FACTORS AFFECTING CARABID (COLEOPTERA: CARABIDAE) ASSEMBLAGES IN SUCCESSIONAL SUB BOREAL SPRUCE FORESTS, WITH SPECIAL REFERENCE TO THEIR INTERACTION WITH ANTS (HYMENOPTERA: FORMICIDAE). by Duncan Andrew McColl B.Sc, University of Northern British Columbia, 2004 THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN NATURAL RESOURCES AND ENVIRONMENTAL STUDIES THE UNIVERSITY OF NORTHERN BRITISH COLUMBIA May 2010 ©Duncan Andrew McColl 1*1 Library and Archives Canada Bibliotheque et Archives Canada Published Heritage Branch Direction du Patrimoine de I'edition 395 Wellington Street OttawaONK1A0N4 Canada 395, rue Wellington Ottawa ON K1A 0N4 Canada Your file Votre reference ISBN: 978-0-494-75133-6 Our file Notre reference ISBN: 978-0-494-75133-6 NOTICE: AVIS: The author has granted a nonexclusive license allowing Library and Archives Canada to reproduce, publish, archive, preserve, conserve, communicate to the public by telecommunication or on the Internet, loan, distribute and sell theses worldwide, for commercial or noncommercial purposes, in microform, paper, electronic and/or any other formats. L'auteur a accorde une licence non exclusive permettant a la Bibliotheque et Archives Canada de reproduire, publier, archiver, sauvegarder, conserver, transmettre au public par telecommunication ou par I'Internet, preter, distribuer et vendre des theses partout dans le monde, a des fins commerciales ou autres, sur support microforme, papier, electronique et/ou autres formats. The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. L'auteur conserve la propriete du droit d'auteur et des droits moraux qui protege cette these. Ni la these ni des extra its substantiels de celle-ci ne doivent etre imprimes ou autrement reproduits sans son autorisation. In compliance with the Canadian Privacy Act some supporting forms may have been removed from this thesis. Conformement a la loi canadienne sur la protection de la vie privee, quelques formulaires secondaires ont ete enleves de cette these. While these forms may be included in the document page count, their removal does not represent any loss of content from the thesis. Bien que ces formulaires aient inclus dans la pagination, il n'y aura aucun contenu manquant. 1+1 Canada Abstract Carabid (Coleoptera: Carabidae) assemblages in west central British Columbia are relatively poorly examined. Additionally, the influence of ants (Hymenoptera: Formicidae) on carabid assemblages is infrequently acknowledged as a factor that affects carabid diversity, distribution, and activity. The purpose of this study was to examine carabid assemblages in successional sub boreal spruce forests in west central British Columbia, and specifically how they are affected by two species of ants; Formica aserva (Forel) and Camponotus herculeanus (L). Data pertaining to carabid and ant activity-abundance were collected over a chronosequence of successional forest stages by pitfall trapping. The data were analyzed for the effects of canopy cover, vegetation, and ant influence on carabid species assemblages. Carabids were shown to be influenced by the presence of ants on the basis of a pattern of avoidance, and the frequency of carabid injuries were significantly related to F. aserva activity-abundance. An experiment where F. aserva nests were introduced into a clearcut was conducted to further examine this relationship confirmed that carabid activity-abundance is affected by ant presence. n TABLE OF CONTENTS Abstract n Table of Contents in List of Tables VI List of Figures vn Acknowledgement IX Chapter 1 Chapter 2 Introduction 1 Reference List 9 Effect of Forest Succession on Carabid (Coleoptera: Carabidae) Assemblage Structure in Post-Harvest Stands in Sub Boreal Spruce Forests of West-Central British Columbia 15 Abstract 16 Introduction 17 Methods 19 Site Selection 19 Pitfall Sampling 21 Vegetation Sampling 23 Data Analyses 24 27 Results Season and Gender 27 Vegetation Cover 31 III Diversity 32 Assemblage Ordination 33 35 Discussion Chapter 3 Seasonal Variation 35 Gender Differences 37 Influence of Canopy Cover / Stand Age and Vegetation 38 Assemblage Ordination 38 Concluding Remarks 41 Reference List 42 Interactions of Formica aserva (Forel), Camponotus herculeanus (L.) (Hymenoptera: Formicidae) and Carabid Beetle (Coleoptera: Carabidae) Assemblages in Sub Boreal Forests 47 Abstract 47 Introduction 48 Methods 52 Study Sites 52 Pitfall Sampling 52 Specimen Identification and Data 54 Collection Data Analyses 55 58 Results Ants and Carabids 58 Injury 64 Discussion 66 Reference List 72 IV Chapter 4 Effect of Addition of Formica aserva (Forel) 76 (Hymenoptera: Formicidae) Nests on a Carabid (Coleoptera: Carabidae) Assemblage Abstract 76 Introduction 77 Methods 80 Site Selection 80 Formica aserva Colony Selection 81 Experimental Design and Data Collection 81 Data Analyses 83 Results 84 Discussion 91 Reference List 93 Synthesis 96 Reference List 103 Appendix I Study Site Locations; 2005 107 Appendix II ANOVA Results Figure 2, Chapter 2 108 Appendix III Non-metric Multidimensional Scaling: PC - Ord Outputs Figure 4 Chapter 2 110 Appendix IV UTM's For Experimental Replicates Chapter 4 113 Chapter 5 LIST OF TABLES Table 1 Summary of carabid species collected in successional sub-boreal spruce forests in west central BC, during the spring and summer of 2005. Catch for each species is standardized to number of individuals per 100 trapping days. 28 Table 2 Diversity measures calculated for carabid assemblages sampled in 2005. Mean carabid diversity is calculated by plot canopy cover classes (no canopy, developing canopy, closed canopy) Diversity measures are defined in the text. 33 Table 3 Summary of x analysis on injury frequency of carabids in stands before and at different successional stages after harvest (yph = years post harvest). (X2crit= 3.841: df=1; a = 0.05 for all tests; Significant %2 results values are given) 65 Table 4 Carabid species collected and percent contribution by each species to the total catch. Bold indicates the most abundant species, which were analyzed individually. 86 VI LIST OF FIGURES Figure 1 Diagram of pitfall trapping transect. Closed dots indicate pitfall traps. 22 Figure 2 Seasonal variation in mean activity-abundance (± SEM) for males and females of the most abundant (>2% of total catch) species of carabid standardized to 14 trap days. Figure 3 Sigmoidal relationship between percent canopy cover and stand age (years post-harvest): adjusted R2 = 0. 91; F(2,149) = 784.86; P < 0.0001. 32 Figure 4 NMS ordination of carabid assemblages. Axes scales are the raw correlation coefficients. Species codes consist of the first 3 letters of genus and the first 3 letters of species names (see Table 1). 34 Figure 5 Relationship between activity-abundance of Pterostichus riparius and canopy cover 35 Figure 6 Scaphinotus marginatus with injured tarsus (left) vs. broken tarsus (right). Photo: Ward B. Strong 56 Figure 7 Injured femur of a captured Scaphinotus angusticollis, showing scleritization at the severed end. Photo: Ward B. Strong 57 Figure 8 Non-linear regression of carabid activity-abundance predicted by the ant Formica aserva y = 37.9587*exp(-0.0032*x); Adjusted R2 = 0.165; F = 17.439; P < 0.0001. 58 Figure 9 Plot of Camponotus herculeanus activity-abundace against Formica aserva activity-abundance showing a negative association between the two species. 59 Figure 10 Carabid activity-abundance in mid-successional sub boreal spruce stands plotted against Formica aserva or Camponotus herculeanus activity-abundance. 60-63 Figure 11 Relationship between mean (±SEM )activity-abundance of carabids and activity-abundance Formica aserva: Absent (0) — F. aserva were not observed at the plot, Low (1) - between 1 and 50 workers were collected at the plot, Moderate (2) - between 51 and 150 workers were collected at the plot and High (3) - more than 150 workers were collected at the plot. Means with the same letter designations are not significantly different as determined by Tukey's tests. 64 29-30 VII Figure 12 Relationship between proportion of injured carabids (±SEM ) and activity-abundance of Formica aserva: Absent (0) — F. aserva were not observed at the plot, Low (1) - between 1 and 50 workers were collected at the plot, Moderate (2) - between 51 and 150 workers were collected at the plot and High (3) - more than 150 workers were collected at the plot. Means with the same letter designations are not significantly different as determined by a posteriori Tukey's tests. 66 Figure 13 Diagram of trap layout for all treatment types. Closed dots indicate pitfall locations and the open dot indicates control or treatment. 83 Figure 14 Pooled mean (± SEM) activity-abundances of carabids for: control (CTRL), and two treatments coarse woody debris (CWD) and Formica aserva nest (ANT). 87 Figurel5 Mean (± SEM) activity-abundance of Formica aserva for: control (CTRL), and two treatments coarse woody debris (CWD) and F. aserva nest (ANT). 87 Figure 16 Mean (± SEM) activity-abundance of carabids, at five distances from treatment center, for: control (CTRL), and two treatment coarse woody debris (CWD) and Formica aserva nest (ANT). 89-90 VIII Acknowledgements I would like to thank: my thesis supervisor B. Staffan Lindgren, and the members of my thesis committee; Russ Dawson and David Langor, for their valuable knowledge, constructive criticisms, and patience; Deanna Danskin, Jovan Simic, Nick Bartell, Jeff Selesnic and Angela Gatt, and members of the UNBC Forest Insect Research Group for their assistance in the field and the lab; Melissa Todd, the GIS department and all the biology summer students from Houston Forest Products; Robert Higgins for assistance and taxonomic training regarding ants; and Greg Pohl from the Canadian Forest Service, George Ball, and Danny Shpeley from the E.A. Strickland Entomological Museum at the University of Alberta for their assistance in carabid taxonomy. The research was funded by a NSERC CRD as well as an NSERC Discovery Grant, a grant from the British Columbia Forest Science Program, as well as cash and in-kind contributions from Houston Forest Products. IX Chapter 1: Introduction Throughout forested regions of Canada, forestry is a dominant economic, social, political and ecological force. It accounted for $36.3 billon in gross domestic product in 2006, and the alteration through harvesting of 903,009ha of land in 2005 (Canadian Forest Service 2007). Numerous ecological effects initiated by harvesting persist for long periods of time as the stand recovers, and influence the trajectory of stand recovery. Landscape-level alteration of the distribution, shape, species composition, size and juxtaposition of clearcuts on the landscape will likely persist for several stand rotations (DeLong 2002). Within-stand changes initiated by harvesting can affect coarse woody debris (CWD) cycles, as well as distribution, quantity and quality of CWD immediately following harvest (Lloyd 2003), and these effects may persist for many years (Densmore et al. 2004). In addition, there may be changes in soil properties through mechanical disturbance (exposure of mineral soil, compaction etc.) associated with harvest (Ballard 2000, Page-Dumroese et al. 2005), chemical changes associated with altered soil biota (Ballard 2000), and shifts in forest hydrology (Moore and Wondzell 2005). Many of these effects impact forest invertebrates directly through habitat alteration such as depletion of CWD, removal of organic layer and soil compaction (Ehnstrom 2001), loss of suitable habitat (Pettersson et al. 1995), isolation and fragmentation (Debinski and Holt 2000, Komonenet et al. 2000), or indirectly by influencing the distribution and or abundance of competitors/predators (Orjan et al. 2007). Within an organism's environment specific conditions are optimal for survival and reproduction. The role a species plays within this biotic and abiotic realm constitutes that species' niche (Whittaker et al. 1973, Rejmanek and Jenik 1975). The niche concept can encompass, in whole or in part, individual organisms, populations of organisms and species. 1 Two early theories in the concept of niche, which have since been refined (e.g., Whittaker et al. 1973), include the fundamental niche, the total possible environmental space in which a species can exist in the absence of competition, and the realized niche, the niche that an organism occupies due in part to the influence of other species (McGill et al. 2006). Competition and other interactions, including predation, that occur between species define a species' realized niche (Alley 1982). Biotic interactions that affect the distribution, abundance, and diversity of organisms can be considered to be beneficial, deleterious or neutral in terms of the effect of one species on another (Odum 1969). Competition has to be considered an important factor in determining the distribution, abundance and diversity of the populations that occupy a specific habitat (Diamond and Case 1986). It is important to note that niche overlap does not always result in competition (Alley 1982). As noted by den Boer (1979) in the case of carabids, taxonomically related species also tend to be ecologically related and thus tend to be found in the same types of habitat. One method to measure interference competition in a community is to perform experiments that alter the composition of the community, i.e., experiments that perturb the community (examples summarized in Connell 1983, Schoener 1983). A frequently used technique to achieve observable change in a measurable factor of a community is to remove one or more competitors. The theoretical response in this situation is ecological release: the increase of remaining competitors; a measurable indication of an increase in the use of resources that were negatively impacted by the removed competitor(s), and/ or; occupation of the habitat that was formerly occupied by the removed competitor(s) (Bender et al. 1984). Removing one species from a community to observe the response of another species is not 2 always possible; therefore, other methods are used to observe, infer, or otherwise detect interspecific competition (Keddy 1989). Altering a resource (frequently food/prey) that is hypothesised to be limiting in a competitive interaction is another method for experimentally determining if resource competition is occurring. In order to observe competition for a resource, that resource must be utilized by the competing organisms and limiting (Keddy 1989). Additional difficulties in interpreting the results of perturbation experiments may be important, as noted by Bender et al. (1984). For example, in studies using data on adult carabid numbers, observing the effect of competition may be difficult because competition and predation may be important at the larval stage of the life cycle but not readily observed in adults (e.g., Currie et al. 1996). In the absence of experimental evidence, observed trends in the distribution of species have been hypothesised to be the result of possible interference competition. Species that interfere with each other tend to be distributed in what has been described as a checkerboard distribution in contrast to a random distribution (Diamond 1975). Another technique that has been used to infer competition is character displacement, i.e., the examination of physical characters of several species that occur in a delineated habitat. If competition among the species over evolutionary time was important in structuring the community and the resources that each species utilize, then there should be observable morphological differences between similar species related to their niche. This idea, however, has been examined and dismissed (Connell 1980). Anthropological utilization of forested land occurs across several spatial scales from within stands to forests and landscapes, and extends temporally as dynamic forest processes, including anthropogenic disturbance, occurring over time (Burton et al. 2003). 3 Development of tools to monitor how forested ecosystems respond to disturbances is important for ecologically sustainable forest management. Biological indicators, or bioindicators, are potentially useful in this regard. Several requirements have been identified as desirable for biological indicators (Dale and Beyeler 2001). Invertebrates, particularly arthropods, have several attributes that make them suitable for such a role (Weaver 1995, Maleque et al. 2006). In particular, ground beetles (Coleoptera: Carabidae) have received much attention as useful indicators of impacts of anthropogenic disturbance in many forested ecosystems (Rainio and Niemela 2003). Ground beetles or carabids are appealing research subjects for several reasons. They represent a well known taxonomic group that responds to environmental change, and is relatively easy and inexpensive to collect (Refseth 1980, Niemela et al. 2000). Carabid research has spanned all continents except Antarctica, where carabids do not currently occur (Ashworth 2001), and they have been studied in all habitat types (Lovei and Sunderland 1996). Furthermore, they occupy a wide spectrum of ecological niches and trophic levels (Lovei and Sunderland 1996), and have been demonstrated to have potential utility in indicating variations in biodiversity (Butterfield 1997) and in ecological and environmental conditions. Carabids generally are considered to occur in assemblages rather than in communities. Although interpretations differ (Morin 1999), communities are generally considered a group of populations that interact, while assemblages are populations that co-exist but do not necessarily interact. As carabids are a portion of the larger invertebrate community as a whole (Lovei and Sunderland 1996), the term assemblage will be used, except where cited authors use the term community or when referring to the larger invertebrate community. 