UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Notes on species coexistence, invasion and ecosystem function Gilbert, Benjamin 2008

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

Item Metadata

Download

Media
24-ubc_2008_fall_gilbert_benjamin.pdf [ 12.68MB ]
Metadata
JSON: 24-1.0067972.json
JSON-LD: 24-1.0067972-ld.json
RDF/XML (Pretty): 24-1.0067972-rdf.xml
RDF/JSON: 24-1.0067972-rdf.json
Turtle: 24-1.0067972-turtle.txt
N-Triples: 24-1.0067972-rdf-ntriples.txt
Original Record: 24-1.0067972-source.json
Full Text
24-1.0067972-fulltext.txt
Citation
24-1.0067972.ris

Full Text

Notes on Species Coexistence, Invasion and Ecosystem Function by Benjamin Gilbert B.Sc, University of British Columbia, 2001 M.Sc, McGill University, 2003 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Botany) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) September 2008 © Benjamin Gilbert, 2008  Abstract Species coexistence theory attempts to explain and predict the mechanisms that promote species diversity and the ecological consequences of this diversity. In this thesis I used observational and experimental field studies to test the predictions of several coexistence theories, and developed specific predictions in a theoretical study. The observational study was used to test general predictions made by two mechanisms, neutral interactions and spatial niche partitioning, using bromeliad-dwelling mosquito larvae in Costa Rica. Results from this study were only consistent with spatial niche partitioning, and showed how local, within-bromeliad interactions could scale up to meso-scale (among-bromeliad) distributions. The experimental study, based in the boreal forest understory, used the standard rank-abundance relationships of plant species to test both competitive and facilitative coexistence mechanisms that differentiate between the effects of dominant species and species diversity. In particular, removals of a consistent biomass that targeted one dominant or many low-abundance species were used in conjunction with seedling additions to test the roles of different species, and species diversity, in limiting the establishment of new species. High mortality of new seedlings in completely cleared areas indicated that facilitation was important. However, small-scale disturbances (7% of community biomass removed) either had no effect on seedling survival, or increased survival, indicating competitive effects. These competitive effects were limited to a single dominant species, and were inconsistent with current models of resource niche partitioning. The theoretical study used computer simulations to investigate the effects of regional habitat heterogeneity on local diversity in communities that differed in their connectivity (dispersal among patches) and neutrality (niche overlap among species). The model suggested that dispersal and niche overlap have synergistic effects on local diversity by increasing the size of sink populations, and likewise destabilizing coexistence regionally. However, they have opposite effects on resource-use, causing either positive or negative diversity to resource-use relationships. Together, the three studies illustrate that different processes can scale up to general patterns of species distributions, but that these processes have very different implications for understanding and protecting species diversity and the functions this diversity may provide.  ii  Table of contents Abstract  ii  Table of contents  hi  List of tables  vi  List of figures  vii  List of plates  ix  Acknowledgements  x  Co-authorship statement  xi  1. Species coexistence, invasion and ecosystem function  1  Introduction  1  Theoretical framework  1  Approaches to studying coexistence  4  Thesis overview  6  Literature cited  11  2. Niche partitioning at multiple scales facilitates coexistence among mosquito larvae  18  Introduction  18  Methods  22 Study 1: Observation of species' spatial niches within containers  22  Study 2: Sampling species distributions  22  Analysis  23  Results  26  Discussion  28  Literature cited  33  3. Coexistence and mechanisms of invasion in a boreal forest understory  38  Introduction  38  Study area and methods  43  Statistical methods  45  Results  Seedling responses  45  Extant community  46 49  iii  Seedling responses  49  Extant community  51  Discussion  52 Facultative effects  53  Competitive effects  55  Conclusion  57  Literature cited  58  4. Consequences of regional environmental heterogeneity on local diversity and ecosystem function  62  Introduction  62  The model  64 Simulations  68  Results  70  Discussion  74 Dispersal, niches and diversity  75  Diversity and ecosystem function  76  Environmental heterogeneity  78  Summary  79  Literature Cited  81  5. Conclusions  85 Unique insights into the study communities: Bromeliad-dweiling mosquito larvae  85  Unique insights into the study communities: Boreal forest understory plants  86  Unique insights into the study communities: Computer simulations of an idealized community  89  General insights: Integrating the results from three approaches  90  Literature cited Appendix A: Competitive interactions amongst mosquito larvae  92 94  Introduction  94  Methods  94  IV  Analysis  95  Results  96  Discussion  97  Literature cited Appendix B: Principal facilitator expectations from a Lotka-Volterra model Literature cited Appendix C: Species in extant community and transplanted into plots Literature cited  100 102 104 105 108  v  List of tables Table CI: Extant species in sampled aspen stands in Kluane area, Yukon  105  Table C2: Species used as transplants in Chapter 3  107  VI  List of figures Figure 1.1: General framework for species coexistence based on competition for common resources  4  Figure 1.2: Map of Costa Rica with arrow indicating approximate position of Pitilla station where the bromeliad-mosquito research described in Chapter 2 was done  6  Figure 1.3: Map of land-use histories surrounding the Pitilla field station, Costa Rica  7  Figure 1.4: Map of Canada with the arrow marking the approximate location of the field study done in the southwestern Yukon, Canada  8  Figure 1.5: The effect of the plant community on resource availability and invasibility  9  Figure 2.1: Hypothetical trends in mosquito community depending on whether neutral dynamics (solid lines) or niche partitioning (dashed lines) determine co-occurrence patterns  21  Figure 2.2: Mean depth (+ standard error) of individuals below water surface for Anopheles neivai (A. nei.), Culex rejector (C. rej.), Wyeomyia melanopus (W. mel.) and Wyeomyia circumcincta (W. cir.)  26  Figure 2.3: Tests of the spatial niche hypothesis (A) and two neutral hypotheses (B & C)  27  Figure 2.4: Histograms of average species densities (individuals per litre of water holding capacity) along a gradient of bromeliad water holding capacities  29  Figure 3.1. The rank-abundance relationship among species and its relationship to frequency within aspen stands  40  Figure 3.2. Predictions from different coexistence mechanisms about the diversity of seedlings resulting from removal of all plants (complete), no plants (none), or an equivalent amount (7% of total biomass) of dominant species or low-abundance species Figure 3.3. Total and species-specific survival  42 48  Figure 3.4. The difference in diversity among treatments and its relationship to the number of survivors  50  Figure 3.5. Differences in resource availability among removal treatments  51  Figure 3.6. Extant species' colonization rates and competitive abilities  53  Figure 4.1. Species R* values and their relationship to local abundance  69  Figure 4.2. The relationship between dispersal, local species richness and resource use  71  Figure 4.3. The effect of niche width on species richness and resource use  72  vii  Figure 4.4. The affect of species pool (initial regional species richness) on local species richness, and the resulting effect of local richness on unused resources Figure A.l: The effect of competitors on the growth (mm) of 2nd instar Anopheles neivai  74 97  vm  List of plates Plate 1. Bromeliads in the Guanacaste conservation area  17  Plate 2. Aspen stands in Kluane, Yukon  37  IX  Acknowledgements This thesis incorporates over four years of work and would not have been possible without the help, support and encouragement of many people over that time. I would first like to thank my supervisors, Roy Turkington and Diane Srivastava, for the many insights and skills that they have offered. They have been invaluable mentors and critics, and I feel lucky to have had the opportunity to work with them. Andrew MacDougall has also acted as a mentor, colleague and friend. I have learned a great deal from his passion for plant ecology and his ability to mingle field observations with ecological theory. The Canon National Parks Science Scholars Program provided substantial funding for my research, for which I am grateful. The work presented here, in particular in Kluane, could not have been completed without their support. The Natural Sciences and Engineering Research Council, Northern Scientific Training Program, Li Tze Fong Memorial Fellowship and University of British Columbia also provided personal or research funding. Dr. William Laurance, the Smithsonian Tropical Research Institute, Andy Williams and the Arctic Institute of North America all provided logistical support. I would also like to thank the Champagne and Aishihik First Nations for access to their land near Kluane National Park. A number of other people have helped me academically by discussing ideas or commenting on drafts of specific papers. In particular, Trevor Lantz, Jackie Ngai, Jon Shurin, Brian Starzomski, Kurtis Trzcinski, and Mark Vellend have contributed to my work through specific comments and ongoing encouragement. Guillermo Chaverri played a key role in helping with mosquito larvae identification. Numerous field assistants were essential to completing my research in Kluane. I would like to thank Adrian Leitch, Karen Miller, Jenn Mundy, Aimee Pelletier, Katy Pieczora and Momo Price in particular. Beyond doing the necessary field work, each one was a pleasure to work with, and I enjoyed discovering and sharing the magic of the Yukon with them. I would like to thank Justine Karst for her encouragement and feedback on all stages of my research. She has been a good friend, and our work together has been both fun and productive. My family has, as always, been unfaltering in their support for my endeavors. Finally, I thank Kate Kirby for being both helpful and supportive with all aspects of my academic work. She is a star.  x  Co-authorship statement Chapter 2 was co-authored by Diane Srivastava and Kathryn Kirby. I designed the main sampling protocol, created the framework of the manuscript, analyzed the data and wrote all drafts. Kathryn helped with all field work and species identifications, and we jointly decided on a methodology for observing species' spatial niches, which she then monitored. Diane gave critical feedback on the design of the sampling, and also on the theoretical framework of the paper. Diane and Kathryn assisted with manuscript revision.  Chapter 3 was co-authored by Roy Turkington and Diane Srivastava. The experimental design was mainly my creation, but it evolved considerably over the course of several discussions with both Roy and Diane. I was responsible for the field work, data analysis, and the framework and writing of the manuscript. Roy and Diane contributed to many lengthy discussions and assisted with manuscript revision.  XI  1. Species coexistence, invasion and ecosystem function Introduction Research on the conservation of species diversity has recently gained prominence in the ecological literature, both as an ethical goal and as a mechanism for maintaining ecosystem functions, such as nutrient cycling, resilience following disturbance and resistance to invasion by exotic species (Schwartz et al. 2000, Srivastava 2002, Hooper et al. 2005, Srivastava and Vellend 2005, Cardinale et al. 2006). Species diversity results from mechanisms that promote species coexistence; these mechanisms have been the subject of numerous classic papers in the ecological literature (Tilman 1980, Tilman et al. 1997b, Chesson 2000, Loreau et al. 2003, Tilman 2004), and continue to be actively debated (Callaway and Walker 1997, Chesson 2000, Hubbell 2001, Leibold et al. 2004). Although any number of coexistence models could explain patterns of species distributions in a given community (Chave et al. 2002, McGill et al. 2006, McGill et al. 2007), the various mechanisms that allow species to coexist determine the means by which species diversity may be conserved (Hubbell 2001, Amarasekare et al. 2004, Gilbert et al. 2006, Adler et al. 2007, MacDougall et al. 2008). In addition, the different coexistence mechanisms that can promote diversity predict differing relationships between this diversity and the ecosystem services that a community provides (Tilman et al. 1997b, Levine and D'Antonio 1999, Hooper et al. 2005, Srivastava and Vellend 2005). A first, basic step in conserving species diversity or mediating the impacts of species loss on ecosystem function therefore lies in identifying the mechanisms that allow species to coexist in a given area.  Theoretical framework The underlying basis of most theories of coexistence is the principle of competitive exclusion (Gause 1934, Hardin 1960, Chesson 2000, Chase and Leibold 2003). This principle states that no two species competing for a single limiting resource can coexist locally over the long-term unless a mechanism exists that limits interspecific competition. Coexistence theories must therefore contain mechanisms that promote assemblages of multiple species in a single area. This can be done in two ways. First, competition between species can be limited, leading to the stable coexistence of species. For example, limited overlap between species' niches (Hutchinson 1959) and trade-offs between competitive ability and resistance to predation (Chase and Leibold  1  2003) are mechanisms that allow competing species to coexist stably by limiting competitive interactions. Second, processes that cause rapid input (immigration or speciation) of species to an area relative to initial rates of competitive exclusion can be invoked, which results in high species diversity despite numerous extinction events (Chesson 2000, Figure 1.1). The neutral theory of biodiversity (Hubbell 2001) is a recent example of this second type of model; species must be created (via speciation) more quickly than they initially go extinct in order for an area to have more than a single species present over the long-term. Mathematically, these two processes have been simplified by Chesson (2000) as shown in the following example of resource competition:  b;  b„  b.  In this model, i is an invading species and s the resident species, with the invader being defined as any species at a very low density, either due to new introduction or to a fluctuation in population size. The low-density per capita growth rate of the invader (F,-) depends on the per capita growth rate of both species in the absence of resource limitation (w) and the decline in this rate as the resource declines (b) for both the invader and resident. The degree of niche overlap that the invader has with the resident species is denoted by p, which varies from 1 (complete overlap) to 0 (completely distinct fundamental niches). This model is used to determine if an invading species' population growth rate is positive when it occurs at low densities, that is, if there are mechanisms that cause it to persist stably in its new area. The model has both an equalizing portion and a stabilizing portion. The first, or equalizing, portion of the model shows that as the competitive difference between species (——-) approaches zero the rate at which b, bs competitive exclusion occurs will decrease (Fig. 1.1). The stabilizing portion of the model, determined by the magnitude of p, is necessary for stable coexistence; when complete resource competition occurs (i.e. p = 1) the model will only allow a single species to exist (Chesson 2000). Thus, the strength of the stabilizing portion of the model can be calculated as 1-p. It is important to note that models based solely on stabilizing forces (MacArthur 1957) or solely on equalizing forces (Hubbell 2001) can explain natural patterns of species abundance reasonably  2  well, and that models may combine both stabilizing and equalizing mechanisms (Chesson 2000). The basic model (Eq. 1) forms the basis of the hypothesis that a community with more species has a higher level of ecosystem function, such as resource utilization and resistance to invasion (Levin 1970, Tilman 1980, 1982, Tilman et al. 1997b, Levine and D'Antonio 1999, Chesson 2000). However, certain coexistence mechanisms, such as the competition-colonization tradeoff, rely on different assumptions about species' niche and fitness asymmetries (Tilman 1994, Chesson 2000). The competition-colonization trade-off is discussed in more detail in Chapter 3. Despite the numerous models of species coexistence that have been developed (Chesson 2000), results from field studies often conflict with predictions of coexistence theory. For example, facilitation has been repeatedly reported in field studies, and yet the incorporation of facilitation into coexistence theory has been lacking (Bertness and Callaway 1994, Callaway and Walker 1997, Bruno et al. 2003, Bruno 2005, Callaway 2007, Brooker et al. 2008). Although recent developments have begun exploring facilitation (Gross 2008b), the most extensive and complete review of the coexistence literature to date does not have a single mention of facilitation (Chesson 2000). Similarly, most coexistence theory relies on mathematical solutions to equations that specify species interactions, and thus generate expected relationships among species in general, but rarely generate predictions for common community metrics. For example, a number of conceptual models suggest that dispersal into an area can increase diversity beyond what would be supported if that area were spatially isolated, and that this increased diversity can have either no effect or a negative effect on ecosystem function (Shmida and Wilson 1985, Leibold and Miller 2004a). This hypothesis has been explored to some degree by models (Mouquet and Loreau 2003), but the overall effect of such processes on community metrics and ecosystem function remains unresolved. In this thesis I have used contemporary coexistence theory, and its inconsistencies with results from field research, to identify three themes that could advance our understanding of species coexistence. First, many ecological communities have not been studied sufficiently to generate even basic hypotheses on the mechanisms that promote coexistence, especially in tropical ecosystems (SCBD 2006). This first step to understanding diversity requires basic descriptions of communities in which the natural histories of extant species are used to generate predictions about coexistence mechanisms. Second, once basic descriptions of communities have been made and different hypotheses for species coexistence have been proposed, these  3  mechanisms can be tested. Again, understudied communities provide more information for species coexistence in general, as current studies are biased towards relatively few communities (Duffy 2003, Srivastava and Vellend 2005, SCBD 2006). Third, based on the first two steps, we can then further advance and develop species coexistence theory, and in particular how different coexistence mechanisms affect observed community metrics such as alpha and beta diversity and species abundance distributions (Chave et al. 2002, Mouquet and Loreau 2002).  Mechanisms for coexistence in a local community with common resource use  Migration/ speciation  Competition between species is limited  Species input  Leads to stable coexistence e.g. Niche segregation (spatial or temporal) e.g. Tilman's R* with multiple limiting resources (Tilman 1980, 1982, 2004, Tilman et al. 1997), Chesson's storage effect and relative nonlinearity (Chesson 2000) Generalized niche model (Levin 1970)  Competitive differences between species are minimized  Leads to unstable coexistence if rate of extinction < rate of speciation or immigration e.g. Neutral model (Hubbell 2001) Lottery models or similar priority effect models (Chave et al. 2002) Source-sink models (Mouquet and Loreau 2002)  Figure 1.1: General framework for species coexistence based on competition for common resources. Concepts adopted from Chesson (2000a) and Hubbell (2001).  Approaches to studying coexistence Studies on species diversity and coexistence have traditionally followed three approaches. First, theoretical studies have acted as a cornerstone for predicting how species interactions can promote or prevent coexistence (MacArthur and Levins 1967a, Tilman 1980, Chesson 2000, Amarasekare and Nisbet 2001). Specific theories that often involve two or a few species can be scaled up to predict how a number of species could coexist while constrained by a specific set of interactions (MacArthur 1957, Levin 1970, MacArthur 1970, MacArthur 1972, Levin 1976,  4  Tilman 1982, 1994, Chesson 2000, Hubbell 2001, Gross 2008c, Gross 2008a). For example, the classic work by MacArthur and colleagues provided many predictions about species coexistence through competition, niche partitioning and dispersal limitation, many of which form the basis of theoretical models still used today (MacArthur and Levins 1967a, MacArthur and Wilson 1967, MacArthur 1970, MacArthur 1972, Chesson 2000). Second, researchers use observational studies of ecological communities either to test specific theories (Harrison 1999, Ennos 2001, Gilbert and Lechowicz 2004b, Davies et al. 2005, Bell et al. 2006, Gilbert et al. 2006, Harrison et al. 2006b), or to inform new theoretical or conceptual models (Tilman 1994, Callaway and Walker 1997). Early studies that described species relative abundances, for example, were used first to test models of niche partitioning (MacArthur 1957) and were later used as a platform for developing and testing neutral theory (Hubbell 2001, Chave 2004). Third, there have been tests of theoretical predictions in controlled experiments (Srivastava and Lawton 1998, Fargione et al. 2003b, Cadotte and Fukami 2005, Bezemer and van der Putten 2007, Cadotte 2007, Harpole and Suding 2007). Many of the central axioms to coexistence theory, such as the competitive exclusion principle (Gause 1934, Hardin 1960) and the importance of predators in allowing prey coexistence (Paine 1966), have been tested and supported using controlled experiments. More recently, theories of coexistence have been used to predict the impacts of high species diversity on community resistance to invasion and other ecosystem properties (Tilman et al. 1997b, Levine and D'Antonio 1999, Schwartz et al. 2000), and this theory has been matched by hundreds of studies in this field (Levine and D'Antonio 1999, Schwartz et al. 2000, Srivastava and Vellend 2005, Fridley et al. 2007). Each of these three broad approaches have specific strengths and weaknesses. At one extreme, observational studies capture a complete picture of the community, but can only be used to infer causation because unobserved processes are not controlled. At the other extreme, theoretical models tend to provide very specific predictions, but at the expense of reality. Most models require many simplifying assumptions in order to be analysed, and these assumptions ignore potentially important processes in communities (Chesson 2000, Gross 2008c). Similarly, experimental approaches are limited by their controlled nature, and it is possible to have statistically significant experimental results that do not test those processes that are key to the species in a community (Bell 2001, Hubbell 2001).  