :.0 ' ;:t~v T f T f CN . m CN . CS o o • m ©. © .o"-. © " .. © tn • r- r-• O ren: vO :AO:-O ; CS .: © O; — • Oi-a — Ov CN — 0 0 —• oo CN - H r-::—•:: in oo • oo d d d CN d cn i; T cn o o d d 3 -OH" — tn := — fN r- ; —-'• CS '.C\ : — O O — <N := O (S: \q :oi; o ir>; O O. O O i i(N :ivo^  m :.—• cn 1 i O ^ ' O CO: : CO o o o o \0 —* — O: . O o S.JI o CD a. w 601 £ <l CD 1 S31 c a e D. s it--;: cn r~. vo cn cn rl oo 0 ; i cn :cn; 13 „ , 5 0 3 3 * = " 5 , 5 5 a ~a a 3 a ct. a 3 S £ Z . ' 3 : 3 > r* *^  S ? 5 i ' c 3 - 5 S 3 3 ^ : L i ^ . : C3 :^: a-:.2:: So: . G . Cj tX,:r^:: 3 c c - ~ 3 ' SU; 1, , 3 <C OS « O 2 :"« S . : « 3 ' i O vj 'C .i2.: •o •: Pi-b a s 3 3 S - 3 - s < - c 3 3 3 a. = .a .a .a -a -a s s -3 3 « „ V ^* V - 3 -S ^3 3 u >i 15 15 S 5 g5 3 3 S ^ N W vj <u <u CJ ,:Cj ; SU a.:: co i.en : c vj 3 CJ : 3 «•: ca o. * * l 44 1 J 1 • 1 1 June July August September Figure 10a: Engelmann Spruce Subalpine Fir 1994 seasonal richness calculated by rarefaction with 10 as the standard number of specimens. 7 -i 1 1 1 1 , June July August September Figure 10b: Interior Cedar Hemlock 1994 seasonal richness calculated by rarefaction with 10 as the standard number of specimens. 45 5 i 4 1 uncut partial cut clear cut June July August September Figure 11a: Alpha values of the logarithmic series with standard error for 1994 ESSF collections. Alpha is a measure of the dominance structure of the species assemblage. 8 i 7 6 H 4 -I 3 2 1 0 -•— uncut - O — partial cut -•— old clear cut new clear cut June July August September Figure 1 lb: Alpha values of the logarithmic series with standard error for 1994 ICH collections. Alpha is a measure of the dominance structure of the species assemblage. 46 values indicate greatest differences in dominance structures in July and trends in the dominance of the uncut and partial cut collections are identical throughout the summer. The structure of the assemblages did not change significantly, but in terms of elucidating differences amongst treatments, greater distinction was seen in the early months for the ICH and in the later months for the ESSF. The heterogeneity indices also varied from month to month (Appendix 6 a, b). There was little generality with respect to month, for either forest type or treatment. Intraspecific comparisons The abundance of several species varied greatly over the course of the summer. S. angusticollis was most abundant in August (Figure 12a). C. advena, in both forests, was found in much greater numbers in June or July: the June data indicate that it does not necessarily prefer uncut habitat (Figure 12b). P. adstrictus was most abundant in uncut sites of the ESSF stands in June 1994 (in contrast to the pooled August samples where it was most abundant in clear cuts) (Figure 12c). P. riparius, appeared to prefer clear cuts rather than being a generalist (Figure 12d) (see August collections above). Ten species, while they were few in numbers, were found in the early months in some treatments, but were absent in the August collections. Abundance and in some cases, the presence of individual species varied over the course of the season. While general patterns of diversity can be deciphered, collection of samples in August alone is not a reliable way to elucidate responses of individual species to timber harvesting. 47 12 l r-i |J E S S F uncut E S S F clear cut E S I ICH uncut h:-»yy.i ICH old clear cut H i ICH new clear cut June July August September Figure 12a: Mean number of individuals of Scaphinotus angusticollis trapped in old growth and clear cuts of both forests during the summer of 1994. 12 4 E S S F uncut E S S F clear cut ICH uncut ICH old clear cut ICH new clear cut June July August September Figure 12b: Mean number of individuals of Calathus advena trapped in old growth and clear cuts of both forests during the summer of 1994. 48 ESSF uncut ESSF clear cut ICH uncut ICH old clear cut ICH new clear cut I L June July I I Ip n W August September Figure 12c: Mean number of individuals of Pterostichus adstrictus trapped in old growth and clear cuts of both forests during the summer of 1994. 1 I ESSF uncut 3 ESSF clear cut1 I ICH uncut 1 ICH old clear cut I ICH new clear cut J T June July August September Figure 12d: Mean number of individuals of Pterostichus riparius trapped in old growth and clear cuts of both forests during the summer of 1994. 49 2. Shrub Removal In the two years between the time of shrub removal, and the final collection period, forb cover regrew to be almost 4 times as dense as it was when shrubs were first removed. At that time, in the ESSF forbs were almost twice as thick in treatment plots as they were in control plots. In the ICH, treatment and control plot forbs were equally dense (Appendix 7). No species were found in the control plots that were not found in the shrub removal plots (Table 7). The ICH had more species in total. There were more species in the treatment plots than in the controls of both forests. In terms of overall abundance, there were 1.7 individuals per trap in both the treatment and control of the ICH, but in the ESSF control and treatment there were 6.8 and 9.4 respectively. Richness, evenness and non-parametric measures Richness was the same in control and treatment sites in the ESSF, but in the ICH it was higher in the treatment plots than in controls (Figure 13). The shape of Whittaker plots (Figure 14a) matched the alpha values which did not vary between control and treatment for either forest type (Figure 14b). Shrub removal did not change the dominance structure of the carabid assemblage in either forest type. The heterogeneity index values were higher for treatment than for control in both forests (Table 8). The Shannon-Weiner values for control sites of both forests were similar. As in the tree harvest experiment, the Simpson 50 •-o E CU JS u -a CU *c c tu £ 'u tU "3 > o XI c/3 <U fi !&•• -D s co (U fi (O fi s % 6X3s fi <u ft c/l c E ' u tu & CU X E 3 c c crj j u X ctj H Ov s\D:: cn m slO-v; — J'm's co is* cn SCO:: vo —, CN HMDs CN vo - t O O o CO! CN o O o : © i m :!©; OV CN oo vO oo P©: 00 i r o o O o o o O o o o O O CO CN m © o © d o d d d d - d d d d d d <d" d d d d d o »—i !:C: vo ;"asiii vo i-i'S! CO CO m CO CN CN .•p|Mj;. m CO CO wo »—I 1.3 :';,: O B'-i: ; .r-r<S O ;;*:.:.! o © CN CO •oo. oo CN oo o o CN o;:':'! O f — O ::' O o o - CO o o T i n o O O d d is •;;;.;: d d it3; d d d o d d d sds d d vosi co co CN CN :c©ii; ' i ' CO o r~ CO iKSl :;OV oo o CN o —' CO oc 1 Mir" O co- d d r d d d d d d ca < •2 < 3 ^ Q • e y . O ;.:•';:•-• tJJ I f f a ym ^ i i ^ i O 5^ .-.to &o 'jr .3 ;•<>)!->'S? CO-:; cs:.! • S : •:cj;; tu :V3:;; s to to CJ a htol —^^  ; 1 a • to ; S -c .Jo to . s. :::.to? ^ ! ;:«L: :Ch •cs., '• taT: s .Jo • ; •• •yu-to S tu C -2 a Co •-to; i::2S:: S::C; :;::gi; • .©!:! tu ~.-<m •St •:.-•«!!' to • : .Jo ; to ; '•.Si-: CO rCS -m S :m £ ;:s$: £ :^;:i ^ j i ^ J ^ O O K | ^ i^S, O, ,QS; ft, :;Cti:;: a. to •2 c, cs. •S-: to ; S j -s; i .jo -G ^5 i s; I'CS .Jo ' tO:: ; a <50 S a o o to a •StO:' :::3; cs CS : «o : •S : co on ::tov ^a1 •:£. Ss: s: to to : a tu I f ! cs to ! a ;:;^ o; CO s a :^ :::v VO o ^ CM VO CO:! Ov ?:?:: i co m VO — CN oo s-COs; ss; o Cv • <n CO :-i S O o m o m ''553;:;: CO o ;'os?; -::ra:; CN o o CO —i CN Ov d :;-l?- d ••:' fi d d d O s CN d d i l f l VO ' 51 io H shrub removal i control control shrub removal ESSF ICH Figure 13: Corrected richness estimates for shrub removal experiment. Rarefaction is used to calculate richness based on a standard sample size of 580 specimens. 52 100 n 0.01 J 1 1 1 1 1 , , , , 2 4 6 8 10 12 14 16 18 species rank Figure 14a: Species rank abundance (Whittaker) plots for shrub removal experiment. Percent abundance of each species is plotted on a logarithmic scale against species rank ordered from most abundant to least abundant species. 6 i 4H 3 53 2H I1 control • shrub removal control • shrub removal ESSF ICH Figure 14b: Alpha values of logarithmic series of shrub removal experiment. Alpha is a measure of the dominance structure of the species assemblage. 