4 Carabids have also been the subject of interest as an indicator group in forested ecosystems, grasslands and agroscapes (see Rainio and Niemela 2003 for a review). For example, in a study examining the effect of forest succession on carabids, Baguette and Gerard (1993) found that the carabid assemblage composition in Belgian spruce plantations varied with stand structure and age. Brumwell et al. (1998) and Lemieux and Lindgren (2004) found that carabids in British Columbia also respond to successional changes in forests, and Koivula et al. (2002) found differences in species richness in regenerating stands of differing ages in Finland. Studies examining landscape level effects on carabid communities provide evidence of response to environmental conditions; for example, Halme and Niemela (1993) examined the effect of fragmentation on carabids and found that largebodied carabids were more abundant in contiguous forests in Finland than in forest fragments. Burke and Goulet (1998) had similar results when examining the response of carabids to forest fragmentation in Ontario; large-bodied species were more abundant, and species richness was higher in large fragments. Similarly, Abildsnes and Tommeras (2000) found that different species of carabids respond differently to fragmentation of an old growth Norwegian forest. Many carabid ecology studies ignore other surface-dwelling invertebrates (Lovei and Sunderland 1996). Within the epigaeic arthropod community, ants can play a particularly significant role (Lovei and Sunderland 1996; Laakso and Setala 1998, Laakso and Setala 2000, Punttila et al. 2004). They can reshape the landscape by mixing soil and distributing plant seeds. Additionally, ants affect community composition by exerting a strong predatory pressure on other invertebrates (Holldobler and Wilson 1990). Established colonies can dominate their territories through aggressive behaviour which aids in their ability to acquire, 5 exploit and defend resources (Holldobler and Wilson 1990). In many habitats ants may also numerically dominate the invertebrate community (Laakso and Setala 1998, 2000). Ants are thermophilic (Holldobler and Wilson 1990) and are therefore influenced by canopy cover (Punttila et al. 1991), which influences how much solar radiation reaches the forest floor (Huber and Baumgarten 2005). As forests grow, the amount of light penetration through the canopy is reduced. Reduction of solar radiation may influence dominance hierarchies among ant species (Cerda et al. 1998), and has been shown to lead to colony failure or abandonment when thermal requirements cease to be met (Higgins 2010). The change in the thermal environment of a stand likely influences the interactions between ants and other invertebrates, including carabids. Ant colonies tend to occupy delineated territories, and many ant species defend these against both conspecifics from different colonies and heterospecifics (Holldobler and Wilson 1990). Ants may alter the behaviour of other species, e.g., spiders (Halaj et al. 1997) and even birds (Haemig 1996). Consequently, it is logical to expect that ants may significantly affect carabids. While some studies make cursory mention of negative correlation between ants and carabids (Niemela et al. 1992, Koivula et al. 1999, Koivula 2002, Koivula and Niemela 2003), only a few studies have examined this interaction closely. Carabid adults appear to be non-significant as food items for red wood ants of the Formica rufa group (Skinner 1980), but Reznikova and Dorosheva (2004) demonstrated that the presence of ants can alter the behaviour of carabids. They showed that ants act aggressively towards carabids, and that different species of carabids may respond differently to the presence of ants by altering movement patterns and/or protecting limbs. Hawes et al. (2002) showed that the presence of 6 ants affects species composition and distribution of carabids, although the mechanisms causing these changes are not clear. The overall objective of my research was to examine the potential effect of ants on carabid assemblages at different successional stages of forest development after harvesting. Several collection techniques have been employed in epigaeic invertebrate research, depending on the goal of the research. The predominant technique utilized in ecological studies of carabids is passive capture in pitfall traps of various designs, a technique that also can be used to collect ants. Passive capture relies on invertebrates falling into a neutrally attractive trap, which gives a measure of abundance that is generally inseparable from activity. This is due to the two requirements of pitfall trapping: 1) carabids are present (abundance), and; 2) they are able to move into the pitfall trap (activity), resulting in the commonly used measure "activity-abundance" (Spence and Niemela 1994). The "standard" or "conventional" pitfall trap is any container with a round opening placed with the opening flush with the ground (Greenslade 1964). Numerous variations of pitfall trap designs have been tested and compared for their effectiveness in collecting carabids (Greenslade 1964, Epstein and Kulman 1984, Spence and Niemela 1994, Lemieux and Lindgren 1999, Abensperg-Traun and Steven 1995, Work et al. 2002, Koivula et al. 2003, and Pearce et al. 2005). Variation among trap types was observed in the quantity of carabids collected, with larger traps tending to catch more beetles (Work et al. 2002, Koivula et al. 2003) and ants (Abensperg-Traun and Steven 1995), although this relationship is not linear (Work et al. 2002). Nordlander traps provided a better reflection of species richness (Pearce et al. 2005), while mitigating other pitfall trap associated difficulties. 7 Pitfall traps do not provide data pertaining to the absolute density of carabids (Andersen 1995, Lang 2000), but on activity abundance (Spence and Niemela 1994). Perner and Schueler (2004), however, propose that using a nested cross array trapping pattern, and fitting catch data to a single hyperbolic function may provide density estimates. Maehara (2004) found that pitfall catches of Carabus insulicola insulicola Chaudoir in an enclosed population correlated with the population density. Carabids tend to move at random across the landscape (Baars 1979; Drach and Cancela da Fonseca 1990, Lovei and Sunderland 1996; Firle et al. 1998), and pitfall trapping is therefore an appropriate technique for sampling. Movement is influenced by habitat, ambient temperature (Baars 1979) and hunger, since carabids utilize random search to locate prey items (Wallin and Ekbom 1994; Lovei and Sunderland 1996). The movement tends to result in a linear increase in the area covered by beetles over time, but not necessarily in point to point distance traveled (Firle et al. 1998). Carabids respond to encounters with inhospitable or unfavourable environments by fleeing, and they then frequently move in a more linear direction than observed in more favourable environments (Baars 1979). In favourable environments it is unlikely that carabids moving exclusively by ambulatory means travel distances greater than 100m in a season (den Boer 1990). Movement patterns may be highly variable between species that fill different habitat niches, e.g., forest generalists vs. forest specialists (Brouwers and Newton 2009). The area covered by individuals may be at a scale of hectares as suggested by movement modeling (Firle et al. 1998), but experimental and observational data are currently lacking (Brouwers and Newton 2009). 8 Pitfall trapping is a cost effective and repeatable method of capture which has proven particularly suitable for the collection of carabids (Spence and Niemela 1994) and ants (Melbourne 1999) provided that biases and limitations are understood (Koivula et al. 2003). Thus, pitfall trapping has been extensively utilized for the purpose of collecting and examining carabid activity abundances, carabid assemblage composition (Lovei and Sunderland 1996) and the activity abundance and species composition of ants (Melbourne 1999). The purpose of this study is to: (1) examine changes in carabid assemblages in postharvest systems at different successional stages in west central British Columbia sub boreal spruce stands (Chapter 2); (2) examine the effect of ants on carabid abundance (Chapter 3); and (3) to experimentally examine the relationship between carabids and the ant Formica aserva Forel in a young regenerating stand (Chapter 4). Reference List Abensperg-Traun, M. and Steven, D.E. 1995. The effects of pitfall trap-diameter on ant species richness (Hymenoptera: Formicidae) and species composition of the catch in a semiarid eucalypt woodland. Australian Journal of Ecology 20: 282-287. Abildsnes, J., and Temmeras, B.A. 2000. Impacts of experimental habitat fragmentation on ground beetles (Coleoptera, Carabidae) in a boreal spruce forest. Annales Zoologici Fennici 37:201-212. Alley, T. 1982. Competition theory, evolution, and the concept of an ecological niche. Acta Biotheoretica31: 165-179. Andersen, J. 1995. A comparison of pitfall trapping and quadrat sampling of Carabidae (Coleoptera) on river banks. Entomologica Fennica 6: 65-77. Ashworth, A.C. 2001. Perspectives on quaternary beetles and climate change. American Association of Petroleum Geologists Studies in Geology, Tulsa, Oklahoma, pp. 153-168. Baars, M.A. 1979. Patterns of movement of radioactive carabid beetles. Oecologia 44: 125140. 9 Baguette, M., and Gerard, S. 1993. Effects of spruce plantations on carabid beetles in southern Belgium. Pedobiologia 37: 129-140. Ballard, T.M. 2000. Impacts of forest management on northern forest soils. Forest Ecology and Management. 133: 37-42. Bender, E.A., Case, T.J., and Gilpin, M.E. 1984. Perturbation experiments in community ecology: theory and practice. Ecology. 65: 1-13. Brouwers, N.L., and Newton, A.C. 2009. Movement rates of woodland invertebrates: a systematic review of empirical evidence. Insect Conservation and Diversity. 2: 10-22. Brumwell, L.J., Craig, K.G, and Scudder, G.G.E. 1998. Litter spiders and carabid beetles in a successional Douglas-fir forest in British Columbia. Northwest Science 72: 94-95. Burke, D., and Goulet, H. 1998. Landscape and area effects on beetle assemblages in Ontario. EcographyH: 472-479. Burton, P.J., Messier, C, Weetman, G.F. Prepas, E.E., Adamowicz, W.L., and Tittler, R.. 2003. The current state of boreal forestry and the drive for change. Chapter 1. In Towards Sustainable Management of the Boreal Forest. Edited by P.J. Burton, C. Messier, D.W. Smith and W.L. Adamowicz. NRC Research Press, Ottawa, Ontario, Canada, pp. 1 -40. Butterfield, J. 1997. Carabid community succession during the forestry cycle in conifer plantations. Ecography 20: 614-625. Canadian Forest Service. 2007. The state of Canada's forests: Annual Report 2007. Natural Resources Canada, Canadian Forest Service, Headquarters, Ottawa. 28 pp. Cerda, X., Retana, J., and Manzaneda, A. 1998. The role of competition by dominants and temperature in the foraging of subordinate species in Mediterranean ant communities. Oecologia. 117: 404-412. Connell, J.H. 1980. Diversity and the coevolution of competitors, or the ghost of competition past. Oikos 35: 131-138. Connell, J.H. 1983. On the prevalence and relative importance of interspecific competition: Evidence from field experiments. American Naturalist. 122: 661-696. Currie, C.R., Spence, J.R., Niemela, J. 1996. Competition, cannibalism and intraguild predation among ground beetles (Coleoptera: Carabidae): a laboratory study. Coleopterists Bulletin. 50: 135-148. Dale, V.H., and Beyeler, S.C. 2001. Challenges in the development and use of ecological indicators. Ecological Indicators. 1: 3-10. Debinski, M.D., and Holt, D.R. 2000. A survey and overview of habitat fragmentation 10 experiments. Conservation Biology. 14: 342-355. DeLong. C. 2002. Using nature's template to best advantage in the Canadian boreal forest. Silva Fennica. 36:401-408. den Boer, P.J. 1979. Exclusion or coexistence and the taxonomic or ecological relationship between species. Netherlands Journal of Zoology. 30: 278-306. den Boer, P.J. 1990. Density limits and survival of local populations of 64 carabid species with different powers of dispersal. Journal of Evolutionary Biology. 3: 19-48. Densmore, N., J. Parminter, and V. Stevens. 2004. Coarse woody debris: inventory, decay modelling, and management implications in three biogeoclimatic zones. BC Journal of Ecosystems and Management 5:14-29. Diamond, J. M. 1975. The island dilemma: lessons of modern biogeographic studies for the design of natural reserves. Biological Conservation 7: 129-146. Diamond, J., and Case, T.J. 1986. Community Ecology. New York: Harper and Row. 704 pp. McGill, B.J., Enquist, B.J., Weiher, E., and Westboy, M. 2006. Rebuilding community ecology from functional traits. Trends in Ecology and Evolution. 21: 175-185 Drach, A., and Cancela da Fonseca, J.P. 1990. Experimental and theoretical approach to forest carabid beetle movements. Revue D'Ecologie et de Biologie du Sol. 27: 61-71. Ehnstrom, B.. 2001. Leaving dead wood for insects in boreal forests - suggestions for the future. Scandinavian Journal of Forest Research Supplement 3: 91-98. Epstein, M.E., and Kulman, H.M. 1984. Effects of aprons on pitfall trap catches of carabid beetles in forests and fields. Great Lakes Entomologist 17: 215-221. Firle, S., Bommarco, R., Ekbom, B., and Natiello, M. 1998. The influence of movement and resting behavior on the range of three carabid beetles. Ecology 79: 2113-2122. Greenslade, P.J.M. 1964. Pitfall trapping as a method for studying populations of Carabidae (Coleoptera). Journal of Animal Ecology 33: 301-310. Greenslade, P.J.M. 1973. Sampling ants with pitfall traps: digging-in effects. Insectes Sociaux. 20: 343-353. Haemig, P.D. 1996. Interference from ants alters foraging ecology of great tits. Behavioral Ecology and Sociobiology 38: 25-29. Halaj, J., Ross, D.W., and Moldenke, A.R. 1997. Negative effects of ant foraging on spiders in Douglas-fir canopies. Oecologia 109: 313-322. Halme, E., and Niemela, J. 1993. Carabid beetles in fragments of coniferous forest. Annales 11 Zoologici Fennici 30: 17-30. Hawes, C, Stewart, A.J.A., and Evans, H.F. 2002. The impact of wood ants (Formica rufa) on the distribution and abundance of ground beetles (Coleoptera: Carabidae) in a Scots pine plantation. Oecologia 131: 612-619. Higgins, R.J. 2010. The ant (Hymenoptera: Formicidae) communities of the central interior of British Columbia: Adaptations to a temperature-constrained environment. Ph.D. Dissertation, University of Northern British Columbia, Prince George, BC, 184 pp. Holldobler, B.; Wilson, E.O. 1990. The Ants. Cambridge, MA: Harvard University Press. 732 pp. Huber, C, and Baumgarten, M. 2005. Early effects of forest regeneration with selective and small scale clear-cutting on ground beetles (Coleoptera, Carabidae) in a Norway spruce stand in southern Bavaria (Hoglwald). Biodiversity and Conservation 14: 1989-2007. Keddy, P.A. 1989. Competition. London: Chapman & Hall, 202 pp. Koivula, M. 2002. Boreal carabid-beetle (Coleoptera, Carabidae) assemblages in thinned uneven-aged and clear-cut spruce stands. Annales Zoologici Fennici 39: 131-149. Koivula, M., and Niemela, J. 2003. Gap felling as a forest harvesting method in boreal forests: responses of carabid beetles (Coleoptera, Carabidae). Ecography 26: 179-187. Koivula M., Kotze D.J., Hiisivuori L., and Rita H. 2003. Pitfall trap efficiency: do trap size, collecting fluid and vegetation structure matter? Entomologica Fennica 14: 1-14. Koivula, M., Kukkonen, J., and Niemela, J. 2002. Boreal carabid-beetle (Coleoptera, Carabidae) assemblages along the clear-cut originated succession gradient. Biodiversity and Conservation 11: 1269-1288. Koivula, M., Punttila, P., Haila, Y., and Niemela, J. 1999. Leaf litter and the small-scale distribution of carabid beetles (Coleoptera, Carabidae) in the Boreal Forest. Ecography 22: 424-435. Komonen, A., Penttila, R., Lindgren, M., and I. Hanski. 2000. Forest fragmentation truncates a food chain based on an old-growth forest bracket fungi. Oikos. 90: 119-126. Laakso, J., and Setala, H. 1998. Composition and trophic structure of detrital food web in ant nest mounds of Formica aquilonia and in the surrounding forest soil. Oikos 81: 266-278. Laakso, J., and Setala, H. 2000. Impacts of wood ants (Formica aquilonia Yarr.) on the invertebrate food web of the boreal forest floor. Annales Zoologici Fennici 37: 93-100. Lang, A. 2000. The pitfalls of pitfalls: a comparison of pitfall trap catches and absolute density estimates of epigeal invertebrate predators in arable land. Anzeiger fur Schadlingskunde- Journal of Pest Science 73: 99-106. 12 Lemieux, J.P., and Lindgren, B.S. 1999. A pitfall trap for large-scale trapping of Carabidae: comparison against conventional design, using two different preservatives. Pedobiologia 43: 245-253. Lemieux, J.P., and Lindgren, B.S. 2004. Ground beetle responses to patch retention harvesting in high elevation forests of British Columbia. Ecography 27: 557-566. Lloyd, R. 2003. A comparison of coarse woody debris in harvested and unharvested sites. Morice-Lakes IFPA rep. no. 442.01. Lovei, G.L., and Sunderland, K.D. 1996. Ecology and behavior of ground beetles (Coleoptera: Carabidae). Annual Review of Entomology 41: 231-256. Maehara, T. 2004. The efficiency of pitfall traps in relation to a density and activity of the carabid beetle, Carabus insulicola insulicola Chaudoir (Coleoptera: Carabidae), under controlled population density. Japanese Journal of Applied Entomology and Zoology 48: 115-121. Maleque, M.A., Ishii, H.T., and Kaoru, M. 2006. The use of arthropods as indicators of ecosystem integrity in forest management. Journal of Forestry. 104: 113-117. Melbourne, B.A. 1999. Bias in the effect of habitat structure on pitfall traps: an experimental evaluation. Australian Journal of Ecology 24: 228-239. Moore, R.D., and Wondzell, S.M. 2005. Physical hydrology and the effects of forest harvesting in the Pacific Northwest: a review. Journal of the American Water Resources Association. 41: 763-784. Morin, P.J. 1999. Community Ecology. Blackwell Publishing, Maiden, MA, 432 pp. Niemela, J., Kotze, J., Ashworth, A., Brandmayr, P., Desender, K., New, T., Penev, L., Samways, M., and Spence, J. 2000. The search for common anthropogenic impacts on biodiversity: a global network. Journal of Insect Conservation 4: 3-9. Niemela, J., Spence, J.R., and Spence, D.H. 1992. Habitat associations and seasonal activity of ground-beetles (Coleoptera, Carabidae) in central Alberta. Canadian Entomologist 124: 521-540. Odum, E.P. 1969. The strategy of ecosystem development. Science. 164: 262-270. Orjan 6., Nicholas W.G, Jared L. Strasburg, J.A. Brisson, A.R. Templeton, T.M. Knight, J., and Chase M. 2007. Habitat area affects arthropod communities directly and indirectly through top predators. Ecography. 30: 359-366. Page-Dumroese, D.S., Jurgensen, M.F., Tiarks, E.A., Ponder, F.Jr., Sanchez, F.G., Fleming, R.L., Kranabetter, J.M., Powers, R.F., Stone, D.M., Elioff, J.D., and Scott, D.A. 2005. Soil physical property changes at the North American long-term soil productivity study sites: 1 and 5 years after compaction. Canadian Journal of Forest Research. 36: 551-564. 13 Pearce, J.L., Schuurman, D., Barber, K.N., Larrivee, M., Venier, L.A., McKee, J., and McKenney, D. 2005. Pitfall trap designs to maximize invertebrate captures and minimize captures of nontarget vertebrates. Canadian Entomologist 137: 233-250. Perner, J., and Schueler, S. 2004. Estimating the density of ground-dwelling arthropods with pitfall traps using a nested-cross array. Journal of Animal Ecology 73: 469-477. Pettersson, R.B., Ball, J.P., Renhorn, K., Esseen, P., and Sjoberg K. 1995. Invertebrate communities in boreal forest canopies as influenced by forestry and lichens with implications for passerine birds. Biological Conservation. 74: 57-63. Punttila, P., Haila, Y., Pajunen, T., and Tukia, H. 1991. Colonisation of clearcut forests by ants in the southern Finnish taiga: a quantitative survey. Oikos. 61: 250-262. Punttila, P., Niemela, P., and Karhu, K. 2004. The impact of wood ants (Hymenoptera: Formicidae) on the structure of invertebrate community on mountain birch (Betula pubescens ssp czerepanovii). Annales Zoologici Fennici 41: 429-446. Rainio, J., and Niemela, J. 2003. Ground beetles (Coleoptera: Carabidae) as bioindicators. Biodiversity and Conservation 12: 487-506. Refseth, D. 1980. Ecological analysis of Carabid communities potential use in biological classification for nature conservation. Biological Conservation 17: 131-141. Rejmanek, M., and Jenik, J. 1975. Niche, habitat, and related ecological concepts. Acta Biotheoretica. 24: 100-107. Reznikova, Z., and Dorosheva, H. 2004. Impacts of red wood ants Formica polyctena on the spatial distribution and behavioural patterns of ground beetles (Carabidae). Pedobiologia 48: 15-21. Schoener, T.W. 1983. Field experiments on interspecific competition. American Naturalist. 122: 240-284. Skinner, G.J. 1980. The feeding habits of the wood-ant, Formica rufa (Hymenoptera: Formicidae), in limestone woodland in north-west England. The Journal of Animal Ecology 49: 417-433. Spence, J.R., and Niemela, J.K. 1994. Sampling carabid assemblages with pitfall traps - the madness and the method. Canadian Entomologist 126: 881-894. Wallin, H., and Ekbom, B. 1994. Influence of hunger level and prey densities on movement patterns in 3 species of Pterostichus beetles (Coleoptera, Carabidae). Environmental Entomology 23: 1171-1181. Weaver, JC. 1995. Indicator species and scale of observation. Conservation Biology. 9: 939942. 14 Whittaker, R.H., Levin, S.A., and Root, R.B. 1973. Niche, habitat, and ecotope. American Naturalist. 107: 321-338. Work, T.T., Buddie, CM., Korinus, L.M., and Spence, J.R. 2002. Pitfall Trap Size and Capture of Three Taxa of Litter-Dwelling Arthropods: Implications for Biodiversity Studies. Environmental Entomology 31: 438-448. 