5  Thesis overview My general approach in this thesis was to test hypotheses that further our understanding of species coexistence, with a particular focus on the three themes identified above. This approach led to three distinct studies.  CARIBBEAN SEA  Liberia  %*i ~%  Laaun  &A.V CARLOS  ~#m Tilaran  SEMMSQLI.  #G«ipil«s C San i s Cf*2 fticaqm.^ ^ R a m W Miramar. * * f ? A l a j u e l a g n , ^ , , •Siquirras  Pwitar'enas Desamparidos*S8nJos4 CHIRM ptimtiiowu. PARK Oiifthff €i Cape Btaws> /:•. ..-v. Sanlsidro ^  PACIFIC  OCEAN  101  • Puerto Limoo  flay gs COICOTAIJ^^ NATOKALPARK ^ ^  Golfito  9t  PGNNSLIA ®  Wffli 25  Sikttl  Point S u i t e 02001. Efltr^lafHMia iril»*]r*ie*. Inc.  Figure 1.2: Map of Costa Rica with arrow indicating approximate position of Pitiila station where the bromeliad-mosquito research described in Chapter 2 was done.  6  / \ / / \ / / \ / | •  B  Ben.dbf Probablestream.shi Certainstream.shp Alltrails.shp Laguna2.shp Lagunal.shp Pasturel.shp Pasture2.shp Pasture3.shp Pasture4.shp Pitilla_field.shp Bosque_16yrs.shp Bosque_30-50yrssh Bosque_select_log, shp Bosque_30yrs.shp Bosque_primary.shIP  1.2 Kilome  a  Figure 1.3: Map of land-use histories surrounding the Pitilla field station, Costa Rica. Points with numbers mark locations of sampled bromeliads. Codes refer to Pasture, Pond (Laguna), Forest (Bosque) and field station (Pitilla). In the first study (Chapter 2), I explored the general patterns of diversity of bromeliaddwelling mosquito larvae in Costa Rica (Fig. 1.2), and related these back to broad hypotheses about species coexistence. In particular, I used two observational studies to test co-occurrence patterns of mosquito larvae of different species, and tested for consistency between the observed patterns and two coexistence hypotheses: neutral theory and spatial niche partitioning. This approach was particularly relevant to the community studied for a number of reasons. First, mosquito communities in Costa Rica are poorly studied, with only biogeographic and broad habitat data available for most species. Thus, except for knowledge about the species that may occur in a given community, little is known about community dynamics. Second, bromeliads form distinct container habitats, and are excellent model systems for testing species interactions (Srivastava et al. 2004). In particular, the natural replication of many natural microcosms within a study site allowed for strong tests of species co-occurrence patterns (Fig. 1.3). Finally, most mosquito species, and all those encountered in my study, are filter feeding in the larval stage, and thus use similar resources. This makes mosquitoes ideal for testing basic predictions about  7  coexistence, many of which were developed for communities competing for one or few resources (Chase and Leibold 2003), as appears to be the case for these species (Appendix A).  Figure 1.4: Map of Canada with the arrow marking the approximate location of the field study done in the southwestern Yukon, Canada. For the second study (Chapter 3), I developed and used an experimental approach that could discriminate amongst a number of specific coexistence mechanisms. This study was based on plants in the northern boreal forest understory near Kluane Lake, Yukon (Fig. 1.4). The forest understory in Kluane was particularly relevant for this study because previous research showed both positive and negative species interactions among plants, and also that negative interactions were due to resource competition, not mediated by herbivores (Callaway et al. 2002, Turkington et al. 2002). These general interactions are represented conceptually in Figure 1.5. This background information, in combination with the competitive responses of certain species (Turkington et al. 2002), allowed me to generate specific predictions for each coexistence mechanism. In addition, because species in the plant community have fairly typical patterns of relative abundances (McGill et al. 2007), an experimental approach could be planned a priori that was suitable for testing numerous competing hypotheses of species coexistence. In the experiment, I used species removals (i.e. disturbing the extant community) in conjunction with  8  seedling additions to test whether establishment patterns were consistent with the predictions of six commonly-cited coexistence mechanisms. Three of these mechanisms have already been discussed (neutral, niche segregation, and competition-colonization trade-off) and are based on the underlying assumption of competitive exclusion. The other three were based on conceptual models of facilitation, in which facilitation was either neutral (Callaway et al. 2002), increased with increasing species richness (Bruno et al. 2003, Hooper et al. 2005), or relied on a 'principal facilitator' that facilitated many other species (Bertness and Callaway 1994, Callaway 2007). Each hypothesis was used to generate predictions about the impact that a specific species should have on the establishment of new species, based on the extant species' abundance.  * * * & »  Invader  Resource Supply  Extrinsic environmental factors (climate, soil bedrock, etc.)  Figure 1.5: The effect of the plant community on resource availability and invasibility. In this diagram members of the plant community affect each other, and potentially the invader, directly through facilitation or interference competition (vectors A and B). The plant community is regulated through competition for available resources (E), which are impacted by the members of the community (C) and also supplied extrinsically (D). The invader must be able to increase from low densities in this resource environment (E) to successfully establish.  9  My third study (Chapter 4) was theoretical, and was designed to generate predictions about the role of dispersal and niche overlap in increasing local diversity, and the resulting effects of this diversity on resource use. In particular, researchers have suggested that species richness in a locality may be higher than would be supported if the locality were isolated, simply because they disperse in from adjacent areas (Shmida and Wilson 1985, Allouche et al. 2008). This is termed a mass effect by some (Shmida and Wilson 1985), and more generally a sourcesink effect when a species only exists locally because its population is bolstered from an outside source (Amarasekare and Nisbet 2001, Mouquet and Loreau 2003, Mouquet et al. 2006). This phenomenon is interesting from the point of view of testing coexistence theories because it suggests that there may be a disconnect between the mechanism promoting coexistence, the resulting large-scale effect of diversity on ecosystem function (resource use), and the local pattern of diversity and ecosystem function. For example, if two species coexist regionally because they are each a better competitor in different soil conditions, but co-occur locally due to source-sink effects, then the measured effect of species diversity may depend on the scale at which they are considered. Early theoretical work has suggested that source-sink dynamics can cause a negative trend between diversity and resource use (Mouquet and Loreau 2003), but did not link population dynamics to resource levels, making conclusions about the effects of populations on resources questionable. In Chapter Four I develop a metacommunity model that links population dynamics with resource exploitation and use simulations to generate predictions on the effect of source-sink dynamics on diversity and resource use. Overall, the research conducted for this thesis is important both for applied ecology and theoretical ecology. For example, the field research (Chapters 2 and 3) was designed to advance our understanding of the processes that promote diversity in understudied communities. This knowledge is important for predicting the impacts of different disturbances or management options, and also for falsifying and advancing theory. The implications that these studies have for specific ecosystems and coexistence theory in general is described in the Chapter Five.  10  Literature cited Adler, P. B., J. HilleRisLambers, and J. M. Levine. 2007. A niche for neutrality. Ecology Letters 10:95-104. Allouche, O., O. Steinitz, D. Rotem, A. Rosenfeld, and R. Kadmon. 2008. Incorporating distance constraints into species distribution models. Journal of Applied Ecology 45:599-609. Amarasekare, P., M. F. Hoopes, N. Mouquet, and M. Holyoak. 2004. Mechanisms of coexistence in competitive metacommunities. American Naturalist 164:310-326. Amarasekare, P., and R. Nisbet. 2001. Spatial heterogeneity, source-sink dynamics, and the local coexistence of competing species. American Naturalist 158:572-584. Bell, G. 2001. Neutral Macroecology. Science 293:2413-2418. Bell, G., M. J. Lechowicz, and M. J. Waterway. 2006. The comparative evidence relating to functional and neutral interpretations of biological communities. Ecology 87:1378-1386. Bertness, M. D., and R. Callaway. 1994. Positive interactions in communities. Trends in Ecology and Evolution 9:191-193. Bezemer, T. M., and W. H. van der Putten. 2007. Ecology: Diversity and stability in plant communities. Nature 446:E6-E7. Brooker, R. W., F. T. Maestre, R. M. Callaway, C. L. Lortie, L. A. Cavieres, G. Kunstler, P. Liancourt, K. Tielborger, J. M. J. Travis, F. Anthelme, C. Armas, L. Coll, E. Corcket, S. Delzon, E. Forey, Z. Kikvidze, J. Olofsson, F. Pugnaire, C. L. Quiroz, P. Saccone, K. Schiffers, M. Seifan, B. Touzard, and R. Michalet. 2008. Facilitation in plant communities: the past, the present, and the future. Journal of Ecology 96:13-17. Bruno, J. F., Fridley, J.D., Bromberg, K. and Bertness, M.D. 2005. Insights into biotic interactions from studies of species invasions. Pages 13-40 in D. F. Sax, Stachowicz, J.J., Gaines, S.D., editor. Species Invasions: Insights into Ecology, Evolution, and Biogeography. Sinauer, Sunderland, MA. Bruno, J. F., J. J. Stachowicz, and M. D. Bertness. 2003. Inclusion of facilitation into ecological theory. Trends in Ecology and Evolution 18:119-125. Cadotte, M. W. 2007. Competition-colonization trade-offs and disturbance effects at multiple scales. Ecology 88:823-829.  11  Cadotte, M. W., and T. Fukami. 2005. Dispersal, spatial scale, and species diversity in a hierarchically structured experimental landscape. Ecology Letters 8:548-557. Callaway, R. M. 2007. Positive interactions and mterdependence in plant communities. Springer, Dordrecht, The Netherlands. Callaway, R. M., R. W. Brooker, P. Choler, Z. Kikvidze, C. J. Lortie, R. Michalet, L. Paolini, F. I. Pugnaire, B. Newingham, E. T. Aschehoug, C. Armas, D. Kikodze, and B. J. Cook. 2002. Positive interactions among alpine plants increase with stress. Nature 417:844-848. Callaway, R. M., and L. R. Walker. 1997. Competition and facilitation: a synthetic approach to interactions in plant communities. Ecology 78:1958-1965. Cardinale, B. J., D. S. Srivastava, J. E. Duffy, J. P. Wright, A. L. Downing, M. Sankaran, and C. Jouseau. 2006. Effects of biodiversity on the functioning of trophic groups and ecosystems. Nature 443:989-992. Chase, J. M., and M. A. Leibold. 2003. Ecological Niches: Linking Classical and Contemporary Approaches. University of Chicago Press, Chicago. Chave, J. 2004. Neutral theory and community ecology. Ecology Letters 7:241-253. Chave, J., H. C. Muller-Landau, and S. A. Levin. 2002. Comparing classical community models: Theoretical consequences for patterns of diversity. American Naturalist 159:1-23. Chesson, P. 2000. Mechanisms of maintenance of species diversity. Annual Review of Ecology and Systematics 31:343-366. Davies, K. F., P. Chesson, S. Harrison, B. D. Inouye, B. A. Melbourne, and K. J. Rice. 2005. Spatial heterogeneity explains the scale dependence of the native-exotic diversity relationship. Ecology 86:1602-1610. Duffy, J. E. 2003. Biodiversity loss, trophic skew and ecosystem functioning. Ecology Letters 6:680-687. Ennos, R. A. 2001. Inferences about spatial processes in plant populations from the analysis of molecular markers. Pages 45-72 in J. Silvertown and J. Antonovics, editors. Integrating Ecology and Evolution in a Spatial Context. Blackwell Scientific Publications, Oxford. Fargione, J., C. S. Brown, and D. Tilman. 2003. Community assembly and invasion: An experimental test of neutral versus niche processes. Proceedings of the National Academy of Sciences 100:8916-8920.  12  Fridley, J. D., J. J. Stachowicz, S. Naeem, D. F. Sax, E. W. Seabloom, M. D. Smith, T. J. Stohlgren, D. Tilman, and B. Von Holle. 2007. The invasion paradox: Reconciling pattern and process in species invasions. Ecology 88:3-17. Gause, G. F. 1934. The Struggle For Existence. Williams and Wilkins, London, England. Gilbert, B., W. F. Laurance, E. G. Leigh, and H. E. M. Nascimento. 2006. Can neutral theory predict the responses of amazonian tree communities to forest fragmentation? American Naturalist 168:304-317. Gilbert, B., and M. J. Lechowicz. 2004. Neutrality, niches, and dispersal in a temperate forest understory. Proceedings of the National Academy of Sciences 101:7651-7656. Gross, K. 2008a. Fusing spatial resource heterogeneity with a competition-colonization trade-off in model communities. Theoretical Ecology 1:65-75. Gross, K. 2008b. Positive interactions among competitors can produce species-rich communities. Ecology Letters In Press. Hardin, G. 1960. The competitive exclusion principle. Science 131:1292-1297. Harpole, W. S., and K. N. Suding. 2007. Frequency-dependence stabilizes competitive interactions among four annual plants. Ecology Letters 10:1164-1169. Harrison, S. 1999. Local and regional diversity in a patchy landscape: Native, alien, and endemic herbs on serpentine. Ecology 80:70-80. Harrison, S., H. D. Safford, J. B. Grace, J. H. Viers, and K. F. Davies. 2006. Regional and local species richness in an insular environment: serpentine plants in California. Ecological Monographs 76:41-56. Hooper, D. U., F. S. Chapin, J. J. Ewel, A. Hector, P. Inchausti, S. Lavorel, J. H. Lawton, D. M. Lodge, M. Loreau, S. Naeem, B. Schmid, H. Setala, A. J. Symstad, J. Vandermeer, and D. A. Wardle. 2005. Effects of biodiversity on ecosystem functioning: A consensus of current knowledge. Ecological Monographs 75:3-35. Hubbell, S. P. 2001. The Unified Neutral Theory of Biodiversity and Biogeography. Princeton University Press, Princeton. Hutchinson, G. E. 1959. Homage to Santa Rosalia or why are there so many kinds of animals? American Naturalist 93:145-159. Leibold, M. A., M. Holyoak, N. Mouquet, P. Amarasekare, J. M. Chase, M. F. Hoopes, R. D. Holt, J. B. Shurin, R. Law, D. Tilman, M. Loreau, and A. Gonzalez. 2004. The  13  metacommunity concept: a framework for multi-scale community ecology. Ecology Letters 7:601-613. Leibold, M. A., and T. E. Miller. 2004. From metapopulations to metacommunities. Pages 133150 in I. Hanski and O. Gaggiotti, editors. Ecology, Genetics, and Evolution of Metapopulations.. Elsevier Academic Press., Amsterdam. Levin, S. A. 1970. Community equilibria and stability, and an extension of the competitive exclusion principle. American Naturalist 104:413. Levin, S. A. 1976. Population dynamic-models in heterogeneous environments. Annual Review of Ecology and Systematics 7:287-310. Levine, J. M., and C. M. D'Antonio. 1999. Elton revisited: a review of evidence linking diversity and invasibility. Oikos 87:15-26. Loreau, M., N. Mouquet, and A. Gonzalez. 2003. Biodiversity as spatial insurance in heterogeneous landscapes. Proceedings of the National Academy of Sciences 100:1276512770. MacArthur, R. 1970. Species packing and competitive equilibrium for many species. Theoretical Population Biology 1:1-11. MacArthur, R., and R. Levins. 1967. The limiting similarity, convergence, and divergence of coexisting species. American Naturalist 101:377-385. MacArthur, R. H. 1957. On the relative abundance of bird species. Proceedings of the National Academy of Sciences 43:293-295. MacArthur, R. H. 1972. Geographical Ecology; Patterns in the Distribution of Species. Harper & Row, New York. MacArthur, R. H., and E. O. Wilson. 1967. The Theory of Island Biogeography. Princeton University Press, Princeton, N. J. MacDougall, A. S., S. D. Wilson, and J. Bakker. 2008. Climatic variability alters the outcome of long-term community assembly. Journal of Ecology In Press. McGill, B. J., R. S. Etienne, J. S. Gray, D. Alonso, M. J. Anderson, H. K. Benecha, M. Dornelas, B. J. Enquist, J. L. Green, F. He, A. H. Hurlbert, A. E. Magurran, P. A. Marquet, B. A. Maurer, A. Ostling, C. U. Soykan, K. I. Ugland, and E. P. White. 2007. Species abundance distributions: moving beyond single prediction theories to integration within an ecological framework. Ecology Letters 10:995-1015.  14  McGill, B. J., B. A. Maurer, and M. D. Weiser. 2006. Empirical evaluation of neutral theory. Ecology 87:1411-1423. Mouquet, N., T. E. Miller, T. Daufresne, and J. M. Kneitel. 2006. Consequences of varying regional heterogeneity in source-sink metacommunities. Oikos 113:481-488. Mouquet, N., and M. Loreau. 2002. Coexistence in metacommunities: the regional similarity hypothesis. American Naturalist 159:420-426. Mouquet, N., and M. Loreau. 2003. Community patterns in source-sink metacommunities. American Naturalist 162:544-557. Paine, R. T. 1966. Food web complexity and species diversity. American Naturalist 100:65-75. SCBD. 2006. Global biodiversity outlook 2. Secretariat of the Convention on Biological Diversity, Montreal, Canada. Schwartz, M. W., C. A. Brigham, J. D. Hoeksema, K. G. Lyons, M. H. Mills, and P. J. van Mantgem. 2000. Linking biodiversity to ecosystem function: implications for conservation ecology. Oecologia 122:297-305. Shmida, A., and M. V. Wilson. 1985. Biological determinants of species-diversity. Journal of Biogeography 12:1-20. Srivastava, D. S. 2002. The role of conservation in expanding biodiversity research. Oikos 98:351-360. Srivastava, D. S., J. Kolasa, J. Bengtsson, A. Gonzalez, S. P. Lawler, T. E. Miller, P. Munguia, T. Romanuk, D. C. Schneider, and M. K. Trzcinski. 2004. Are natural microcosms useful model systems for ecology? Trends in Ecology and Evolution 19:379-384. Srivastava, D. S., and J. H. Lawton. 1998. Why more productive sites have more species: an experimental test of theory using tree-hole communities. American Naturalist 152:510529. Srivastava, D. S., and M. Vellend. 2005. Biodiversity-ecosystem function research: Is it relevant to conservation? Annual Review of Ecology Evolution and Systematics 36:267-294. Tilman, D. 1980. Resources - a graphical-mechanistic approach to competition and predation. American Naturalist 116:362-393. Tilman, D. 1982. Resource Competition and Community Structure. Monographs in population biology 17:1-296. Tilman, D. 1994. Competition and biodiversity in spatially structured habitats. Ecology 75:2-16.  15  Tilman, D. 2004. Niche tradeoffs, neutrality, and community structure: A stochastic theory of resource competition, invasion, and community assembly. Proceedings of the National Academy of Sciences of the United States of America 101:10854-10861. Tilman, D., C. L. Lehman, and K. T. Thomson. 1997. Plant diversity and ecosystem productivity: Theoretical considerations. Proceedings of the National Academy of Sciences 94:1857-1861. Turkington, R., E. John, S. Watson, and P. Seccombe-Hett. 2002. The effects of fertilization and herbivory on the herbaceous vegetation of the boreal forest in north-western Canada: a 10-year study. Journal of Ecology 90:325-337.  16  Plate 1. Bromeliads in the Guanacaste conservation area. Top: large bromeliad with experimental tubes in leaf wells (Appendix A). Bottom: Sampled bromeliad (Chapter 2). Background illustrates rainforest.  2. Niche partitioning at multiple scales facilitates coexistence among mosquito larvae1 Introduction Determining the mechanisms that allow species to coexist is important for understanding community ecology and conserving biodiversity (e.g., Chave et al. 2002, Amarasekare et al. 2004). Ecological theory distinguishes between mechanisms that stabilize coexistence among species (sensu Chesson 2000), and those that lead to unstable coexistence and species 'drift' by solely equalizing species' competitive abilities (Chesson 2000, Hubbell 2001). Understanding how each mechanism contributes to coexistence in a community allows for predictions on both the long-term persistence of community members and the processes that will lead to the extirpation of a given species (Chesson 2000). For these reasons, tests of the importance of stabilizing mechanisms and neutral drift have recently gained focus in research on diverse organisms, including forest plants (e.g., Gilbert and Lechowicz 2004), intertidal animals (Wootton 2005), and stream insects (Thompson and Townsend 2006). The scale at which stabilizing mechanisms and neutral drift are expected to function is particularly important to species coexistence (Holyoak and Loreau 2006). Neutral drift may be apparent in local interactions among species in a metacommunity, and when these interactions are considered at the spatial scale of the metacommunity they lead to spatial clustering so long as individuals have limited dispersal (Chave and Leigh 2002, Gilbert and Lechowicz 2004, Holyoak and Loreau 2006). Stabilizing mechanisms may also lead to scale-dependent coexistence, whereby species that cannot coexist locally due to asymmetric competitive ability or predator avoidance can co-exist at larger temporal or spatial scales that are environmentally heterogeneous (e.g., Chesson 2000, Davies et al. 2005). In both scenarios, local coexistence mechanisms may be nested within larger, landscape-level processes that are dominated by environmental heterogeneity (Chave 2004). Tests of coexistence mechanisms should therefore include members of a single metacommunity and test across spatial scales that capture the environmentally heterogeneity experienced by those species (Amarasekare 2003). 