53 Table 8: Heterogeneity indices for the shrub removal experiment. The Shannon-Weiner is sensitive to changes in abundance of rare species while the Simpson marks changes in the abundance of common species. Engelmann Spruce Subalpine Fir Interior Cedar Hemlock control treatment control treatment Shannon-Weiner (H') 1.25 1.39 1.27 1.51 1/Simpson (1/H) 2.63 2.91 3.06 3.41 54 values varied more than the Shannon-Weiner values, indicating greater changes in the abundance of common species than of rare species. Intraspecific comparisons The following species were more abundant in the plots in which shrub cover remained: in the ICH, C. advena, S. angusticollis ((Figure 15 a), P. ecarinatus (Figure 15b), S. marginatus, P. neobrunnus (Figure 15c), and P. herculaneous, and in the ESSF, S. angusticollis. However, the differences in abundance were not significant according to the Wilcoxon signed rank test. The test did find significant differences in the ESSF for N. crassicornis, P. ecarinatus (Figure 15b), P. neobrunnus (Figure 15d) and probably S. marginatus: abundances for all of these species were higher in the treatment plots. Probabilities are listed in Table 9. 55 1 CD P-co 13 3 T3 ' > '•3 c CD control ESSF shrub removal control ICH . shrub removal Figure 15a: Mean number of individuals (with standard error) of Scaphinotus angusticollis trapped in control and treatment sites of each forest during the shrub removal experiment. control ESSF -shrub removal control ICH — shrub removal Figure 15b: Mean number of individuals (with standard error) of Pterostichus ecarinatus trapped in control and treatment sites of each forest during the shrub removal experiment. 56 1 CD OH 3 "2 c c CD control ESSF shrub removal control ICH shrub removal Figure 15c: Mean number of individuals (with standard error) of Pterostichus neobrunnus trapped in control and treatment sites of each forest during the shrub removal experiment. CU CL, co -3 > •3 c c CD control ESSF shrub removal control ICH shrub removal Figure 15d: Mean number of individuals (with standard error) of Nebria crassicornis intermedia trapped in control and treatment sites of each forest during the shrub removal experiment. 57 Table 9: P values for differences found using the Wilcoxon signed rank test in abundance of species in control and shrub removal sites. Probability values less than 0.05 indicate a significant preference for indicated habitat. Where "preferred habitat" is N/A. , there were too few specimens to determine a preference. Species Forest Preferred habitat Probability (oc = 0.05) Calathus advena ESSF shrub removal 0.31 ICH control 0.17 Leistus ferruginosus ESSF ICH shrub removal N/A. 0.31 Nebria crassicornis ESSF ICH shrub removal N/A. 0.03 Pterostichus ecarinatus ESSF shrub removal 0.01 ICH control 0.18 Pterostichus herculaneous ESSF N/A. ICH control 0.31 Pterostichus neobrunnus ESSF shrub removal 0.01 ICH control 0.12 Pterostichus riparius ESSF shrub removal 0.17: ICH shrub removal 0.12 Scaphinotus angusticollis ESSF control 0.17 ICH shrub removal 0.95 Scaphinotus marginatus ESSF shrub removal 0.05 ICH control 0.68 Trechus chalybeus ESSF shrub removal 0.67 ICH shrub removal 0.31 58 3. Coarse Woody Debris The number of species was higher for the entire ESSF collection than for that of the ICH. This is opposite to the trends for the tree harvest and shrub removal collections (Tables 11 and 12). There was no obvious pattern for numbers of species or specimens amongst locations relative to the logs. Richness and Evenness Richness values calculated by rarefaction (Appendix 8 a, b) revealed no consistant patterns in differences between locations within treatments. Similarly, the Whittaker plots (Appendix 9 a-d) revealed no trends in diversity amongst locations with respect to the log. 59 o O o o CN o in TH o o o ID 3 o u. ca © *o o © • © : O C N © . in © : © l-H © ID 3 O .2 t2 © CNS OH 0 5 l> C U -4-1 C U c c C •..cj.. OX _© c o u CM C U u C 3 -sip O N N O O N o © cn r- O O CN cn cn cn cn CN O N cn in © N O in © O N 00 © so •r © CN CN rt o © d © d © © © © © CN © © © cn oo O cn 00 O N O in o CN CN CN 00 CN O CN CN CN CM O N r t cn ' sc © in CN cn O N N O CN oo CN © c-N O o O N ' cn r t © | © © d — d © © © d d © © o © oo CN q N O m o CN r f © N C 00 cn cn © cn in cn cn o 00 CN © CN r t r t CN O N 00 rt ^t oo CN © d © © d © © © © © © © ©' S O in o . cn cn •r. C O CN © ,rt r t CN S O © CN r t r f CN CN CN © © CN O N cn © © © d © d © © © S O I / -o CN"! 00 cn. cn 00 r-r f r t C N r t 00 CN CN CN CN N O © 00 t-~ cn S O in S O m © rt ^t rt o d ° © © © © © © oi °' d d so m o 00 CN O N O in 00 r-CN cn rt r t © O r t oo CN © o o CN O N cn t-; CN S O m © 00 CN CN © © d ° © © © ° © © d d d © S 21 © cn •n cn m oo © NO ~ O N 9. -I cn m r-© o CN ,,;cni;! • cnK cn ,,;,.; cn -',,': cn cn cn, cn cn cn cn r-NO rt SO © d CN' © j;™ 'sp;: © © © ©-©"•: /.so", :-:.©: © © CN • i->:. •'so;;' © r-so :::; © .'.-.•: f-::i so •,;:. © ©;! • ©;• cn cn cn so "mV. j.cn': rt SO © M., .-di; d in d ,-.'.[ '.' ° ' in d © d C— so: >n' so © -cnV cn.-CN SO ••'©. r— i so .,' '©• "'©. top' o-so CN r ••so.' ':©:;-d i © cn ?'-d:' © ' ' : d © cn;,; cn* r-- SO CN cn cn ,r. r~ '.vo-' CN r~ so CN ,:'SO-l ,'rt!' T" ; CN © f; rt d ' d 4 ©III O l rt so o O -sO 00 !' :J;-:;ii, r-~ §? so , , :Y © 1 ; ; •©; • ' o o cn r SO rt dji © B i l d ip © ;|d| ol © 00::: Vc*Si3 ;;m;;! ;Cnj t— 11! so £d ;l:0:i © SO cn cn oo r-SO rt SD © ? ' © ff CN d d d =5 S •sii; .:s;:; -:,oa l-^ !;! 5 s'JC'sl cu .5 ;!RS]i! :!IC;;; :;;5U!,! •!»pi S O i: ;!::;,.-:;:;:;; cu mi § js3:! ''I !;!6<i! t^ !!^!;! r...6 : Ooi'!!:!fc:: 2Kl t3 !!:':«!;; !!0!i O COV; :.6o-:| ;:S!i ;!:a:i ;!;60! is. e s c -a s:;*s; 3 Kg! S ^ !;!Q5!:! cq isa a «!;! Cr-,' ESi ,Q i:!s2! 3^ ! S ' c«l : O -: •S 1 Sol s : !^ V . '! S -2 .a ^ •n ! S ' .co? r!S;: .'Ig'!1 :;:!;S!-'sis;-;,-.;oi!. s •S 0 s -« o 1 . 1 : | U ! csi;i .LSB .?.OLi S .!!0!:! .:!!Si-; :::Cj, . O ;; .: to -: S ! 00;: S i , =o S .; !!!sa :;!5S:", fiS!:; : S 3 3 Q !;!!^  CNJ :,:::;CNj;: so . Si CNJ -ft, Ifti;;: ft, iftl; •S 03 SO s a :•• ca !!; . . . . P ,!,fc;! ?>!!! . - S O ! ! 53 !' •:• 3 ;;: ;•:]*•;!! T N : : I •;coi ^ ;• C so •~ si;; :T<3 -^ ', iS:;; ca mm J- % 00 E-H m 00 m 00 © r» O N 00 cn co C 1) O 00 60 o © co o o «N ss "O" iO —, — 3 O s, © T 3 "o o o ro o • © © © o ^ 3 o U5 C 3 tU o tu c l/J U <U <u c S 3 e cu u : ^ H ^ C5J3 i o u CM CU •i O e dis O : O : o d' o o vo O ' o ; vo d IO o o VO T f o d vo o o d o d : 0 •O: vo vo vo m O r o d d vo — , io — •oi: vo i IO i o : VO i O d o d _ - H ^O O ' d • vo • o I O : O ' d : o d o\ CO IO CO 0 0 CN oo tN o • o : VO VO o coi: CO o VO CO d d d d d d oo IO oo CN o Ov o oo CN o oo CN O CN s oo CN o r -CN Ov d d d d d d d OV CN VO CO o •oo-CN . o . IO o CN © vo: o CN CN oo d d d rO d ,d: : 0 . CO o VO VCO;. CO O' vo VO CO :co:: O o CO d d d d d CN r~ oo CN Ov CO CN CO m d d d d vo o 0 0 CN o Ov d d d d s s a-.: £ : CU" O : VJ s a c a C3;; •vj: 53 *i -a Oo :g g» • ' CU': 00' 2 a: 2 :Ct3 to oo S S S •< •< "j; vj tU ... & « §•« 1 to C a ¥ o oo ss, § a a a • i -2 -2 ^ -I § cu cu CS oa oa vj : 13 ; : 5> : « 1 I .§ a. c to <3 .a to C a y : 2-tO to • a • £• 2 :'tO:: a? to •Oi! is;. .§», | =0 ..:jju-to a ta ig a g to a tj 'a a c r- -Cj:; -«HH S3 " a <•> la,-2 s: * •2 g | : a to a 5S •5 a t-tu to a -a •2 ra-to 2 to a tu S a a 2 tu -a to a -a •2 a to : 2; 5 .=v a: C. to a •c •2 a to:: 2: to • a i •a • •2: to VJ )«. : VO I. t. V :::*«.:; . : : :*-» : : 1H» ft, °- a. a. o •2 to: a oo a to a •S - c Ig 15 a, a. to a' • a : c: Si S: V J i K a; .a, ^ ' a 3 -§ a •2 tu to 00 to a vj to a ? •a a " •£ a o a -a iS,. vj a « Co t~. CO CO C tu a r OH * 61 DISCUSSION Limitations of pitfall trapping Pitfall trapping was used in this study to assess habitat effects on carabid diversity and to compare relative abundances of individual species between habitats. This collection method has been widely used in arthropod research to collect specimens, particularly those of the orders Coleoptera, Arachnida, and Collembola ( Marshall et al. 1994, Southwood 1966). The popularity of pitfall trapping stems from its convenience: it barely disturbs habitat, it is inexpensive and labour-efficient, and it allows collections of arthropods in numbers suitable for rigorous statistical analyses (Spence and Niemala 1994). While the advantages of pitfall trapping are known, its shortcomings must be considered: Southwood (1966) says that pitfall traps are of little value for the direct estimation of populations, or for the comparison of communities. The basis of these shortcomings rests on the fact that the method relies on both the abundance and activity of insects, and activity is difficult to predict or quantify, since it is influenced by a wide range of factors, which are not equal for all species or individuals. Factors influencing activity may relate to gender or phenology, or may be associated with trapping conditions such as weather, surrounding vegetation, material used for the construction of the trap, or as preservative, or the size, shape and arrangement of traps (Halsell and Wratten 1988, Holmes and Boyce 1993, Drift 1951, Topping and Sunderland 1992, Hayes 1970, Mitchell 1963, Ericson 1979, Greenslade 1964, Spence and Niemala 1994). 