15 Chapter Two Effect of forest succession on carabid (Coleoptera: Carabidae) assemblage structure in post-harvest stands in sub boreal spruce forests of west-central British Columbia I examined carabid beetle (Coleoptera: Carabidae) assemblages in post-harvest and unharvested sub boreal spruce (SBS) stands in west central British Columbia, Canada. To obtain a description of carabid assemblages in successional SBS forest, carabid species composition, the seasonal activity of abundant carabid species and the differences in activityabundance of males and females, I installed 750 Nordlander pitfall traps in 10 stands encompassing a gradient in canopy cover from 0% to 100%. A total of 4801 individual carabids, representing 31 species, were collected over 12 weeks. Carabid assemblages in SBS stands are influenced by vegetation and structure. Non-metric multidimensional scaling revealed four carabid assemblages, each associated with different vegetative characteristics. Variables associated with stand succession (canopy cover, vegetation diversity, and correlations along a gradient of grass-dominated ground cover) accounted for most of the variation in the ordination. Seasonal activity and sex ratios of eight common species showed significant within-species variation in standardized mean activity-abundance for seven species. In the SBS, seasonal activity of individual species, as well as forest type associations related to canopy cover, differed from findings of other studies with similar species assemblages. Canopy cover, which influences temperature and relative humidity, shows strong influence over the species composition of carabid assemblages in regenerating stands although other factors not examined here may also influence the distribution of species within assemblages. 16 Introduction Lindroth's (1961-69) comprehensive taxonomic monograph on the carabid beetles of Canada and Alaska provides an excellent taxonomic base for carabid studies in British Columbia (B.C.), however, knowledge of the carabid assemblages in west-central B.C. is poor. A study undertaken near Smithers, B.C. in harvested, high-elevation stands in the Engelmann Spruce-Subalpine Fir biogeoclimatic zone (Meidinger and Pojar 1991), to examine the immediate (2 to 4 years post-harvest) response of carabids to partial harvesting (Lemieux and Lindgren 2004), is the only example of carabid research in this region. Species inventories are to understand the distribution of species, relationships among species, effects of landscape structure and effects of anthropogenic disturbances, such as forestry (Niemela et al. 1994), on biodiversity (Jenkins 1988). Baseline inventories of carabids have yet to be completed for different successional stages of forest regeneration in west-central B.C. Without this knowledge it is impossible to assess the possible long-term effects that harvesting may have upon carabids. Comparing the assemblage in regenerating stands shortly after harvest to unharvested stands provides information pertaining to the immediate effect of forest harvesting on carabids. Examining carabid communities in regenerating forests at different intervals after harvesting provides insights into ecological processes involved in managed forests (Niemela et al. 1993, Atlegrim et al. 1997, Ings and Hartley 1999, Koivula et al. 2002, Heyborne et al. 2003, Vance and Nol 2003, Brouat et al. 2004, de Warnaffe and Lebrun 2004). Hence, knowledge of the fauna that inhabits regenerating and mature forested stands is paramount to the understanding of ecological processes that shape the boreal forests (Korpilahti 1996). 17 Carabids are distributed non-randomly across the landscape (Niemela and Halme 1992, Niemela et al. 1992). Therefore, microhabitat associations provide information as to where carabids tend to be found. Variation in carabid communities has been linked to light penetration (Niemela et al. 1988, Abildsnes and T0mmeras 2000) or solar radiation (Huber and Baumgarten 2005). Penetration of light in forested stands is directly related to canopy closure (Huber and Baumgarten 2005), which in turn has been identified as a variable which influences carabid activity-abundance and assemblage composition (Magura and Tothmeresz 1997, Brumwell et al. 1998, Humphrey et al. 1999, Jukes et al. 2001, Koivula 2002, Koivula et al. 2002, Magura 2002, Magura et al. 2002, Heyborne et al. 2003, Lassau et al. 2005). Several studies have found shifts in carabid assemblage structure post-harvest (Niemela et al. 1993, Atlegrim et al. 1997, Beaudry et al. 1997, Butterfield 1997, Koivula 2002, Koivula et al. 2002, Heyborne et al. 2003). Niemela et al. (1992) proposed a conceptual model of carabid species succession. The model predicts a loss of mature-forest species, a decrease in forest-generalist species and an increase in open-ground species in early succession post-harvest. Open-ground species then decline as the forest matures, while forest-generalists increase. Eventually the open-ground species will drop out of the community and forest-generalists will be dominant. As the forest regains a mature structure, mature-forest specialists will reappear, although this has not been readily observed (Niemela et al. 1993, Spence et al. 1996, Heyborn et al. 2003) and may not occur if source populations are no longer available to provide individuals to colonize what may appear to be suitable habitat (Niemela et al. 1993, Spence et al. 1996). Most research has occurred relatively early in the forest succession after disturbance, however. In general the model suggested by Niemela et al. (1992) has been supported by several studies (Niemela et al. 1992, Niemela et 18 al. 1993, Atlegrim et al. 1997, Koivula et al. 2002, Koivula and Niemela 2003). Relating carabid faunal groupings specifically to forest canopy closure gradients has been suggested as a more appropriate method, however, since it reflects the habitat requirements of forest species, and coincides with patterns observed in other studies (Koivula 2002). The pattern of carabid response to deforestation and subsequent regeneration is not universal. Heyborn et al. (2003) noted that carabids associated with mature forest conditions were not re-establishing when the vegetation succession was moving towards becoming a closed forest. Also, carabids considered mature-forest species were not lost in young postharvest stands at high elevations (Pearsall et al. 2003, Lemieux and Lindgren 2004), indicating that factors other than vegetation structure may be important. The objectives of this study were to: characterize the carabid assemblage in a cool forest ecosystem classified in the sub-boreal spruce biogeoclimatic zone (SBS) (Meidinger and Pojar 1991), examine the effect of clearcut harvesting on carabids, and document how carabid communities change in post-harvest stands as forest succession proceeds and stand canopies close. Methods Site Selection Ten stands (study sites) were selected in the SBS biogeoclimatic zone in west central B.C. near the village of Houston. This biogeoclimatic zone generally occurs between 800 m to 1300 m above sea level in central B.C. between latitudes 51° 30' and 59° N. Mean annual temperatures range between 1.5°C and 5°C, with 2-5 months having average temperatures below 0°C Picea glauca (Moench) Voss x Picea engelmanni Parry ex Engelm. and Abies lasiocarpa (Hook.) Nutt. are the dominant climax tree species, with Pinus contorta Dougl. ex 19 Loud. var. latifolia Engelm a common serai species (Meidinger and Pojar 1991). Natural grasslands are rare and only occur scattered in dry valley bottoms. Within the SBS, ten subzones have been described. This study was conducted in the moist-cold variant (mc), which is typified by a shrub/herb layer dominated by Cornus canadensis L., Vaccinium membranaceum Dougl. ex Torr., Lonicera involucrata (Richards.), Viburnum edule (Michx.) Raf., Rubus pedatus Sm., Petasites frigidus var. palmatus (Ait.) Cronq., and the mosses Ptillium crista-castrensis (Hedw.) De Not., Pleurozium schreberi (Brid.) Mitt, and Hylocomium splendens (Hedw.) Schimp. (Meidinger and Pojar 1991). Study sites were required to be within the operating area of West Fraser's Houston Forest Products Division (HFP West Fraser) for logistic reasons associated with funding in 2005. Possible sites were identified using HFP West Fraser's database, and were restricted to lodgepole pine-leading stands within the SBS. Additionally, sites were selected in the following post-harvest age classes: 2 years post-harvest (yph), 12 yph, 16 yph, >25 yph and non-harvested. Approximate age classes were selected as they represent shifts in vegetation structure from open ground to shrub, shrub to low canopy, low canopy and closed canopy, encompassing a gradient ranging from no canopy (0% coverage) to closed canopy (100% coverage). Post-harvest sites with a developing high canopy did not exist or were inaccessible in the SBS within the operating area of HFP West Fraser. Within each site, a one hectare plot was randomly positioned at least 50m away from an edge (i.e., road, stream, abrupt elevation change, forest or cut block). The exact study site locations are listed in Appendix I. 20 Pitfall Sampling Plots were placed within cut blocks to avoid inclusion of anthropogenic features such as spur roads, skid trails and landings, and natural features such as streams, swamps or substantial bodies of standing water, that would exclude trapping. An intense sampling protocol was used to attempt to provide fine scale data regarding variation in activityabundance and habitat use. Three 80m long transects were established, each starting at a randomly selected point within each lha plot. Each transect ran at a randomly determined bearing not constrained by the plot boundary. Five trapping clusters were established along each transect, each consisting of a central pitfall trap at 20m intervals, two satellite traps 3m from the central trap along the transect, and an additional two traps 3m from the transect at 90° and 270° from the transect (Figure 1). 21 6m I 1 • • • A - • <> • 80m 20m • * • Figure 1: Diagram of pitfall trapping transect. Closed dots indicate pitfall traps. Thus, trapping clusters were separated by a minimum of 14m, as a distance between 10m and 20m is required to mitigate trapping effects between traps (Digweed et al. 1995) while providing fine-scale resolution. Pitfall traps were a Nordlander type (Nordlander 1987) modified after Lemieux and Lindgren (1999). Each trap was constructed out of an 8oz translucent multipurpose container (VWR catalogue number 4333-002) with a diameter of 8cm and depth of 7.5cm. Fourteen 12mm long x 6mm high entrance holes were punched below the lid of the container using a hole punch. Traps contained a 90mL of 25% propylene glycol solution. Samples were collected and the fluid replaced every 14 days from May 25, 2005 until August 28, 2005. 22 Specimens were transferred to specimen cups for transport to the lab, where they were rinsed in a soap and water solution to remove debris, then rinsed in distilled water to remove soap residue. Carabids and ants from each individual pitfall trap were identified and then placed in labelled vials containing 70% ethanol. Several specimens from each species (a minimum of 10 where possible) were pinned and labelled. Voucher specimens have been deposited at the Strickland Entomological Museum, University of Alberta, Edmonton, Alberta, and the Royal BC Museum, Victoria, British Columbia. Vegetation Sampling The composition of vegetation was recorded at each pitfall trap in an approximate lm 2 area using digital photographs taken from a height of approximately lm directly above the trap. From these images the dominant ground cover and the vegetation density was assessed. Vegetation density was categorized as sparse, low, moderate, or high. Additionally, the presence of coarse woody debris (CWD) (pieces of dead wood with diameters greater than or equal to 10cm), stumps, large rocks, litter, fine woody debris (downed wood less than 10cm and greater than or equal to 0.5cm in diameter), wood in advanced decay and exposed mineral soil were recorded as they appeared in the photos. Data pertaining to canopy cover and vegetation vertical cover were collected for fifteen sampling points located in each of the ten study stands. Vegetation species were recorded as present or absent at each trap. The vegetation presence/absence data were then summed for all five traps at each sampling point. Thus a score of zero to five was possible for each species at each sampling point; five being locally ubiquitous and zero being locally absent. 23 Canopy cover (the percent of the ground area shaded by overhead foliage) was assessed using a convex spherical densiometer (Forest Densiometers, Model - A, Arlington, Virginia), which have been shown to be reliable for assessing canopy cover (Lemmon 1957). From the center of each sampling point, four readings were taken from waist height (approximately lm), facing north, east, south and west (Lemmon 1957). Mean values were calculated for each sampling point. Due to the ease and speed of use, low cost and transportability, vertical vegetation cover was assessed using a forest cover pole 2m in height divided into ten 20cm sections. The cover pole was positioned at the sampling point center. Cover was assessed from a distance of 4m and a height of lm from north, east, south and west aspects. Sections more than or equal to 25% obscured by vegetation were considered covered and counted as such (Griffith and Youltie 1988). Data Analyses Carabids were identified according to Lindroth (1961-1969). Each individual carabid was also sexed. Confirmation of species identification was undertaken at the Canadian Forest Service, Northern Forestry Centre, and the Strickland Entomological Museum, University of Alberta, both in Edmonton, Alberta. Statistical analyses were done using the SYSTAT (v. 11) (SYSTAT Software, Inc., Richmond, CA) software package, the only exception being non-metric multidimensional scaling which were done using the PC-ORD (PC-ORD v.5, MjM Software, Gleneden Beach, OR) software package. Seasonal variation was examined for the most abundant species, defined as those that comprised at least 2% of the total carabid catch. For these species the actual catch per 14 day trapping period, and five-trap trapping cluster, was standardized to a single day and then 24 multiplied by 14 to give the standardized catch for a single trapping cluster for a two week trapping period. Repeated measures ANOVAs were performed to examine differences in the seasonal activity-abundance of male and females. Post-hoc tests to determine seasonal differences in the mean activity-abundance between males and females over the trapping period were performed when the null hypothesis was rejected using univariate F tests. Alpha levels were corrected using a Bonferroni adjustment in order to account for multiple comparisons (Tabachinick and Fidell 2001). Carabid data were standardized to 100 trap-days prior to analysis. Standardization was the sum of the five traps in each trapping cluster divided by total number of days the trap cluster was operational multiplied by 100. Missing, destroyed, or damaged traps for each cluster were accounted for by multiplying the summed total of the trap cluster by 1, plus 0.2 for each missing, damaged or destroyed trap. This reduced the comparative influence of lower trap days for traps that were destroyed, damaged or otherwise disturbed during a 14 day sample period. Relationships among vegetative structure parameters were assessed by examining correlations among the vegetation cover and canopy cover. Relationships were then used to group the vertical vegetation data into three groups: Low - cover measured on the lower 20cm - 60cm of the cover pole representing herbaceous vegetation and forbs, Shrub - cover measured between 60cm - 100 cm on the cover pole representing shrubs, and High - cover measured above 100cm on the cover pole, representing mature shrubs and low branches/crowns. Three measures of diversity were calculated for carabids in three canopy cover groups: no canopy (0 to 10% coverage), developing canopy (11 to 89% canopy coverage) and closed canopy (90 to 100%) canopy coverage). Measures of gamma diversity (y) 25 (landscape level diversity; the sum of all species collected), and alpha diversity (a) (species richness per trap cluster; sum of the species collected at a single trap cluster), were calculated. Beta diversity (P) (heterogeneity in the data=a divided by y), was also calculated (McCune and Grace 2002). ANOVA was used to examine differences in alpha diversity of the three canopy cover groups. Assemblage data were visualized using NMS (McCune and Mefford 1999) with the goal of examining clusters of species, or species-groups and to explore variation in the assemblages. All NMS ordinations were run using distance matrices constructed using Sorensen distance measures. Each NMS ordination involved 50 runs, each with random and real data to ensure reliable ordination. Ordinations were considered reliable if they were significantly different than random, as determined by Monte Carlo analysis, but similar to the other ordinations run on the same data set. Determination of the number of dimensions appropriate for the data was achieved by examining NMS scree plots and selecting the number of axes beyond which reductions in stress is small (McCune and Grace 2002). Once the number of axes for interpretation had been determined, a final ordination was run as recommended by McCune and Grace (2002). Non-metric multidimensional scaling was run to produce an ordination of sample points in carabid species space. Ordinations were compared to elucidate differences based on species selections, i.e., does species selection influence the picture provided though NMS. Second matrix (explanatory variables) correlations were examined, with structural variables, to attempt to provide possible explanations for observed axes correlations. 26 Results The total carabid catch, collected from May 25, 2005 until August 28, 2005, consisted of 31 species of carabids and 4801 individual specimens (Table 1). Nine species, Scaphinotus marginatus (Fischer von Waldheim), S. angusticollis (Fischer von Waldheim), Calathus ingratus Dejean, C. advena (LeConte), Synuchus impunctatus (Say), Elaphrus lapponicus Gyllenhal, Pterostichus adstrictus Eschscholtz, P. riparius Dejean and Trechus chalybeus Dejean, comprised 77.0% of the total catch. The remaining 22 species each contributed less than one percent of total catch. Season and Gender The only species where the variation in mean activity-abundance between sexes over the trapping season was not significant was T. chalybeus (Figure 2h, Appendix II). Scaphinotus species (Figure 2a,b; Appendix II) have higher activity-abundance in late summer, while three species (Figure 2d, e, f; Appendix II) have higher activity-abundances in spring. Synuchus impunctatus (Figure 2g; Appendix II) had a peak in activity-abundance in the early summer and P. adstrictus (Figure 2c; Appendix II) activity-abundance increased from spring to late summer with a large number of males collected in the first trapping period. 