1  A version of this chapter has been published as: Gilbert, B., D.S. Srivastava and K.R. Kirby. 2008. Niche partitioning at multiple scales facilitates coexistence among mosquito larvae. Oikos 117:944-950.  18  Here we focus on larvae of bromeliad-dwelling mosquito species in the Guanacaste region of northern Costa Rica. Previous research on mosquito larvae has provided contrasting evidence for coexistence mechanisms. Although several studies of container-dwelling mosquitoes have identified potential stabilizing mechanisms (e.g., Livdahl and Willey 1991, Sunahara and Mogi 2002b, Costanzo et al. 2005), others have been either unable to find a stabilizing mechanism of coexistence (Bradshaw and Holzapfel 1983; Schneider et al. 2000), or have cited priority effects as drivers of species distributions whereby larger or early-arriving individuals are able to outcompete late-arriving individuals for a particular niche space (e.g., Livdahl 1982, Sunahara and Mogi 2002a, Lounibos et al. 2003). When priority effects occur without other life history trade-offs, they are consistent with species drift (Hubbell 2001, Amarasekare et al. 2004). Indeed, zero-sum neutral models (e.g., Hubbell 2001) rely on a very strong priority effect: pre-emptive competition for space. Such contrasting results from previous studies may have resulted from the scales at which the mosquito community was considered. The most common case for which a stabilizing mechanism was identified involves two species that have different environmental niches, as occurs when species perform best in different habitat types, such as tree holes and tires (e.g., Lounibos 1981, Livdahl and Willey 1991). Likewise, temporal fluctuations in environmental conditions, such as drought and rainy periods, have been shown to favour different species at different times (Juliano et al. 2002, Costanzo et al. 2005). However, these mechanisms of habitat specialisation generally lead to species coexisting at large spatial and/or temporal scales (i.e. regional coexistence), by favouring a single species within a single habitat (e.g., Constanzo et al. 2005, but see Yee et al. 2007). This pattern is consistent with Leibold and McPeek's (2006) hypothesis that stable coexistence is likely to occur at larger spatial and temporal scales, as observed in mosquito communities, and that neutral dynamics are more likely to occur within single habitats. The mechanisms promoting the coexistence of mosquito larvae within a single habitat are particularly interesting because all filter feeding species consume the same resources (small particles of detritus and microscopic organisms), with little evidence for differences in feeding preferences or behaviour among species (Kesavaraju et al. 2007, Yee et al. 2007, Yee et al. 2004 and references therein). Recently, Yee et al. (2004) proposed that mosquito species create 'spatial niches' by specialising in specific vertical strata within a given container habitat. This  19  mechanism could potentially promote within-habitat coexistence by restricting interactions among species, such as resource competition or interference competition (Lounibos et al. 2003). To our knowledge, the spatial niche mechanism for coexistence has not been tested in natural mosquito communities. In this study we use sampling data to test the degree to which the bromeliad mosquito community shows evidence of niche segregation or neutral dynamics. If partitioning of spatial niches is important in this community, species with similar spatial niches should co-occur less in natural bromeliads, as these species will have the strongest potential for competition (Fig 2.1 A). In contrast, if neutral processes are dominant and species do not differ in their competitive abilities, we expect species distributions to be random with respect to spatial niches. Neutral dynamics may produce different patterns of co-occurrence, depending on whether dispersal limitation and priority effects are important. In the case of dispersal limitation, species tend to be spatially clustered, and the similarity between any two bromeliads decreases linearly with logarithm of distance (Fig. 2. IB; Chave and Leigh 2002). Priority effects in mosquito communities are expected to cause high mortality in young cohorts (late-arriving mosquitoes), which would decrease the co-occurrence of different cohorts (Fig. 2.1C). Depending on the processes that are most important at local scales, we expect differences in the relationship between species' distributions and the environmental characteristics of bromeliads. If niche processes are important, species' distributions should correspond to environmental differences among bromeliads, allowing specific species to be optimal competitors at distinct points along an environmental gradient. In contrast, if neutral processes determine local coexistence, there will either be no correspondence between species and environment, or environmentally distinct bromeliads will house neutral species groups (Chave 2004). Here we test 1) if there is evidence for species partitioning of spatial niches within containers, 2) if sampled mosquito communities yield species co-occurrence patterns that are consistent with predictions based on species' spatial niche overlap, priority effects (size differences), or dispersal limitation (spatial clustering), and 3) if among-bromeliad habitat partitioning is consistent with local co-occurrence patterns.  20  U.f CD CD  on  0.60.5-•-  o E  u> c cc o  a.  E CD  +*  0.30.2 0.1 «•"* -•*  0 -  **  <<  0.4 =3 O" CO  s "  ** ** **  -"•*  «***-*  -*•  S>  *•"•  -**  ^^ *" -** *-*  *> Difference in spatial niche of mosquito pair  500  1000  1500  2000  Distance between pair of bromeliads (m) 0.7  1  2  3  Difference in instar size  Figure 2.1: Hypothetical trends in mosquito community depending on whether neutral dynamics (solid lines) or niche partitioning (dashed lines) determine co-occurrence patterns. A) neutral prediction: there is no relationship between the areas of the water column that species tend to occupy and their co-occurrence; niche prediction: species that occupy different areas of the water column co-occur, while those with similar spatial niches do not; B) neutral prediction: the mosquito community changes in species composition as the distance between bromeliads increases due to universal dispersal limitation; niche prediction: species composition is determined by environmental heterogeneity such that geographic distance is not important once environmental heterogeneity is accounted for; C). neutral prediction: individuals of a given instar do not co-occur with individuals of different instars because of priority effects; niche prediction: inter-specific co-occurrence is not solely related to instar size. 21  Methods Two studies were conducted from 6 October to 28 November, 2004, in the area immediately surrounding Pitilla biological station (10°59'N, 85°26'W) in the Area de Conservation Guanacaste (http://www.acguanacaste.ac.cr/). The station is at an altitude of 700 m in the premontane tropical wet forest life zone (Holdridge 1967) and approximately 4000 mm of rain falls between May and February. The landscape surrounding the station is made up of primary forest, regenerating (1-80 year old) forest and pasture. All experiments and sampling were performed in primary or secondary forest. Further details on the area are given in Srivastava et al. (2005). Study 1: Observation of species' spatial niches within containers An observational study was designed to detect differences in the spatial positions of the four common mosquito species in the area {Anopheles neivai (Howard, Dyar and Knab), Wyeomyia melanopus (Dyar), Culex rejector (Dyar and Knab), and Wyeomyia circumcincta (Dyar and Knab)). Two 3rd instar individuals of a species were put into a transparent 50 ml centrifuge tube to which 0.4 g (dry weight) of fine detritus (1mm < diameter < 5mm) from the wells of bromeliads in nearby forest and 35 mL of water had previously been added. The top of each tube was covered with mesh. The depth in the water column of each individual larva was recorded after allowing sufficient time for the mosquitoes to acclimatize to the presence of the researcher. This observational study was designed to replicate that of Yee et al. (2004), but using an observer instead of a camera. In total, the experiment consisted of 6 replicates (tubes) of A. neivai and 3 of each other species, each of which was monitored twice a day over 4 days. Each monitoring period was considered a single observation, with observations nested within experimental units (see statistical methods, below). Because of the consistent and clear spatial positioning of the species, we deemed it unnecessary to monitor the larvae beyond this time. Study 2: Sampling species distributions A total of 71 bromeliads were sampled to determine species patterns of co-occurrence, and whether these patterns correlated with environmental conditions within or around the bromeliad. Individuals from two genera of bromeliad, Vriesia and Guzmania, were sampled at <2.5 meters  22  from the forest floor. Sampling was done with a turkey baster and all water from each leaf well in each bromeliad was removed. More water was added twice to completely flush each leaf well. Measures of leaf well pH, water holding capacity, and actual water level at time of sampling were taken when possible. Bromeliads were measured for characteristics that predict their water holding capacity (i.e. their maximum volume) based on relationships generated from previous sampling of the two genera encountered in our survey (r2 = 0.95, d.f.=22 and r2 = 0.88, d.f.=14, both p < 0.001; D. Srivastava, unpub.). These characteristics included the bromeliad genus, number of leaves, and the maximum width and length of leaves. In addition, bromeliad species, date sampled, geographic location, forest age, forest canopy openness (on a 5 point scale), elevation, and local bromeliad density on the tree sampled and within a 3m radius (to 10 m height) were quantified. All aspects of leaf wells (species and environmental variables) were grouped within each bromeliad prior to analysis, as leaf wells often share water and therefore are not independent. All water collected while sampling was scanned using a dissecting scope (10X magnification) to ensure that small mosquitoes were not missed. Mosquito larvae were identified at Pitilla with a dissecting scope, with some 1st instar individuals raised in rearing cups to 2nd or 3rd instar for positive identification. Larval traits for identification did not change after the 2nd instar for any of the species. Larvae were identified using a key for Costa Rican mosquitoes generated by G. Chaverri at the Instituto National de Biodiversidad (INBio), and voucher specimens were collected and stored in the INBio Culicidae Collection. Mosquito sizes were measured as the length of the body from the tip of the head to the base of the siphon. Sizes were measured for all common species at each instar except C. rejector, which had its first instar length estimated from the lengths of larger instars (exponential equation, r2 = 0.97, p<0.0001, 10 d.f.) because we were unable to distinguish between C. rejector and C. jenningsi (Dyar and Knab) in the first instar.  Analysis The observational study of the spatial niches of mosquito larva was analysed using a general linear model (GLM) with a multinomial distribution and a cumulative logit link. The multinomial distribution was used because the spatial distribution of individuals was not normal, with larvae occurring either at the water surface, at the bottom of the water column (on the  23  detritus) or in a small band in the centre of the water column. Observations were aggregated within experimental units and the dispersion parameter was set as the Pearson's x2 statistic divided by the degrees of freedom for computing standard errors and likelihood ratio statistics. Aggregating within experimental units corrects for multiple observations from a single unit by treating them as a single observation when calculating standard errors and likelihood ratios. Sampling data was used to evaluate two niche-based (spatial partitioning of the water column and habitat partitioning among bromeliads) and two neutral (priority effects and dispersal limitation) mechanisms. The two neutral mechanisms could not be tested simultaneously because of different predictor variables in each relationship; however, because of the different predictor variables, these effects should both be detectable if they occur. In each of these tests, co-occurrence patterns or community similarity were calculated using the Steinhaus coefficient (Legendre and Legendre 1998); this similarity coefficient is the complement of the Bray-Curtis distance, and measures the relative co-occurrence of two species based on their individual abundances across all bromeliads and their joint abundances in bromeliads in which they co-occur. It can also be used to measure community similarity by comparing species relative abundances on a per bromeliad basis. The Steinhaus coefficient ranges from 0 (no cooccurrence or no similarity) to 1 (two species have identical abundances in all bromeliads, or 2 bromeliads have identical abundances of all species). Uncommon species are not appropriate for these analyses, thus species that occurred in less than 5 bromeliads were removed from the analysis. These species were Wyeomyia abebela (Dyar and Knab), W. pseudopecten (Dyar and Knab), Culexjenningsi, and an unidentified species of Wyeomyia. The role of priority effects in determining coexistence patterns were evaluated by correlating size differences between any two instars with the co-occurrence of these two instars. If priority effects alone determine competitive outcomes, then we would expect that early instar individuals (1st and 2nd instar) would occur less with later instar individuals (4th instar) than with each other (Sunahara and Mogi 2002a; Fig. 2.1C). When priority effects are considered a neutral mechanism, species identity should not matter. Therefore we ignored species identity in this analysis, with one exception: we did not include comparisons among instars within a species, as different growth rates among individuals from the same oviposition event could cause spurious differences. Instar co-occurrences were compared with two metrics: differences among instars based on mosquito size measurements from samples, and differences among instars treating a  24  given instar of all species as equivalent (i.e. all Is instar individuals had a size of 1, 2n instars were 2, etc.). To test this relationship, a randomization was performed by generating correlation coefficients for the pairwise relationship between instar size differences and a randomized instar co-occurrence matrix. To minimize the risk of type 1 error, randomizations were constrained by keeping the total abundances of all instars of each species constant (Gotelli 2000). This and all other randomization tests were run 9999 times. This analysis was subsequently repeated with 1st instar individuals removed, and then with 1st and 2nd instar individuals removed, to insure that the lack of effect shown was not due to a time lag in mortality. These subsequent analyses did not differ qualitatively (all had p>0.05), and we therefore only present results from the first analysis. We evaluated the role of dispersal limitation by testing for spatial clustering of species with a Mantel test (Legendre and Legendre 1998). The neutral hypothesis was that species similarity between bromeliads decreases with geographic distance (Chave and Leigh 2002; Fig. 2.1b). Both linear distances and logarithmically transformed distances were used for this test. We then evaluated whether species co-occurrences increase with differences in spatial niches, (Fig. 2.1 A) to test for a niche-based explanation of species co-occurrence. Differences in spatial niches were measured as the absolute difference in depth between each species pair from the observational study, and species co-occurrences were calculated from the sampling data. As with the test for priority effects, a randomization test was used to evaluate the pairwise correlation between species co-occurrence patterns and species spatial niche differences. Following the analysis of similarities, we tested whether species partition niches at larger scales (i.e. between bromeliads) by examining species' distributions among bromeliads. In particular, we tested for correlates of species densities using a canonical correspondence analysis (CCA), which is able to model linear or unimodal species distributions along environmental gradients (Legendre and Legendre 1998). Densities were used to prevent large bromeliads with more individuals from dominating the ordination, with density calculated as density = # individuals / bromeliad water holding capacity. The densities of species that occurred in 5 or more bromeliads were used as the dependent matrix, and the explanatory matrix consisted of the following environmental variables: water pH, Julian date, water holding capacity of bromeliad, forest canopy openness, forest canopy height, forest age, local bromeliad density and bromeliad species. Explanatory variables were included through a forward selection process with an a =  25  0.05 cut-off. Because total species density showed a slight decrease with bromeliad size (r2 =0.08, p<0.04), sampling analyses were rerun with this trend removed statistically to test their sensitivities to this effect; the results did not change, so the initial test is reported. Analyses were conducted using SAS (SAS 1999) for GLMs, CANOCO for the CCA (ter Braak and Smilauer 1998), and PC-ORD (McCune and Mefford 1999) and our own code in visual basic for Mantel tests, randomization tests and distance matrices. A  I  B  -1  > 0) • =  *J  B  -V  $  o  an  I  » t T3 3 -3 C « (0 g  1°.5  -4  E  ^o_ -5 -6  A. nei  C, rej  W. mel  W. cir  Species  Figure 2.2: Mean depth (± standard error) of individuals below water surface for Anopheles neivai (A. nei.), Culex rejector (C. rej.), Wyeomyia melanopus (W. mel.) and Wyeomyia circumcincta (W. cir.). Different letters represent statistically significant differences (Bonferroni corrected a = 0.05) in vertical habitat use based on frequency of occurrence at the water surface, middle, or bottom of the container.  Results Species showed significant differences in their vertical distributions in the water column (multinomial GLM, F3;25 = 18.74, p<0.0001), with A. neivai occurring mainly at the water surface, W. circumcincta occurring in or on the detritus at the base of the container, and the other species occurring in the middle of the water column (Fig. 2.2). All species appeared to feed by both filtering the water column and browsing the detritus or the sides of the container.  26  A r = 0.95 p < 0.001  0.7 ,  $ o  0.60.5  An-Wc  An-Wm 0.4 Wc-Wm*  0.3 to  0.2  a.  0.1  Cr-Wc An-Cr  Cr-Wm  1  2  3  4  5  Pairwise differences in species spatial niches  B  r = 0.04 NS *•  !  0.81 • * * •  •  ••%, .  •  # * M»*»  #  «  Distance between pair of bromeliads (m) r = 0.06 NS  a) o r (/) m 0M -i  U)  o o m o •F 1 O (/> o t/5  R  ^ £  TO 10  0. 1  2  Pairwise differences in instar sizes  Figure 2.3: Tests of the spatial niche hypothesis (A) and two neutral hypotheses (B & C). A) The relationship was significant between species' pairwise difference in spatial niches (vertical distributions) and their pairwise co-occurrence. Species with different niches score high on the x-axis, and frequently co-occurring species score high on the y-axis. Abbreviations for both species in a species pair are labelled with the genus (first letter) and species (second letter). B & C) Both tests of neutral hypotheses were not significant (NS, p>0.05).  27  Species' pairwise co-occurrence patterns were strongly correlated with differences in their spatial niches (r = 0.95, p=0.0005; Fig. 2.3A). The correlation is consistent with predictions of local niche segregation, with those species most likely to interact within a water column (C. rejector and W. melanopus) least likely to co-occur, and species least likely to interact (W. circumcincta and A neivai) co-occurring most frequently. By contrast, there was no evidence of consistent priority effects. Individuals from small instars were as likely to co-occur with individuals from larger instars as with other individuals of small instars, and this result was consistent whether actual sizes were used or all individuals of a given instar were included in a single rank (r = 0.06 and 0.03, both p>0.05; Fig. 2.3C). Likewise, there was no evidence of dispersal-limitation at the distances measured (distances ranged from 0.015 m to 2128 m); spatial clustering in community composition was not significant using linear or log-transformed distances (r = 0.01 and 0.04, both p>0.05; Fig. 2.3B). The CCA indicated that bromeliad water holding capacity was the only significant correlate of species-distributions amongst bromeliads, and it explained 36% of the total variation in common species' densities (pO.OOOl, Fig. 2.4). Overall, species that had high overlap in vertical niches (e.g., C. rejector and W. melanopus, Fig. 2.2) showed segregation in the size of bromeliad in which they were each most common (C. rejector occurred primarily in bromeliads > 60ml, W. melanopus primarily < 60 ml; Fig. 2.4).  Discussion The mosquito species studied here are restricted to bromeliads for the larval stage of their life cycles (e.g., Dyar 1928), and their coexistence is therefore mediated by their interactions within this habitat. For these species to coexist stably over the long term, species pairs with strong overlap in one niche axis must compensate by segregating along a second niche axis (reviewed in Chesson 2000). The community studied here followed this principle; species had well-defined spatial niches, and species pairs with the most overlap in their spatial niches co-occurred least (Fig. 2.3). In contrast, a neutral community would show random patterns of species cooccurrence with respect to spatial niches, and species co-occurrence patterns would instead reflect priority effects and dispersal limitation. Our results reject a purely neutral model of coexistence both locally and in the larger metacommunity.  