62 Baars (1979) and Niemala et al. (1993) demonstrated ways to overcome problems associated with species differences: a reliable measure of the relative density of carabid species in different environments can be obtained by trapping continuously, and by keeping aspects of the trap itself (materials for construction or preservative, or the size, shape and arrangement of traps) constant. While it cannot be argued that traps were set continuously in this study, it is true that over the course of the years of the three experiments, traps were set at least once at all times during the summer, as well as during most parts of winter. On some occasions in the summer of 1994, in the tree harvest experiment, not all traps were set. However, a comparison of richness estimates using all data versus the subset which included only traps that had been set continuously, revealed no difference in trends amongst treatments. Trap materials were the same for all sites, and they were kept in the same arrangement throughout the collection period. Potential problems associated with varying climate conditions were considered. There were no significant differences in the richness estimates of the five consecutive Augusts collections, in spite of inevitable differences in climate from year to year. Greenslade (1964) found that different species of carabids travel in different layers of the vegetation and soil, and concluded that relative carabid abundances should not be compared from different habitats. She found that large ground moving species are more susceptible to being caught in traps cleared of grass. Craig (1995) however, found that microhabitats, even within a site, vary in terms of vegetation density. She found that 63 setting a large number of traps within a site, offset the variation between sites. Furthermore, she compared numbers of Scaphinotus angusticollis in traps surrounded by different kinds and densities of vegetation and found no correlation between the two. While the coarse woody debris experiment had a limited number of traps, the tree harvest and shrub removal experiments had an ample number of traps set (45 for each treatment) through most trapping sessions and were left in place throughout the years. The problem of varying vegetation density from trap to trap was probably overcome, at least in those two experiments, by the use of many traps. Most problems associated with pitfall trapping have been dealt with in this study. However, while year round trapping may allow intraspecific comparisons amongst habitats, no condition equalizes individual species activity such that the interspecific comparisons can be made within an environment. Each species is active to a particular degree and, as such falls into the trap at a particular frequency. There is a problem insofar as both evenness measures and heterogeneity measures (which includes an evenness factor) rely on interspecific measures of abundance, within an environment. Care must be taken when interpreting these indices, as they may be confounded by this problem. Richness estimates, which make no assumption about interspecific abundances, are most reliable. 64 Forest differences It has been hypothesized that temperature is the main factor affecting carabid species distribution in temperate systems, since mean annual soil temperature has a close negative linear relationship to altitude (Butterfield 1996). This suggests that species assemblages from the two forest types should differ, since the ICH is lower in altitude than the ESSF. In terms of richness the ESSF assemblage was basically a subset of the ICH assemblage. There were more species in the ICH, but most species found in the ESSF were also found there. In terms of dominance structure, the old growth assemblages of both forests were similar. In terms of overall abundance, numbers per trap for each forest type, for the pooled August collections of the tree harvest experiment were close. However, over the course of the three studies, many more individuals were collected in the ESSF than in the ICH. The majority of these were trapped in June and July. Some argue that arthropod abundance is positively correlated with rainfall (Nummelin 1989), but Niemala and Spence (1994) stated that the abundance of carabids in particular, is negatively correlated with rainfall. While there was more precipitation in the ESSF throughout the studies, overall, the ICH sites wetter than those in the ESSF (Huggard, pers. comm. 1998). Perhaps the ESSF has a higher abundance of carabids due to drier soil conditions. In summary, ESSF carabid assemblages are less diverse than ICH assemblages, however, there are more individuals in the ESSF. 65 1. Tree Harvest - Treatment effects In the ESSF there was a slight increase in generic diversity and in the ICH, there were almost twice as many genera in the old clear cut sites as there were in uncut areas. This is contrary to the results of a study by Lenski (1982), in which a decrease in intergenic diversity was found with cutting. Several authors have found a significant increase in species diversity after forest cutting (Bultman et al. 1982, Lenski 1982, Niemala et al. 1992, 1993, Boyle 1991, Craig 1995). This was certainly the case with this study. Richness, evenness and heterogeneity indices for both forest types, all indicated higher diversity in clear cuts than old growth stands. Explanations have been offered as to why there are more species in clear cuts: canopy thinning caused by silvicultural practices results in an increase of species diversity due to higher habitat heterogeneity and a reduced competitive exclusion (Martel et al. 1991). Also, if sites are isolated, competition for resources is reduced, since few populations survive at maximum density (Mader 1984). A l l of the Whittaker plots resembled the log series model. This model results if the intervals between the arrival of species in an unsaturated habitat is random rather than regular (Magurran 1988). Alpha, the statistic that describes the shape of the curve, increases as dominance becomes equitable. Lower values for the alpha statistic of uncut and partial cut assemblages indicate that these are assemblages dominated by only a few species. This is similar to the study by Niemala et al. (1993) in which the relative 66 abundance of species in a boreal forest was more evenly distributed in regenerating stands than it was in mature forests. Lower diversity in mature than in clear-cut or regenerating forests seems to be a general pattern for carabids in both temperate and boreal forest. The higher number of species and the more even dominance structure in the regenerating sites produced higher values of heterogeneity indices. While trends in these were similar across treatments, differences were greater between values of the Simpson index than the Shannon-Weiner. The former is known to be sensitive to abundances of the common species while the opposite is true for the latter (Magurran 1988). The very minimal changes in the Shannon-Weiner value probably reflects the fact that no old growth species disappeared completely in the clear cuts. The large differences in diversity amongst treatments, detected by the Simpson index, indicate changes in abundance for many of the more common species. The ESSF clear cuts had been harvested several years before the ICH new clear cuts. The carabid assemblages of the two were very similar in terms of all of the diversity measures, and were closest to one another in the cluster analysis: they had the most species and fewest individuals. This may reflect the similarity in the successional stage of the vegetation: both lacked shrub cover. One important conclusion derived from this similarity, is that the impact of cutting in the ESSF, is manifested over a much longer time than it is in the ICH. 67 There were more individuals in the old clear cut than in the new clear cut in the ICH, which may indicate that the assemblage as a whole is recovering. Unlike the new clear cuts, the old clear cuts had dense shrub and deciduous tree cover. Why the Shannon-Weiner values for the old clear cut and uncut assemblages are so similar, is unknown, especially