27 Table 1: Summary of carabid species collected in successional sub-boreal spruce forests in west central BC, during the spring and summer of 2005. Catch for each species is standardized to number of individuals per 100 trapping days Species Trachypachus holmbergi Mannerheim Scaphinotus angusticollis (Fischer von Waldheim) Scaphinotus marginatus (Fischer von Waldheim) Carabus taedatus Fabricius Notiophilus sylvaticus Eschscholtz Elaphrus clairvillei Kirby Elaphrus lapponicus Gyllenhal Patrobus fossifrons (Eschscholtz) Trechus chalybeus Dejean Bembidion grapii Gyllenhal Bembidion fortestriatum (Motschulsky) Amerizus oblongulus (Mannerheim) Pterostichus herculaneus Mannerheim Pterostichus adstrictus Eschscholtz Pterostichus riparius Dejean Pterostichus brevicornis (Kirby) Pterostichus castaneus (Dejean) Stereocerus haematopus (Dejean) Calathus ingratus Dejean Calathus advena (LeConte) Synuchus impunctatus (Say) Agonum gratiosum (Mannerheim) Agonum affine Kirby Agonum cupreum Dejean Amara sinuosa (Casey) Amara erratica (Duftschmid) Harpalus animosus Casey Harpalus somnulentus Dejean Bradycellus conformis Fall Trichocellus cognatus (Gyllenhal) Lebia moesta LeConte Total Standardized Catch Species codes TRAHOL SCAANG SCAMAR CARTEA NOTSYL ELACLA ELALAP PATFOS TRECHA BEMGRA BEMFOR AMEOBL PTEHER PTEADS PTERIP PTEBRE PTECAS PTEHAE CALING CALADV SYNIMP AGOGRA AGOAFF AGOCUP AMASIN AMAERR HARANI HARSOM BRACON TRICOG LEMMOE Females Males 2.38 370.36 412.02 3.57 17.14 7.14 87.98 15.48 476.79 4.64 5.71 4.05 1.43 373.57 202.98 8.81 7.62 5 87.62 653.33 514.29 2.62 8.57 0 19.76 41.26 4.05 9.05 5.71 0 7.38 2.38 411.31 655.48 1.19 18.69 1.43 80.71 20.00 301.55 5.71 2.86 0 1.43 198.93 114.40 0 3.81 2.62 31.19 322.26 558.21 0 14.29 2.86 17.14 24.88 0 10.48 8.57 1.43 2.62 6112.74 Examination of the seasonal trends in activity-abundance for male and female carabids indicated male activity-abundance differed significantly from females (or vice versa) during at least one trapping period for all species except T. chalybeus (Figure 2; Appendix II). Additionally seasonal effect on activity-abundance was significant in five of the eight abundant species (Figure 2; Appendix II). Post-hoc (F test) examination indicated that seasonality significantly affected both males and females for S. marginatus, C. advena 28 and S. impunctatus. Male S. angusticollis were significantly influenced by seasonality, as were female P. adstrictus, P. riparius and C. ingratus (Figure 2; Appendix II). M Female Scaphinotus marginatus ** • • • Female Scaphinotus i Male Scaphinotus marginatus ** - ^ Male Scaphinotus angusticollis angusticollis ** 1.4 I 1.2 i I c 1.0 nj "O „ c 0.8 - CO 0.4 - •-• .£> < <? 0.2 .Q co o O CO c co CO c 0.2 CO CD CD 0.1 I 0.1 0.0 LL 0.0 Figure 2 a - d: Seasonal variation in mean activity-abundance (± SEM) for males and females of the most abundant (>2% of total catch) species of carabid standardized to 14 trap days. *= significant difference between sexes for a trapping period (a=0.025) ** = significant effect of season (a=0.05). 29 Female Calathus ingratus' Male Calathus ingratus _ • • Female Calathus advena ' i-*v I Male Calathus advena " 1.2 u.ia 0.16 - CD CJ 1.0 CO •o 0.8 c XI XI CO .£> . 0.14 - c CO 0.12 0.10 - 0.6 0.08 - o 0.4 CO c CO CD 0.2 0.06 - I 0.04 - 2 li ii i; 0.02 - g) • _ • Female Synuchus impunctatus ' ifs'. i Male Synuchus impunctatus ** h) 0.0 ?i 1.0 1.4 I h T- • _ • Female Trechus chalybeus i .- i Male Trechus chalybeus 1.2 0.8 CO I 1.0 0.8 c co "O c XI 0.6 CO 0.4 o CO 0.4 0.2 0.0 I CO CD _ 0.2 0.0 Figure 2 continued e - h: Seasonal variation in mean activity-abundance (± SEM) for males and females of the most abundant (>2% of total catch) species of carabid standardized to 14 trap days. * = significant difference between sexes for a trapping period (a=0.025) ** = significant effect of season (a=0.05). 30 Vegetation Cover There was no correlation between mean canopy cover and vertical vegetation cover classes of 0 - 20cm, 21 - 40cm, and 41 -60cm (r = -0.053, -0.114, -0.010, respectively), and only weak positive correlations between mean canopy cover and vertical vegetation cover classes of 61 - 80cm and 81 - 100cm (r = 0.210 and 0.254, respectively). For mean canopy cover, and the five remaining vertical vegetation cover classes >101cm in height, there was a stronger correlation (r = 0.335, 0.399, 0.364, 0.327, 0.407, respectively). All correlations among the vertical vegetation cover classes were positive, with adjacent classes tending to be strongly correlated, i.e., r > 0.7. Based on these results, vegetation height classes were pooled into four classes, 0 - 20cm, 21 - 60cm, 61 - 100cm, and >101cm. Values for vertical vegetative cover were then ranked. A sigmoidal regression to model canopy cover and stand age used the function: f=a/(l+exp(-(x-x0)/b)) The model yielded the best fit when a = 96.65; b = 0.37; x0 = 3.04 (adjusted R2 = 0. 91; F(2,149) = 784.86; P < 0.0001) (Figure 3). The plot showed that 16yph stands displayed great variation in canopy cover. 31 100 CD > o o >> a. o c ro O CD o CD Q. Mature Stand Age (years post harvest) Figure 3: Sigmoidal relationship between percent canopy cover and stand age (years post-harvest): adjusted R2 = 0. 91; F(2,149) = 784.86; P < 0.0001. Diversity Significant differences in mean carabid alpha diversity between canopy cover classes were noted (R2 = 0.07; F(2;i48)= 5.8; P = 0.004). Post-hoc examination (Tukey's) indicated that plots with less than 10% canopy cover had significantly greater mean diversity (P = 0.002) than plots with developing canopies (11-89% canopy cover). Plots with closed canopies (>90% canopy cover) had mean alpha diversity values slightly greater than plots with developing canopies, but the differences were not significant. Beta diversity followed the same trend (Table 2). 32 Table 2: Diversity measures calculated for carabid assemblages sampled in 2005. Mean carabid diversity is calculated by plot canopy cover classes (no canopy, developing canopy, closed canopy) Diversity measures are defined in the text. Diversity Measure Open canopy Developing canopy Closed canopy a 4.8 5.39 4.11 4.86 P 0.15 0.17 0.13 0.16 7 31 31 31 31 Assemblage Ordination Data pertaining to species collected in fewer than 5 plots were discarded prior to NMS ordination of the carabids. Removal of species with relatively few collections was done to increase the stability of the ordination (McCune and Grace 2002). This resulted in removal of 10 carabid species (Trachypachus holmbergi, Carabus taedatus, Pterostichus herculaneus, Trichocellus cognatus, Harpalus animosus, Agonum gratiosum, Amerizus oblongulus, Agonum cupreum, Elaphrus clairvillei, Patrobus fossifrons) leaving 21 carabid species in the ordination. Examination of the NMS scree plot indicated that two-axis solutions could be obtained for the ordination with a final stress of 20.34. Species points indicate the relative orientation of a given species in relation to carabid assemblages, it is important to note that the position should not be considered an absolute point, but rather a central point within a cloud of points representing the species distribution relative to all other species in the ordination (Figure 4). The percent of the total variance in the NMS ordination explained by structural variables on the x axis was 18.8%, while the y axis explained 37.0% (Figure 4). NMS correlations (Figures 4) can be found in Appendix III. The x axis includes variation contributed by positive correlations with a high grass component of ground cover. 33 The y axis includes variation contributed by positive correlations with canopy cover, high needle component of ground cover, diversity in vegetation and a negative correlation with slash; residual fine woody debris left on the ground after harvest, consisting primarily of fine twigs, small branches and cones. 0.8 • STEHAE SCAANG • • CALADV SCAMAR • • 0.4CALING PTEBRE PTECAS * NOTSYL • * • 1 -0.6 1 U.U " 1 • „ „ PTERIP -0.2 1 0.6 0.2. TRECHA * SYNIMP • -0.4AMASIN ..,„„. BEMFOR AMAERR E L A I ^p • • • BRACON • PTEADS AGOAFF AARSOM • LEBMOE • BEMGRA • -0.8« Figure 4: NMS ordination of carabid assemblages. Axes scales are the raw correlation coefficients. Species codes consist of the first 3 letters of genus and the first 3 letters of species names (see Table 1). Examination of the association between canopy cover and carabid activityabundance yielded little information that was not captured in the ordination (Figure 4). In the case of P. riparius, however, the highest activity-abundance was in the highest canopy 34 cover plots, with lower activity-abundance in the lowest canopy cover classes. Zero individuals of this species were collected in plots with 43% - 79% cover (Figure 5). 25 • O Females Males 20 25 yph and non-harvested to represent successional shifts in vegetation structure from open ground to low canopy, and a gradient ranging from no canopy (0% coverage) to closed canopy (100% coverage). Within each site, a one hectare sampling plot was randomly positioned at least 50m away from an edge (i.e., road, stream, abrupt elevation change, forest or cutblock edge). The exact study site locations are listed in Appendix I. Pitfall Sampling Trap-clusters were placed within sites to avoid inclusion of any feature that would exclude trapping, e.g., anthropogenic structures and wet areas. Each of three 80m long transects running at randomly determined bearings from randomly selected points of commencement within the plots, and not constrained by the plot boundary, were established. 52 Traps were established as described in Chapter 2 (Chapter 2, Figure 1) with trapping clusters separated by a minimum of 14m. Modified Nordlander pitfall traps (Lemieux and Lindgren 1999) as described in Chapter 2 were filled with 90mL of 25% propylene glycol solution in water. Samples were collected and the fluid replaced every 14 days from May 25, 2005 until August 28, 2005. Captured carabids were processed as described in Chapter 2. Carabids and ants from each individual pitfall trap were identified (see below) and then placed in labelled vials containing 70% ethanol. Several specimens from each species; a minimum of 10 where possible, were pinned and labelled. Voucher specimens have been deposited at the Strickland Entomological Museum, University of Alberta, Edmonton, Alberta, and the Royal BC Museum, Victoria, British Columbia. Following identification and sexing, carabids were examined for physical deformities and injuries (e.g., amputation of extremities, cuts or breaks in elytra, abnormal fusing of body or limb/antennae segments). Injuries were determined to be pre-collection if they showed evidence of sclerotization (i.e., healing) (Figures 6 & 7). Carabids with injuries showing no sclerotization, or where determination of sclerotization was impossible, were tallied as uninjured. As carabids tend to move and search for food using a random walk pattern (Wallin and Ekbom 1994), a significantly higher frequency of injuries at higher activity-abundances could be the result of intra-specific or inter-specific encounters. If the majority of injuries inflicted on conspecifics occur in contests for mates, it is reasonable to assume that males competing for mates would have a higher proportion of injuries than females of the same species. 53 Specimen Identification and Data Collection Carabids were identified according to Lindroth (1961-1969) and sexed. Confirmation of species identification was undertaken at the Canadian Forest Service, Northern Forestry Centre, and the Strickland Entomological Museum, University of Alberta, both in Edmonton, Alberta. Camponotus and Formica ants were identified to species using Wheeler and Wheeler (1963, 1986) and Naumann et al. (1999). Carabid and ant data were standardized to 100 trap-days prior to analysis. Standardization was the sum of the five traps in each trapping cluster divided by total number of days the trap-cluster was operational multiplied by 100. Missing, destroyed, or damaged traps for each cluster were accounted for by multiplying the summed total of the trap cluster by 1, plus 0.2 for each missing, damaged or destroyed trap. This reduced the comparative influence of lower trap days for traps that were destroyed, damaged or otherwise disturbed during a 14 day sample period. Formica aserva and C. herculeanus activity-abundances were then categorized. Activity-abundance of F. aserva at each trap-cluster was assigned to one of four categories based on the standardized activity-abundance at the trap-cluster. Categories were as follows: Absent - F. aserva were not observed at the trap-cluster, Low - between 1 and 50 workers were collected at the trap-cluster, Moderate - between 51 and 150 workers were collected at the trap-cluster and High - more than 150 workers were collected at the trap-cluster. F. aserva colonies in the SBS do not occur in mature stands or young post-harvest stands (Higgins 2010), thus trap-cluster occurring in mature and 2 yph stands were excluded from analysis. Camponotus herculeanus activity-abundances at the trap-cluster were similarly categorized into four groups. The ratio of C. herculeanus to F. aserva (0.225:1) was used to 54 generate C. herculeanus activity-abundance categories proportionally similar to those of F. aserva. Thus, C. herculeanus activity-abundance was categorized as: Absent - C. herculeanus were not observed at the trap-cluster, Low - between 1 and 11 workers were collected at the trap-cluster, Moderate - between 12 and 34 workers were collected at the trap-cluster and High - more than 34 workers were collected at the trap-cluster. Data Analyses Statistical analyses were preformed using SYSTAT 11 (Systat Software, Inc., Chicago, IL) except where otherwise noted. Non-linear regressions were used to examine the effect of F. aserva on C. herculeanus, on carabid activity-abundance (Sigmaplot v.l 1). Environmental effects were limited by truncating the data set including in the analyses only stands where F. aserva, C. herculeanus and carabids occurred, i.e., 12, 16, and 25 years post harvest (yph). ANOVA was used to examine the effect of F. aserva activity-abundance groups on carabids activityabundance. The truncated carabid activity-abundance was square root transformed prior to analysis. Where one-tailed ANOVAs were significant, post-hoc examination using Tukey's correction were performed. Carabid injury data were evaluated for all sites and stand ages, and then truncated as above prior to ANOVA to examine the influence of ants on injury proportions. Data plots of carabid and ant activity-abundance at the trap-cluster were used to visualize the relationship between carabids and ants in stands where they co-occurred. Species that accounted for at least 2% of the total carabid catch were examined individually. Trap-cluster by trap-cluster activity-abundance of all carabids, F. aserva, and C. herculeanus was constructed to examine the interrelationship among all species. 55 Figure 6: Scaphinotus marginatus with injured tarsus (left) vs. broken tarsus (rig Phot©! Ward B. Strong Figure 7: Injured femur of a captured Scaphinotus angusticollis, showing sclerotization at the severed end. Photo: Ward B. Strong Carabids with injuries were tallied, and frequencies of occurrence were determined for each species, gender and stand age. Observed frequencies of injury were then compared, using Chi squared tests, against expected injury frequencies. Injury proportions were calculated for each of the four F. aserva and C. herculeanus activity-abundance categories. Proportional data were then transformed as x' = loglO (x + 0.01) to achieve a normal distribution. Carabid injury proportions, grouped by F. aserva and C. herculeanus activity-abundance categories, respectively, were examined using ANOVA. For C. herculeanus, F. aserva activity-abundance was used as a covariate to account for its effect on carabids injury proportions. 57 Results Ants and Carabids Residual plots indicated that the relationship between carabids and F. aserva is curvilinear in form. Comparison of residual standard deviations and standard deviations of the data set indicated that the non-linear regression fit to a decay curve was a better fit than the linear form as did regressions examining the relationship between C. herculeanus and F. aserva. Non-linear regression showed a significant effect of F. aserva activity-abundance on carabid activity-abundance (Figure 8), although only 16.5% of the variation was explained by F. aserva activity-abundance. All of the regression coefficients were significant (a = 0.05). 100 CD o c 80 TO 60 c JQ TO 40 O I co O 20 * • • • . < * • • 200 400 600 Formica aserva activity-abundance Figure 8: Non-linear regression of carabid activity-abundance predicted by the ant Formica aserva y = 37.9587*exp(-0.0032*x); Adjusted R2 = 0.165; F = 17.439; P < 0.0001. There was no significant influence of C. herculeanus activity-abundance on carabid activity-abundance (P = 0.081) examined using non-linear regression. Activity -abundance 58 of C. herculeanus, however, shows a negative and curvilinear relationship with the activityabundance of F. aserva (Figure 9). These data failed to improve with transformation and thus failed tests for normality and homoscedasticity. o CD CO 100 C CO -2 50 t co to 3 o £c: V 30 - >> 25 > t3 20 - 20 - co V V & tr, ¥ • CO to .2 35 -i "O 15 co 15 - c 10 5 - V & co E 10 - vjfi 'v^kO• to 5 - JsFw 5o 0 - 0- • Jtfj% "\7fjm mtittQ V V V V *C?& • • • Q. to o o co 0 100 200 300 400 500 600 700 co Formica aserva activity-abundance XI co 8 30 c 25 XI co 3 CD C to to > t5 co to 15 ' 200 30 25 20 15 o o 10 •g 5 CO to 10-1 c 8 o •c Q. CD O CO 1 150 i o o •g 1 100 co •o 20 o co to 1 50 Camponotus herculeanus activity-abundance CD o c co •o c ' » 5 - c 0 s 9S7 7®C& O 9 300 400 CO CO 3 "to 0 100 200 500 600 100 150 200 14 CD O c • •CO a c3 XI 10 - • 12 10 CO > => > ^u 8 > CO to 3 6 - 2 4 - to 3 2 - o> 50 Camponotus herculeanus activity-abundance CO o <7? 0 14 12 - V «UUAW «? > CO 20 - * J • TO 0 - • o • <7 J= 8 •*-t S» • a2 V CD • • V $7 w CO CO •W • to 3 • • c CGWK? TOSOTWCOT? V V "to O :£ .to is "to O 0 100 200 300 400 500 600 Formica aserva activity-abundance 700 o • • • V •iJJ.> •? V o • <7 C E T O / i ? W C VVO 0 50 100 150 200 Camponotus herculeanus activity-abundance Figure 10 continued, e-h: Carabid activity-abundance in mid-successional sub boreal spruce stands plotted against Formica aserva (e, g) or Camponotus herculeanus (f, h) activity-abundance. Male carabids: V , Female carabids • 61 CD O CD o c c CO CO T> T> 3 XI CO c 3 XI CO .&• 4 .&• 4 "0 o CO to 3 CO CO 3 C: c to 3 •C O -G CO 50 100 150 200 u- • 8 - V 6 - • 4 V CO CL to Q. c: 0 Camponotus herculeanus activity-abundance CO 1 V cfHXSO? CCC* © • W to s rf Formica aserva activity-abundance o c 2- to 3 •C O 3 •C O 0 • c 2 <7» t? ^is»w» to 3 -C • 2 <3» V V» V o s .2 CO s 0 100 200 300 400 500 600 Formica aserva activity-abundance 700 a. 0 - AAMW C*Wg7 «CC 0 50 100 9 150 200 Camponotus herculeanus activity-abundance Figure 10 continued, i-1: Carabid activity-abundance in mid successional sub boreal spruce stands plotted against Formica aserva (i, 1) or Camponotus herculeanus (j, 1). Male carabids: V , Female carabids • 62 m CD 30 - o 1 n CD o c CO •o 25 - • 25 -^ 30 - c3 XI XI s > Tj to to s co X! 15 - co V -~ 10- • 3 £ £ _ _v umMMiX^y oI 0 100 to v •5 V «s?