28  Wyeomyia melanopus  o CD  o. CD O -i—i CD  CO  Z3  -q > C  c CD  T3 3 CT W  o E CD  I  300  Bromeliad capacity (ml)  Figure 2.4: Histograms of average species densities (individuals per litre of water holding capacity) along a gradient of bromeliad water holding capacities. Bromeliad water holding capacity explained approximately 36% of species densities (p<0.001). Note that the last size category ranges from 120 to 300 ml, and the range of densities (y axis) differs among species.  29  Interference competition has been documented as an important regulator of mosquito density (Broadie and Bradshaw 1991), and may be especially relevant within the limited volume of bromeliad tanks (Lounibos et al. 2003). Spatial partitioning among species has been proposed as a mechanism that limits interference and potentially resource competition and thus facilitates species coexistence (Yee et al. 2004). The strong correlation between species co-occurrence patterns and overlap in spatial niches supports this hypothesis, and suggests not only that the abundance of each species is affected by the abundance of other species in a bromeliad, but that the strengths of interspecific interactions are determined by species' spatial niches. It has been hypothesized that coexistence among species in many communities is determined by one or two important niche axes (Chase and Leibold 2003). The wide applicability of spatial niche partitioning, not only for mosquitoes but across taxa (e.g., MacArthur 1958), suggests that this mechanism may be an important and common stabilizing mechanism for species coexistence (Chesson 2000). If within-bromeliad co-occurrence relies on differences in species' spatial niches, what causes these different niches? In some cases the answer to this question is clear: species from the genus Anopheles lack respiratory siphons, and thus would be expected to occur more at the surface of the bromeliad. However, for the other species present there is no such limitation and yet even the two Wyeomyia species have different spatial niches. These different niches may have evolved from negative interspecific interactions or may instead be related to some other aspect of the biology of the individual species, as appears to be the case with A. neivai. However, the cause of these differences may be unimportant to the outcome of ecological interactions: if a difference in life-history traits decreases negative interactions among species, it increases the probability that those species will coexist stably (Chesson 2000). The strong relationship between spatial niche differences and co-occurrence (Fig. 2.3) suggest that this is indeed the case. Species with overlapping spatial niches within bromeliads occur in bromeliads of different sizes, effectively changing the scale at which niche partitioning occurs. For example, species with large microhabitat overlap (W. melanopus and C. rejector) occur in bromeliads at opposite ends of a size spectrum, while species with little microhabitat overlap (W. circumcincta and A. neivai) occupy similarly sized bromeliads (Fig. 2.4) and co-occur frequently (Fig. 2.3). These differences in occurrence can result from one of two mechanisms; differential survival  30  rates or oviposition differences. Although much of the literature on niche partitioning tests differential survival among species (e.g., Costanzo et al. 2005), oviposition preferences can also determine mosquito distributions (Edgerly et al. 1998). For example, the larger scale of spatial partitioning (Fig. 2.4) is consistent with other species that cue oviposition choices to bromeliads of a certain size (e.g., Srivastava et al. 2005), and may be enhanced by females detecting competing species at the time of oviposition (e.g., Edgerly et al. 1998). Although our data could not test between these possibilities, patterns of species segregation were fairly consistent across instars, suggesting that much of the observed partitioning among differently sized bromeliads was in fact due to oviposition choice. Selecting different bromeliad sizes, however, likely represents a trade-off for some of the species considered, with C. rejector (large bromeliads) and W. melanopus (small bromeliads) occurring in sub-optimal habitats. Small bromeliads are more likely to dry out (D. Srivastava, unpub.), which reduces mosquito survival (e.g., Juliano et al. 2002). On the other extreme, bromeliads with a water-holding capacity greater than 100 ml usually host the top odonate predator Mecistogaster modesta (Srivastava et al. 2005), which consumes mosquitoes (D. Srivastava and J. Ware, fecal dissections) and is likely to have a strong negative impact on mosquito densities (Fincke et al. 1997, Chase and Knight 2003). Neutral theory assumes that other trophic levels do not impact species distribution patterns, and further study would need to be undertaken to test this assumption directly. Nonetheless, from the data presented here we speculate that the effects of predators (C. rejector) and drought (W. melanopus) could impact competitive outcomes at different bromeliad sizes (e.g., Chase and Knight 2003, Costanzo et al. 2005), and may represent trade-offs between resistance to stress-induced mortality (predation or drought) and competitive ability (Chase and Leibold 2003). As with many studies that correlate pattern with process (e.g., Karst et al. 2005, Thompson and Townsend 2006), we did not experimentally test the effects of competition among species across the range of bromeliad sizes to ensure that the observed patterns match the predicted mechanism. Although such a test is beyond the scope of this study, we suggest that future research focus on the strength of interspecific competition amongst species with different spatial niches and how their competitive asymmetries change with bromeliad volume. In summary, the bromeliad mosquito community studied shows local co-occurrence patterns that are consistent with spatial niche partitioning, and these patterns scale up to  31  environmental segregation among species. Although partitioning at the local and amongbromeliad scale is related to species richness and species turnover respectively, competitive interactions among species are consistent with the partitioning at both scales. Further studies should test the underlying mechanism of spatial niche partitioning and its generality in other mosquito communities.  32  Literature cited Allison, P.D. 1995. Survival analysis using SAS: a practical guide. SAS institute Inc. Cary, NC, USA. Amarasekare, P. 2003. Competitive coexistence in spatially structured environments: A synthesis. Ecology Letters 6: 1109-1122. Amarasekare, P., M.F. Hoopes, N. Mouquet, M. Holyoak. 2004. Mechanisms of coexistence in competitive communities. American Naturalist 164: 310-326. Bell, G. 2001. Neutral macroecology. Science 293: 2413-2418. Bradshaw, W.E., and CM. Holzapfel. 1983. Predator-mediated, non-equilibrium coexistence of tree-hole mosquitoes in southeastern North America. Oecologia 57: 239-256. Broadie, K.S. and W.E. Bradshaw. 1991. Mechanisms of interference competition in the western tree-hole mosquito, Aedes sierrensis. Ecological Entomology 16: 145-154. Chase, J.M. and T.M. Knight. 2003. Drought-induced mosquito outbreaks in wetlands. Ecology Letters 6: 1017-1024. Chase, J.M. and M.A. Leibold. 2003. Ecological Niches: Linking Classical and Contemporary Approaches. University of Chicago Press, Chicago. Chave, J. 2004. Neutral theory and community ecology. Ecology Letters 7: 241-253. Chave, J. and E. G Leigh, EG. 2002. A spatially explicit neutral model of P-diversity in tropical forests. Theoretical Population Biology 62: 153-168. Chave, J., H. C. Muller-Landau and S. A. Levin. 2002. Comparing classic community models: theoretical consequences for patterns of diversity. American Naturalist 159: 1-23. Chesson, P. 2000. Mechanisms of maintenance of species diversity. Annual Review of Ecology and Systematics 31: 343-366. Clark, J.S. 2003. Uncertainty in ecological inference and forecasting. Ecology 84: 1349-1350. Clark, S.J. and J.S. McLachlan. 2003. Stability of forest biodiversity. Nature 423: 635-638. Costanzo, K.S., B.Kesavaraju and S.A. Juliano. 2005. Condition-specific competition in container mosquitoes: the role of non-competing life-history stages. Ecology 86: 3289-3295. Davies, K. F., P. Chesson, S. Harrison, B. D. Inouye, B. A. Melbourne, and K. J. Rice. 2005. Spatial heterogeneity explains the scale dependence of the native-exotic diversity relationship. Ecology 86: 1602-1610.  33  Dyar H. G. 1928. The Mosquitoes of the Americas. Carnegie Institution, Washington. Publication 387. Edgerly, J.S., M. McFarland, P. Morgan and T. Livdahl. 1998. A seasonal shift in egg-laying behaviour in response to cues of future competition in a treehole mosquito. Journal of Animal Ecology 607: 805-818. Fincke, O.M., S.P. Yanoviak and R.D. Hanschu. 1997. Predation by odonates depresses mosquito abundance in water-filled tree holes in Panama. Oecologia 112: 244-253. Gaston, K.J. and Chown, S.L. 2005. Neutrality and the niche. Functional Ecology 19: 1-6. Gilbert, B. and M.J. Lechowicz. 2004. Neutrality, niches, and dispersal in a temperate forest understory. Proceedings of the National Academy of Sciences 101: 7651-7656. Gotelli, N.J. 2000. Null model analysis of species co-occurrence patterns. Ecology 81: 26062621. Gravel, D., C. D. Canham, M. Beaudet, and C. Messier. 2006. Reconciling niche and neutrality: The continuum hypothesis. Ecology Letters 9: 399-409. Holdridge, L.R. 1967. Life zone ecology. Tropical Science Center, San Jose, Costa Rica. Holyoak, M, and M. Loreau. 2006. Reconciling empirical ecology with neutral community models. Ecology 87: 1370-1377. Hubbell, S.P. 2001. The Unified Neutral Theory of Biodiversity and Biogeography. Princeton University Press, Princeton, N.J. Juliano, S. A., G. F. O'Meara, J. R. Morrill, and M.M. Cutwa 2002. Desiccation and thermal tolerance of eggs and the coexistence of competing mosquitoes. Oecologia 130: 458-469. Karst, J., B. Gilbert and M.J. Lechowicz. 2005. Assembling fern communities: the roles of chance and environmental determinism at local and intermediate spatial scales. Ecology 86: 2473-2486. Kesavaraju, B., D.A. Yee and S.A. Juliano. 2007. Interspecific and Intraspecific Differences in Foraging Preferences of Container-Dwelling Mosquitoes. Journal of Medical Entomology . 44:215-221. Legendre, P. and L. Legendre. 1998. Numerical Ecology. 2nd English Edition. Developments in Environmental Modeling, 20. Elsevier Science, Amsterdam, Netherlands. Leibold, M.A., and M.A. McPeek. 2006. Coexistence of the niche and neutral perspectives in community ecology. Ecology 87: 1399-1410.  34  Livdahl, T.P. 1982. Competition within and between hatching cohorts of a treehole mosquito. Ecology 63: 1751-1760. Livdahl, T.P. and W.S. Willey. 1991. Prospects for an invasion: competition between Aedes albopictus and native Aedes triseriatus. Science 253: 189-191. Lounibos, L.P. 1981. Habitat segregation among African treehole mosquitoes. Ecological Entomology 6: 129-154. Lounibos, L.P., G.F. O'Meara, N. Nishimura and R.L. Escher. 2003. Interactions with native mosquito larvae regulate the production of Aedes albopictus from bromeliads in Florida. Ecological Entomology 28: 551-558. MacArthur, R.H. 1958. Population ecology of some warblers of northeastern coniferous forests. Ecology 39: 599-619. McCune, B. and M.J. Mefford. 1999. Multivariate Analysis of Ecological Data Version 4.17, MjM Software, Gleneden Beach, Oregon, U.S.A. McGill, B.J. 2003. A test of the unified theory of biodiversity. Nature 422: 881-885. McGill, B.J., B. A. Maurer, and M. D. Weiser. Empirical evaluation of neutral theory. Ecology 87:1411-1423. McPeek, M.A. 2004. The growth/predation trade-off: so what is the mechanism? American Naturalist 163: E88-E111. SAS Institute. 1999. The SAS System for Windows. Version 8.02. SAS Institute, Cary, North Carolina, USA. Schneider, P., W. Takken and PJ. McCall. 2000. Interspecific competition between sibling species larvae of Anopheles arabiensis and An. gambiae. Medical and Veterinary Entomology 14: 165-170. Shea, K. and P. Chesson. 2002. Community ecology theory as a framework for ecological invasions. Trends in Ecology and Evolution 17: 170-176. Srivastava, D.S., M.C. Melnychuk and J.T. Ngai. 2005. Landscape variation in the larval density of a bromeliad-dwelling zygopteran, Mecistogaster modesta (Odonata: Pseudostigmatidae). International Journal of Odonatology 8: 67-79. Sunahara, T. and M. Mogi. 2002a. Priority effects of bamboo-stump mosquito larvae: influences of water exchange and leaf litter input. Ecological Entomology 27: 346-354.  35  Sunahara T. andM. Mogi. 2002b. Variability of intra- and interspecific competitions of bamboo stump mosquito larvae over small and large spatial scales. Oikos 97: 87-96. ter Braak, C. J. F., and P. Smilauer. 1998. CANOCO for Windows. Version 4.0. Centre for Biometry, CPRO-DLO, Wageningen, The Netherlands. Thompson, R.M. and C.R. Townsend. 2006. A truce with neutral theory: local deterministic factors, species traits and dispersal limitation together determine patterns of diversity in stream invertebrates. Journal of Animal Ecology 75: 476 - 484. Wooton, J.T. 2005. Field parameterization and experimental test of the neutral theory of biodiversity. Nature 433: 309-312. Yee, D.A., B. Kesavaraju and S.A. Juliano. 2004. Larval feeding behavior of three co-occurring species of container mosquitoes. Journal of Vector Ecology 29: 315-322. Yee, D. A., M. G. Kaufman, and S. A. Juliano. 2007. The significance of ratios of detritus types and micro-organism productivity to competitive interactions between aquatic insect detritivores. Journal of Animal Ecology 76: 1105-1115.  36  JLJ,/*  A- #*&***  N;^ '  •  i » W ' ^ ' * ' \ *' -'-' '""-""^V ••>:.*. ^ » >  Plate 2. Aspen stands in Kluane, Yukon. Top: Landscape view of aspen stands amidst grassy shrubland and spruce forest. Bottom: Field assistants Aimee Pelletier and Jenn Mundy sample understory vegetation.  37  3. Coexistence and mechanisms of invasion in a boreal forest understory2 Introduction One of the fundamental goals of ecology is to understand the processes that determine species abundances and diversity, and in turn to understand how these patterns affect ecosystem processes (Chesson 2000, Srivastava and Vellend 2005). Over the past two decades, threats of species loss and species invasions have motivated numerous empirical studies that quantify the effects of species loss on ecosystem processes, such as resistance to invasion (Levine and D'Antonio 1999, Schwartz et al. 2000, Srivastava and Vellend 2005, Fridley et al. 2007). Despite the broad range of mechanisms that are predicted to promote coexistence and determine diversity, these studies have focused almost exclusively on a niche-saturation hypothesis as a common theoretical framework (Levine and D'Antonio 1999, Fridley et al. 2007). The nichesaturation hypothesis posits that diversity is maintained through each species having a distinct resource niche, and as a consequence, more diverse communities are predicted to use resources more completely and thus better exclude invaders (i.e. limit diversity). While many tests of the niche-saturation hypothesis have shown that it is important (Lyons and Schwartz 2001, Kennedy et al. 2002, Levine et al. 2004), studies that explicitly test different components of the invaded community often find that a competitively dominant species is equally or more important than diversity per se (Crawley 1999, Fargione et al. 2003a, Emery and Gross 2007). Given the large number of mechanisms that can promote diversity, it is not surprising that different studies find differing effects of diversity on invasibility. For example, Hubbell's neutral model proposes that the competitive equivalence of all species, both newly arriving species and residents, enables the maintenance of large numbers of rare species that typically cause a community to be diverse (Hubbell 2001). Thus, in a neutral world there should be no relationship between native diversity and success of invaders. Other theories, such as the competition-colonization trade-off, predict that species identity, not diversity, determines the resistance of a community to invaders (Chesson 2000). In particular, this trade-off posits that good competitors will be poor dispersers, and this competitive ranking determines if an invader  " A version of this will be submitted for publication with the following authorship: B. Gilbert, R. Turkington and D.S. Srivastava  38  can establish in the presence of the resident species (Tilman 1994). Consequently, communities dominated by good dispersers should be more invasible because they have less competitive resistance than communities dominated by strong competitors. A more extreme divergence from the niche-saturation framework comes from studies that demonstrate positive interactions among species, which may increase species' realized niches (Bruno et al. 2003, Bulleri et al. 2008). Numerous studies have shown that one or more species in a community may facilitate other species, which has led to the hypothesis that facilitation is more important than competition in certain communities, especially those that are particularly stressful (Callaway and Walker 1997, Callaway 2007). These studies have produced three distinct facilitation hypotheses. The first, which we term the 'neighbor facilitation' hypothesis, predicts that all extant species within a community facilitate their neighbors. An example was shown in alpine sites around the world, where facilitation was common regardless of the identity of the focal (facilitated) species or its neighbors (Callaway et al. 2002). A second facilitationbased hypothesis proposes that more diverse communities may have higher levels of invasibility because each extant species has the potential to facilitate new species (Bulleri et al. 2008). Indeed, this diversity-facilitation hypothesis forms one of the arguments for the role of diversity in maintaining ecosystem function (Hooper et al. 2005). The third hypothesis, which we term a 'principal facilitator' hypothesis, describes a single species that facilitates others through unique functions, such as habitat stabilization, nutrient fixation, or hydraulic lift, while the beneficiary species often have a negative impact on the facilitator (reviewed in Callaway 2007). The numerous and distinct mechanisms that may underlie species coexistence in any community make general tests about the causes and consequences of diversity difficult. Nonetheless, each of the individual theoretical frameworks can be understood by its underlying predictions about specific species' effects on other species in a community. Here we consider the effects of disturbances that remove a fixed amount of biomass from a typical community that is characterized by few dominant species and many low-abundance species (McGill et al. 2007). Such equal-sized disturbances could remove either a proportion of individuals of the dominant species or completely remove many rare species, and the effect of these different disturbances on invaders will depend on the coexistence mechanism that operates. In particular, we used the average rank-abundance curve of a naturally-occurring plant community to generate three removal treatments that would be equal in the proportion of biomass removed but differ in the  39  number and identity of species removed (Fig. 3.1). We also created two controls (no removal and 100% removal) and added transplanted seedlings to assess the effects of different extant species on invasibility.  I  0) to O CO  1 2  II  3  4  10  -Q O < C CO o  £ i  > =E  sa  x  6  7  12  I  CO o  5  11  13 14  i  8  Species Rank  B •  0.8 -  •  •  rs=0.75 p=<0.0001  *  0.6-  •  • •  0.4 -  S. LL  •  0.2-  • •  n-  *  * 0.0001  ••  • 0.001  ••• 4* .  • • •  • 0.01  0.1  Relative Abundance (mean proportion of biomass)  Figure 3.1. The rank-abundance relationship among species and its relationship to frequency within aspen stands. A) The average rank-abundance relationship for aim 2 plot in the study area. Patterns illustrate the removal treatment, with each removal treatment consisting of 7% of the total community biomass. Biomass was removed either by removing as many of the low biomass species as necessary (low-abundance removal treatment, striped) or by removing most of the 2n rank species (herbaceous dominant removal, white), or by removing a small proportion of the 1st rank species (woody dominant removal, dots). The inset shows the same graph with a linear y-axis. B) The relationship between each species' average relative biomass when it was present and the frequency with which it occurred in lm2 plots (n=75).  40  The framework we developed (Fig. 3.1) can be used to generate specific invasion predictions for each coexistence mechanism considered. For example, neutral theory and the neighbor facilitation hypothesis predict equal effects for removals of the same biomass, but differ in whether the underlying interaction is competitive or facilitative (Fig. 3.2A,B). The niche saturation hypothesis posits that, on average, the relative abundance of each species reflects the availability of its resource niche (McKane et al. 2002). Thus, a disturbance that eliminates many low-abundance species should allow a greater number of invading species than a disturbance of similar size that affects only a single dominant species (Fig. 3.2C). The diversity-facilitation hypothesis also relies on the importance of many low-abundance species, but predicts the opposite effect of the niche-saturation hypothesis (Fig. 3.2D). In contrast, because of the greater importance of species identity in a competition-colonization trade-off, a disturbance that affects a single dominant species would cause a greater increase in invasibility if this species is competitively superior (Fig. 3.2E), as is often the case in late successional communities (Tilman 1994, Harpole and Tilman 2006). However, unlike the other theories, the link between dominance and competitive ability needs to be tested independently for a competitioncolonization trade-off. We are unaware of specific predictions for principal facilitators, but a Lotka-Volterra model can be used to generate predictions for such a species that facilitates others and in turn is negatively impacted by its beneficiaries (Appendix B). Under the assumptions of the Lotka-Volterra model considered, dominant species are more likely to be principal facilitators, and thus a disturbance that eliminates the dominant species should decrease invasibility more than a similarly sized disturbance that eliminates many low-abundance species (Fig. 3.2F). We applied our design in the boreal forest understory in northern Canada. Previous work in the area has shown that plant community composition is driven by both facilitation and competition for resources, and that herbivores have little effect on plant abundances (Callaway et al. 2002, Turkington et al. 2002). Despite these general findings, no studies have examined the roles of extant species in limiting or promoting diversity in this area. Invasibility was assessed with transplanted seedlings that consisted of both exotic and native herbs and grasses that were functionally similar to the low-abundance species already present. In addition, we used differences in resource availability among treatments and extant species' competition and colonization abilities to discriminate among potential coexistence mechanisms.  