since the cluster analysis indicated that the ICH old clear cut was dissimilar to all other treatments. There were fewer individuals in partial cut sites than in old growth areas. Hoekstra et al. (1995), found that partially cut areas had fewer individuals than old growth areas and Nummelin (1989) found that in selectively felled areas, arthropod numbers were positively correlated with the percentage of ground vegetation cover. There was some difference between ground vegetation cover in partial cuts and old growth sites: most of the partial cuts had been harvested as long ago as 1979 and 1960 (ICH) or 1980 (ESSF), and were densely covered with shrubs, while the old growth sites were not. Diversity values of the partial cut site assemblages were slightly higher than those of old growth areas, but basically there was little difference between the two. The cluster analysis emphasized their similarities. Partial cutting reduced numbers and increased richness slightly, but it had no dramatic effect on the diversity of carabid assemblages. In terms of carabid diversity, treatment effects were at least equally, if not more conspicuous in August, than they were in other months. This means that analysis of the 68 pooled August collections has probably yielded reasonable conclusions about the responses of carabid assemblages to tree harvesting. Intraspecific comparisons Forest species Forest species are almost all hygrophilic and nocturnal, they avoid direct sunshine as a result of their preference for darkness, and they show little resistance to dryness (Thiele 1977). Lindroth (1962) called Scaphinotus angusticollis a true woodland species. In the pooled August collections, it was far more abundant in uncut stands than in cut blocks in both forests. In a similar study in a Coastal Douglas-Fir forest, Craig (1995) found that this species preferred mature forest, but seemed to recover over time after harvest: she called it a "forest species". S. angusticollis is one of the Cychrines, and is known to be a specialist predator of gastropods, particularly snails (Hengeveld 1980, Digweed 1993). Gastropods originated as marine organisms, and while some have adapted to terrestrial environments, they are basically restricted to moist or humid habitats (Russell-Hunter 1979). Scarcity of this Scaphinotus species in the cutblocks may indicate a lack of available prey. Lemieux (1998), collected carabids in a drier, more northern ESSF forest. He found that S. angusticollis did not respond strictly as a "forest species", since it appeared to use 69 harvested areas successfully. He may have confounded his results as he used a pitfall trap in which insects, rather than falling over the edge, had to crawl through 1.3 cm holes. These holes may not have been large enough for some S. angusticollis which are wider than that with spread legs. S. angusticollis peaked in abundance in August. Thiele (1977) reported that forest dwelling carabids live under cool conditions, but require high temperatures for reproduction. Therefore, reproduction does not take place until late summer and autumn when the nights are warmer. Forest species would thus overwinter as larvae and tend to peak in abundance late in the season. S. marginatus was most abundant in uncut stands, but evidently recovered over time, because the old clear cut abundance was significantly higher than that of the new clear cut. The pattern for this species in the 1994 seasonal collection emphasizes this, which likens it to its congener, S. angusticollis. Calathus advena was the most common species overall. In the pooled August collections, there was more than an 8 fold difference in its abundance in the uncut and clear cut areas of the ESSF and a virtual absence of them in the ICH clear cuts. This species has been called a forest specialist (Niemala and Halme 1992, Spence and Spence 1988, Lindroth 1966), a generalist (Lemieux 1998), and a clear cut specialist (Duschene et al. 1992). In most treatments, in both forests, the greatest number of specimens was found in June or 70 July: if June data alone had been available for this species, it would have appeared to prefer clear cut habitat. Nebria crassicornis intermedia was an old growth specialist. Niemela et al. (1993) found that this species crashed immediately after harvest. They noted its persistance at recently cut sites, which indicates that they survive the physical disturbance of logging and scarification, and maintain dwindling populations in the open sites for a few years. According to Hengeveld (1980) species of this genus are strict predators which specialize on the mites, springtails, spiders and small flies which live at the soil surface. Perhaps, their absence indicates the absence of some or all of these groups. Niemala et al. (1992 , 1993) categorized Pterostichus adstrictus as a habitat generalist, since it was not only the dominant carabid in the young regeneration forests, but was also relatively common in some types of mature forest. In this study, P. adstrictus appeared to be a clear cut specialist upon analysis of the August data. However, its abundance in June 1994, in ESSF uncut stands far exceeded its abundance in any of the clear cut areas at any time. It actually preferred old growth habitat. Clear cut species The removal of the canopy by clear cutting changes the understory vegetation and alters the light, temperature and water conditions in the soil (Martel et al. 1991). Field dwellers 71 are usually xerophilic or euryhygric, are indifferent to light and are thus diurnal and tend to be resistant to dryness (Thiele 1977). The pooled August collections indicated that Pterostichus riparius was a generalist. The 1994 data revealed however, that this species probably preferred clear cuts. In the ESSF it was roughly 8 times more abundant in clear cut sites in June than it was in uncut sites. Niemela et al. (1992, 1993) found that Agonum, Amara, Bembidion and Harpalus species were almost entirely restricted to regeneration sites and called them meadow species. Almost all specimens of these genera were found in clear cuts. Sericoda quadripunctata and Harpalus species are known to colonize burned sites (Holliday 1991b). In these collections, Harpalus species were found in most clear cuts, but S. quadripunctata was found exclusively in cutblocks that had been burned. Insects that have lost the ability to fly are often associated with habitats that have remained stable through time (Lattin 1993). Bembidion jocylyn is brachypterous and yet, surprisingly, it was found exclusively in clear cuts. The dispersal power of brachypterous species is often thought to be low, but studies on spatial behaviour have shown that some species are able to cover hundreds of metres within a few days (Baguette et al. 1993). Thiele (1977) concurred with this notion when he said, carabids are capable of such feats of movement on the ground, that their distribution over wide areas appears to be possible without flight. Although dispersal by air is possibly the most important method of 72 colonization, brachypterous individuals may move anthropocorously along the roads (Kinnunen et al. 1996). The presence of this flightless animal in clear cuts therefore is not surprising. Habitat generalist species Pterostichus species tend to be polyphagous and as such, may be habitat generalists (Digweed 1993). Most of the Pterostichus species were able to survive in all treatments (except new clear cuts) according to the August collections and the data from 1994 did not suggest otherwise. Species in partial cuts Significant differences were seldom detected between the abundances of species in old growth stands and partial cut blocks. However, as with several of the species above, many of these species were more abundant in June or July, and in those months, clearly were more abundant in old growth sites than in partial cuts. An important conclusion that is drawn from this work, is that assessments of species' responses to tree harvest, based solely on August collections can not be trusted. 