•§ V CO 5 U- • %j&MiM C«OTO c 50 w c 100 «7 150 200 Camponotus herculeanus activity-abundance Figure 10 continued, m & n: Carabid activity-abundance in mid successional sub boreal spruce stands plotted against Formica aserva (p) or Camponotus herculeanus (m) activity-abundance. Male carabids: V , Female carabids I ANOVA showed a significant impact of F. aserva on carabid activity-abundance (F(3,86) = 9.15; P < 0.001), and a posteriori tests showed that trap-clusters with no or low F. aserva activity-abundance had significantly higher carabid activity-abundance than trapclusters with moderate or high F. aserva activity-abundance (Figure 11). 63 40 -| (1) o c: :ifi - CO a 30 X) (0 .& ?5 > o (0 •D X} (0 (0 o c 20 - is - CO u> 10 - i> 5 - Formica aserva activity-abundance catagories Figure 11: Relationship between mean (±SEM )activity-abundance of carabids and activity-abundance Formica aserva: Absent (0) - F. aserva were not observed at the plot, Low (1) - between 1 and 50 workers were collected at the plot, Moderate (2) - between 51 and 150 workers were collected at the plot and High (3) - more than 150 workers were collected at the plot. Means with the same letter designations are not significantly different as determined by Tukey's tests. Injury A total of 774 carabids, 16.12% of the total catch, possessed some damage that was categorized as pre-capture injury. For most common carabid species significant stand age associated differences in frequency of injury were observed (Table 3). 64 Table 3: Summary of y? analysis on injury frequency of carabids in stands before and at different successional stages after harvest1 (yph = years post harvest). ()C2crit = 3.841: df =1; a = 0.05 for all tests; Significant / 2 results values are given) Species Scaphinotus angusticollis Scaphinotus marginatus 2 yph 0 Successional stage 12 yph 16 yph 25 yph +++ 0 0* Mature 17.432 24.628 + 0 0 0 0 0 ++ n/a + 0 7.856 5.795 4.795 5.543 Pterostichus adstrictus 0 0 Pterostichus riparius 0 0 Calathus advena 0* 0* Calathus ingratus n/a 0 0 0 n/a 0 0 0 n/a 0* 0 Synuchus impunctatus Trechus chalybeus 4.609 7.138 8.197 ++ 6.807 + +++ 5.838 23.190 0 u.( 95 significantly fewer (P < 0.05) " significantly fewer (P < 0.005) - - -" significantly fewer (P < 0.001) +" significantly more (P < 0.05) + +" significantly more (P < 0.005) + + +" significantly more (P < 0.001) * low sample size n < 10 Proportions of injured carabids per trap-cluster differed significantly with F. aserva activity-abundance category (FQ, 86)= 3.215; P = 0.027) with the significantly greatest proportion of injured carabids at moderate F. aserva activity and the fewest at high F. aserva activity. C. herculeanus activity-abundance did not significantly affect carabid injury proportions (FQ, g5) = 1.738; P = 0.165), although the proportion of injured carabids was lowest at moderate C. herculeanus activity. 65 0.35 •S 0.30 XI CO L. CO o X3 0.25 CD t_ 0.20 o ° 0.15 o a. o 0.10 0.05 0 1 2 3 Formica aserva activity-abundance catagories Figure 12: Relationship between proportion of injured carabids (±SEM ) and activityabundance of Formica aserva: Absent (0) - F. aserva were not observed at the plot, Low (1) - between 1 and 50 workers were collected at the plot, Moderate (2) - between 51 and 150 workers were collected at the plot and High (3) - more than 150 workers were collected at the plot. Means with the same letter designations are not significantly different as determined by a posteriori Tukey's tests. Discussion Assemblage structure can be attributed in part to competitive interactions (e.g., Morse 1970), and has been shown to be important in structuring ant communities (Punttila et al. 1994). Inter-guild predation, i.e., predation between species that occupy similar niches, on carabid larvae has been demonstrated experimentally, and may play a role in carabid communities (Currie et al. 1996). Competition among carabid species in the absence of other ground-dwelling invertebrates has been viewed as minor for structuring carabid assemblages based on adult carabid data, however (Lovei and Sunderland 1996). 66 While not directly examined, morphological differences among species of carabids reflect differing ecological functions and specific adaptations that likely influence habitat selection (Forsythe 1987) and niche differentiation (Loreau 1988). As forests recover from anthropogenic disturbance, competitive abilities of epigaeic invertebrates influenced by stand structure can shift. Competitive shifts likely influence the ability of carabids to occupy habitats and possibly manifest as observed shifts in activity-abundance (see Chapter 2) or changes in dominance hierarchies. Shifts in competitive ability have been observed for ants (Punttila et al. 1994) and in carabids (Niemela et al. 1993) in stands recovering from disturbance. The trend towards lower carabid activity-abundance in stands with established ant colonies has been observed in numerous studies (Niemela et al. 1992, Oliver and Beattie 1996, Karhu 1998, Laakso and Setala 1998, Koivula et al. 1999, Laakso and Setala 2000, Hawes et al. 2002, Koivula 2002, Foord et al. 2003, Koivula and Niemela 2003, Mody and Linsenmair 2004 (Camponotus species), Punttila et al. 2004, Reznikova and Dorosheva 2004). Observations of increased activity-abundance of carabids with relatively low ant activity-abundance (Figure 11) may indicate that a threshold of ant activity must be reached before a decrease in carabid activity-abundance results. This pattern may be an artefact of carabids modifying their behaviour, e.g. increased movement (Reznikova and Dorosheva 2004) in the presence of potential threats (e.g., F. aserva), as opposed to an increase in the number of carabids present per se. Clustering of data points along axes in plots of carabid versus ant activity abundance (Figure 10) indicates an aversion between most abundant carabids and F. aserva, and to a much lesser degree between some species of carabids and C. herculeanus. Innate aversion of ants is thought to exist in some jumping spider species 67 (Nelson and Jackson. 2006), but may be a learned response in predominantly insectivorous carabids. Learned avoidance does not seem to occur as readily, if at all, in seed predators (Reznikova and Dorosheva 2002). Avoidance may not be universal as some species may be keying in on ant colonies as a potential food source (Kolbe 1969 cited in Reznikova and Dorosheva 2004). Thus response of carabids to the presence of ants is not necessarily similar between species, which likely contributes to the high variation observed in tests examining the carabid assemblages a whole and F. aserva. While not a member of the red wood ants (Formica rufa group), aggressive behaviour of F. aserva is likely similar to what has been observed in other dominant ant species (Savolainen et al. 1989, Punttila et al. 1994), and may result in avoidance behaviours by other invertebrates. Such avoidance behaviour has been observed in carabid and spider interactions with ants (Reznikova and Dorosheva 2004, Nelson and Jackson. 2006) and is suggested by negative regression values for both carabids and C. herculeanus relative to F. aserva. Generally when ants encounter a potential prey item the initial physical encounter involves the seizing of prey appendages with its mandibles (personal observation). Similar species of ants show aggression in the form of biting (Fellers 1987), and it is quite likely that physical interactions between carabids and F. aserva or C. herculeanus result in deleterious impacts on the injured carabid such as loss of limb segments. Carabids may attempt to avoid this type of physical interaction with formicine ants (specifically ants in the Formica rufa group) to protect their appendages (Reznikova and Dorosheva 2004). Variation in the frequency of injury among carabid species suggests different competitive relationships among predatory epigaeic invertebrates in different successional 68 stands. If observed injuries were the result of carabids competing with carabids, both interand intra- specifically, then trends in injury frequency when examined without consideration of F. aserva fit the hypothesis that competition among carabids does not play a role in structuring their communities, a conclusion also suggested by Loreau (1988). It is unlikely that the injuries observed in carabids were the result of sexual competition as variation in frequency of injury among males and females within the most abundant species was not significant. The possibility that injuries observed in carabids are the result of interactions among carabids is also unlikely, as in stands with high carabid activity-abundance observed injury frequencies were significantly lower than expected, or no different than those in stands with the low carabid activity-abundance. For example S. angusticollis (Table 4) mean activity-abundance in mature and 25 yph stands did not differ significantly, but observed injury frequency was higher than expected in the 25yph stand, but significantly lower in the mature stand, indicating carabid injury frequency is not dependant on activity-abundance, especially for S. angusticollis. Significantly higher than expected carabid injury frequencies were observed (summarized in Table 4) where ants, particularly F. aserva, appear to be the dominant epigaeic predator, while lower than expected injury frequencies occurred in stands without of with few or no ants present. Interaction with the aggressive F. aserva resulted in fewer carabids being present where the activity-abundance of this ant is high, and a greater proportion of those carabids present possess some form of injury. This relationship was not observed between carabids and C. herculeanus, although it is possible that the relatively low activity -abundance for C. herculeanus, in comparison to F. aserva, may mask their interactions with carabids. 69 Schoener (1979) developed a model for relating frequency of injury of lizards to predation pressure. He determined that if predation (by the assumed predator) is the primary cause of mortality then decreased injury frequencies indicate increased predation pressure. Subsequent studies have demonstrated that high injury frequencies also indicate inefficient predation (e.g. Medel et al. 1988). For the ant-carabid system I examined, Schoener's (1979) model likely is inappropriate as predation by ants is unlikely to be the primary agent of mortality, and injuries are not likely to be an evolutionary escape mechanism (autotomy) as seen in many lizards. The predators (ants) in my study are social, so predation efficiency is in part a function of colony size, location, and recruitment ability. Thus, a proportion of the injuries observed in carabids may be due to interference rather than predation. My results show a positive density effect of F. aserva on carabids (Figure 11); at low ant activityabundance level fewer carabids are collected suggesting avoidance. At moderate ant activity-abundance levels, the interaction between carabids and F. aserva may manifest as interference or failed predation attempts, resulting in an increase in injury frequency. At high ant activity-abundance, F. aserva is able to recruit a sufficient force rapidly enough for successful predation (decrease in injury) (Figure 12). The differences in influence exerted on carabid assemblages by C. herculeanus and F. aserva may be due to behavioural differences between the two ant species, as Formica have been shown to be more territorial and aggressive than Camponotus in Europe (Savolainen and Vepsalainen 1988). Complex interplay between stand structure, competitive abilities, assemblage composition and activity-abundance can all be inferred from the data presented here. Acknowledging the possibility of multiple interactions within the epigaeic arthropod 70 community is necessary if advancement in our understanding of forest arthropods is to be achieved. The knowledge that ants may influence the behaviour, distribution, species composition and diversity of carabid assemblages (Hawes et al. 2002, Reznikova and Dorosheva 2004), needs to be integrated into studies examining carabid beetles. Integration is particularly needed where managers seek to monitor forest invertebrates, where the relative influence of ants on carabid assemblages vary as carabid and ant assemblages shift as stand succession proceeds. The level of ant activity-abundance needed to elicit effects on carabid injury frequency and carabid activity-abundance are influenced by the environment, carabid assemblage composition, and the dominant species of ant. Since the effect of ants on carabids is density dependant, and ant densities can vary greatly by species and are also dependent on environmental variation (Holldobler and Wilson 1990), the effect that ants have on carabid communities will likely differ somewhat based on species composition and species-specific adaptation to avoiding ants (Reznikova and Dorosheva 2004). Therefore, prior to drawing broad conclusions across ecosystems and community composition based on the levels of activity presented here, further examination of interactions within epigaeic invertebrate communities that have an aggressive resident ant population are needed. While this study showed interactions between carabids, F. aserva and C. herculeanus, additional studies of mechanisms, identification of the resources subject to competition, and of various invertebrate interactions are needed to understand the dynamics of the epigaeic predatory arthropod community. 71 Reference List Cristobal-Azkarate, J., Americo, P., Dias, D., and Joaquim J. Vea. 2004.Causes of intraspecific aggression in Alouatta palliata mexicana: evidence from injuries, demography, and habitat. International Journal of Primatology 25: 939-953. Currie, C.R., Spence, J.R., and Niemela, J. 1996. Competition, cannibalism and intraguild predation among ground beetles (Coleoptera: Carabidae): a laboratory study. Coleopterists Bulletin 50: 135-148. Fellers, J.H. 1987. Interference and exploitation in a guild of woodland ants. Ecology 68: 1466-1478. Foord, S.H., Ferguson, J.W.H., and Van Jaarsveld, A.S. 2003. Coleopteran assemblages in afromontane grasslands reflect fine-scale variation in vegetation. Environmental Entomology 32: 797-806. Forsythe, T.G. 1987. Common Ground Beetles. Naturalists Handbooks 8, Richmond Publishing, Richmond, UK. 74 pp. Hawes, C , Stewart, A.J.A., and Evans, H.F. 2002. The impact of wood ants (Formica rufa) on the distribution and abundance of ground beetles (Coleoptera: Carabidae) in a Scots pine plantation. Oecologia 131: 612-619. Higgins, R.J. 2010. The ant (Hymenoptera: Formicidae) communities of the central interior of British Columbia: Adaptations to a temperature-constrained environment. Ph.D. Dissertation, University of Northern British Columbia, Prince George, BC, 184 pp. Holldobler, B., and Wilson, E.O. 1990. The Ants. Cambridge, MA: Harvard University Press. 732 pp. Karhu, K.J. 1998. Effects of ant exclusion during outbreaks of a defoliator and a sap-sucker on birch. Ecological Entomology 23: 185-194. Koivula, M. 2002. Boreal carabid-beetle (Coleoptera, Carabidae) assemblages in thinned uneven-aged and clear-cut spruce stands. Annales Zoologici Fennici 39: 131-149. Koivula, M., and Niemela, J. 2003. Gap Felling as a Forest Harvesting Method in Boreal Forests: Responses of Carabid Beetles (Coleoptera, Carabidae). Ecography 26: 179-187. Koivula, M., Punttila, P., Haila, Y., and Niemela, J. 1999. Leaf litter and the small-scale distribution of carabid beetles (Coleoptera, Carabidae) in the boreal forest. Ecography 22: 424-435. 72 Laakso, J., and Setala, H. 1998. Composition and trophic structure of detrital food web in ant nest mounds of Formica aquilonia and in the surrounding forest soil. Oikos 81: 266-278. Laakso, J., and Setala, H. 2000. Impacts of wood ants (Formica aquilonia Yarr.) on the invertebrate food web of the boreal forest floor. Annales Zoologici Fennici 37: 93-100. Lemieux, J.P., and Lindgren, B.S. 1999. A pitfall trap for large-scale trapping of Carabidae: comparison against conventional design, using two different preservatives. Pedobiologia 43: 245-253. Lessard, J., Dunn, R., and Sanders, N. 2009. Temperature-mediated coexistence in temperate forest ant communities. Insectes Sociaux 56: 149-156 Lindroth, C. H. 1961-1969. The ground beetles (Carabidae excl. Cicindelinae) of Canada and Alaska. Parts 1-6. Opuscula Entomologica xlviii + 1192 pp. Loreau, M. 1988. Determinants of the seasonal pattern in the niche structure of a forest carabid community. Pedobiologia 31: 75-87. Lovei, G.L., and Sunderland, K.D. 1996. Ecology and behavior of ground beetles (Coleoptera: Carabidae). Annual Review of Entomology 41: 231-256. Maginnis, T.L. 2006. The costs of autotomy and regeneration in animals: a review and framework for future research. Behavioural Ecology 17: 857-872. McCune, B., and Grace, J.B. 2002. Analysis of Ecological Communities. MjM software designs, Glenden Beach, OR. Medel, R.D., Jimenez, J.E., Fox, S.F., and Jaksic, F.M. 1988. Experimental evidence that high population frequencies of lizard tail autotomy indicate inefficient predation Oikos. 53: 321-324. Meidinger D, and Pojar J (compilers and editors). 1991. Ecosystems of British Columbia. B.C. Min. For. Special Report Series No. 6. 330 pp. Melbourne, B.A. 1999. Bias in the effect of habitat structure on pitfall traps: An experimental evaluation. Austral Ecology 24: 228-239. Mody, K., and Linsenmair, K.E. 2004. Plant-attracted ants affect arthropod community structure but not necessarily herbivory. Ecological Entomology 29: 217-225. Morse, D. H. 1970. Ecological aspects of some mixed-species foraging flocks of birds. Ecological Monographs 40: 119-168. 73 Murray, M. G. 1987. The closed environment of the fig receptacle and its influence on male conflict in the old world fig wasp, Philotrypesispilosa. Animal Behaviour 35: 488-506. Naumann, K., Preston, W.B., and Ayre, G.L. 1999. An annotated checklist of the ants (Hymenoptera: Formicidae) of British Columbia. Journal of the Entomological Society of British Columbia 96: 29-68. Nelson, X. J., Jackson, R. R. 2006. Vision-based innate aversion to ants and ant-mimics. Behavioural Ecology 17: 676-681. Niemela, J., Haila, Y., Halme, E., Pajunen, T., and Punttila, P. 1992. Small-scale heterogeneity in the spatial-distribution of carabid beetles in the southern Finnish taiga. Journal of Biogeography 19: 173-181. Niemela, J., Langor, D., and Spence, J.R. 1993. Effects of clear-cut harvesting on boreal ground-beetle assemblages (Coleoptera, Carabidae) in western Canada. Conservation Biology 7: 551-561. Oliver, I., and Beattie, A.J. 1996. Designing a cost-effective invertebrate survey: a test of methods for rapid assessment of biodiversity. Ecological Applications 6: 594-607. Punttila, P., Haila, Y., Niemela, J., and Pajunen, T. 1994. Ant communities in fragments of old-growth taiga and managed surroundings. Annales Zoologici Fennici 31: 131-144. Punttila, P., Haila, Y., and Tukia, H. 1996. Ant communities in taiga clearcuts: habitat effects and species interactions. Ecography 19: 16-28. Punttila, P., Niemela, P., and Karhu, K. 2004. The impact of wood ants (Hymenoptera: Formicidae) on the structure of invertebrate community on mountain birch (Betula Pubescens Ssp Czerepanovii). Annales Zoologici Fennici 41: 429-446. Reznikova, Z., and Dorosheva, H. 2004. Impacts of red wood ants Formica polyctena on the spatial distribution and behavioural patterns of ground beetles (Carabidae). Pedobiologia 48: 15-21. Ribera, I., Doledec, S., Downie, I.S., and Foster, G.N. 2001. Effect of land disturbance and stress on species traits of ground beetle assemblages. Ecology 82: 1112-1129. Savolainen, R., and Vepsalainen, K. 1988. A competition hierarchy among boreal ants: impact on resource partitioning and community structure. Oikos 51: 135-155. Savolainen, R., Vepsalainen, K., and Wuorenrinne, H. 1989. Ant assemblages in the taiga biome: testing the role of territorial wood ants. Oecologia 81: 481-486. 74 Schoener T.W. 1979. Inferring the properties of predation and other injury-producing agents from injury frequencies. Ecology 60: 1110-1115. Spence, J.R., and Niemela, J.K. 1994. Sampling carabid assemblages with pitfall traps - the madness and the method. Canadian Entomologist 126: 881-894. Taylor, P.W., and Jackson, R.R. 2003. Interacting effects of size and prior injury in jumping spider conflicts. Animal Behaviour 65:787-794. Wallin, H., and Ekbom, B. 1994. Influences of hunger level and prey density on movement patterns in 3 species of Pterostichus beetles (Coleoptera, Carabidae). Environmental Entomology 23: 1171-1181. Wheeler, G. C, and Wheeler, J. 1963. The Ants of North Dakota. Grand Forks, North Dakota, The University of North Dakota Press. 326 pp. Wheeler, G. C , and Wheeler, J. N. 1986. The Ants of Nevada. Los Angeles, Natural History Museum of Los Angeles County. 138 pp. 75 Chapter Four Effect of addition of Formica aserva (Forel) (Hymenoptera: Formicidae) nests on a carabid (Coleoptera: Carabidae) assemblage Evaluation of carabids (Coleoptera: Carabidae) as indicators of forest ecosystem health has largely ignored the influence of ants (Hymenoptera: Formicidae) on carabid distribution, abundance, diversity, assemblage composition, and behaviour. To examine the influence of Formica aserva (L.), a dominant aggressive ant, on carabids, F. aserva nests were added to a clearcut stand where it was absent. A significant effect of distance from plot center on the activity-abundance of carabids was detected. This effect was possibly influenced by colony "choice" and viability in the ant treatment. Behavioural response of carabids in the presence of ants (increased movement) and as a result of variation in habitat quality, as well as trap interaction at the 0.5m distance from the treatment may also have influenced this result. No significant treatment effects were found for individual carabid species, however. Possible explanations for the weak experimental results and the subsequent difficulties in interpretation may be due to trapping depletion due interaction between 0.5m traps, influence of habitat patch differences between the treatments and control, differences in colony size and persistence over the experiment, and issues with colony transplanting. Loss of foragers and difficulties in ensuring the presence of a queen made establishment success of transplanted colonies poor. Nevertheless, this study indicates that accounting for interactions between carabids and ants may be valuable in the development and implementation of models using carabids as indicators. 