41  Neighbor Facilitation  Neutral  12 CD U CO C  n 14  Niche Saturation  Diversity-Facilitation  CD  >  X3  2  T3 CD  o  .9 "° CD  14  E i  Competition-colonization * ^ « M trade-off  Principal Facilitator  Removal Treatment Figure 3.2. Predictions from different coexistence mechanisms about the diversity of seedlings resulting from removal of all plants (complete), no plants (none), or an equivalent amount (7% of total biomass) of dominant species or low-abundance species. All predictions assume that each seedling is only interacting with the extant flora, not other seedlings. A) Neutral model. B) Neighbor facilitation that assumes no species-specific effects, but that facilitation underlies invader establishment. C) Niche saturation model whereby higher diversity increases resource use and thereby excludes invaders. D) Diversity-facilitation model, which posits that high diversity facilitates more species. E) The competition-colonization trade-off predicts that a species' competitive rank will determine the number of invading species that it excludes. Dominance is often correlated with competitive ability in late successional communities, as modeled here. F) Principal facilitation occurs when a single species facilitates many others, which is more likely to occur with an abundant species. Only the two dominant species in this study occurred in all plots and were therefore likely to act as principal facilitators.  42  Study area and methods The study area, near Kluane Lake in the south-western Yukon in northern Canada (61 00 32 N, 138 14 49 W), has been described in long-term studies of the area (Krebs et al. 2001, Turkington et al. 2002). The area receives a mean annual precipitation of c. 230 mm, mostly falling as rain during the summer months, but including an average annual snowfall of about 100 cm (Turkington et al. 2002). The vegetation at lower elevations is a patchwork of spruce forest, aspen forest, and shrubby grasslands. Five replicate sites were selected from a number of aspen stands on the landscape, with the selected sites ranging in distance from 0.8 - 9km from each other and separated by different habitat types. The central site was located at Christmas creek near the Alaska Highway (138° 13.9' W, 61° 00.5' N). In each aspen stand we established ten 1 m2 plots with similar plant communities, based on the relative cover of dominant species. We first developed allometric relationships for all understory plant species occurring in the plots that incorporated specific traits for each species, such as height or leaf area (Appendix C). These individual plant relationships were used on each plant within each plot, which allowed us to calculate total and relative species biomass for each plot. We used these calculations of species biomass within each plot to generate an average rank-abundance curve for all the 1 m2 plots (Fig. 3.1). Plots in each site were randomly assigned a treatment (below). We used the average rank-abundance curve to generate three removal treatments that would be equal in proportional biomass removed but differ in number of species and the dominance of the species removed (Fig. 3.1). All treatments had 7% of the total plot biomass removed, which was the average relative biomass of all the low-abundance species in a plot (rank 3 or higher, Fig. 3.1). The species removed in each treatment were determined by their rank, with Arctostaphylos uva-ursi (woody dominant) always the 1st rank and Epilobium angustifolium (herbaceous dominant) always the 2nd rank in removal plots. In the low-abundance removal plots we started with the species at the highest rank (lowest biomass) and moving progressively downward. If a target species had more biomass than the total biomass to be removed, plants were randomly selected within plots until the desired removal amount was attained. As many species in the area are clonal, the individuals to be removed were separated from conspecifics by severing any root connections. Two control treatments, 0 and 100% removal, were also created.  43  Removals were performed by brushing a general herbicide, glyphosate (6.8 g active ingredient L"1) to the leaves of individual plants. After the initial application, follow-up applications were done on target plants as needed. Previous research suggests that the application method and timing of glyphosphate additions would not impact the non-target native vegetation and the seedlings that were transplanted into the plots the following year (Coupland and Lutman 1982, Torstensson et al. 1989, Cornish and Burgin 2005), and no non-target effects were observed. Target plants were left in the plot, as this was the only way to insure complete plant mortality. Target plants were spatially referenced so that their continued absence could be ensured in subsequent years. Initial removals were completed in late July 2004, and were maintained until June 2006. In addition to these removal treatments, root connections were severed with a spade 15 cm outside of plot edge and the buffer between the edge and spaded line was given the same removal treatment. In mid-May 2005, bare-root seedlings were transplanted from sand into one plot of each removal treatment per site. Twelve species of seedling were used (Appendix C). These species were selected from a pool of native and exotic species that were expected to survive in this region, with some species occurring at low densities in sampled stands: Anemone multifida, Elymus trachycaulus, Taraxacum officinale (Appendix C). All scientific names follow the published flora for the area (Cody 2000). Eleven of the species were planted at 7 seedlings per plot, and one species, A. multifida, at 10 seedlings per plot because of the small initial size of these seedlings. Seedlings were planted at randomly assigned locations within each plot, with a 10 cm buffer (plot edge) excluded and the randomization constrained so that no more than two seedlings were in any 10 x 10 cm section of the plot. Seedlings were watered after transplanting (2 L m"2). Resource availability was monitored in plots with seedlings and in 'control' plots that had the same removals but no seedlings. We used ion exchange membranes that assess ion availability in the soil (PRS™-probes ,Western Ag Innovations Inc., Saskatoon, Canada). Four anion and four cation probes were systematically placed in all plots for the 2005-2007 growing seasons, for total burial times of 67, 61 and 60 days respectively. Volumetric soil water content was measured with a Hydrosense™ soil water meter (Campbell Scientific), which uses timedomain reflectometry for standardized soils to estimate soil water from 0-8cm depth. Measurements were taken up to twice per season, but only following at least one week without  44  rain to avoid rapid changes that can occur when soil moisture is above field capacity. To avoid sunflecks, light availability was measured on overcast days using a Spectrum 6-sensor bar that detects photon flux density within the photosynthetically active range (400-700 nm). Light was measured at seedling height (10 cm) and directly above the plot (1.1 m), with this ratio indicating the difference in light availability that is due to the treatment effect.  Statistical methods Seedling responses The experiment was designed as a blocked split-plot, with the survival time of each seedling treated as a sub-replicate of the species within a plot (Allison 1995). Certain removal treatments and seedling species had high mortality, which made it impossible to consider all data in the full experimental design both due to violation of analysis assumptions and model convergence problems. We therefore first analyzed the main effect of the experiment (i.e. removal effects without species effects or interactions) by grouping all seedlings within a plot, regardless of species identity. This acts as an unbiased test of the main effect if there are no statistical interactions present, and will underestimate the significance of the main effects if there is an interaction (Koch 1969). After finding a significant difference between the 'complete' removal and all other treatments, this treatment was removed from the analysis as it clearly influenced the effect of time in the model. We then analyzed the effect of the other removal treatments, again grouping all seedlings within each plot. Both of these initial analyses were done using generalized linear mixed models (GLMM) with a logistic link function. Experimental block was considered a random effect and temporal autocorrelation in the error term was modeled using a spatial power function because of unequal sampling intervals (Littell et al. 2006). The type of autocorrelation function for all analyses was chosen based on visual analysis of the residuals and their correlation over time, assessing the sensitivity of the statistical results to alternate models, and by comparing model fit statistics (AIC, BIC). For the first tests, temporal autocorrelation was modeled with the plot acting as the subject, as all individuals were grouped within plots. Degrees of freedom were determined with the Kenward-Roger correction (Littell et al. 2006), and post-hoc comparisons were performed using the 'simulate - stepdown' option in SAS. Because there was unexplained heterogeneity in the experiment (Pearson %2/d.f. >1), we considered covariates that showed no relationship with the treatments to avoid confounding  45  covariate and treatment effects (we used a cut-off of p>0.6 from the resource results, described below). Three covariates, S, NH4 and the initial community biomass (before removal treatment), met this requirement and were considered. Only NH4 was significant and included in analyses. A scale parameter was also included as needed to model the heterogeneity (Littell et al. 2006). To analyze the full model, we first removed the 'complete' removal treatment, as high mortality made it impossible to analyze. We then removed all species that had 100% mortality in some removal treatments, namely Anemone multifida, Hedysarum mackenzii, Medicago sativa and Agrostis scabra. Sampling data was then combined within each year (spring and fall census), which was necessary for model convergence. Differences in mortality rates were tested in a GLMM. Temporal autocorrelation was again modeled with a spatial power function, with each species*plot combination acting as a subject. Experimental blocks and block*treatment combinations were included as random effects. Three diversity indices were generated for the transplants in each plot: species richness, Simpson's index and the Shannon-Weiner index. Mixed models, with an unstructured autocorrelation function, were used to test the effect of removal treatment on diversity. The complete removal treatment was removed from this analysis because the high mortality in these plots made comparisons meaningless (see Results). Extant community Treatment effects on resource levels were tested for all macronutrients, soil moisture and light levels. Mixed models were used for all tests, and variables were log-transformed as necessary to homogenize variances. The residuals of the macronutrients were modeled with an autoregressive model or compound symmetry model depending on residual autocorrelation. An unstructured variance model was used for soil moisture and light due to more complex seasonal patterns in these variables. For nutrients and soil moisture, if the global test for treatment effects was significant, pre-planned contrasts were used to compare partial-removal treatments with each control (complete removal and no removal), and differences amongst partial-removal treatments were compared with a post-hoc test. Significant interactions were examined graphically, and comparisons made with least-squared means. Because we hypothesized that removing the woody dominant species would not affect light levels (it is <10 cm tall), we compared light levels among all treatments with a post-hoc test. Resource data included plots with seedlings added or  46  no seedlings, but tests indicated that seedling addition made no difference to resource availability (p values varied from 0.14 to 0.9) and this variable was dropped. We also tested whether partialremoval plots with greater amounts of biomass removed had higher resource availability, as removing 7% of biomass in a plot resulted in different absolute amounts of biomass removed. Neither biomass nor a biomass*removal interaction were statistically significant and therefore are not reported further. Species' natural colonization rates were estimated by calculating the percentage of plots that a species colonized if it were absent after initial removal treatments. For the dominant species, this included only the complete removal plots, while many low-abundance species had the potential to colonize more plots (up to 40). To see if the different denominators caused a bias, we redid the estimates, restricting all species to invasion in complete removal plots. These measures were strongly correlated (r=0.94, p<0.0001) and gave similar results in subsequent tests, so we only report the first. Species' competitive abilities were not measured directly, but rather inferred from a 10 year fertilization study by Turkington et al. (2002). Previous work has shown that a species' R* (the concentration to which a species lowers free nutrients when grown in monoculture, with a low R* indicating a high competitive ability) is positively correlated to its response to fertilization (Harpole and Tilman 2006). We therefore used the change in percent cover with fertilization for all species measured in the Turkington study that also occurred in our plots. Relative competitive ability (RCA) was calculated as the decrease in abundance that occurred with fertilization: RCA = (abundanceCOntroi - abundancefertiiiZed plots) / m a x Aabundance across species-  47  Herbaceous dominant « _ _ Woody dominant —— Low abundance spp. No removal  H> 0.7  Complete removal  • >  •g  0.6  | 0.5 O '•C 0.4  o Q.  S 0.3 CL 0.2-  B/B/B  100  200  300  400  500  700  600  800  Days since start of experiment  B Woody dominant Herbaceous dominant Low abundance spp.  •  No removal  Jj  II  X0"  I  .#  Species Figure 3.3. Total and species-specific survival. A) Cumulative survival of all seedlings in each removal treatment. Points represent sampling dates, with lines between points included for visual clarity only. Treatments with different letters (right side of figure) are significantly different (oc=0.05). B) Survival of the eight most successful species in the partial-removal and no-removal treatments after 3 growing seasons.  48  Results Seedling responses The survival rate of all seedlings differed among treatments (F4j74=19.7, pO.0001; Fig. 3.3A). Survival was significantly higher in plots where the herbaceous dominant was removed compared to all other treatments (all p values < 0.02), and the complete removal treatment had significantly lower seedling survival than all treatments (all p values O.0001). The other three treatments did not differ significantly (all p values > 0.29). When we excluded the complete removal treatment from our analysis and considered only those species whose seedlings had > 0% survival in all other treatments, the analysis showed two trends. First, there was a significant interaction between removal treatment and year (F6,297=2.23, p=0.04). This interaction occurred because seedlings in the low-abundance removal treatment had a similar survival rate as seedlings in the herbaceous dominant removal treatment in the final year of the study (78 and 77% respectively), whereas the herbaceous dominant treatment had the highest average survival in the other years. This interaction, along with the reduction of the species and sampling periods considered, made the difference among the partialremoval and no-removal treatments non-significant (p=0.35). Second, the species identity of transplanted seedlings had an important effect on survival (Fytui=\23, p<0.0001; Fig. 3.3B). This species effect did not change with removal treatment (interaction p=0.23), but did change over time (year*species interaction Fi4,303=7.22p<0.0001). This interaction occurred because two species (Phleum alpinum and Poa compressa) had much lower survival in the third year than previous years, whereas most species' survival rates did not change significantly in the second and third year. Differences in species richness among the partial-removal and no-removal treatments emerged over time (removal*time interaction, Fi2,76=2.58 p=0.006), with the herbaceous dominant removal treatment maintaining a slower rate of species loss than the other treatments. This trend was similar for all diversity indices, as these indices were strongly correlated (rs ranging from 0.95 to 0.99, all p<0.0001). Differences in diversity appeared to be driven mainly by the number of surviving seedlings, both at a plot level and treatment level (Fig. 3.4). When these numbers were standardized through ratification, the interaction term became insignificant (p=0.4), indicating that treatment effects were indeed driven by differences in total survival. The  49  seedlings in the complete removal treatment had a much lower species richness than all other treatments, with three of five replicates each containing one surviving seedling of differing species.  Complete  Woody Herb. dominant dominant  Low None abundance  Complete  Woody Herb. Low None dominant dominant abundance  D  in O) C 12  c T5 <D  C  *o  co "  Q>  »  CD CD  • •  .  CO M~  O  CO •  0) c o  in to  'o a> a. co  X <D "D C  Dno  <1>  c Herbaceous dominant  CD  or  >  - - Woody dominant  o  —~ Low abundance spp.  100  200  C  o c c  <D Q.  No removal  CO 300  400  500  600  700  Days since transplanting  CD  10  15  20  25  30  35  40  45  co  N u m b e r of seedlings surviving  Figure 3.4. The difference in diversity among treatments and its relationship to the number of survivors. A & B) Average differences in species richness among treatments reflect differences in survival. C) Differences in species richness of seedlings among treatments increase over time. D) The relationship between number of seedlings, species richness (squares) and the Shannon-Weiner index (diamonds). The 'complete removal' treatment was not included, in C&D as it had only 1 survivor in each of 3 plots by the final sampling date.  50  b b b  Complete  b  lill  Complete  Woody Herb. Low None dominant dominant abundance  b  • III  Woody dominant  Herb. Low dominant abundance  None  D 16  " o  -  ^60 -  L  \  \ N  - - ——- - -  Herbaceous dominant Woody dominant Low abundance spp. No removal Complete removal  12 'o o CO  s  I  s ^ j v S ^ - T K ? ^ aw «*•?—"-  2005  2006  T  2007  July 2005  June 2006  July 2006  June 2007  July 2007  Figure 3.5. Differences in resource availability among removal treatments. A) Nitrate availability. Different letters show significant differences (a=0.5) based on pre-planned comparisons (see Methods). B) Proportion of photosynthetically active radiation available at seedling height (0.10 m) compared to above all plants in a plot (1.1 meters). C) Phosphorus availability (mean ± SEM) shows a time by removal interaction, with no significant difference two or more years after treatment. D) Soil moisture (mean ± SEM) showed a time by treatment interaction. Extant community Resource availability differed among treatments, with the various resources showing different patterns (Fig. 3.5). Three macronutrients, Ca, K, S and the ammonium form of nitrogen (NEU) did not differ significantly between any removal treatments. Nitrate availability was significantly greater in the complete removal treatment than in the partial-removal treatments (p=0.004), but these latter treatments did not differ from each other, nor from the plots with no removal treatment (all p values >0.13). Phosphorus showed a similar pattern in the first year with the complete removal having the most available phosphorus (all p values < 0.02), but this difference disappeared in subsequent years (all p values<0.18). Light availability within a plot was lowest  51  when no plants were removed or when the woody dominant was removed, and greatest with complete removal (Fig. 3.5B, all differences p<0.02). Soil moisture was highest in complete and herbaceous dominant removals at the outset of the experiment (July 2005-June 2006), after which the herbaceous dominant was not significantly different from any of the other treatments (all p> 0.15; Fig. 3.5D). Magnesium showed a marginal difference among treatments (F4,i3i=2.49, p=0.046) with no removal plots having a higher Mg concentration than the partialremoval treatments (p=0.04). No other differences were statistically significant, and the direction of the magnesium trend suggests that this effect was due to a spurious correlation. The colonization rates of extant species showed a general increase with relative abundance (rs=0.36, p=0.04; Fig. 3.6A). However, the woody dominant was a clear exception to this rule, as it did not successfully colonize any plots. We compared our colonization results with published data on the competitive response of species (Table 1 from Turkington et al 2002). A strong negative correlation between colonization ability and competitive ability was found (Fig. 3.6B), as is predicted for a competition-colonization trade-off.  Discussion After testing a number of competing mechanisms that could potentially control diversity in this forest understory, we found that no single mechanism explains the patterns of species establishment. Indeed, although several mechanisms make predictions that are partially supported by our data, none predict the outcome of the positive and negative interactions shown here. Nonetheless, our experimental design separates biomass removal effects from species effects, and therefore allows us to compare the net effect of species of differing abundance. This approach allows us to infer which factors limit seedling establishment in the community, and in particular to evaluate predictions for both facilitative and competitive hypotheses of species invasion.  