73 2. Shrub removal According to the results of vegetation surveys, the number of insect species is influenced by the heterogeneity of vegetation (Southwood et al. 1979, Refseth 1980, Parmenter and MacMahon 1984, Webb et al. 1984). The complexity and areal extent of vegetation architecture contributes to the diversity and availability of arthropod feeding sites, oviposition sites, overwintering sites and enemy free space. In addition, the vertical and horizontal components of vegetation structure interact with abiotic environmental factors to produce microclimes suitable for arthropod colonization and survival (Parmenter and MacMahon 1984). If shrubs are an important part of the microenvironment of carabids, then their removal should result in changes in carabid diversity and relative species abundances. Indeed there were changes in carabid diversity in both forests with shrub removal. In the ESSF, there was a change in overall abundance, although proportional abundance of species did not change; in the ICH, the change in diversity was in the form of increased richness. Heterogeneity measures were higher in treatment sites for both forests. The results of this study are not consistant with other shrub-beetle interaction studies. By manipulating habitat in a shrub-steppe ecosystem, Parmenter and MacMahon (1984) found that the shrub canopy was of no detectable importance in determining the abundance and distribution of ground beetles. In a study in the Sahara, Larmuth (in Parmenter and MacMahon 1984) found no beetles under shrubs, but 57 individuals (3 74 species) in clumps of grass. The lack of correlation was attributed to the fact that the animals may be physiologically adapted to the extremes of their environment and do not require refugia: the small body size of a beetle permits a forb or grass clump to provide the animal with a suitable microenvironment. If it were true that carabids did not require shrubs, their response to shrub removal in this study, is correlative rather causative. The increase in carabid diversity should perhaps be attributed to the increase in the horizontal components of the vegetation structure which came with increased forb cover. Parmenter and MacMahon (1984) conceded that while shrub architecture per se does not directly influence the ground beetle community, shrubs may be important to beetles in the long run by indirectly providing food resources and a greater diversity of herbaceous vegetation. Huggard (pers. comm. 1998) hypothesized that this response of carabid beetles indicates differences in characteristics of the respective biogeoclimatic zones, akin to the difference between tropical and temperate systems. If the growth and diversification of forbs represents a display of increased productivity, the ESSF carabid assemblage has responded like a temperate system by increasing the number of individualsof constituant species; the ICH assemblage has, on the other hand, responded like a tropical system by increasing in richness. 75 3. Coarse woody debris Large and well-rotted snags and logs are conspicuous features of structurally complex, old growth forests (Stelfox 1993). This coarse woody debris has many ecological functions. It provides a substrate for fungus and tree seedling establishment in many forest types. Both vertebrates and invertebrates use coarse woody debris for feeding, reproduction and cover. Coarse woody debris plays an important role in nutrient cycling, and it has received attention recently because of the potential impact on the global carbon cycle of the large quantities of carbon stored within it (Wells and Trofymow 1997, Parmenter and MacMahon 1984). The amount and distribution of coarse woody debris and fine litter are changed by harvesting, slash burning and subsequent mechanical site preparation (Thomas 1979). Haila et al. (1994) found a dramatic difference between carabid assemblages of young and mature forest-stages, particularly in groups dependant upon decaying wood. Carabid species for which coarse woody debris is important ought to be found in close proximity to deadfall. The results of this experiment showed no particular patterns in terms of species' proximity to dead logs. However, there were many aspects of this study that may have confounded results. It is possible that the size of debris chosen was inappropriate. Also, while it is reported that some species (Goulet 1974) are often found in decaying logs, it does not necessarily follow that these species will be found on the ground, near a log. Maybe 76 species with an affinity for decaying matter would be found inside the debris, or near either smaller fragments or debris with more advanced decay. The data from 1994 (above) showed that species assemblages of early summer were different than those of late summer. Data from the coarse woody debris experiment were pooled in order to have large enough sample sizes for analysis. Because the traps in this experiment were set once in the spring and twice in August, there may have been a paucity of spring abundant species. Some of these may have had more distinct habitat preferences with respect to woody debris: early spring collections may have revealed patterns for species which were not present in these collections. Finally, Simberloff (1978) explained that using large sample sizes for rarefaction produces reliable richness estimates, while the use of small ones may not. The sample size for the coarse woody debris richness estimation was 10. The richness estimates may therefore have been less accurate than the richness estimates for the shrub removal or the tree harvest experiments, which were based on sample sizes of 580 and 40, respectively. It is possible that the evenness measures are unreliable as well, since pitfall trapping is not conducive to interspecific comparisons. 77 CONCLUSIONS 1. As predicted, the ESSF and ICH sites had many species in common. Furthermore, there were differences in the assemblages: generally ESSF sites had higher numbers of individuals and fewer species than ICH sites. 2. The prediction that logging has a positive effect on species diversity was supported by this study. Furthermore, the total number of carabid individuals was reduced by tree harvest. Interestingly, the impact of clear cutting in the ESSF endured longer than that in the ICH. In the ICH, the abundance of individuals seemed to be recovering in older clear cuts, but many forest species were still sparse. 3. As expected, partial cutting reduced the abundance of individuals, but had little effect on overall diversity. 4. As predicted, forest species decreased, but did not disappear in clear cuts; species typical of open habitats increased in clear cuts and many species appeared (in small numbers) that had not been present in old growth habitat; habitat generalists were common in old growth, partial and clear cut sites, but were rarely found in ICH new clear cuts. (The use of August collections yielded reasonable results in terms of the responses of carabid assemblages to tree harvest. However, information gained from the August 78 collections was not complete in determining effects of harvest practices on individual species. Seasonal differences in abundance were needed to draw conclusions.) 5. As predicted, carabid diversity was effected by shrub removal. Surprisingly, the response was positive overall. In the ESSF, there was an increase in the number of individuals; in the ICH, the number of species increased. Apparently very few species depend on shrubs directly: most individual species were either unaffected or responded favorably to the removal of shrubs. An increase in the herb layer following shrub removal may have been the actual stimulus for their response, and should be considered in drawing conclusions about these results. 7. Contrary to my prediction, the results of this study do not indicate that carabid beetles have any particular affinity for, or aversion to coarse woody debris. Sample size and diversity measures, duration of the collection period or nature of debris may have been inappropriate for this study. 79 L I T E R A T U R E C I T E D Adis, J. and O.R. Schubart. 1985. Ecological research on arthropods in central Amazonian forest ecosystems with recommendations for study procedures. In Trends in ecological research in the 1980s. Eds. J.H. Colley and F.B. Golley. Natio Conference Series I: Ecology. Plenum Press, New York. pp. 111-114. Anderson, T., S. Ligaard, T. Pederson and G.E.E. Sol. 1990. Pitfall catches of Carabidae and Staphlinidae (Coleoptera) in a temporarily protected forest area on the Eidanger peninsula, Telemark, S.E. Norway. Fauna norv. Ser. B. 37: 13-22. Baars, M.A. 1979. Catches in pitfall traps in relation to mean densities of carabid beetles. Oecologia 41: 25-46. Baguette, M . and S. Gerard. 