76 Introduction Interference competition is suggested to explain interactions between ants (Hymenoptera: Formicidae) of different species. For example, Formica polyctena Foerster (a Eurasian red wood ant) affected prey retrieval and selection in Formica fusca L. (Savolainen 1991), and Leptothorax sp. and Lasius flavus (F.) nest numbers increased following the decline of Lasius niger (L.), which in turn resulted from the introduction of the red wood ant Formica truncorum (F.) (Rosengren 1986). Interference competition has also been explored between ants and birds. For example, the presence of the red wood ant, Formica aquilonia Yarrow influenced Parus major (L.) foraging on trees (Haemig 1996). Interference competition has also been suggested between carpenter ants (Camponotus sp.) and spiders (Halaj et al. 1997), and between red wood ants (F. rufa group) and carabids (Hawes et al. 2002). Most species of European red wood ants tend to be arboreal in their habits, foraging more intensively on foliage than on the ground (Skinner 1980, Lenoir 2003). As such there have been several studies examining the potential utility of Formica species as biological control agents in European forests. For example, Karhu (1998) found that as distance from F. aquilonia nests increased the impact of ants on defoliators decreased during the end of an autumnal moth, Epirrita autumnata Borkhausen (Lepidoptera: Geometridae) outbreak and during an outbreak of birch aphid, Euceraphis punctipennis (Zetterstedt) (Homoptera: Aphididae). A similar association was noted in mid-elevation birch (Betula spp.) forests during an outbreak of E. autumnata (Puntilla et al. 2004). Some species of North American red wood ants also primarily forage in trees, e.g., Formica obscuripes Forel, a species which 77 may have had a significant effect on populations of defoliating insects during a spruce budworm outbreak in 1980-1992 (Mclver et al. 1997). Studies examining the impact of ants on epigaeic fauna are less common, and most of these examine the impact of arboreal-foraging ant species on epigaeic invertebrates. Laakso (1999) examined the soil fauna in a Finnish mixed Norway spruce (Picea abies (L.) Karst), silver birch (Betula pendula Roth) and Scots pine (Pinus sylvestris L.) forest, and concluded that it was unlikely that arboreal F. aquilonia impact epigaeic animals. Also grounddwelling invertebrates, with the exception of Linyphiidae spiders, were not impacted when red wood ants, F. polyctena, were excluded from trees (Lenoir 2003). In Sweden Lenoir et al. (2003) suggested that F. polyctena is not the dominant predator of epigaeic invertebrates, likely because a large proportion of these ants forager on trees (Skinner 1980), and will increase their foraging range to access trees rather than resort to foraging on the ground (Lenoir 2003). The impact of aggressive Formica rw/a-group ants vary within forest fragments in Finland. Of five aggressive territorial species, F. aquilonia, F. polyctena, F. lugubris Zetterstedt, F. rufa (L.), and F. pratensis Retzius, only F. aquilonia exerted a consistent impact on epigaeic invertebrates over the sampled area (Savolainen et al. 1989), and the other species had low impact based on their lower activity-abundances and uneven occurrence (Savolainen et al. 1989), assuming that encounters with ant foragers drive other ground-dwelling invertebrates out of the area. Studies examining carabid activity-abundance and distribution have, in cases where red wood ants have been examined, shown negative associations between ant presence and carabid beetle activity-abundance (see Niemela et al. 1992, Oliver and Beattie 1996, Karhu 1998, Laakso and Setala 1998, Koivula et al. 1999, Laakso and Setala 2000, Hawes et al. 78 2002, Koivula 2002, Foord et al. 2003, Koivula and Niemela 2003, Mody and Linsenmair 2004 (Camponotus species), Punttila et al. 2004, Reznikova and Dorosheva 2004). An experimental study by Reznikova and Dorosheva (2004) indicated that carabid responses to red wood ants are species-specific. Activity-abundance of carabids of different sizes was affected to differing degrees by red wood ant presence and density (Hawes et al. 2002). The least affected carabids were the diurnally active Notiophilus biguttatus (F.) (Niemela et al. 1992, Hawes et al. 2002). The effect of wood ant presence on carabids may be sex-specific within a species as well; Hawes et al. (2002) found a stronger negative correlation between red wood ant densities and female Abaxparallelepipedus Piller & Mitterpacher than with males of the same species. Ant nest removal experiments have resulted in the increase in biomass of predatory invertebrate mesofauna (predatory invertebrates other than ants) in Finland, this increase indicates the presence of a nearby source populations (Laakso and Setala 1998, 2000). Since thatch-mound-building ant species in Europe use foraging trail systems that are closely followed, areas exist between foraging trails that are patrolled by relatively few territorial ants (Skinner 1980, Holldober and Wilson 1990). Areas of low or high ant density may contain low numbers of predatory invertebrate mesofauna suppressed by the ant colony. Removal of the colony may result in release from competition which allows suppressed predatory invertebrate mesofauna to increase in numbers, as observed by Laakso and Setala (1998, 2000). Knowledge pertaining to ant ecology and distribution in British Columbia (B.C.), and in Canada in general, is sparse (Jurgensen et al. 2005). In particular, there is very little information pertaining to ant ecology and behaviour in northern B.C. Naumann et al. (1999) 79 summarized what is known about ants in B.C. Lindgren and Maclsaac (2002) studied the importance of dead wood as a nest substrate for forest ants in the central interior of B.C. The dominant ant that occurs in regenerating stands of 8-30 years post-harvest in west central B.C. is almost exclusively Formica aserva Forel (= F. subnuda Emery; Higgins and Lindgren in prep). Formica aserva is a polygynous species in the F. sanguinea group (Savolainen and Deslippe 1996, Savolainen and Deslippe 2001). Ants in this group are facultative slave-making ants that lack morphological specialization for the acquisition of slaves, instead relying on aggression to overwhelm other Formica colonies and capture pupae, a portion of which become slaves (Savolainen and Deslippe 1996). Formica aserva colonies produce sexual offspring later in the spring than red wood ants. Production of the sexual caste requires large amounts of protein-rich food, which F. aserva foragers and their slaves procure in the form of insect prey (Savolainen and Deslippe 1996). Like the red wood ants (F. rufa group), F. aserva is aggressive and tends to occupy coarse woody debris in states of moderate decay (Lindgren and Maclsaac 2002). It probably competes with carabids for resources in habitats where they occur together. Colonies of F. aserva are generally patchy in distribution, but can be locally abundant with high colony densities (Francoeur 1997). The objective of this study is to examine the effect that introduced colonies of F. aserva have on the carabid assemblage in young regenerating forests. Methods Site Selection The site was selected based on the following criteria: absence of F. aserva, sufficient size to contain 40 experimental trials and a post regeneration age of 2-4 years. The study site 80 was located in the south Nadina, Houston Forest Products West Fraser (HFP) cutblock number 045-1. Information obtained from HFP in 2005 indicated that the block was sub boreal spruce moist cold variant (SBS mc2) with pine leading prior to harvest sub boreal spruce stands are typified by a shrub/herb layer dominated by Cornus canadensis L., Vaccinium membranaceum Dougl. ex Torr., Lonicera involucrata (Richards.), Viburnum edule (Michx.) Raf., Rubus pedatus Sm., Petasites frigidus var. palmatus (Ait.) Cronq., and the mosses Ptillium crista-castrensis (Hedw.) De Not., Pleurozium schreberi (Brid.) Mitt, and Hylocomium splendens (Hedw.) Schimp. (Meidinger and Pojar 1991). Formica aserva colony selection Colonies of F. aserva in pieces of coarse woody debris (CWD) with a maximum length of 2 m and diameters between 20 cm and 40 cm were located in clearcuts between 10 and 20 years post-harvest. All colonies were collected within 50 km of the experimental plot. Colony vigour was assessed by response to disturbance; only vigorous colonies, i.e., colonies that respond to disturbance with greater than 25 workers, were selected. Selected individual colonies, in CWD, were placed into body bags. Body bags prevented loss of individuals and nest material in transit. Colony selection and transport to the experimental site was undertaken on 3-9 June 2006. Colonies and their nesting substrate (CWD) were moved by truck to the study area, carried to the randomly assigned treatment replicate, and placed perpendicular to the pitfall transect. Experimental Design and Data Collection Ten replicates of the control and coarse woody debris treatment, and 19 ant treatments were established in a randomized complete block design. UTM coordinates for 81 each replicate are given in Appendix IV. Plots were separated by a linear distance of 100m from edges and neighbouring plots as determined by GPS positioning. The control and treatments were: 1) Control - CTRL - (10 replicates) - No coarse woody debris or ants added. 2) Coarse woody debris - CWD - (10 replicates) - Single pieces of class 3 CWD, free of ants and approximately 2 m long with a diameter between 10 and 30 cm, were placed at plot center with its long axis perpendicular to randomly selected transect bearings. 3) Colony introduction - ANT - (19 replicates) - Single piece of CWD with a maximum length of 2m and a diameter between 20 and 40 cm containing a single F. aserva colony was placed at plot center with its long axis perpendicular to the randomly selected transect bearing. Linear transects consisting of 10 pitfall traps were each run perpendicular to the long axis of the CWD at plot center. Pitfall traps were a modified Nordlander type after Lemieux and Lindgren (1999). Each trap was constructed out of an 8oz translucent multipurpose container (VWR catalogue number 4333-002) with a diameter of 8cm and depth of 7.5cm. Fourteen 12mm long x 6mm high entrance holes were punched below the lid of the container using a hole punch. Traps contained 90mL of 25% propylene glycol solution. Collection of samples occurred every two weeks from June 30, 2006 to September 7, 2006. At all plots, a transect bearing was randomly selected. The treatment (CWD, ANT) or control (CTRL) was located at the transect midpoint. Pitfall traps were placed along the 82 transect at 0.5m, 5m, 10m, 15m and 20m in both directions (Figure 13) for a total of 10 traps per treatment plot. • Pitfall trap CXTreatment or Control £0n 15m 1 0.5m -Q 10m Figure 13: Diagram of trap layout for all treatment types. Closed dots indicate pitfall locations and the open dot indicates control or treatment. A total of 390 pitfall traps were installed over 3 days, from May 31st, 2006 to June 2nd, 2006. All control and CWD replicates were started on June 2nd and 3rd, 2006, whereas the ant treatments were started on the dates that the colony was added as described above. Data Analyses Carabids were identified according to Lindroth (1961-1969) and sexed. Confirmation of species identification was undertaken at the Canadian Forest Service, Northern Forestry Centre, and the Strickland Entomological Museum, University of Alberta, both in Edmonton, Alberta. Formica aserva was identified using Wheeler and Wheeler (1963, 1986) and 83 Naumann et al. (1999). As some traps were occasionally disturbed or destroyed during a trapping period all catches were standardized by dividing the total seasonal catch by the total number of trapping days that the trap was active. Catches for within plot traps at equivalent distances were summed. The standardized catch per trap day was then multiplied by 96 to give a value representing the experimental trapping season. A repeated measures ANOVA (SYSTAT 11, SYSTAT Software, Inc., Richmond, CA), with carabid activity-abundance as the dependent variable over trap distance from plot center as the independent variable, was used to examine the effect of trap distance from treatment center and treatment type on carabid activity-abundance. Pooled carabid data for all species were logio +1 transformed prior to analysis to achieve normal distribution. Data for species that comprised at less than 1 % of the total tended to be very nonnormal and had zero values for most traps, making the data unsuitable for transformation; these data, and data for the large-bodied species are graphically displayed but not statistically analysed. Species were pooled according to mean body lengths into 3 categories (Table 4) (as per Hawes et al. 2002). Data were then transformed to achieve a more normal distribution prior to analysis of variance. As Trechus chalybeus Dejean contributed nearly all the data pertaining to small species, analysis of this species can be considered the small species category. Formica aserva data for the ANT treatment were log transformed prior to analysis (there were no F. aserva in the CWD and CTRL). Results Of the 19 colonies of F. aserva relocated to randomly selected treatment plots, eight were abandoned in the first trapping period and three more by the end of the field season. In addition three colonies relocated themselves to other pieces of CWD and one colony was 84 consumed by a bear. Thus, a total of 11 colonies, representing 58% of those introduced, failed and were not used in the analyses. The relocated colonies and the colony that was destroyed by a bear were included, however. Samples collected during sample periods when the relocated colonies moved were discarded as it is likely that during the period of relocation the influence of the ants differed from that of established colonies. Consequently, the ANT treatments were highly variable in total sampling time varying from a low of 21 trap days to a high of 96 trap days. Control and CWD treatments were all sampled for 96 trap days. The total raw catch consisted of 25 species and 3716 specimens of carabids (Table 6). 85 Table 4: Carabid species collected and percent contribution by each species to the total catch. Bold indicates the most abundant species, which were analyzed individually. Species Individuals Percent of catch Trachypachus holmbergi Mannerheim" 0.161464 6 Carabus taedatus Fabricius * 0.430571 16 Scaphinotus marginatus (Fischer von Waldheim)" 101 2.717976 Scaphinotus angusticollis Mannerheim * 25 0.672766 Patrobus fossifrons (Eschscholtz) ^ 0.134553 5 Trechus chalybeus Dejean "* 168 4.52099 Pterostichus castaneus (Dejean) 0.161464 6 Pterostichus adstrictus Eschscholtz" 18.59526 691 Pterostichus riparius Dejean** 403 10.84499 0.053821 Pterostichus brevicornis (Kirby) " 2 Calathus ingratus Dejean ** 32 0.861141 Calathus advena Leconte" 2.421959 90 54.60172 Synuchus impunctatus Say" 2029 Agonum gratiosum (Mannerheim) " 1 0.026911 Agonum cupreum Dejean 1 0.026911 Bembidion grapii Gyllenhal *** 5 0.134553 Amara hyperborea Dejean " 1 0.026911 Amara sinuosa (Casey) " 0.215285 8 Amara errata Kirby " 0.968784 36 Harpalus animosus Casey * 0.941873 35 0.861141 Harpalus somnulentus Dejean 32 0.053821 Trichocellus cognatus (Gyllenhal) "* 2 Bradycellus conformis Fall ^ 0.134553 5 0.053821 Lebia moesta Leconte *" 2 0.161464 Cymindis cribicollis Dejean " 6 0.215285 Unknown Carabidae* 8 100 Total 3716 * damaged specimens % = Large carabid group; species with a mean length > 13mm J J = Medium carabid group; species with a mean length between 6mm and 13mm J|J = Small carabid group; species with a mean length < 6mm A large proportion of the catch consisted of three species of carabids: 2029 Synuchus impunctatus Say (54.6% of the catch), 691 Pterostichus adstrictus Eschscholtz (18.6%), and 403 Pterostichus riparius Dejean (10.8%). While not significant (F(2,24)= 1-3; P- 0.288), mean activity-abundances of carabids (Figure 14) were consistently higher for the ANT 86 treatment than for the CWD treatment and CTRL at all trap distances except at 0.5m. Formica aserva activity-abundance was significantly affected by distance from treatment center (F(4,96) = 10.626; P < 0.0001) (Figure 15). 0.5 m 5m 10m 15 m • • ANT t- 1 ClHL 20 m Distance from treatment (m) Figure 14: Pooled mean (± SEM) activity-abundances of carabids for: control (CTRL), and two treatments coarse woody debris (CWD) and Formica aserva nest (ANT). 160 o c to T3 120 c 100 co 80 CJ CO c co CD M Ii 140 ANT I CTRL CWD 60 40 20 0 0.5 m i 5m 10m 15 m 20 m Distance from treatment (m) Figure 15: Mean (± SEM) activity-abundance of Formica aserva for: control (CTRL), and two treatments coarse woody debris (CWD) and F. aserva nest (ANT). ANOVA indicated a significant effect of distance from the plot center on the mean carabid activity-abundance (F(4;24)= 10.6; P < 0.001), and no significant effect of treatment or treatment x distance. 87 Activity-abundance data pertaining to individual species, as well as carabids in the small, medium and large size categories did not meet the assumptions required for statistical analysis; therefore the following data summaries are descriptive rather than statistical in nature. High activity-abundance variation within species and between treatments and control, and a lack of obvious or unifying trends contribute to the inability to draw conclusions, however, activity-abundance at the 0.5m distance for all treatments was low, except for T. chalybeus (Figure 16). Variation in activity-abundance at other sampling distances cannot be explained by the data collected in this experiment. 88 a) Pterostichus adstrictus 16 CD U c CO •o 14 H c 12 3 10 CO O CO c CO CD 8 6 4 • i I i I I l a Ini 2 05m 5m 10 m 15 m 20 m Distance from treatment b) Synuchus impunctatus CD CJ • • ANT r ~ 1 CTRL c CO •o 300 c CO 200 O CD c 100 - CO CD _Il 0- 05m n B3 5m 1 . I M .. s i 10m 15m • i i 20 m Distance from treatment 25 c) Calathus advena CD o c CO •o c CO 20 15 H 1 0 05 oo Ha XL 05m 5m 10 m in! 15 m 20 m Distance from treatment Figure 16: Mean (± SEM) activity-abundance of carabids at five distances from treatment center for control (CTRL), coarse woody debris with no ants (CWD) and coarse woody debris with a Formica aserva nest (ANT). 89 40 d) Pterostichus riparius CD O c CO -° 30 .a CO >• 20 o CO 10 c CO CD An 5m 0.5 m 10m a^a • — • 15m 20 m Distance from treatment e) Trechus chalybeus CD O • • c I T S pyni 1... v....•<:•...i O I r\|_ ANT CO 73 C XI CO I > 2 - o CO c II • CO CD 0.5 m 1 5m i i i 10 m ii i 111 —r 15 m 20 m Distance from treatment f) Large carabids 5 -i CD O c CO "O c XI CO I > •2 2 o CO c CO CD .6. 10M •H i 0.5 m 5 m 10m W .—. B 15m 20 m Distance from treatment Figure 16 continued: Mean (± SEM) activity-abundance of carabids at five distances from treatment center for control (CTRL), coarse woody debris with no ants (CWD) and coarse woody debris with a Formica aserva nest (ANT). 