52  0.9  rs=0.35 p = 0.04  0.8 CD CD  0.7  DC c o  0.6  +-« CO N  E  _g o O  -Herbaceous dominant  0.5 0.4 0.3 0.2  Woody dominant 0.1  | ffi  I  l 0.00001  0.0001  -••  •  «*i — 0.001  I  \  0.01  0.1  1  Relative Abundance (Average Proportion of Biomass) B  0.6-  -Q CD CD .>  •  0.4-  0.2 -  » •  CD  Woody dominant  1 ° O « 0  rs=0.83 p = 0.02  -0.2-  Herbaceous dominant  >  "S -0.4CD  • -0.6 -  0  0.1  0.2  0.3  0.4  0.5  0.6  0.7  0.8  Colonization Rate  Figure 3.6. Extant species' colonization rates and competitive abilities. A) Colonization rates of species increased with mean abundance, except in the case of the woody dominant. B) Colonization was negatively correlated to species' competitive abilities, with competitive ability taken from published literature on changes in abundance following fertilization (greater increase following fertilization = poorer competitor; see Methods for details). Facilitative effects Theories that assume that competition for resources is the sole limit on the establishment of new species are not supported by our data (Fig. 3.2A,C,E). When all potential competitors are removed from a community, resources do become more available (Fig. 3.5), but invaders fail to establish. This facilitative effect was evident in both summer and winter, suggesting that neighboring vegetation reduces multiple types of stress for new seedlings. However, despite this strong facilitative effect when all extant species are removed, removal of 7% of community 53  biomass does not result in an incremental facilitative effect. Instead, when a small portion of the community is removed, new seedlings either respond positively or neutrally (Fig. 3.3). Together, these results indicate that facilitation is a non-linear function of plant density that becomes important for seedling establishment between 7% and 100% biomass removal. There have been both conceptual and theoretical models that propose that facilitation dominates at lower plant densities but switches to neutral or competitive dynamics at high densities (Callaway and Walker 1997, Hernandez 1998). Our results support this hypothesis, and suggest that facilitative processes may be very important in successional dynamics but play a relatively minor role when small-scale disturbances occur in intact communities. These results also offer insights into the hypothesis that through facilitation, high species diversity may lead to further increases in diversity (Bruno et al. 2003, Hooper et al. 2005, Bulleri et al. 2008). Seedling survival in the plots from which low-abundance species were removed indicates that over 80% of plant species (representing 7% of plant biomass) can be completely removed without causing net facilitative effects. Thus, when species loss from this community consists of many low-abundance species, which are often considered to be most vulnerable (Lyons and Schwartz 2001, Srivastava and Vellend 2005), the diversity - facilitation hypothesis is not supported. Instead, it appears that the non-linear pattern of facilitation seen here is either a general effect of having neighbors (Fig. 3.2B) or instead results from a dominant species (Fig. 3.2F). In either scenario, dominant species are important to facilitative dynamics either through species-specific effects or because of their high density. Previous research on facilitation by dominant species (Smith et al. 2004) or by neighbors of differing species (Callaway et al. 2002) suggests that either hypothesis could be accurate. Although it has previously been argued that facilitation needs to be incorporated into theories of species coexistence (Bertness and Callaway 1994, Callaway and Walker 1997, Bruno et al. 2003, Callaway 2007), there has been relatively little progress. Many models that incorporate facilitation predict when facilitation or competition will be most important (Bertness and Callaway 1994, Callaway and Walker 1997), but such models do not make predictions about how these processes determine the diversity of a community, or what limits a community from continually growing. More recent models, such as the facilitation-R* model (Gross 2008b) provide new hypotheses about facilitation, but remain difficult to test. For example, Gross's model requires tests of both mortality and reproduction rates, as well as information on species'  54  competitive ranks. Similarly, density-dependent facilitation models (Hernandez 1998) can only be tested empirically with a priori knowledge of which species are likely to facilitate others, or through removals of various densities of all species within a community. Thus, although these models extend our conceptual understanding of how facilitation can occur within communities, further work needs to be done to develop feasible tests. Competitive effects Despite the importance of facilitation when all neighbors are removed, the competitive effect of removing the herbaceous dominant was large given the disturbance level (10% increase in survival for 7% of biomass removed, versus 24% decrease in survival for 100% biomass removed for facilitative effect). This competitive effect informs us about the processes that limit diversity when the community is subjected to minor disturbances, even though these processes change at high levels of disturbance. In particular, the strong competitive effect of the herbaceous dominant, compared to the other species, suggests that a competitive hierarchy of the extant flora limits seedling establishment. In addition, the composition of invading species was consistent across treatments (Fig. 3.3B). This resulted in seedling diversity being best described by a sampling effect, where more surviving individuals led to greater diversity (Fig. 3.4). Together, these results are sufficient to test and reject a number of coexistence theories (Fig. 3.2). For example, neutral theory posits that all species behave equally both in terms of their impacts on other species and in their recruitment probabilities (Hubbell 2001). Our results clearly reject this theory. Removing low-abundance species also did not appear to create open niches for seedling species, as predicted by a niche-saturation hypothesis. If grass species, for example, fill a specific functional niche, we would expect the removal of low abundance species to favour grass establishment, as 28% of the biomass of low abundance species is made up of grasses. Our data reject this hypothesis, instead showing that both alpha diversity (Fig. 3.4) and the composition of establishing species (Fig. 3.3B) are unaffected by removing numerous low-abundance species. These results are inconsistent with a number experimental communities with artificial gradients in diversity (Fargione et al. 2003a, Levine et al. 2004), which may not be as well suited to testing invasion hypotheses because of sampling effects (Wardle 2001, Fridley et al. 2007). However, our results are also inconsistent with one experiment that also removes low-abundance species  55  (Lyons and Schwartz 2001), suggesting that the role of low-abundance species varies across communities, as has been seen in many biodiversity-ecosystem function studies (Schwartz et al. 2000, Srivastava and Vellend 2005). Differences in seedling survival following partial removals, and the finding that diversity is simply caused by a sampling effect, suggests that a coexistence mechanism that depends on a competitive hierarchy may best explain diversity in this community. One such mechanism, the competition-colonization trade-off, appears at first glance to operate within the community (Fig. 3.6). If such a mechanism determines invasion success, we would expect the removal of the competitive dominant to have the highest impact on seedling survival. We see the opposite trend, however, with the best colonizer limiting seedling establishment most. This result is clearly unstable from a coexistence viewpoint, as any species that is both competitively dominant and a superior colonizer would quickly displace all others in a community (Tilman 1994). Indeed, our results suggest that a competition-colonization trade-off alone cannot account for the results found here. More complex versions of competition-colonization trade-offs that incorporate different stress-tolerances among species may be relevant to this community (Levine and Rees 2002), especially if such differences underlie a different trade-off axis. Just as the herbaceous dominant appears to 'break the rules' by having a high colonization rate and also suppressing invaders more than other species, it is apparent that the other dominant species, A. uva-ursi, plays a very different role in the community. Although many theories suggest that the dominant should be best at suppressing other species in a given environment (Tilman 1980, Sala et al. 1996), others suggest that avoiding competition altogether by occupying a distinct niche may lead to dominance and even promote facilitative interactions. A number of studies support this latter hypothesis by showing that one species or group of species is facilitated by a functionally distinct species (Levine 2000a, Smith et al. 2004, Valiente-Banuet et al. 2006). More generally, the diverse impacts of dominant species on invasibility and ecosystem function indicate that dominant species are often critically important to community dynamics, but that their roles are not always predictable a priori (Smith and Knapp 2003, Smith et al. 2004, Emery and Gross 2007).  56  Conclusion Establishment of seedlings is a key stage in population growth (Emery and Gross 2007), with early establishment patterns often forming long-term patterns (Foster and Tilman 2003). Our study, which shows large species-specific effects on seedling survival, has important implications for diversity in this community and for the study of plant community dynamics more generally. For the Kluane area, it highlights the importance of facilitation in early succession, and illustrates that early and late successional dynamics may be influenced by different processes as species interactions switch from facilitative to competitive. It also illustrates the importance of one dominant species in limiting diversity within the community, and suggests that the other dominant species likely facilitates other species when at low densities. Overall, our approach has provided insight into the types of coexistence mechanisms that are likely acting in the community, and highlights specific species that warrant further study. More generally, the experimental design that we have developed offers a framework for discriminating amongst a number of coexistence theories, and also for broadly identifying the net importance of different species groups in a community. Although the importance of dominant species has been hypothesized (Sala et al. 1996, Smith and Knapp 2003, Smith et al. 2004, Emery and Gross 2007), previous studies have not linked the role of dominant species with specific predictions from coexistence theory. Our work demonstrates both the theoretical and practical importance of considering dominant species even though they contribute little to a community's diversity. This general approach is stronger than many experiments that test single hypotheses, and may act as a starting point for broader tests of theory in community ecology.  57  Literature cited Allison, P. D. 1995. Survival analysis using SAS: A practical guide. SAS Institute Inc., Cary, NC. Bertness, M. D., and R. Callaway. 1994. Positive interactions in communities. Trends in Ecology and Evolution 9:191-193. Bruno, J. F., J. J. Stachowicz, and M. D. Bertness. 2003. Inclusion of facilitation into ecological theory. Trends in Ecology and Evolution 18:119-125. Bulled, F., J. F. Bruno, and L. Benedetti-Cecchi. 2008. Beyond competition: Incorporating positive interactions between species to predict ecosystem invasibility. PLoS Biology 6:el62. Callaway, R. M. 2007. Positive Interactions and Interdependence in Plant Communities. Springer, Dordrecht, The Netherlands. Callaway, R. M., R. W. Brooker, P. Choler, Z. Kikvidze, C. J. Lortie, R. Michalet, L. Paolini, F. I. Pugnaire, B. Newingham, E. T. Aschehoug, C. Armas, D. Kikodze, and B. J. Cook. 2002. Positive interactions among alpine plants increase with stress. Nature 417:844-848. Callaway, R. M., and L. R. Walker. 1997. Competition and facilitation: a synthetic approach to interactions in plant communities. Ecology 78:1958-1965. Chesson, P. 2000. Mechanisms of maintenance of species diversity. Annual Review of Ecology and Systematics 31:343-366. Cody, W. J. 2000. Flora of the Yukon Territory. Second edition. National Research Press, National Research Council of Canada, Ottawa. Cornish, P. S., and S. Burgin. 2005. Residual effects of glyphosate herbicide in ecological restoration. Restoration Ecology 13:695-702. Coupland, D., and P. J. W. Lutman. 1982. Investigations into the movement of glyphosate from treated to adjacent untreated plants. Annals of Applied Biology 101:315-321. Crawley, B. H. E. 1999. Invasion-resistance in experimental grassland communities: species richness or species identity? Ecology Letters 2:140-148. Emery, S. M., and K. L. Gross. 2007. Dominant species identity, not community evenness, regulates invasion in experimental grassland plant communities. Ecology 88:954-964.  58  Fargione, J., C. S. Brown, and D. Tilman. 2003. Community assembly and invasion: An experimental test of neutral versus niche processes. PNAS 100:8916-8920. Foster, B. L., and D. Tilman. 2003. Seed limitation and the regulation of community structure in oak savanna grassland. Journal of Ecology 91:999-1007. Fridley, J. D., J. J. Stachowicz, S. Naeem, D. F. Sax, E. W. Seabloom, M. D. Smith, T. J. Stohlgren, D. Tilman, and B. Von Holle. 2007. The invasion paradox: Reconciling pattern and process in species invasions. Ecology 88:3-17. Gross, K. 2008. Positive interactions among competitors can produce species-rich communities. Ecology LettersTn Press. Harpole, W. S., and D. Tilman. 2006. Non-neutral patterns of species abundance in grassland communities. Ecology Letters 9:15-23. Hernandez, M.-J. 1998. Dynamics of transitions between population interactions: a nonlinear interaction a-function defined. Proceedings of the Royal Society B: Biological Sciences 265:1433-1440. Hooper, D. U., F. S. Chapin, J. J. Ewel, A. Hector, P. Inchausti, S. Lavorel, J. H. Lawton, D. M. Lodge, M. Loreau, S. Naeem, B. Schmid, H. Setala, A. J. Symstad, J. Vandermeer, and D. A. Wardle. 2005. Effects of biodiversity on ecosystem functioning: A consensus of current knowledge. Ecological Monographs 75:3-35. Hubbell, S. P. 2001. The Unified Neutral Theory of Biodiversity and Biogeography. Princeton University Press, Princeton. Kennedy, T. A., S. Naeem, K. M. Howe, J. M. H. Knops, D. Tilman, and P. Reich. 2002. Biodiversity as a barrier to ecological invasion. Nature 417:636-638. Koch, G. G. 1969. Some Aspects of the Statistical Analysis of "Split Plot" Experiments in Completely Randomized Layouts. Journal of the American Statistical Association 64:485-505. Krebs, C. J., S. A. Boutin, and R. Boonstra. 2001. Ecosystem Dynamics of the Boreal Forest: The Kluane Project. Oxford University Press, New York. Levine, J. M. 2000. Complex interactions in a streamside plant community. Ecology 81:34313444. Levine, J. M., P. B. Adler, and S. G. Yelenik. 2004. A meta-analysis of biotic resistance to exotic plant invasions. Ecology Letters 7:975-989.  59  Levine, J. M., and C. M. D'Antonio. 1999. Elton revisited: a review of evidence linking diversity and invasibility. Oikos 87:15-26. Levine, J. M., and M. Rees. 2002. Coexistence and relative abundance in annual plant assemblages: The roles of competition and colonization. American Naturalist 160:452467. Littell, R. C , G. A. Milliken, W. W. Stroup, R. D. Wolfmger, and O. Schabenberger. 2006. SAS for Mixed Models. 2nd edition. SAS Institute Inc., Cary, NC. Lyons, K. G., and M. W. Schwartz. 2001. Rare species loss alters ecosystem function - invasion resistance. Ecology Letters 4:358-365. McGill, B. J., R. S. Etierme, J. S. Gray, D. Alonso, M. J. Anderson, H. K. Benecha, M. Dornelas, B. J. Enquist, J. L. Green, F. He, A. H. Hurlbert, A. E. Magurran, P. A. Marquet, B. A. Maurer, A. Ostling, C. U. Soykan, K. I. Ugland, and E. P. White. 2007. Species abundance distributions: moving beyond single prediction theories to integration within an ecological framework. Ecology Letters 10:995-1015. McKane, R. B., L. C. Johnson, G. R. Shaver, K. J. Nadelhoffer, E. B. Rastetter, B. Fry, A. E. Giblin, K. Kielland, B. L. Kwiatkowski, J. A. Laundre, and G. Murray. 2002. Resourcebased niches provide a basis for plant species diversity and dominance in arctic tundra. Nature 415:68-71. Sala, O. E., W. K. Lauenroth, S. J. Mcnaughton, G. Rusch, and X. Zhang. 1996. Biodiversity and ecosystem functioning in grasslands. Pages 129-149 in H. A. Mooney, J. H. Cushman, E. Medina, O. E. Sala, and E.-D. Schulze, editors. Functional Roles of Biodiversity: A Global Perspective. John Wiley & Sons Ltd, Chichester. Schwartz, M. W., C. A. Brigham, J. D. Hoeksema, K. G. Lyons, M. H. Mills, and P. J. van Mantgem. 2000. Linking biodiversity to ecosystem function: implications for conservation ecology. Oecologia 122:297-305. Smith, M. D., and A. K. Knapp. 2003. Dominant species maintain ecosystem function with nonrandom species loss. Ecology Letters 6:509-517. Smith, M. D., J. C. Wilcox, T. Kelly, and A. K. Knapp. 2004. Dominance not richness determines invasibility of tallgrass prairie. Oikos 106:253-262. Srivastava, D. S., and M. Vellend. 2005. Biodiversity-ecosystem function research: Is it relevant to conservation? Annual Review of Ecology Evolution and Systematics 36:267-294.  60  Tilman, D. 1980. Resources - a graphical-mechanistic approach to competition and predation. American Naturalist 116:362-393. Tilman, D. 1994. Competition and Biodiversity in Spatially Structured Habitats. Ecology 75:216. Torstensson, N. T. L., L. N. Lundgren, and J. Stenstrom. 1989. Influence of climatic and edaphic factors on persistence of glyphosate and 2,4-D in forest soils. Ecotoxicology and Environmental Safety 18:230-239. Turkmgton, R., E. John, S. Watson, and P. Seccombe-Hett. 2002. The effects of fertilization and herbivory on the herbaceous vegetation of the boreal forest in north-western Canada: a 10-year study. Journal of Ecology 90:325-337. Valiente-Banuet, A., A. V. Rumebe, M. Verdu, and R. M. Callaway. 2006. Modern Quaternary plant lineages promote diversity through facilitation of ancient Tertiary lineages. Proceedings of the National Academy of Sciences 103:16812-16817. Wardle, D. A. 2001. Experimental demonstration that plant diversity reduces invasibility evidence of a biological mechanism or a consequence of sampling effect? Oikos 95:161170.  61  4. Consequences of regional environmental heterogeneity on local diversity and ecosystem function3 Introduction Despite years of research, there is still no consensus on the relative importance of processes that create and maintain diversity, nor the role of diversity in maintaining ecosystem function (Chesson 2000, Schwartz et al. 2000, Raffaelli et al. 2005, Cardinale et al. 2006). Communityresource dynamics models (CRDMs) predict that species diversity is maintained by each species being an optimal competitor in a particular environment or set of environmental conditions (Tilman 1980, Chesson 2000). An extension of this prediction is that, in a heterogeneous environment, a large number of species will utilize resources more completely by maximizing resource use in each of those species' optimal niche spaces (Tilman et al. 1997a, Cardinale et al. 2006). However, this prediction of species complementarity is often based on the assumption that competitive dynamics reflect equilibrium conditions in a closed community (Tilman 1980, Tilman et al. 1997a), or that local competitive dynamics are not influenced by dispersal (Gross and Cardinale 2007). Other CRDMs that avoid this assumption describe general rules for stable coexistence but do not generate explicit predictions for ecological metrics such as alpha diversity or local resource-use (Chesson 2000). Indeed, one of the fundamental challenges of CRDMs is that of scale, and in particular linking the scales of environmental heterogeneity to scales of species occurrence, dispersal and competition (Snyder and Chesson 2003). Two areas of field research have challenged the general applicability of CDRMs to ecological communities. First, recent meta-analyses of ecosystem function experiments have shown that a diverse community is usually no better than the 'best' species in that community, with diverse communities generally decreasing function at the outset of experiments, but eventually equalling the dominant competitor after about five years (Cardinale et al. 2007). In other words, the prediction that a large number of species is beneficial because they exploit more available niches is controversial. However, many of these studies are subject to criticisms of scale, as they tend to test the effects of diversity and function in small, homogenous plots, ' A version of this chapter will be submitted for publication.  