1993. Effects of spruce plantations on carabid beetles in southern Belgium. Pedobiologia 37: 129-140. Boyle, T.J.B. 1991. Biodiversity of Canadian Forests: Current status and future challenges. The Forestry Chronicles 68(4): 444-453. Bultman, T.L., G.W. Uetz and A.R. Brady. 1982. A comparison of cursorial spider communities a along a successional gradient. J. Arachnol. 10:23-33. Burton, P.J., A.C. Balisky, L.P. Coward, S.G. Cumming and D.D. Kneeshaw. 1992. The value of managing for biodiversity. The Forestry Chronicle 68(2): 225-236. Butterfield, J. 1996. Carabid life-cycle strategies and climate change: a study on an altitude transect. Ecological Entomology 21: 9-16. Callicott, J.B. 1986. On the intrinsic value of non-human species. In The Preservation of species. The Value of Biological Diversity. Ed. B.G. Norton. Princeton University Press, Princeton, N.J. pp. 138-172. Coates, K.D., S. Haessler, S. Lindeburgh, R. Pojar, A.J . Stock. 1994. Ecology and silviculture of interior spruce in British Columbia. FRDA Rep. 220. Forestry Canada and B.C. Ministry of Forestry, Victoria, original not consulted Colwell, R.K. and J.A. Coddington. 1994. Estimating terrestrial biodiversity through extrapolation. Philosophical Transactions of the Royal Society (Series B) 345: 101-118. 80 Craig, K . G . 1995. Variation in Carabid Community Structure Associated with Coastal Douglas-Fir Forest Successional Stages. M . Sc.. Thesis, University of British Columbia, pp. ii to 56. den Boer, P.J. 1986. Carabids as Objects of Study. In Carabid Beetles: Their Adaptations and Dynamics. Gustav Fischer. New York. pp. 539-551. Desender, K. and H.Turin. 1989. Loss of habitat and changes in the composition of the ground and tiger beetles fauna in four west European countries since 1950 (Coleoptera: Carabidae, Cicindelidae). Biological Conservation 48: 277-94. Digweed, S.C. 1993. Selection of terrestrial gastropod prey by Cychrine and Pterostichine ground beetles (Coleoptera: Carabidae). The Canadian Entomologist 125: 463-472. Drift, J. van der (1951). Analysis of the animal community in a beech forest floor. Tijdschr. Ent. 94: 1-168. (Original not consulted) Duschene, L .C . and R.S. MacAlpine. 1994. Using Carabid Beetles (Coleoptera: Carabidae) as a Means to Investigate the Effect of Forestry Practices on Soil Diversity. PNFI Technical Reports. Forestry Canada, pp. 1-10. Ericson, D. 1979. The interpretation of pitfall catches of Pterostichus cupreus and P. melanarius (Coleoptera, Carabidae) in cereal fields: Pedobiologia, 19: 320-328. Fazekas, J., F. Kadar and G.L. Lovei. 1992. Comparison of ground beetle assemblages (Coleoptera: Carabidae) of an abandoned apple orchard and the bordering forest. Acta Phytopathologica Hungarica 27(1-4): 233-238. Gold, W. 1985. Logging As It Was: A Pictorial History of Logging on Vancouver Island. Morriss Publishing Ltd. Victoria, British Columbia. Gould, E. 1975. Logging; British Columbia's Logging History. Hanock House Publishers. Saanichton, British Columbia, Canada. Goulet, H. 1974. Biology and relationships of Pterostichus adstrictus Eschscholtz and P. pensylvanicus Leconte (Coleoptera, Carabidae). Quaest. Ent. 10: 3-33. Greenslade, P.J.M. 1964. Pitfall trapping as a method for studying populations of Carabidae (Coleoptera). Journal of Animal Ecology 33: 301-310. Haila,Y., I.K. Hanski, J.K. Niemela, P. Punttila, S. Raivio and H . Tukia. 1994. Forestry and the boreal fauna: matching management with natural forest dynamics. Ann. Zool. Fennici 31: 187-202. 81 Halsall, N.B. and S.D. Wratten. 1988. The efficiency of pitfall trapping for polyphagous predatory Carabidae. Ecological Entomology 13: 293-299. Hammond, H. 1993. Forest practices: Putting holistic forest use into practice. In Touch Wood: B.C. Forests at the Crossroads. Eds. K. Drushka, B. Nixon, R. Travers. Harbour Publishing Co. Ltd., Madeira Park, B.C. pp. 96-136. Hayes, W.B. 1970. The accuracy of pitfall trapping for the sand-beach isopod Tylos punctatus. Ecology 51(3): 514-516. Hengeveld, R. 1980. Qualitative and quantitative aspects of the food of ground beetles (Coleoptera, Carabidae): a review. Netherlands Journal of Zoology 30(4): 564-584. Hicks, CR. 1993. Fundamental Concepts in the Design of Experiments. Caunders College Publishing. New York. Hoekstra, J.M., R.T. Bell, A.E. Launer and D.D. Murphy. 1995. Soil arthropod abundance in coast redwood forest. Effect of selective timber harvest. Environmental Entomology 24(2): 246-252. Holliday, N.J. 1991. The carabid fauna (Coleoptera: Carabidae) during postfire regeneration of boreal forest: properties and dynamics of species assemblages. Canadian Journal of Zoology 70: 440-452. Holmes, P.R. and D.C. Boyce. 1993. The ground beetles (Coleoptera, Carabidae) fauna on Welsh peatland biotopes: Factors influencing the distribution of ground beetles and conservation implications. Biological Conservation 63: 153-161. Huggard, D. and W. Klenner. 1996. The effects of forest management on shrews and ground-dwelling arthropods: working plan. Wildlife Ecology Research, Forest Sciences Section. Ministry of Forests. Kamloops, B.C. Janzen, D.H. 1987. Insect diversity of a Costa Rican dry forest: why keep it and how? Biol. J. Linn. Soc. 30: 343-356. Jarosik, V. 1991. Are diversity indices of carabid beetle (Coleoptera: Carabidae) communities useful, redundant, or misleading. Acta Entomol. Bohemoslov. 88: 273-279. Jennings, D.T., M.W. Houseweart and G.A. Dunn. 1986. Carabid beetles (Coleoptera: Carabidae) associated with strip clear cut and dense spruce-fir forests of Maine. The Coleopterists Bulletin. 40(3): 251-263. 82 Kavanaugh, D.H., T.L. Erwin and J. McDowell. 1998. A new species of Bembidion subgenus Lionepha Casey (Coleoptera: Carabidae: Bembidiini: Bembidiina) from British Columbia, with a reanalysis of phylogenetic relationships among species of the subgenus. Submitted. ' Kim, K.C. 1993. Biodiversity, conservation and inventory: why insects matter. Biodiversity and Conservation 2: 191-214. Kinnunen, H., K. Jarveleinen, T. Pakkala and J. Tiainen. 1996. The effect of isolation on the occurrence of farmland carabids in a fragmented landscape. Ann. Zool. Fennici. 33: 165-171. Koen, J.H. and T.M. Crowe. 1987. Animal-habitat relationships in the Knysna Forest, South Africa: discrimination between forest types by birds and invertebrates. Oecologia 72:414-422. Krebs, C.J. 1989. Ecological Methodology. Harper Collins, New York. . Kuusipalo, J. and J. Kangas. 1994. Managing Biodiversity in a Forestry Environment. Conservation Biology 8(2): 450-460. Lattin, J.D. 1993. Arthropod Diversity and Conservation in Old-Growth Northwest Forests. American Zoologist 33: 578-587. Laurence, W.F., S.G. Laurance, L.V. Ferreira, J.M. Rankin-de Merona, C. Gascon, T.E. Lovejoy. 1997. Biomass collapse in Amazonian forest fragments. Science 278: 1117-1118. Lehmkuhl, D.M., H.V. Danks, V .M. Behan-Pelletier, D.J. Larson, D.M. Rosenberg, and I.M. Smith. 1984. Recommendations for the appraisal of environmental disturbance: some general guidelines, and the value and feasibility of insect studies. Biological survey of Canada (Terrestrial Arthropods). Lemieux, J.P. 1998. Species and Assemblage Responses of Carabidae (Coleoptera) to forest harvesting: contrasting clear cut and patch retention removals in high-elevation forests of central British Columbia. Masters Thesis, University of Northern British Columbia, pp. 67-106. Lenski, R.E. 1982. The impact of forest cutting on the diversity of ground beetles (Coleoptera: Carabidae) in the southern Appalachians. Ecological Entomology 7: 385-390. Lindroth, C.H. 1961-1969. The ground beetles of Canada and Alaska Opuscula Entomologica, SupplementumXX-XXXIV. pp 1-1192. 83 Luoma, J.R. 1991. A wealth of forest species is found underfoot. New York Times. July2, 1991. Mader, H.J. 1984. Animal habitat isolation by roads and agricultural fields. Biological Conservation 29: 81-96. Magurran, A. E. 1988. Ecological Diversity and its Measurement. Princeton Univ. Press., Princeton, N.J. Marshall, S.A., R.S. Anderson, R.E. Roughly, V. Behan-Pelletier and H.V. Danks, 1994. Terrestrial Athropod Diversity: Planning a study and Recommended Sampling Techniques. Entomological Society of Canada, Supplement. 26(1) pp. 1-33. Marshall, V.G., D. Mckey Fender, J.V. Mathews Jr. and A.D. Tomlin. 1982. Status and research needs of Canadian soil fauna. Bulletin of the Entomological Society of Canada. Supplement 14. pp. 1-5. Martel, J., Y. Mauffette and S. Tousignant. 