90 Discussion Although there was no treatment effect on the activity-abundance of the entire carabid assemblage there was a significant effect of distances from treatment center. These findings are not consistent with the findings of Reznikova and Dorosheva (2004) and Hawes et al. (2002) as the ANT treatment failed to show significant effects on the carabid assemblage or individual species. High variation within species and between replicates, coupled with relatively low sample sizes and non-normal data made statistical analysis for all species difficult. Species specific responses to the introduced nest were noted in shifts in activity-abundances (Figure 16), but none were significant. Depression of carabid activityabundance at 0.5m was observed for all treatments including the control. This depression is likely due to interaction between the paired traps at 0.5m. Trap interactions and "trapping depletion" (Digweed et al. 1995) may have been mitigated by a different sampling protocol. Slightly higher activity-abundances of carabids at distances greater than 0.5m (with the exception of the 5m distance), for the ANT replicates when compared to control and CWD replicates, may be due to habitat differences. F. aserva relocate or abandon colonies when the conditions at the nest site become unfavourable (Higgins 2010). Trends observed for the control treatments (CWD and CTRL) appear to be similar for P. adstrictus, S. impunctatus, T. chalybeus and P. riparius. This suggests an environmental influence that more or less elicits similar responses from carabids. There is a possibility of microhabitat selection by F. aserva colonies, i.e., the location of a nest may be in part determined by favourable microhabitat conditions. The random assignment of treatment or control may have led to numerous nests being placed in poor habitat, leading to abandonement, which in turn may have introduced bias into the random assignment of treatment or control. Carabids are 91 known to have patchy distributions that infer a link to habitat quality or microhabitat availability (Niemela et al. 1993). It is possible that the habitats that were good for relocated F. aserva colonies also happened to be good for carabids; however, higher activityabundance of carabids 5m from F. aserva nests coupled with lower activity-abundances at 0.5m, suggests an interaction between carabids and F. aserva although trap interference is another possibility. Observations of prey collected by foraging Formicinae suggest that carabids are a very minor prey item (Skinner 1980, Lenoir 2003). The increased activity-abundance 5m away from the nest suggest that ant colonies are excluding carabids from the area or causing carabids to alter their behaviour, e.g,. increasing speed of travel and reducing periods of rest (no movement), as observed for several species by Reznikova and Dorosheva (2004). A major difficulty with introducing ant colonies for the purpose observing an effect on the resident assemblage is evenness among replicates. High variation between colonies in terms of numbers of ants and the degree of aggressive behaviour between colonies of different sizes makes interpretation difficult. The impact of colonies that have likely lost large numbers of foragers, been introduced into a novel environment, and are considerably weakened by loss of individual foragers, is probably much lower than the impact of a strong colony that has not been moved and should be considered when examining the influence ant colonies have on carabids. Along with the results of previous research pointing to an effect by ants on carabid assemblages, my study indicates that accounting for interactions between carabids and ants may be valuable in the development and implementation of models using carabids as bioindicators. Further research into the mechanisms and species-specific behavioural 92 responses of carabids to ants would greatly enhance model utility and application in indicator studies Reference List Digweed, S.C., Currie, C.R., Carcamo, H.A., and Spence, J.R. 1995. Digging out the "digging-in effect" of pitfall traps: influences depletion and disturbance on catches of ground beetles (Coleoptera: Carabidae). Pedobiologia 39: 561-576. Foord, S.H., Ferguson, J.W.H., and Van Jaarsveld, A.S. 2003. Coleopteran assemblages in afromontane grasslands reflect fine-scale variation in vegetation. Environmental Entomology 32: 797-806. Francoeur, A. 1997. Ants (Hymenoptera: Formicidae) of the Yukon, ed. Danks, H.V., and Downes, J. A. In Insects of the Yukon. Biological Survey of Canada (Terrestrial Arthropods), Ottawa, pp. 901-910. Haemig, P.D. 1996. Interference from ants alters foraging ecology of great tits. Behavioral Ecology and Sociobiology 38: 25-29. Halaj, J., Ross, D.W., and Moldenke, A.R. 1997. Negative effects of ant foraging on spiders in douglas-fir canopies. Oecologia 109: 313-322. Hawes, C, Stewart, A.J.A., and Evans, H.F. 2002. The impact of wood ants (Formica rufa) on the distribution and abundance of ground beetles (Coleoptera: Carabidae) in a Scots Pine Plantation. Oecologia 131: 612-619. Higgins, R.J. 2010. The ant (Hymenoptera: Formicidae) communities of the central interior of British Columbia: Adaptations to a temperature-constrained environment. Ph.D. Dissertation, University of Northern British Columbia, Prince George, BC, 184 pp. Holldobler, B., and Wilson, E.O. 1990. The Ants. Cambridge, MA: Harvard University Press. 732 pp. Jurgensen, M., Storer, A., and Risch, A. 2005. Red wood ants in North America. Annales Zoologici Fennici 42: 234-242. Karhu, K.J. 1998. Effects of ant exclusion during outbreaks of a defoliator and a sap-sucker on birch. Ecological Entomology 23: 185-194. Koivula, M. 2002. Boreal carabid-beetle (Coleoptera, Carabidae) assemblages in thinned uneven-aged and clear-cut spruce stands. Annales Zoologici Fennici 39: 131-149. Koivula, M., and Niemela, J. 2003. Gap felling as a forest harvesting method in boreal forests: responses of carabid beetles (Coleoptera, Carabidae). Ecography 26: 179-187. 93 Koivula, M., Punttila, P., Haila, Y., and Niemela, J. 1999. Leaf litter and the small-scale distribution of carabid beetles (Coleoptera, Carabidae) in the Boreal Forest. Ecography 22: 424-435. Laakso, J. 1999. Short-term effects of wood ants (Formica aquilonia Yarn) on soil animal community structure. Soil Biology and Biochemistry 31: 337-343. Laakso, J., and Setala, H. 1998. Composition and trophic structure of detrital food web in ant nest mounds of Formica aquilonia and in the surrounding forest soil. Oikos 81: 266-278. Laakso, J., and Setala, H. 2000. Impacts of wood ants (Formica aquilonia Yarr.) on the invertebrate food web of the boreal forest floor. Annales Zoologici Fennici 37: 93-100. Lemieux, J.P., and Lindgren, B.S. 1999. A pitfall trap for large-scale trapping of Carabidae: comparison against conventional design, using two different preservatives. Pedobiologia 43: 245-253. Lenoir, L. 2003. Response of the foraging behaviour of red wood ants (Formica rufa group) to exclusion from trees. Agricultural and Forest Entomology 5: 183-189. Lenoir, L., Bengtsson, J., and Persson, T. 2003. Effects of Formica ants on soil fauna-results from a short-term exclusion and a long-term natural experiment. Oecologia, 134: 423-430 Lindgren, B.S., and Maclsaac, A.M. 2002. A preliminary study of ant diversity and of ant dependence on dead wood in central interior British Columbia. In: Laudenslayer W.F., Jr, Shea, P.J., Valentine, B.E., Weatherspoon, C.P., Lisle, T.E., technical editors. Proceedings of the Symposium on the Ecology and Management of Dead Wood in Western Forests; Nov 2-4 1999; Reno, NV. Albany, CA; Pacific Southwest Research Station, Forest Service, USDA. Gen. Tech. Rep. PSW-GTR-181: p 111-119. Lindroth, C. H. 1961-1969. The ground beetles (Carabidae excl. Cicindelinae) of Canada and Alaska. Parts 1-6. Opuscula Entomologica xlviii + 1192 pp. Mclver, J.D., Torgersen, T.R., and Cimon, N.J. 1997. A supercolony of the thatch ant Formica obscuripes Forel (Hymenoptera: Formicidae) from the blue mountains of Oregon. Northwest Science 71: 18-29. Meidinger, D., and Pojar, J. (compilers and editors). 1991. Ecosystems of British Columbia. B.C. Min. For. Special Report Series No. 6. 330p. Mody, K., and Linsenmair, K.E. 2004. Plant-attracted ants affect arthropod community structure but not necessarily herbivory. Ecological Entomology 29: 217-225. Naumann, K., Preston, W.B., and Ayre, G.L. 1999. An annotated checklist of the ants (Hymenoptera: Formicidae) of British Columbia. Journal of the Entomological Society of British Columbia 96: 29-68. Niemela, J., Langor, D., and Spence, J.R. 1993. Effects of clear-cut harvesting on boreal 94 ground-beetle assemblages (Coleoptera, Carabidae) in western Canada. Conservation Biology 7: 551-561. Niemela, J., Spence, J.R., and Spence, D.H. 1992. Habitat associations and seasonal activity of ground-beetles (Coleoptera, Carabidae) in central Alberta. Canadian Entomologist 124: 521-540. Oliver, I., and Beattie, A.J. 1996. Designing a cost-effective invertebrate survey: a test of methods for rapid assessment of biodiversity. Ecological Applications 6: 594-607. Punttila, P., Niemela, P., and Karhu, K. 2004. The impact of wood ants (Hymenoptera: Formicidae) on the structure of invertebrate community on mountain birch (Betula pubescens ssp. czerepanovii). Annales Zoologici Fennici 41: 429-446. Reznikova, Z., and Dorosheva, H. 2004. Impacts of red wood ants Formica polyctena on the spatial distribution and behavioural patterns of ground beetles (Carabidae). Pedobiologia 48: 15-21. Rosengren, R. 1986. Competition and coexistence in an insular ant community - a manipulation experiment (Hymenoptera: Formicidae). Annales Zoologici Fennici 23: 297302. Savolainen, R. 1991. Interference by wood ant influences size selection and retrieval rate of prey by Formica fusca. Behavioral Ecology and Sociobiology. 28: 1-7. Savolainen R., and Deslippe, R.J. 1996. Facultative and obligate slavery in formicine ants: frequency of slavery, and proportion and size of slaves. Biological Journal of the Linnean Society 57: 47-58. Savolainen, R., and Deslippe, R.J. 2001. Facultative and obligate slave making in Formica ants. Naturwissenschaften 88: 347-350. Savolainen, R., Vepsalainen, K., and Wuorenrinne, H. 1989. Ant assemblages in the taiga biome: testing the role of territorial wood ants. Oecologia. 81: 481-486. Skinner, G.J. 1980. The feeding habits of the wood-ant, Formica rufa (Hymenoptera: Formicidae), in limestone woodland in north-west England. The Journal of Animal Ecology 49: 417-433. Wheeler, G. C , and Wheeler, J. 1963. The Ants of North Dakota. Grand Forks, North Dakota, The University of North Dakota Press. 326pp. Wheeler, G. C , and Wheeler, J. N. 1986. The Ants of Nevada. Los Angeles, Natural History Museum of Los Angeles County. 138 pp. 95 Chapter 5: Synthesis The purposes of my study were to: (1) describe the carabid assemblage in a cool forest ecosystem classified in the sub boreal spruce biogeoclimatic zone (Meidinger and Pojar 1991), and assess the effect of clearcut harvesting, and how carabid communities change in post-harvest stands as forest succession proceeds and stand canopies close (Chapter 2); (2) examine the effect of two abundant species of ants, Formica aserva Forel and Camponotus herculeanus (L.) on carabid activity-abundance, and injury frequency in stands with varying canopy closure (Chapter 3); and (3) measure interactions between carabids and F. aserva in a young regenerating stand through experimental introduction of ant nests (Chapter 4). The theory that an organisms' environment influences its ability to exist, persist, and thrive, is a basic principle of ecology. Determining where a species exists is a prerequisite to exploring more specific questions concerning population and community ecology (Krebs 1972). In west central British Columbia, little base line data pertaining to ants and carabids and their interactions exist, but such data are essential to answer basic ecological questions. To this end, I collected data regarding the composition, and basic habitat associations of carabids within unharvested and harvested sub boreal spruce stands using a chronosequence (Chapter 2). Thirty one species of carabids and 4801 individual specimens were continuously collected over 6 two week sampling periods from May 25, 2005 until August 28, 2005 (Chapter 2; Table 1). For the common carabid species that contributed at least 2% of the total catch, differences in seasonal activity-abundances and sex ratios were examined and found to significantly vary for Scaphinotus marginatus Fischer, Scaphinotus angusticollis 96 Fischer, Pterostichus adstrictus Eschscholtz, Pterostichus riparius Dejean, Calathus advena Leconte, Calathus ingratus Dejean, Synuchus impunctatus Say, but not for Trechus chalybeus Dejean (Chapter 2; Figure 2; Appendix II). For the two Scaphinotus species and S. impunctatus, the trends observed differed from those observed in other geographic locations and in differing forest type associations, suggesting that some species of carabids may be quite plastic in their habitat associations and seasonal activity patterns. It is, however, important to note that information on forest cover associations and seasonality is rare or absent from the literature for some of the species in my study, although information for several species has been summarized by Larochelle and Lariviere (2003), and the University of Alberta Strickland Entomological Museum on line data base 2 . Differences in activity and broad habitat associations are likely driven by coarse-scale environmental differences that occur over large geographic areas (Work et al. 2008). Consideration of variability in carabid species' seasonal activity-abundance is especially important if an indicator value is placed on individual species' proportional contribution to a given assemblage. In addition, it is important to recognize that the response to an indicator value of any given species may differ depending on the ecosystem in which it is being studied, and the composition of the epigaeic community. Many carabid studies have sought to elucidate differences in carabid assemblage composition in differing habitats across broad geographic areas encompassing several different ecosystems. This large-scale examination of variation has demonstrated different habitat associations for carabids (Atlegrim et al. 1997, Ings and Hartley 1999, Heyborne et al. 2003, Vance and Nol 2003, Brouat et al. 2004, de Warnaffe and Lebrun 2004), but has 2 http://www.entomologv.ualberta.ca/index.html. Accessed 2009-11-13. 97 not specifically examined how disturbance influences carabids within a single forest ecotype through the successional progression from clearcut to mature forest. In my study, I was able to examine shifts in carabid assemblage composition associated with succession; similar trends have also been observed in carabid communities associated with stand succession in other forest types (Baguette and Gerard 1993, Niemela et al. 1993, Brumwell et al. 1998, Koivula et al. 2002, Lemieux and Lindgren 2004). Variation in carabid species composition occurred with shifts in canopy cover (analogous to time since disturbance), which is consistent with the findings of other studies (Magura and Tothmeresz 1997, Brumwell et al. 1998, Humphrey et al. 1999, Jukes et al. 2001, Koivula 2002, Koivula et al. 2002, Magura 2002, Magura et al. 2002, Heyborne et al. 2003, Lassau et al. 2005). Of the 21 species included in nonmetric multidimensional scaling analyses, four clusters or habitat association groups, and a single species "group", were observed (Chapter 2; Figure 4). Grouping of species tended to be governed primarily by variation contributed by positive correlations with a high grass component of ground cover and by positive correlations with canopy cover, high needle component of ground litter, and diversity in vegetation, as well as a negative correlation with slash. Forest cover alters the forest floor environment as it develops (Hamilton 1988). As the canopy increases with time since disturbance, the ability of organisms that occupy the forest floor to acquire resources change, as do the resources they utilize and their thermal environment. Niemela et al. (1993) noticed higher diversity and activity-abundance of carabids in stands with developing canopies in Alberta. They attributed this to the occupation of such disturbed habitats by forest generalist species, open ground specialists that have persisted as the canopy cover has increased, and forest specialists that are beginning to re-occupy the as colonists. This pattern was not observed in 98 my study. Carabid diversity and activity-abundance was highest in mature stands and in stands without canopies. As canopy cover increased with time since disturbance, in keeping with general forest succession models, additional factors appeared to reduce either, the quantity of habitat available to carabids, the resources available to them, or some combination of both. This factor appears to be ants, specifically Formica aserva. Exactly how F. aserva influences the habitat or the ability of carabids to acquire resources will require additional research. While forest canopy (Magura and Tothmeresz 1997, Brumwell et al. 1998, Humphrey et al. 1999, Jukes et al. 2001, Koivula 2002, Koivula et al. 2002, Magura 2002, Magura et al. 2002, Heyborne et al. 2003, Lassau et al. 2005) and vegetation (Hawes et al. 2002, Brose 2002, Vanbergen et al. 2007) influence habitat suitability for carabids, it appears that the presence of ants influences their finer scale spatial distribution (Reznikova and Dorosheva 2004). Data displaying negative associations (Chapter 3: Figure 10) shows a trend towards aversion towards F. aserva in most of the abundant species of carabids, with a weaker trend, that may be due to lower ant numbers, towards C. herculeanus. The aversion of carabids to ants, particularly Formica rufa-group (so called red wood) ants, has been seen in numerous carabid ecology studies (Niemela et al. 1992, Oliver and Beattie 1996a, Karhu 1998, Laakso and Setala 1998, Koivula et al. 1999, Laakso and Setala 2000, Hawes et al. 2002, Koivula 2002, Foord et al. 2003, Koivula and Niemela 2003, Mody and Linsenmair 2004 (Camponotus species), Punttila et al. 2004, Reznikova and Dorosheva 2004). Further analysis of the influence of F. aserva on carabids indicated that there is an activityabundance threshold of F. aserva above which carabid activity-abundance is significantly lowered (Chapter 3: Figure 7). This trend was also suggested in the F. aserva colony 99 introduction experiment, where carabid activity-abundance was slightly higher (but not significantly so) near colonies than in control plots although trap interactions were likely a factor as well (Chapter 4: Figure 14). While this may appear to indicate that carabids increase in the presence of ants, a more likely explanation is that carabids alter their behaviour in the presence of ants as observed by Reznikova and Dorosheva (2004). By increasing rates of movement and decreasing resting periods in the presences of ants the probability of carabids getting caught in pitfall traps will increase, even if the absolute number of carabids in the proximity of a trap is no higher. The interaction between carabids and F. aserva may result in detrimental effects on the carabids. While evidence of carabids as prey of ants has not been reported in the literature a significantly higher proportion of carabids tend to be injured in plots with moderate F. aserva activity-abundance (Chapter 3: Figure 9). As F. aserva colonies mature, and grow in size, it is possible that these ants shift from being an ineffective predator that interferes with carabids to an effective predator with a higher capture rate, which results in lower carabid injury frequency (Chapter 3: Figure 9). Species-specific trends in injury frequency support the assertion that in stands with F. aserva present, the frequency of collecting injured carabids was significantly higher than the frequency of injury in stands where F. aserva were absent (Chapter 3: Table 4). These stands also tended to be those with developing canopies and significantly lower carabid diversity (Chapter 2: Table 2). While the evidence indicating that F. aserva is responsible for injuries to carabids is circumstantial, it is nevertheless compelling (Chapter 3). Additional trials, specifically designed to investigate if similar injuries can be replicated, either in laboratory settings or the field, would be required for a definitive determination of cause and effect. 