62  whereas much of the theory relating diversity to resource use depends on heterogeneous environments (Tilman et al. 1997a). Although niche complementarity is possible at these small scales it is not guaranteed (Levin 1976, Tilman 1980), and CRDMs predict much stronger diversity - resource use relationships at larger scales where environmental variation is more pronounced (Davies et al. 2005). The second area of field research uses seed additions to test invasibility of communities. Despite the importance of resource competition in plant communities (Lambers et al. 2004, Harpole and Tilman 2006), studies on species invasions and persistence have indicated that the relative number of seeds of competing plants is often equally or more limiting to plant establishment than competition for resources (Tilman 1997, Seabloom et al. 2003, MacDougall and Turkington 2006). Dispersal of seeds into a patch is likewise an important determinant of alpha diversity (Levine 2000b), even when the species dispersing into the community appears to be an inferior competitor (MacDougall and Turkington 2006). Models of closed communities, which ignore dispersal from outside sources, are clearly not appropriate for these systems. Together, these discrepancies between experimentation and CDRM predictions suggest that mechanisms that increase local diversity, other than local niche partitioning, need to be considered. The results from seed addition experiments suggest that one important determinant of local diversity is dispersal. A number of metacommunity models have been developed to address the effects of dispersal on local diversity when dispersal rates are high enough to affect competitive interactions (Amarasekare 2003, Leibold et al. 2004, Leibold and Miller 2004b). These models often use a lottery or patch occupancy approach, and assume that competitive outcomes are a product of both propagule supply and competitive ability (Amarasekare and Nisbet 2001, Mouquet and Loreau 2002, Mouquet and Loreau 2003, Snyder and Chesson 2003). Such approaches have generated predictions about species coexistence and the conditions necessary for the persistence of source and sink populations. However, because each individual fills a single space in lottery models, this approach may not be suitable for predicting the effect of resource use on population dynamics and vice versa. For example, Mouquet and Loreau (2003) used a lottery model in which established individuals depleted resources, but unused resources did not affect the number of new individuals establishing. Thus, inefficient resource use by poor competitors could not be offset by new individuals establishing and creating larger population sizes, which is an important component of CRDMs (Tilman 1980).  63  A second important mechanism for maintaining diversity is the degree to which species in a community are competitively similar, or alternatively, show strong niche differentiation (Gravel et al. 2006, Adler et al. 2007). The two extreme cases of niche differentiation are described by neutral models, in which niche differences do not affect species distributions, and equilibrium CDRMs in which only niche differences determine distributions (Tilman 1980, Chesson 2000, Hubbell 2001). Between these two extremes, niche differentiation may have variable affects on diversity and ecosystem function. For example, functionally similar species are more likely to persist in sink populations, potentially causing negative diversity-function relationship (Mouquet and Loreau 2003). However, broad niche overlap can also decrease the unexploited niche space between neighbouring species, thus creating a positive diversityfunction relationship (Tilman 1997). Here I introduce a metacommunity CRDM that is designed to model an interaction between competitive dynamics and dispersal. The environment is modeled as locally homogenous (within patches) but heterogeneous across patches, an approach commonly used in metacommunity models (Amarasekare et al. 2004, Leibold et al. 2004). I develop a weighted competition model in which a species' local success is dependent on both the number of seeds that it has in the patch and its access to resources. Patches within the metacommunity are connected through seed dispersal. Competitive interactions are placed on a continuum so that the metacommunity can range from strongly niche-based to neutral. I then use the model to test the degree to which dispersal, niche overlap among species and the regional species pool can augment local diversity, and thereby change resource use and productivity. This approach is novel in that it explicitly incorporates resource dynamics into a metacommunity framework, and allows both niche differentiation and the affect of dispersal on competitive outcomes to vary among communities.  The model I use a model in which species' demographic attributes are identical (e.g. adult mortality rates, seed production rates, average competitive ability, etc.), and vary only those model parameters that affect local diversity and ecosystem function (dispersal, niche width, and regional diversity). This allows the model to become a neutral model when species have complete niche overlap, as species are then competitively equal and functionally redundant. The model is written as a plant  64  model with seed dispersal, but can represent any metacomrnunity of sessile organisms that have dispersing propagules and are resource limited. Dispersal is global, and thus closer sites are not more likely to receive more seeds than distant sites. The model progresses with mature plants first competing for a limiting resource (R), which they use to produce seeds (q). A species' ability to compete in a patch depends on its relative abundance and its species-specific environmental response to that patch (E, described below). Seeds either stay in the locality of the adult plant or disperse to a global pool that is then distributed evenly among patches. For a given patch, the number of surviving offspring (j) of each species depends on its relative seed abundance and its potential reproductive success (g), which is also a function of its environmental response to that patch (E) and the resources left unused by adult plants (R'). The model can be written in matrix form by tracking the change in the population (X) of species i in patchy, which depends on adult mortality (m), and the competitive success of all species locally and in other patches: AX, = /;.(^.(X,E^),g,(i?;,£,))- m X,  4.1a  The vector q denotes values for all species at site j . The vector for resource levels (R) is for all sites, and the matrix of populations (X) and environmental responses (E) are for all species at all sites. The change in resource availability is in turn given as S  S  M , = - £ y j r , , * , ) -£c/;^,(x,Ej),g,(i?;,£,))+ 4.1b  rf*,-£ty...)-2y#<~>) i=l  i=l  Here, h is the amount of resource used by mature plants for maintenance growth and reproduction, and c is the per capita resource cost of producing a new plant. The supply function, p, describes the amount of the resource that is replenished from external sources. I set/? to be proportional to the difference in resources from the resource availability in a site at equilibrium  65  with no species present (Roj) so that p(Rj) = C(Roj-Rj), and C is bounded between 0 and 1 (Tilman 1980). Each time step begins with resource uptake by mature plants:  XifaEif, htj - Min  RjX^j^x^  4.2  Equation 4.2 allows all species to uptake their maximum level of resources unless the sum total for all species exceeds the available resources at the site, in which case a species receives an amount proportional to its relative abundance and its per capita resource uptake rate (the lower portion of Eq. 4.2). The per capita resource uptake rate is dependent on the functional form of resource uptake, a, and the species-specific environmental response (Eg). I use a Monod function as the resource uptake function, so that a = bR/(R+K), with b acting as a scalar that sets the maximum uptake per individual. Environmental responses are modeled as a Gaussian function, with Etj = e  optt  ~'  T  , where Eopt is the niche optima of species i and E is the niche value of sitey.  The environmental response of each species therefore depends on two aspects of the niche: the distance of a species' optimal niche from the site's environment (Eoptj-Ej), and the niche width of species (a; Fig. 4.1 A). Seed production (q) is a linear function of resource uptake, with the proportion of seeds that are globally dispersed to all L patches equal to X. The number of seeds of species i in patchy is then:  1 — qv = (1 - A)uhtj H— V Auhy — (1 - X)uhtj + Auht  4.3  where u is the mean number of seeds per unit resource used. In many communities, mature plants produce many more viable seeds than could possibly survive to reproductive maturity (Seabloom et al. 2003, Tilman 2004). In such communities that are not seed limited, there are two processes that determine the number of  66  successful offsprmg of a species. First, each seedling must compete with other seedlings for the remaining resources, and therefore each species realizes only a portion of its potential reproductive success. Second, species have different potential reproductive success (g). The potential reproductive success (g) of each species represents its maximum number of offspring in the absence of interspecific competition. It is a function of a species' environmental response (E) and its functional response to available resources. I use the integral of the Monod function to model the functional response to resource availability, as this integral incorporates the total response curve as the resource is depleted. Potential reproductive success is thus:  Where q> is a scalar for the number of offspring per unit area of the reproductive success function. In the absence of interspecific competition, the integral would range from 0 toi?^., with Rj being the amount of resource left unused by mature plants. Competition among species is also important, however, which limits the amount of resource available for each species. If a species' access to resources is proportionate to its number of seeds, then the amount of resource that each species will have available for reproduction will be Rj qi/£/ly> with the denominator summed across all species. Species-specific establishment (J) is therefore:  fu=<pEif JlJ  Y ,J  f I  —?—dR R+k  4.5  This model for reproduction makes two important assumptions. First, species have different maximum recruitment abilities, even when no competing species are present. Thus, in the absence of dispersal, this model works as an R* model with species having different equilibrium populations and reducing resources to different levels if their environmental responses vary (Tilman 2004). Second, the seeds of one species will inhibit those of other species through resource competition, even if the first species has a lower reproductive growth potential. This effect is analogous to the real-world case where numerous seedlings of a given species may  67  survive for short periods of time (using resources and occupying space), but die before reaching a reproductive size. Equation 4.3 can be incorporated into 4.5 and simplified to determine seedling establishment in a given patch:  fv=<pE9  V —-—kl°z £([(i-^)VM])  k +-  + &log(A:)  4.6  £aa-^,+^])  Simulations I used simulations to determine the effects of model parameters on local and regional diversity, productivity and resource use. I present those sets of conditions that illustrate the effect of a particular parameter on species richness, productivity and resource use. Varying other parameters (i.e. k, RQ, a, etc.) does not alter the qualitative results presented here.  68  A  a=0.3 C£  a=0.5 o=1 Niche distance (Eoptj-Ej)  Rank  Figure 4.1. Species R* values and their relationship to local abundance. A) R* depends on a species environmental response {EoptrEj) and niche width (c). B) The resulting rank-abundance curve within a site depends on dispersal among sites (X). C) The relative abundances of species are negatively correlated with their R* values (C; curves have the same X values as in B). Values for B & C are averages over 1000 patches within simulations with the above parameter values and o=0.1. In C, R* values were averaged over all species of a given rank with tied ranks (two species with the same number of individuals) assigned randomly to upper and lower ranks. Curves for B & C are fitted power curves with R2 values ranging from 0.88 to 0.99. R* values were solved numerically with R0=200, £=0.3, m=0.3, S=50, k=2, c=0.5, b=0.5, (p =1.  69  Simulations were initiated with all species equally present in the global seed pool and the number of patches in the community was held constant at 2000. As with any such simulation model, the long-term equilibrium solution is for all species to go extinct as there is no mechanism for reinvasion from the regional species pool. I initially ran simulations of communities for 5000 time steps to determine what transient effects were present and how long they persisted. Most parameter conditions produced relatively short transient effects that disappeared within 20 time steps. However, parameter combinations that caused a collapse in regional diversity took variable amounts of time for stochastic effects to cause the collapse, and also to reach a semi-stable state afterwards. Nonetheless, even in these communities, trends in species loss emerged within 400 time steps. Based on these results, I ran all simulations for 500 time steps unless otherwise stated. For all simulations I designed the model so that it became more neutral as niche overlap increased (i.e. as a2 became large). This condition required that species' niche optima (Eoptt) and sites' niche values (£)) were uniformly distributed, and also that the niche axis was circular (wrapped so that 0 = 1) so that all species' niche optima were equidistant on average from all site niche values. Stochasticity was introduced into the model by treating mortality (m) as a binomially-distributed random variable and establishment (/) as a poison-distributed random variable.  Results The model predicted varying relationships between local diversity, productivity and resource use depending on the underlying mechanism that increased diversity. I define each of these terms as follows. Local diversity refers to species richness within a patch, and also to a number of commonly accepted diversity indices (Shannon Weiner index, Fisher's alpha, 1-Simpson's index), as these indices are all positively correlated with species richness in the model output (smallest rs = 0.98). Resource use refers to the proportion of resources used over the course of a time step (1 -RJ/RQJ). Productivity is defined as the function h, which describes seed production by mature plants in a patch. However, because productivity and resource use were strongly correlated (r = 0.99), only resource use is shown in figures. Each figure holds all variables constant except those indicated in order to show their effects on model output.  70  a=0.5  CO CO CD  c SZ  o  a=0.1  5 i  a=0.05  CO 0  o  © Q.  W ro 1 o o  a=0.01 0.2  0.4  0.6  0.E  Dispersal (A) B 100 CO CO CD  c  a: CO CD O CD Q.  10  CO  0.4  0.6  Dispersal (A)  xx  «K  tfft  X: a A: a D a 0: a  3  4  5  6  = 0.5 = 0.1 = 0.05 = 0.01  7  8  Local Species Richness  Figure 4.2. The relationship between dispersal, local species richness and resource use. A) The relationship between dispersal and local species richness at different levels of niche overlap, and B) the resulting relationship between local species richness (black dashed line), regional richness (solid black line) and resource use (red line; a = 0.1 for all lines). C) When considered across different levels of dispersal and niche overlap, the relationship between local richness and resource use is funnel-shaped. Other parameter values were the same as in Fig. 4.1, error bars represent standard deviations. Competitive outcomes within patches depended on the R* values of species present and the degree of dispersal among patches (Fig. 4.1). Changing niche-width values altered the 71  competitive relationships (R*) among species (Fig. 4.1 A), whereas altering dispersal did not change the R* but did change the competitive outcomes (Fig. 4.1B,C). For low to intermediate levels of dispersal, increased dispersal caused more evenly-distributed abundances within communities by flattening the slope of the relationship between local dominance and resourceuse efficiency (Fig. 4.1B,C). However, dispersal did not alter the form of the relationship between competitive ability and species' abundances except when regional richness crashed (not shown). Thus, dispersal decreased the relative abundance of the dominant competitor, but did not change its rank.  Niche width (a)  CD O 0.7 3  O CO CD  0.6-  CCL 0.5  Local Species Richness  Figure 4.3. The effect of niche width on species richness and resource use. A) The affect of species' niche widths (a) on local and regional species richness and resource use. Error bars show standard deviations, with model parameters the same as in Fig. 4.1 (X=0.7). B) When considered across communities with different levels of neutrality, resource use is more variable at low levels of local species richness, but otherwise shows no trend. Increasing dispersal among patches caused an increase in local diversity at low levels, followed by a crash that mirrored a crash in regional diversity (Fig 4.2A,B). Both the increase in diversity at lower levels of dispersal and the subsequent crash at higher levels depended on the  72  degree of niche overlap among species. More neutral communities, marked by high niche overlap, had higher local diversity at low dispersal, but also crashed at lower levels of dispersal (Fig. 4.2A). Resource use and productivity were negatively related to dispersal, regardless of whether increasing dispersal increased or decreased local diversity (Fig. 4.2B,C). When considered across communities with differing degrees of dispersal, resource use was more variable at low levels of local diversity, and showed an average level of resource use at high local diversity (Fig. 4.2C, comparing within a given a). Niche width and dispersal had similar effects on local and regional species richness, but opposite effects on resource use (Figs. 4.2B, 4.3A). More neutral communities invariably used more resources as 'gaps' between niches disappeared in these communities (Fig. 4.1 A, 4.3A). However, the elimination of these gaps caused species to become redundant, which in turn caused stochastic processes to dominate population regulation. When considered across communities, increasing niche width can greatly increase local diversity, but higher local diversity does not always result in an increase in resource use (Fig. 4.3B). Increasing the species pool (regional richness at the outset of the simulation) also increased local diversity, although local diversity saturated at relatively small species pools (Fig. 4.4A). The increase in local diversity created consistent and positive diversity-resource use relationships at relatively low levels of niche overlap (a). However, this positive diversityresource use relationship typically saturated at a local species richness of 2 or 3 species (Fig. 4.4B). At high levels of neutrality (high a, not shown), regional diversity has no impact on resource use, as local and regional diversity collapsed to a single species.  73  tn  CO CD  c  a=0.01 CO  'o CD CL  co "co o o 0.01  20  40  60  80  100  Regional Species Pool  A ^ ^ ^  «%  ^A  AA  CD CO  =3  8 1—  O CO 0  A. a = 0.1 D CT = 0.05 0: CT = 0.01  1  2  3  4  5  Local Species Richness  Figure 4.4. The affect of species pool (initial regional species richness) on local species richness, and the resulting effect of local richness on unused resources A) Relationship between local richness and species pool size. B) The relationship between local species richness and regional species richness. Error bars show standard deviations, with model parameters the same as in Fig. 4.1 (X=0.7). Fitted curves are logarithmic, with R2 values ranging from 0.88 to 0.93.  Discussion By incorporating community-resource dynamics into a metacommunity framework, the model predicts two important trends that are novel to our understanding of plant communities. First, increases in local diversity beyond that supported by local environmental heterogeneity can increase or decrease average ecosystem functioning, depending on the underlying mechanism. Mechanisms that increase diversity by saturating regional niches (increasing niche width, increasing regional diversity) also increase local resource use, whereas increasing diversity without altering regional niche use (dispersal) causes a decrease in resource use. Niche-based models do not allow for such local dynamics except in transitory states (Tilman 2004), and 74  previous metacommunity models with similar assumptions predicted that, for a given regional diversity, increasing diversity beyond the environmental heterogeneity of a patch decreased function (Mouquet and Loreau 2003). The current model predicts instead that a number of factors can determine local levels of diversity (Levin 1976, Shmida and Wilson 1985), and that the effects of this local diversity on resource uptake and productivity can only be predicted when these underlying factors are known. Second, despite the common metacommunity prediction that low dispersal rates increase diversity but that high rates reduce diversity (Mouquet and Loreau 2003, Leibold et al. 2004, Leibold and Miller 2004b), the current model predicts that the effect of dispersal on diversity depends critically on the degree of neutrality (versus niche differentiation) in the community. In communities with high levels of immigration and emigration, the interplay between dispersal and niche dynamics may be much more important than previously suggested by both niche models (Tilman et al. 1997a) and metacommunity models (Hubbell 2001, Mouquet and Loreau 2003). Together, these two broad results suggest that the disparate effects of diversity on ecological processes like resource uptake and invasibility may be resolved through understanding the interplay of dispersal and niche dynamics in natural communities. Dispersal, niches and diversity The prediction that diversity peaks at intermediate level of dispersal is common to many metacommunity models (reviewed in Leibold et al. 2004). For example, in a metacommunity model with similar assumptions about regional environmental heterogeneity, Mouquet and Loreau (2002, 2003) showed that local diversity peaks when approximately 30% of seeds are globally dispersed. That scenario turns out to be a special case amongst a number of competition models that range from neutral to strongly niche-based (Fig. 4.2A). Indeed, it is the level of niche differentiation that determines if a given level of dispersal will increase diversity through source-sink dynamics, or cause regional diversity to quickly crash because of stochastic fluctuations. The degree to which species coexist through strong niche differentiation versus equalizing processes (sensu Chesson 2000) is currently an important debate in ecology (Adler et al. 2007). The model presented here represents a completely equalized community, as all species have identical distributions of optimal environments and symmetric competition. It is  75  nonetheless apparent that weak niche differentiation can lead to a fast collapse in diversity, especially when dispersal is high (Fig. 4.2A, 4.3A). However, it is unclear how applicable this scenario is to natural settings; is niche differentiation in natural communities great enough for most species to persist in the face of high dispersal? To my knowledge, there have not been empirical studies that directly test this question, although species invasions that involve