1991. Secondary Effects of Canopy Dieback: the Epigeal Carabid Fauna in Quebec Appalachian Maple Forests. Canadian Entomologist 123:851-859. McLeod, J.M. 1980. Forests, Disturbances, and Insects. Canadian Entomologist 112: 1185-1192. McNeely, J.A. 1994. Lessons from the past: forests and biodiversity. Biodiversity and Conservation 3: 3-20. Meidinger, D. and J. Pojar. 1991. Ecosystems of British Columbia. Special Report Series number 6. BC Ministry of Forests. Victoria, B.C. Miller, J.C. 1993. Insect natural history, multi-species interactions and biodiversity in ecosystems. Biodiversity and Conservation 2: 233-241. Mitchell, B. 1963. Ecology of two carabid beetles, Bembidion lampros (Herbst) and Trechus quadristriatus (Schrank) ii. Studies on populations of adults in the field, with special reference to the technique of pitfall trapping. Journal of Animal Ecology 32: 377-392. Morris, M.G. and W.E. Rispin. 1988. A beetle fauna on oolotic limestone grassland, and the responses of species to conservation management by different cutting regimes. Biological Conservation 43: 87-105. Myers, N. 1979. The Sinking Ark: a New Look at the Problem of Disappearing Species. Pergamon Press, Oxford. 84 Neave, E. 1996. An analysis of beetle (Coleoptera) diversity in clear cut and old growth black spruce. Masters thesis. University of New Brunswick, pp. ii-88. Niemela, J. and E. Halme. 1992. Habitat associations of carabid beetles in fields and forests on the Aland Islands, S.W. Finland. Ecography 15: 3-11. Niemela, J., J.R. Spence, D.L. Langor, H . Tukia and Y . Haila. 1992. Logging and boreal ground beetle assemblages on two continents: implications for conservation. In Perspectives in Insect Conservation. Eds. Gaston, K.J . et al. Intercept Publishers Limited, Andover, U.K. Niemela, J., D. Langor and J.R. Spence. 1993. Effects of clear-cut harvesting on Boreal ground beetle assemblages (Coleoptera: Carabidae) in western Canada. Conservation Biology 7(3): 551-561. Niemela, J. and J.R. Spence. 1994. Distribution of forest dwelling carabids (Coleoptera): spatial scale and the concept of communities. Ecography 17: 166-175. Noss, R.F. 1990. Indicators for monitoring biodiversity: a hierarchical approach. Conservation Biology 4(4): 355-364. Nummelin, M . 1989. Seasonality and effects of forestry practices on forest floor arthropods in the Kibake Forest, Uganda. Fauna norv. Ser. B. 36: 17-25. Parmenter, R.R. and J.A. MacMahon. 1984. Factors influencing the distribution and abundance of ground -dwelling beetles (Coleoptera) in a shrub-steppe ecosystem: the role of shrub architecture. Pedobiologia 26: 21-34. Pearson, D.L. 1994. Selecting indicator taxa for the quantitative assessment of biodiversity. Phil. Trans. R. Soc. Lond. Biol. Sci. 345(1311): 5-12. Pianka, E.R. 1966. Latitudinal gradients in species diversity: a review of concepts. The American Naturalist 100(910): 33-46. Pickett, S.T.A. and P.S. White. 1985. The Ecology of Natural Disturbance Dynamics. Academic Press, Toronto. Pojar, J. 1993. Terrestrial Diversity of British Columbia. In Our Living Legacy: Proceedings of a Symposium on Biological Diversity. Eds. M . A . Fenger, E .H. Miller, J.F. Johnson and E.J.R. Williams. Royal British Columbia Museum, Victoria, B.C. pp. 177-190. 85 Refseth, D. 1980. Ecological analyses of carabid communities-potential use in biological classification for nature conservation. Biological Conservation 17: 131-141. Russell-Hunter, W.D. 1979. A Life of Invertebrates. MacMillan Publishing Co., Inc. New York. Samways, M.J. 1992. Some comparative insect conservation issues of north temperate, tropical, and south temperate landscapes. Agriculture, Ecosystems and Environment 40: 137-154. Samways, M.J. 1993. Insects in biodiversity conservation: some perspectives and directives. Biodiversity and Conservation 2: 258-282. Scudder, G.G.E. 1993. Biodiversity over time. In Our Living Legacy: Proceedings of a Symposium on Biological Diversity Eds. M . A . Fenger, E .H. Miller, J.A. Johnson and E.J.R. Williams, R.B.C.M. , Victoria, B.C. pp. 109-126. Scudder, G.G.E. 1994. Terrestrial Arthropod Biodiversity: Planning a Study and Recommended Sampling Techniques. Biological Survey of Canada. Shafer, C.L. 1995. Values and shortcomings of small reserves. Bioscience 45(2): 80-88. Simberloff, D. 1978. Use of rarefaction and related methods in ecology. In Biological Data in Water Pollution Assessment: Quantitative and Statistical Analyses. American society for testing and materials. Eason, Ma. Smith, S.M. 1990. The greening of the forest: forest pest management into the 21st century. Proceedings of the Entomological Society of Ontario. 121: 49-60. Southwood, T.R.E. 1966. Ecological Methods with Particular Reference to the Study of Insect Populations. Methuen, London. Southwood, T.R.E., V . K . Brown, and P .M. Reade. 1979. The relationships of plant and insect diversities in succession. Biological Journal of the Linnean Society 12: 327-348. Spence, J. R. and J. K. Niemela. 1994. Sampling Carabid Assemblages with pitfall traps: the madness and the method. The Canadian Entomologist, 126. 881-894. Spence, J.R. and D.H. Spence. 1988. Of Ground-Beetles and Men: Introduced Species and the Synanthropic Fauna of Western Canada. Mem. Entomological Society of Canada 144: 152-168. 86 Stelfox, J.B. 1993. Boreal forests, Biodiversity, and Logging. In Forestry on the Hi l l . Special Issue #5. Ottawa, Ontario. Pp. 33-36. Stork, N.E. 1988. Insect diversity: facts, fiction and speculation. Biological Journal of the Linnean Society. 35: 321-337. Thiele, H .U. 1977. Carabid Beetles in their Environments. Springer-Verlag, Berlin. Thomas, J.W. 1979. Wildlife habitats in managed forests in the blue mountains of Oregon and Washington. Amer. Midi. Nat. 106: 119-125. Topping, C.J. and K.D. Sunderland. 1992. Limitations to the use of pitfall traps in ecological studies exemplified by a study of spiders in a field of winter wheat. Journal of Applied Ecology 29: 485-491. Weaver, J.C. 1995. Indicator species and scale of observation. Conservation Biology 9(4): 939-942. Webb, N.R., R.T. Clarke, and J.T. Nicholas. 1984. Invertebrate diversity on fragmented Calluna-heathland: effects of surrounding vegetation. Journal of Biogeography 11: 41-46. Wells, R.W. and J.A. Trofymow. 1997 Coarse woody debris in chronosequences of forests on southern Vancouver island. Information report BC-X-375 Canadian Forest Service Pacific Forestry Center, Victoria, British Columbia. Whitehead, R. and G.J. Harper. 1998. A comparison of four treatments for weeding Engelmann spruce plantations in the Interior Cedar Hemlock Zone of British Columbia: ten years after treatment. Information Report BC-X-379. Canadian Forest Service. Pacific Forestry Center, Victoria, British Columbia. Williams, J.T. 1991. International aspects of biodiversity. The Forestry Chronicle 68(4): 454-458. Wilson, E.O. 1984. Biophilia. Harvard University Press, Cambridge, Mass. Wilson, E.O. 1985. The Biological Diversity Crisis. Bioscience. 35(11): 700-706. Wilson, E.O. 1988. The coming pluralization of biology and the stewardship of systematics. Bioscience 39: 242-245. Wilson, E.O. 1989. The coming pluralization of biology and the stewardship of systematics. Bioscience 39: 242-245. 87 Wilson, E.O. 1992. Biodiversity: Challenge, science, opportunity. American Zoologist 32: 1-7. Winchester, N . N . and G.G.E. Scudder.1993. Methodology for Sampling Terrestrial Arthropods in British Columbia. Resources Inventory Committee. B.C. Zar, J .H. 1996. Biostatistical Analysis. Prentice-Hall Inc., Englewood Cliffs, N J . U.S.A. 88 APPENDICES Appendix 1: Collection dates for all experiments Experiment Year Dates Tree harvest 1991 17 August - 14 September 1992 31 July - 27 August 1993 19 February - 25 March * 4 August - 2 September 4 December - 4 February * 1994 30 May - 27 June (1/3 rings of traps set) 27 June - 25 July (2/3 rings of traps set) 8 August - 6 September 6 September - 5 October 15 December - 13 February * 1995 7 June - 5 July 10 August - 7 September 1996 Jan - March * 10 June - 9 July (ICH) 24 June - 22 July (ESSF) Shrub removal 1994 8 August - 1 September 15 December - 13 February * 1995 7 June - 5 July 10 August - 7 September 1996 January - March * 10 June - 9 July (ICH) 24 June - 22 July (ESSF) Coarse woody debris 1994 8 August - 5 September 1995 10 August - 7 September 1996 10 June - 9 July (ICH) 24 June - 22 July (ESSF) N o adult carabids were trapped in winter sessions. 