100 In examining the data presented in Chapter 4, one cannot help but notice the overall higher (but not significantly so) carabid activity-abundance in the ant treatment compared to the controls. While colony failure was anticipated in the experimental design (58% of the relocated colonies were unsuccessful), the possibility that within stand variation in the treatment plots would influence colony success was not. The failure of colonies tended to occur in spots that could possibly be considered poor habitat, not just for F. aserva but also for the carabids. Within the sub boreal spruce stands examined, and in other habitats, carabids have been shown to be selective in their habitat choices (Niemela and Halme 1992, Niemela et al. 1992). In my study, habitat-use by differing groups of carabids (Chapter 2: Figure 4) seems to be driven by canopy cover, as well as vegetative structure and composition, although there is also an aversion to the presence of F. aserva in carabid assemblages (Chapter 3: Figure 8) as well as in most of the abundant species collected (Chapter 3: Figure 10). Habitat selectivity has also been demonstrated to occur in ants. Selection can be influenced by thermal requirements (Higgins in press), prey species availability; including competitive exclusion (Nonacs and Dill 1990), or nesting substrate/host species. Therefore it is possible that in the colony-introduction experiment, F. aserva that persisted "chose" to remain at the location where they were placed because of a favourable habitat, while the other nests relocated to more favourable sites or failed. These favoured sites may also have been higher quality habitats for the carabids. Therefore, it is likely that carabids, in the absence of F. aserva, would select similar if not the same habitat patches that ant colonies essentially exclude them from during stand successional stages favouring ants, as shown in the negative associations in Chapter 3 (Figure 10). 101 Use of similar habitats, exploitation of similar resources and co-occurrence at the same time and in the same space likely leads to interaction. The presence of a larger proportion of injured carabids, and fewer carabids leads to the conclusion that competition to the detriment of carabids occurs where F. aserva and individual carabids come into contact, and predation is a possible outcome when F. aserva activity is high. Use of species or species groups as indicators of various environmental conditions or to observe the effects of management on ecosystems has proceeded as a cost and time effective technique to observe a variety of effects (Weaver 1995, Rainio and Niemela 2000, Maleque et al. 2006, Work et al. 2008). An issue with indicator species has been interpretation of observed responses. Are the indicator organisms responding to the environmental condition in question (Dale and Beyeler 2001), e.g., forest health condition, coarse woody debris, habitat fragmentation etc., or are they being influenced by unknown variables? While carabids have be shown to be useful as indicators (Rainio and Niemela 2000, Work et al. 2008), the need for understanding the influences and interactions that occur within the broader epigaeic invertebrate community has been demonstrated by my study. The negative association with ants has been noted in several studies, and confirmed in this study. This relationship has, however, been under-acknowledged in carabid studies (Lovei and Sunderland 1996) and is a potentially major influencing factor of carabid activityabundance. The level of influence exerted by ants may influence the interpretation of findings where carabids are used as indicators, particularly when aggressive ant species are abundant. In examining the interaction between carabids, and the ants F. aserva and C. herculeanus, I have provided compelling evidence that single taxon studies examining 102 invertebrate epigaeic community ecology should be re-examined, as they only paint a partial picture of the community as a whole. Broadening the scope of invertebrate species examined in community ecology is not a novel idea, as research into the feasibility of using recognizable taxonomic units, or morphospecies, for rapid assessment of invertebrate diversity have been tested in many areas, e.g., Australia (Oliver and Beattie 1996b) and New Zealand (Derraik et al. 2002). While use of the morphospecies concept should not be viewed as a replacement for species level taxonomic expertise, it may be useful for accounting for the diversity in forest arthropod communities and may provide a great deal of initial information pertaining to the interactions therein. Reference List Atlegrim, O., Sjoberg, K., and Ball, J.P. 1997. Forestry effects on a boreal ground beetle community in spring: selective logging and clear-cutting compared. Entomologica Fennica 8: 19-26. Baguette, M., and Gerard, S. 1993. Effects of spruce plantations on carabid beetles in southern Belgium. Pedobiologia 37: 129-140. Brouat, C , Meusnier, S., and Rasplus, J.Y. 2004. Impact of forest management practices on carabids in European fir forests. Forestry 77: 85-97. Brumwell, L.J., Craig, K.G., and Scudder, G.G.E. 1998. Litter spiders and carabid beetles in a successional douglas-fir forest in British Columbia. Northwest Science 72: 94-95. Dale, V.H., and Beyeler, S.C. 2001. Challenges in the development and use of ecological indicators. Ecological Indicators. 1: 3-10. Derraik, J.G.B., Closs, G.P., Dickinson, K.J.M., Sirvid, P., Barratt, B.I.P., and Patrick, B.H. 2002. Arthropod morphospecies versus taxonomic species: a case study with Araneae, Coleoptera, and Lepidoptera. Conservation Biology 16: 1015-23. de Warnaffe, G.D.B., and Lebrun, P. 2004. Effects of forest management on carabid beetles in Belgium: implications for biodiversity conservation. Biological Conservation 118: 219234. 103 Hamilton, E. 1988. A system for the classification of serai ecosystems within the biogeoclimatic ecosystem classification. First approx. B.C. Min. For. Lands, Victoria, B.C. Res. Rep. RR87004-HQ. Hawes, C , Stewart, A.J.A., and Evans, H.F. 2002. The impact of wood ants (Formica rufa) on the distribution and abundance of ground beetles (Coleoptera: Carabidae) in a Scots Pine Plantation. Oecologia 131: 612-619. Heyborne, W.H., Miller, J.C., and Parsons, G.L. 2003. Ground dwelling beetles and forest vegetation change over a 17-year-period, in western Oregon, USA. Forest Ecology and Management 179: 123-134. Higgins, R.J. 2010. The ant (Hymenoptera: Formicidae) communities of the central interior of British Columbia: Adaptations to a temperature-constrained environment. Ph.D. Dissertation, University of Northern British Columbia, Prince George, BC, 184 pp. Humphrey, J.W., Hawes, C , Peace, A.J., Ferris-Kaan, R., and Jukes, M.R. 1999. Relationships between insect diversity and habitat characteristics in plantation forests. Forest Ecology and Management 113: 11-21. Ings, T.C., and Hartley, S.E. 1999. The effect of habitat structure on carabid communities during the regeneration of a native Scottish forest. Forest Ecology and Management 119: 123-136. Jukes, M.R., Peace, A.J., and Ferris, R. 2001. Carabid beetle communities associated with coniferous plantations in Britain: the influence of site, ground vegetation and stand structure. Forest Ecology and Management 148: 271-286. Karhu, K.J. 1998. Effects of ant exclusion during outbreaks of a defoliator and a sap-sucker on birch. Ecological Entomology 23: 185-194. Koivula, M. 2002. Boreal carabid-beetle (Coleoptera, Carabidae) assemblages in thinned uneven-aged and clear-cut spruce stands. Annales Zoologici Fennici 39: 131-149. Koivula, M., and Niemela, J. 2003. Gap felling as a forest harvesting method in boreal forests: responses of carabid beetles (Coleoptera, Carabidae). Ecography 26: 179-187. Koivula, M., Kukkonen, J., and Niemela, J. 2002. Boreal carabid-beetle (Coleoptera, Carabidae) assemblages along the clear-cut originated succession Gradient. Biodiversity and Conservation 11: 1269-1288. Koivula, M., Punttila, P., Haila, Y., and Niemela, J. 1999. Leaf litter and the small-scale distribution of carabid beetles (Coleoptera, Carabidae) in the boreal forest. Ecography 22: 424-435. Krebs, C.J. 1972. Ecology. The Experimental Analysis of Distribution and Abundance. Harper and Row, New York. 694 pp. Laakso, J., and Setala, H. 1998. Composition and trophic structure of detrital food web in ant nest mounds of Formica Aquilonia and in the surrounding forest soil. Oikos 81: 266-278. Laakso, J., and Setala, H. 2000. Impacts of wood ants (Formica aquilonia Yarr.) on the invertebrate food web of the boreal forest floor. Annales Zoologici Fennici 37: 93-100. Larochelle, A., and Lariviere, M.-C. 2003. A Natural History of the Ground-Beetles (Coleoptera: Carabidae) of America North of Mexico. Pensoft Publishers, Sophia, Bulgaria. Lassau, S.A., Hochuli, D.F., Cassis, G., and Reid, C.A.M. 2005. Effects of habitat complexity on forest beetle diversity: do functional groups respond consistently? Diversity and Distributions 11: 73-82. Lemieux, J.P., and Lindgren, B.S. 2004. Ground beetle responses to patch retention harvesting in high elevation forests of British Columbia. Ecography 27: 557-566. Lovei, G.L., and Sunderland, K.D. 1996. Ecology and behavior of ground beetles (Coleoptera: Carabidae). Annual Review of Entomology 41: 231-256. Magura, T. 2002. Carabids and forest edge: spatial pattern and edge effect. Forest Ecology and Management 157: 23-37. Magura, T., and Tothmeresz, B. 1997. Testing edge effect on carabid assemblages in an oakhornbeam forest. Acta Zoologica Academiae Scientiarum Hungaricae 43: 303-312. Magura, T., Elek, Z., and Tothmeresz, B. 2002. Impacts of non-native spruce reforestation on ground beetles. European Journal of Soil Biology 38: 291-295. Maleque, M.A., Ishii, H.T., and Kaoru, M. 2006. The use of arthropods as indicators of ecosystem integrity in forest management. Journal of Forestry. 104:3. 113-117. Meidinger. D., and Pojar, J. (compilers and editors). 1991. Ecosystems of British Columbia. B.C. Min. For. Special Report Series No. 6. 330pp. Mody, K., and Linsenmair, K.E. 2004. Plant-attracted ants affect arthropod community structure but not necessarily herbivory. Ecological Entomology 29: 217-225. Niemela, J., and Halme, E. 1992: Habitat associations of carabid beetles in fields and forests on the Aland Islands, SW Finland. Ecography 15: 3-11. Niemela, J., Spence, J.R., and Spence, D.H. 1992. Habitat associations and seasonal activity of ground-beetles (Coleoptera, Carabidae) in central Alberta. Canadian Entomologist 124: 521-540. Niemela, J., Langor, D., and Spence, J.R. 1993. Effects of clear-cut harvesting on boreal ground-beetle assemblages (Coleoptera, Carabidae) in western Canada. Conservation Biology 7: 551-561. 105 Nonacs, P., and L.M. Dill. 1990. Mortality risk vs. food quality trade-offs in a common currency: ant patch preferences. Ecology. 71: 1886-1892. Oliver, I., and Beattie, A.J. 1996a. Designing a cost-effective invertebrate survey: a test of methods for rapid assessment of biodiversity. Ecological Applications 6: 594-607. Oliver, I., and Beattie, J.A. 1996b.Invertebrate morphospecies as surrogates for species: a case study. Conservation Biology. 10: 99-109. Punttila, P., Niemela, P., and Karhu, K. 2004. The impact of wood ants (Hymenoptera: Formicidae) on the structure of invertebrate community on mountain birch (Betula pubescens ssp czerepanovii). Annales Zoologici Fennici 41: 429-446. Rainio, J., and Niemela, J. 2003. Ground beetles (Coleoptera: Carabidae) as bioindicators. Biodiversity and Conservation 12: 487-506. Reznikova, Z., and Dorosheva, H. 2004. Impacts of red wood ants Formica polyctena on the spatial distribution and behavioural patterns of ground beetles (Carabidae). Pedobiologia 48: 15-21. Vanbergen, A.J., Watt, A.D., Mitchell, R., Truscott, A., Palmer, S.C. F., Ivits, E., Eggleton, P., Jones, T. H., and Sousa, J.P. 2007. Scale-specific correlations between habitat heterogeneity and soil fauna diversity along a landscape structure gradient. Oecologia. 153: 713-725. Vance, C.C., and Nol, E. 2003. Temporal effects of selection logging on ground beetle communities in northern hardwood forests of eastern Canada. Ecoscience 10: 49-56. Weaver, J.C. 1995. Indicator species and scale of observation. Conservation Biology. 9: 939942. Work, T.T., Koivula, M., Klimaszewski, J., Langor, D., Spence,, J., Sweeney, J., and Hebert, C. 2008. Evaluation of carabid beetles as indicators of forest change in Canada. Canadian Entomologist 140: 393-414. 106 Appendices Appendix I Study Site Locations 2005 1) Non-harvested plots (n=2). Site One: Tanglechain (Houston Forest Products) Nearest road access at UTM 09U 658641E 6092206N. This is not far off of the Tanglechain road near the beginning of the NSR cutblock 428. 0,0 of plot is 9U 620813E 59841 ION 2) Post-harvest sites (2 years) n=2. Site One: Chisholm (Houston Forest Products) Cutblock number 631-TH2 09U613828E6007491N Site Two: Nadina South (Houston Forest Products) Cutblock number 045-1 09U 630054E 5969832N Elevation 1018m 2) Post-harvest sites (8-10 years) n=2. These sites were adjacent to the mature sites under 1 above Site One: Tanglechain (Houston Forest Products) Cutblock number 452-2 09U 0664241 608754IN Site Two: Nadina West (Houston Forest Products) Cutblock number 021-1 09U 621437E 5984074N 3) Post-harvest sites (14-18 years) n=2. Site One: Nadina West 15 (NW15): Block 011-1 -off Duel Lake Road 09U 0623613E5982200N Elevation 1051 m Site Two: Tanglechain (Houston Forest Products) Cutblock number 451 -2 09U 669405E 6089080N 4) Post-harvest sites (>25 years) n=2 Site One: Pimpernel 25 (P25): Immediately to the north of block 342-110 09U 0630078E 6003705N Elevation 875m Site Two: Morice River (MR25): -along Morice river road just after 48 km 09U 06223826E 6002697N Elevation 816m 107 Appendix II ANOVA Results Figure 2 Chapter 2 ANOVA results for variation in activity-abundance for the sexes and over sample periods, seasonal effect, /"-value indicates significant variation in activity-abundance between sexes and over sample periods (a = 0.05). Gender SS df MS F P Scaphinotus marginatus 74.1 4 18.53 7.71 0.000 745 Error 2.4 17.89 Scaphinotus angusticollis 4 8.35 5.25 33.4 0.000 Error 11.86 745 1.59 Pterostichus adstrictus 16.38 4 4.09 3.35 0.010 Error 745 1.22 911.37 Pterostichus riparius 5.94 4 1.48 3.97 0.003 Error 745 0.37 278.86 Calathus advena 4 33.18 17.85 0.000 132.7 Error 745 1384.57 1.86 2.07 4 Calathus ingratus 0.52 6.28 0.000 Error 745 0.08 61.44 4 Synuchus impunctatus 186.45 46.61 9.15 0.000 Error 745 5.09 3796.29 Season Scaphinotus marginatus Error Scaphinotus angusticollis Error Pterostichus adstrictus Error Pterostichus riparius Error Calathus advena Error Calathus ingratus Error Synuchus impunctatus Error 18.9 430.54 14.1 350.5 6.27 271.75 0.82 128.27 5.69 367.72 0.57 35.29 3.99 564.49 4 745 4 745 4 745 4 745 4 745 4 745 4 745 4.72 0.58 3.53 0.47 1.57 0.37 0.66 0.84 1.42 0.49 0.14 0.047 1 0.76 8.19 0.000 7.5 0.000 4.3 0.002 1.19 0.312 2.88 0.022 3 0.018 1.32 0.262 F- Test analysis of the influence of seasonal variation on abundant male and female carabid activity-abundance. P-value indicates a significant effect of season on the mean activity-abundance of either male or female carabids (a = 0.025) SS df MS Female Scaphinotus marginatus 12.34 4 3.09 3.21 0.012 Error 745 0.96 715.25 4 20.16 9.99 0.000 Male Scaphinotus marginatus 80.66 Error 1504.41 745 2.01 4 1.08 0.362 Female Scaphinotus angusticollis 3.55 0.89 Error 609.62 745 0.82 43.94 4 10.99 8.83 0.000 Male Scaphinotus angusticollis Error 926.51 745 1.24 17.41 4 4.35 4.35 0.002 Female Pterostichus adstrictus Error 744.51 745 1 4 2.23 0.065 Male Pterostichus adstrictus 5.25 1.31 Error 438.81 745 0.59 4 1.32 3.53 0.007 Female Pterostichus riparius 5.29 Error 278.77 745 0.37 2.14 0.074 4 0.37 Male Pterostichus riparius 1.47 Error 745 0.17 128.36 4 13.42 0.000 Female Calathus advena 96.37 24.09 Error 745 1.8 133.74 4 Male Calathus advena 10.51 18.86 0.000 42.03 Error 745 0.56 41.49 4 Female Calathus ingratus 2.34 0.58 6.1 0.000 Error 745 0.1 71.46 4 0.08 2.22 0.065 Male Calathus ingratus 0.3 745 0.03 Error 25.27 4 26.58 10.08 0.000 Female Synuchus impunctatus 106.34 Error 745 2.64 1964.57 84.1 4 6.54 0.000 Male Synuchus impunctatus 21.01 Error 2396.2 745 3.22 Appendix III Non metric Multidimensional Scaling: PC - Ord Outputs Coefficients of determination for the correlations between ordination distances and distances in the original n-dimensional space: R2 Axis Increment Cumulative X .188 .188 .170 .358 Y .370 .728 Increment and cumulative R-squared were adjusted for any lack of orthogonality of axes. Orthogonality,% = 100(1- r2) Axis pair r XvsY -0.201 96.0 Number of entities =150 Number of entity pairs used in correlation = 11175 Distance measure for ORIGINAL distance: Sorensen (Bray-Curtis) Final Stress = 20.34 Final instability = 0.00569 Pearson Kendall Species Correlations with NMS axes: Chapter 2 Figure 4 Axis: X R2 R Variables NOTSYL SCAMAR SCAANG PTEADS PTEBRE CALADV CALING SYNIMP AMASIN AMAERR HARSOM TRECHA BEMGRA PTECAS BEMFOR PTERIP ELALAP AGOAFF BRACON STEHAE LEBMOE 0.069 0.131 -0.103 0.118 -0.092 -0.153 -0.092 0.653 0.05 0.351 0.072 0.432 -0.008 -0.159 0.267 0.248 0.282 0.131 0.238 -0.184 -0.025 tau 0.005 0.017 0.011 0.014 0.008 0.023 0.008 0.426 0.003 0.123 0.005 0.187 0 0.025 0.071 0.061 0.08 0.017 0.057 0.034 0.001 0.078 0.114 -0.177 0.076 -0.11 -0.245 -0.023 0.636 0.036 0.272 0.027 0.252 -0.018 -0.152 0.211 0.176 0.316 0.14 0.218 -0.179 -0.08 Y R2 R 0.122 0.624 0.706 -0.468 0.061 0.449 0.18 -0.293 -0.206 -0.284 -0.358 -0.183 -0.244 0.049 -0.13 -0.061 -0.173 -0.116 -0.149 0.219 -0.197 tau 0.015 0.389 0.498 0.219 0.004 0.201 0.032 0.086 0.042 0.081 0.128 0.033 0.06 0.002 0.017 0.004 0.03 0.013 0.022 0.048 0.039 0.083 0.61 0.686 -0.396 0.038 0.488 0.185 -0.364 -0.232 -0.297 -0.299 -0.233 -0.196 0.046 -0.105 -0.118 -0.178 -0.115 -0.12 0.185 -0.188 Pearson and Kendall Variable Correlations: Chapter 2 Figure 4 Axis: X R2 R Variables CCOVER LCOVER HCOVER CWD MOSS SLASH GRASS BAREGR REDROT NEEDLE LITTER LINBOR VEGDEN VEGDIV COVDIV -0.394 -0.102 -0.365 -0.166 -0.397 0.031 0.486 0.022 0.212 -0.066 0.088 0.074 0.238 -0.044 -0.012 tau 0.156 0.01 0.133 0.028 0.158 0.001 0.237 0 0.045 0.004 0.008 0.005 0.056 0.002 0 -0.27 -0.066 -0.233 -0.119 -0.314 0.11 0.398 0.036 0.178 -0.064 0.089 0.089 0.127 -0.047 0.016 Y R2 R 0.838 0.074 0.22 0.052 0.374 -0.553 -0.335 -0.278 -0.026 0.569 -0.019 0.052 0.001 0.502 -0.115 tau 0.702 0.005 0.048 0.003 0.14 0.305 0.112 0.077 0.001 0.324 0 0.003 0 0.252 0.013 0.569 0.024 0.175 0.028 0.275 -0.465 -0.26 -0.254 -0.036 0.474 -0.017 -0.014 -0.026 0.329 -0.079 Appendix IV UTM's For Experimental Replicates Chapter 4 Replicate 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Treatment ANT CWD CTRL ANT ANT ANT CWD CTRL CTRL ANT ANT ANT CWD ANT CWD ANT CTRL CWD CWD ANT UTM 09 629105E5968708N 09 629216E 5968722N 09 629270E 5968796N 09 629294E 5968895N 09 629362E 5968865N 09 628493E 5968808N 09 628577E 5968861N 09 628644E 5968936N 09 628713E5969014N 09 628791E5969079N 09 628678E 5969139N 09 628617E5969031N 09 628551E5968957N 09 628462E 5968901N 09 628395E 5968822N 09 628305E 5968872N 09 628377E 5968942N 09 628483E 5968997N 09 628536E 5969084N 09 629178E5968861N Replicate 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Treatment CWD ANT CTRL ANT CWD ANT CTRL CWD CWD ANT ANT CTRL CWD ANT ANT CTRL ANT ANT ANT CTRL UTM 09 629091E 5968806N 09 629008E 5968735N 09 630264E 5970275N 09 630212E5970189N 09 630157E5970104N 09 630060E 5970084N 09 629962E 5970109N 09 630047E5970185N 09 630126E5970246N 09 630188E5970321N 09 630247E 5970442N 09 630140E 5970419N 09 630064E 5970347N 09 629975E 5970301N 09 629848E 5969635N 09 629817E5969731N 09 629819E5969836N 09 630010E5969836N 09 630015E5969730N 09 630117E5969744N