89 Appendix 2: Corrected richness estimates (40) for each August collection. ESSF ICH uncut partial cut clear cut uncut partial cut old clear cut new clear cut 1991 5.58 7 9.80 4.92 5.35 5.38 6.75 1992 5.58 7.5 8.19 4.43 5.59 7.93 8.93 1993 4.86 6.34 9.52 4.38 5.13 4.87 10.69 1994 3.94 5.64 8.7 4.80 4.13 8.23 8.49 1995 5.48 7.33 8.7 5.11 5.59 6.11 8.67 F ratio = 0.107, p < 0.979, °= = .05, d. f. = 4, 30 90 uc ESSF pc cc uc ICH pc occ ncc Appendix 3 a: Mean number of individuals (with standard error) of Scaphinotus marginatus trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type, uc = uncut, pc = partial cut, cc = clear cut, occ = old clear cut, ncc = new clear cut ESSF H = 1.966, p< 0.374 ICH H = 25.89, p < 0.001, uc^tpc, pc ;£occ, occ £ ncc 91 1 s-l S-. U OH -a -a c <U J L J L uc pc cc uc pc occ ncc ESSF ICH Appendix 3b: Mean number of individuals (with standard error) of Pterostichus neobrunnus trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type, uc = uncut, pc = partial cut, cc = clear cut, occ = old clear cut, ncc = new clear cut ICH H = 15.21, p < 0.002, uc * ncc, pc * ncc 92 0.3 0.2 H 0.1 0.0 uc E S S F pc cc uc I C H pc occ T ncc _I_ Appendix 3c: Mean number of individuals (with standard error) of Pterostichus ecarinatus trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type. uc = uncut, pc = partial cut, cc = clear cut, occ = old clear cut, ncc = new clear cut ESSF H =0.570, p < 0.752 ICH H = 9.823, p < 0.020, uc ± pc 93 0.04 •2 0.02 uc pc cc uc pc occ ncc ESSF ICH Appendix 3d: Mean number of individuals (with standard error) of Leistus ferruginosus trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type, uc = uncut, pc = partial cut, cc = clear cut, occ = old clear cut, ncc = new clear cut ESSF H = 1.633, p< 0.422 94 0.1 A 0.0 uc ESSF pc cc uc ICH pc occ ncc Appendix 3e: Mean number of individuals (with standard error) of Synunchus impunctatus trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type, uc = uncut, pc = partial cut, cc = clear cut, occ = old clear cut, ncc = new clear cut ICH H = 9.545, p < 0.023, pc * ncc 95 uc ESSF Appendix 3f: Mean number of individuals (with standard error) of Trechus chalybeus trapped in each forest and treatment during pooled August collection periods (1991-1995). The Kruskal Wallis test statistic, H , and the probability, p, are the result of treatment comparison within each forest type, uc = uncut, pc = partial cut, cc = clear cut, occ = old clear cut, nc = new clear cut ESSF H = 3.28, p< 0.194 ICH H = 9.02. p < 0.029 96 8 . 6 tu c/3 I c •-4 June July August September Appendix 4a: 1994 ESSF uncut species richness calculated by rarefaction with all data (using 30 as the standard number of specimens) and only ring 2 (using 20 as the standard number of specimens). to + C CJ 74 6 H June July August September Appendix 4b: 1994 ICH uncut species richness calculated by rarefaction with all data (using 30 as the standard number of specimens) and only ring 2 (using 20 as the standard number of specimens). 97 100 H o c T3 C P 03 10 0.1 0 —•— June • • < > • • July — T — August —v • • September o- •o-10 12 2 4 6 8 species rank Appendix 5a: ESSF uncut species rank abundance (Whittaker) plots for four summer months of 1994. Percent abundance of each species is plotted against species rank ordered from most abundant to least abundant species. 14 100 -i u c T3 C 10 June O • July -•— August '•• September o 0 8 10 12 species rank Appendix 5b: ESSF partial cut species rank abundance (Whittaker) plots for four summer months of 1994. Percent abundance of each species is plotted against species rank ordered from most abundant to least abundant species. 98 — • - - June • o • July — - August — V September 0 2 4 6 8 10 12 14 16 species rank Appendix 5c: ESSF clear cut species rank abundance (Whittaker) plots for four summer months of 1994. Percent abundance of each species is plotted against species rank ordered from most abundant to least abundant species. 99 100 10 tu o c a T3 c 3 X> C3 0.1 — • - • June • o - July —T— • August —V' September 0 2 4 6 8 10 species rank Appendix 5d: ICH uncut species rank abundance (Whittaker) plots for four summer months of 1994. Percent abundance of each species is plotted against species rank ordered from most abundant to least abundant species. 12 100 O C ctj c 3 H H 0 • o - July • August —v • September O 8 2 4 6 species rank Appendix 5e: ICH partial cut species rank abundance (Whittaker) plots for four summer months of 1994. Percent abundance of each species is plotted against species rank ordered from most abundant to least abundant species. 100 10 100 o § 10 T3 C 3 -§ —•- • June • o - July — • August —v- September 0 2 4 6 8 10 12 species rank Appendix 5f: ICH old clear cut species rank abundance (Whittaker) plots for four summer months of 1994. Percent abundance of each species is plotted against species rank ordered from most abundant to least abundant species. • June ••<>•• July —•— August —v •,; September 0 2 4 6 8 10 12 14 16 species rank Appendix 5g: ICH new clear cut species rank abundance (Whittaker) plots for four summer months of 1994. Percent abundance of each species is plotted against species rank ordered from most abundant to least abundant species. 101 c tu c o c c c/3 - O ESSF uncut • o - ESSF partial cut -•- ESSF clear cut ICH uncut • ICH partial cut ICH old clear cut ICH new clear cut 0 June July August September Appendix 6a: The heterogeneity statistic, the Shannon-Weiner index (H') for Carabid assemblages trapped throughout the summer of 1994. Heterogeneity statistics are based on both richness and evenness. 7 c 4 o H CO t/3 ESSF uncut • o - ESSF partial cut — • - ESSF clear cut - V - ' ICH uncut • ICH partial cut - a - - ICH old clear cut ICH new clear cut June July August September Appendix 6b: The heterogeneity statistic, the Simpson index (1/H) for Carabid assemblages trapped throughout the summer of 1994. Heterogeneity statistics are based on both richness and evenness. 102 Appendix 7: Percent forb cover in control and treatment plots of shrub removal experiment. date shrubs measured % cover collection date control treatment ESSF after shrub removal 1994 24.7 10.3 August August 1995 24.7 42.8 July, August August 1997 24.7 52.8 ICH after shrub removal 1994 59.4 13.2 August August 1995 59.4 51.1 July, August August 1997 59.4 67.2 103 14 12 A 10 6 H 0 0 0 1 * I 0 o clear cut • partial cut 0 150 200 300 50 100 distance (cm) Appendix 8a: Corrected richness for 6 locations with respect to logs in ESSF clear and partial cuts. Rarefaction is used to calculate richness based on standard sample size of 50 specimens. 8 7 6 5 -4 -0 Q O new clear cut • old clear cut 0 3 H 2 H 1 1 1 1 1 1 0 50 100 150 200 300 distance (cm) Appendix 8b: Corrected richness for 6 locations with respect to logs in ICH new and old clear cuts. Rarefaction is used to calculate richness based on standard sample size of 15 specimens. 104 100 ^ 1 0 o c a T3 C 3 JO 03 — • - 0 • o- 50 • • » • 100 • • o • 150 • • -A- • 200 300 0 2 4 species rank 6 8 Appendix 9a: ICH old clear cut species rank abundance (Whittaker) plots for locations relavent to log. Distances measured in centimeters. Percent abundance of each species is plotted on a logarithmic scale against species rank ordered from most abundant to least abundant species. 10 100 cu o c cd -a c 3 10 — • - 0 • o- 50 • • « • 100 • • o • 150 • • -A- • 200 300 4 species rank 6 8 10 Appendix 9b: ICH new clear cut species rank abundance (Whittaker) plots for locations relavent to log. Distances measured in centimeters. Percent abundance of each species is plotted on a logarithmic scale against species rank ordered from most abundant to least abundant species. 105 100 n — • - • 0 • o 50 • • » • 100 • • D • 150 • • -A- • 200 • 300 0 12 14 16 2 4 6 8 10 species rank Appendix 9c : ESSF clear cut species rank abundance (Whittaker) plots for all locations relavent to log. Distances are measured in centimeters. Percent abundance of each species is plotted on a logarithmic scale against species rank ordered from most abundant to least abundant species. 18 100 — • - • 0 • o - 50 100 150 200 • 300 0 8 10 2 4 6 species rank Appendix 9d: ESSF partial cut species rank abundance (Whittaker) plots for all locations relavent to log. Distances are measured in centimeters. Percent abundance of each species is plotted on a logarithmic scale against species rank ordered from most abundant to least abundant species. 12 106 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088536/manifest

Comment

Related Items