massive seed addition of agriculturally desired grasses suggest that dispersal can sometimes overcome niche mechanisms (MacDougall and Gilbert, unpublished). However, seed addition experiments generally indicate that adding seeds is insufficient to displace competitive dominants with inferior competitors (Clark et al. 2007), suggesting that the dispersal levels tested experimentally are not sufficient to cause local extinctions in plant communities. A more common observation in natural communities is that dispersal increases local diversity through source-sink populations, which have long been predicted to affect community properties such as species-area relationships and beta-diversity (MacArthur and Wilson 1967, Shmida and Wilson 1985, Pulliam 2000, Amarasekare and Nisbet 2001). The relative importance of sink populations in augmenting local diversity is unknown, as most source-sink studies follow the demography of single species (reviewed in Pulliam 2000). Nonetheless, a meta-analysis has shown that diversity does generally increase with dispersal (Cadotte 2006), and source-sink populations are now considered to be sufficiently important to include in both theoretical and applied models of species distributions (Pulliam 2000, Allouche et al. 2008). Diversity and ecosystem function Although it is generally accepted that numerous mechanisms can alter diversity (Levine 2000b), the effect of diversity on ecosystem functioning remains controversial (Raffaelli et al. 2005, Balvanera et al. 2006, Cardinale et al. 2006, Cardinale et al. 2007). At first glance, the model presented here appears to add further complexity to the debate. If a simple model of regional heterogeneity causes opposing relationships between local diversity and ecosystem function, how can these relationships be predicted or interpreted? The large amount of research on diversity and ecosystem function seems to point to a similar question. Meta-analyses show that although some average trends exist, communities show both positive and negative relationships between diversity and ecosystem function when compared to the performance of the dominant competitor (for example, 12% positive versus 25% negative in Cardinale et al. 2007). In  76  addition, the average effect of diverse plots is no different than the dominant competitor after successional dynamics have unfolded (Cardinale et al. 2007). It may well be that differences among communities in their diversity-ecosystem function relationships will further our understanding of ecological dynamics more than their mean effects. The model presented here predicts three distinct trends that can help to frame the debate on diversity and ecosystem function. The first trend predicted by the model is that an increase in local diversity should correlate with a decrease in ecosystem function when this diversity is comprised of sink populations formed by increased immigration (Fig. 4.2A). This prediction is analogous to the effect of high dispersal levels in population genetics: immigration can dilute selection in local populations (Spieth 1974), resulting in a higher genetic load and persistence of less fit genotypes in a patch (Levene 1953, Wiens 1976). The presence of sink populations in spatially structured communities has long been noted theoretically and empirically (Levin 1976, Wiens 1976, Shmida and Wilson 1985, Mouquet and Loreau 2003), although the relevance to resource use has not been explored in a recruitment model that depends on resources. The model presented here does depend on resources for population regulation, and has a positive feedback between resource availability and recruitment of new individuals (Eq. 4.5), which is common to all CRDMs. Given this feedback, why do communities with more unused resources not then recruit more individuals, thus eliminating the effects of the sink populations? The answer is that poorly-adapted species still compete for limiting resources, whether or not they are able to utilize these resources fully. This assumption is essential for developing an R* model that allows dispersal to influence competitive outcomes, and is robust to changes in the way resources are divided among species so long as this division is weighted by the number of seeds. If sink populations increase local diversity and thereby decrease resource use and productivity, we would expect to see these trends in experiments that manipulate species diversity. Although biodiversity-ecosystem function experiments do test this possibility, these experiments alter the 'regional pool' of seeds and manipulate both the probability of including the best-adapted competitor and the probability of including sink populations. These different mechanisms should produce opposite diversity-functioning relationships (Fig. 4.2 vs. 4.4), and it is not clear if the net result should be positive or negative. Testing the role of dispersal in creating diversity, and its successive role in ecosystem function, would be a more direct way of testing the effects of sink populations.  77  The second trend predicted by the model is that species' relative abundances should reflect their competitive abilities in communities where regional richness does not collapse (Fig. 4.1B,C). Plant community studies support this prediction by showing a negative correlation between species' abundances and their R*s (Lambers et al. 2004, Harpole and Tilman 2006). In addition, studies that compare the role of dominant and rare species in plant communities have often shown that dominant species have the most influence on resource uptake (Smith and Knapp 2003, Emery and Gross 2007) although this is not always the case in diversity-ecosystem function studies (Cardinale et al. 2007). Further studies that contrast the role of species dominance and diversity will shed light on the importance of such competitive hierarchies. The third trend predicted by the model is that the effect of dispersal on diversity is mediated by the niche overlap among species. For example, the negative effects of sink populations that is apparent when niche overlap is low virtually disappears when niche overlap is high (Fig. 4.2C, a = 0.01 vs. a = 0.5). This prediction can be tested in studies designed to disentangle the importance of niche processes from neutral processes (Adler et al. 2007), but such experiments are currently in their infancy (Harpole and Suding 2007). Environmental heterogeneity Although the model presented here is simple, it nonetheless represents important dynamics that may drive species diversity in natural systems. A number of ecosystems have distinct environmental 'patches' that are semi-permanent structures on the landscape (Harrison 1997, Davies et al. 2005). Indeed, the use of tools such as digital elevation maps to predict environments and ecological communities relies on the permanence of these patches (Gilbert and Lechowicz 2004a, Harrison et al. 2006a). The temporally static structure of the environment in this model is a good approximation of such ecosystems. Despite the importance of spatially-structured environmental heterogeneity, the dynamics described by the model, and in particular the importance of sink populations, result in part from the underlying assumption that niche complementarity is not possible within a patch. In reality, complementarity may be present at small scales and even in homogenous environments, as can occur when species that partition resources spatially or temporally are present in a locality (McKane et al. 2002, Cardinale et al. 2007), or when more than one resource is limiting (Levin 1976, Tilman 1980). Niche complementarity within patches would partially alter model  78  predictions by producing positive correlations between local diversity and resource use (Tilman et al. 1997a). However, the presence of niche complementarity, as with facilitation or other processes not included in this model, would not negate the role of sink populations but rather determine their relative importance to overall ecosystem function. Niche complementarity can be evaluated with this model by considering the effect of regional diversity on average ecosystem function. The positive effect of regional richness on resource use that is observed in the model is a result of species occupying distinct niches in the region (Fig. 4.4). A previous metacommunity model also showed a decrease in productivity with a decrease in regional diversity (Mouquet and Loreau 2003), however, this was due to the vacating of 'niche spaces' that occurred when species went regionally extinct. The model presented here shows a different trend: adding species into a variable region only increases ecosystem function when a small number of species are present (Fig. 4.4B). The flat trend at higher levels of diversity is likely caused by a trade-off between greater saturation of the regional environmental niche space (causing a positive relationship), the increased number of sink populations (causing a negative relationship) and the increased degree of niche overlap that occurs with greater species packing, which makes each species more redundant on average. Overall, even when niche complementarity is built into a metapopulation model, the positive effects of diversity on ecosystem function plateau at relatively low levels of local diversity (2-4 species; Fig. 4.4), a trend also seen in numerous studies (Schwartz et al. 2000, Cardinale et al. 2006). Summary As with all models, the validity of trends predicted here must be tested with empirical studies. A number of such tests are possible. For example, invasion studies have tested the role of increasing propagule pressure on invasion and local diversity (Levine 2000b, MacDougall and Turkington 2006), and could use similar approaches to test the effects of this increased diversity on resource use and productivity. In addition, the degree to which species coexist because of strong niche differentiation (versus competitive similarity) could be tested in a metacommunity framework by altering the propagule pressure of select species to determine the stability of the competitive dominants (Levine and Murrell 2003, Adler et al. 2007). Undertaking such tests will  79  quantify the importance of spatial dynamics in natural systems and elucidate the causes and consequences of diversity. Understanding how dispersal- and niche-based mechanisms are responsible for the generation of local diversity may well resolve debates on the importance of diversity in ecosystem function at different scales, and related issues such as the role of diversity in preventing invasive species from establishing (Raffaelli et al. 2005, Fridley et al. 2007). Many past and current studies on diversity effects manipulate the 'regional' species pool and simultaneously limit habitat heterogeneity, effectively removing much of the potential variation of species responses to the environment that underlie theories of diversity and resource use. This approach has been criticized for its lack of applicability to conservation (Raffaelli et al. 2005), and should also be scrutinized for its inconsistency with theory. Devising methods to test the relationship between diversity and ecosystem properties at a scale that captures environmental heterogeneity is a priority for both conservation and theoretical ecology.  80  Literature cited Adler, P. B., J. HilleRisLambers, and J. M. Levine. 2007. A niche for neutrality. Ecology Letters 10:95-104. Allouche, O., O. Steinitz, D. Rotem, A. Rosenfeld, and R. Kadmon. 2008. Incorporating distance constraints into species distribution models. Journal of Applied Ecology 45:599-609. Amarasekare, P. 2003. Competitive coexistence in spatially structured environments: a synthesis. Ecology Letters 6:1109-1122. Amarasekare, P., M. F. Hoopes, N. Mouquet, and M. Holyoak. 2004. Mechanisms of coexistence in competitive metacommunities. American Naturalist 164:310-326. Amarasekare, P., and R. Nisbet. 2001. Spatial heterogeneity, source-sink dynamics, and the local coexistence of competing species. American Naturalist 158:572-584. Balvanera, P., A. B. Pfisterer, N. Buchmann, J.-S. He, T. Nakashizuka, D. Raffaelli, and B. Schmid. 2006. Quantifying the evidence for biodiversity effects on ecosystem functioning and services. Ecology Letters 9:1146-1156. Cadotte, M. W. 2006. Dispersal and species diversity: A meta-analysis. American Naturalist 167:913-924. Cardinale, B. J., D. S. Srivastava, J. E. Duffy, J. P. Wright, A. L. Downing, M. Sankaran, and C. Jouseau. 2006. Effects of biodiversity on the functioning of trophic groups and ecosystems. Nature 443:989-992. Cardinale, B. J., J. P. Wright, M. W. Cadotte, I. T. Carroll, A. Hector, D. S. Srivastava, M. Loreau, and J. J. Weis. 2007. Impacts of plant diversity on biomass production increase through time because of species complementarity. Proceedings of the National Academy of Sciences 104:18123-18128. Chesson, P. 2000. Mechanisms of maintenance of species diversity. Annual Review of Ecology and Systematics 31:343-366. Clark, C. J., J. R. Poulsen, D. J. Levey, and C. W. Osenberg. 2007. Are plant populations seed limited? A critique and meta-analysis of seed addition experiments. American Naturalist 170:128-142.  81  Davies, K. F., P. Chesson, S. Harrison, B. D. Inouye, B. A. Melbourne, and K. J. Rice. 2005. Spatial heterogeneity explains the scale dependence of the native-exotic diversity relationship. Ecology 86:1602-1610. Emery, S. M., and K. L. Gross. 2007. Dominant species identity, not community evenness, regulates invasion in experimental grassland plant communities. Ecology 88:954-964. Fridley, J. D., J. J. Stachowicz, S. Naeem, D. F. Sax, E. W. Seabloom, M. D. Smith, T. J. Stohlgren, D. Tilman, and B. Von Holle. 2007. The invasion paradox: Reconciling pattern and process in species invasions. Ecology 88:3-17. Gilbert, B., and M. J. Lechowicz. 2004. Neutrality, niches, and dispersal in a temperate forest understory. PNAS 101:7651-7656. Gravel, D., C. D. Canham, M. Beaudet, and C. Messier. 2006. Reconciling niche and neutrality: the continuum hypothesis. Ecology Letters 9:399-409. Gross, K., and B. Cardinale. 2007. Does species richness drive community production or vice versa? Reconciling historical and contemporary paradigms in competitive communities. American Naturalist 170:207-220. Harpole, W. S., and K. N. Suding. 2007. Frequency-dependence stabilizes competitive interactions among four annual plants. Ecology Letters 10:1164-1169. Harpole, W. S., and D. Tilman. 2006. Non-neutral patterns of species abundance in grassland communities. Ecology Letters 9:15-23. Harrison, S. 1997. How natural habitat patchiness affects the distribution of diversity in Californian serpentine chaparral. Ecology 78:1898-1906. Harrison, S., J. B. Grace, K. F. Davies, H. D. Safford, and J. H. Viers. 2006. Invasion in a diversity hotspot: Exotic cover and native richness in the Californian serpentine flora. Ecology 87:695-703. Hubbell, S. P. 2001. The Unified Neutral Theory of Biodiversity and Biogeography. Princeton University Press, Princeton. Lambers, J. H. R., W. S. Harpole, D. Tilman, J. Knops, and P. B. Reich. 2004. Mechanisms responsible for the positive diversity-productivity relationship in Minnesota grasslands. Ecology Letters 7:661-668. Leibold, M. A., M. Holyoak, N. Mouquet, P. Amarasekare, J. M. Chase, M. F. Hoopes, R. D. Holt, J. B. Shurin, R. Law, D. Tilman, M. Loreau, and A. Gonzalez. 2004. The  82  metacommunity concept: a framework for multi-scale community ecology. Ecology Letters 7:601-613. Leibold, M. A., and T. E. Miller. 2004. From metapopulations to metacommunities. Pages 133150 in I. Hanski and O. Gaggiotti, editors. Ecology, Genetics, and Evolution of Metapopulations. Elsevier Academic Press., Amsterdam. Levene, H. 1953. Genetic equilibrium when more than one ecological niche is available. American Naturalist 87:331-333. Levin, S. A. 1976. Population dynamic-models in heterogeneous environments. Annual Review of Ecology and Systematics 7:287-310. Levine, J. M. 2000. Species diversity and biological invasions: Relating local process to community pattern. Science 288:852-854. Levine, J. M., and D. J. Murrell. 2003. The community-level consequences of seed dispersal patterns. Annual Review of Ecology Evolution and Systematics 34:549-574. MacArthur, R. H., and E. O. Wilson. 1967. The Theory of Island Biogeography. Princeton University Press, Princeton, N.J. MacDougall, A. S., and R. Turkington. 2006. Dispersal, competition, and shifting patterns of diversity in a degraded oak savanna. Ecology 87:1831-1843. McKane, R. B., L. C. Johnson, G. R. Shaver, K. J. Nadelhoffer, E. B. Rastetter, B. Fry, A. E. Giblin, K. Kielland, B. L. Kwiatkowski, J. A. Laundre, and G. Murray. 2002. Resourcebased niches provide a basis for plant species diversity and dominance in arctic tundra. Nature 415:68-71. Mouquet, N., and M. Loreau. 2002. Coexistence in metacommunities: the regional similarity hypothesis. American Naturalist 159:420-426. Mouquet, N., and M. Loreau. 2003. Community patterns in source-sink metacommunities. American Naturalist 162:544-557. Pulliam, H. R. 2000. On the relationship between niche and distribution. Ecology Letters 3:349361. Raffaelli, D., B. J. Cardinale, A. L. Downing, K. A. M. Engelhardt, J. L. Ruesink, M. Solan, and D. S. Srivastava. 2005. Reinventing the wheel in ecology research? Response. Science 307:1875-1876.  83  Schwartz, M. W., C. A. Brigham, J. D. Hoeksema, K. G. Lyons, M. H. Mills, and P. J. van Mantgem. 2000. Linking biodiversity to ecosystem function: implications for conservation ecology. Oecologia 122:297-305. Seabloom, E. W., W. S. Harpole, O. J. Reichman, and D. Tilman. 2003. Invasion, competitive dominance, and resource use by exotic and native California grassland species. Proceedings of the National Academy of Sciences 100:13384-13389. Shmida, A., andM. V. Wilson. 1985. Biological determinants of species-diversity. Journal of Biogeography 12:1-20. Smith, M. D., and A. K. Knapp. 2003. Dominant species maintain ecosystem function with nonrandom species loss. Ecology Letters 6:509-517. Snyder, R. E., and P. Chesson. 2003. Local dispersal can facilitate coexistence in the presence of permanent spatial heterogeneity. Ecology Letters 6:301-309. Spieth, P. T. 1974. Gene flow and genetic differentiation. Genetics 78:961-965. Tilman, D. 1980. Resources - a graphical-mechanistic approach to competition and predation. American Naturalist 116:362-393. Tilman, D. 1997. Community invasibility, recruitment limitation, and grassland biodiversity. Ecology 78:81-92. Tilman, D. 2004. Niche tradeoffs, neutrality, and community structure: A stochastic theory of resource competition, invasion, and community assembly. Proceedings of the National Academy of Sciences of the United States of America 101:10854-10861. Tilman, D., C. L. Lehman, and K. T. Thomson. 1997. Plant diversity and ecosystem productivity: Theoretical considerations. Proceedings of the National Academy of Sciences of the United States of America 94:1857-1861. Wiens, J. A. 1976. Population responses to patchy environments. Annual Review of Ecology and Systematics 7:81-120.  84  5. Conclusions The study of species coexistence attempts to explain and predict the mechanisms that promote species diversity, the population, community and ecosystem- consequences of this diversity, and the processes that can lead to the maintenance or breakdown of biodiversity. In this thesis I addressed each of these goals by testing potential mechanisms that underlie patterns of diversity (Chapter 2 and 3), how these mechanisms affect the invasibility and resource-use within a community (Chapter 3 and 4), and when these processes can lead to a collapse of diversity (Chapter 4). In particular, I used a combination of observational and experimental field research and developed a simulation model to undertake three studies that address these goals. In particular, my thesis research: i.  documented broad patterns of diversity in a previously understudied community, and tested for consistency between these observed patterns and hypothesized mechanisms of coexistence  ii.  experimentally tested the predictions of several coexistence mechanisms in a forest understory community where both competitive and facilitative interactions were expected  iii.  developed a theoretical model that explored the importance of habitat heterogeneity and dispersal on local and regional coexistence, and also their effects on ecosystem function at both scales  Together, this combination of observational, experimental and theoretical approaches produced some important insights into the mechanisms of coexistence both locally and at larger scales. These insights can be separated into those that are unique to the communities studied and those that have broader implications. Unique insights into the study communities: Bromeliad-dwelling mosquito larvae In Chapter 2,1 used the co-occurrence patterns of mosquito larvae to discriminate between neutral and niche models of coexistence, both of which had previously been proposed for mosquito communities (Bradshaw and Holzapfel 1983, Yee et al. 2004). As with many invertebrate and tropical communities, these communities have not been studied extensively, and thus this type of broad observational study is a necessary first step, first to document basic  85  knowledge about how they function (SCBD 2006), and second as a prerequisite to experimental studies. Observational data at the individual organism level and the regional level showed that only one mechanism, spatial niche partitioning within bromeliads, was consistent with mosquito distribution patterns. In addition, the study provided an example of how a local process, spatial partitioning within bromeliads, can scale up to the meso-scale distributions of species, an outcome that is predicted by many ecological coexistence mechanisms. Despite these clear results, the conclusions of the study should be verified experimentally. In particular, pairwise competition studies should be performed to verify that that negative interactions are in fact stronger when species occupy similar spatial niches within bromeliads. Likewise, competition studies coupled with oviposition studies could determine how mosquitoes partition bromeliads of different sizes. Beyond experimentally verifying the conclusions stated in Chapter 2, future research could address a number of ecological questions that are raised by this study and that may be important for this community. For example, the high likelihood of drought in smaller bromeliads, and strong predation effects in larger bromeliads, suggest that there may be specific attributes that allow the two mosquito species that occupy these bromeliad sizes to persist. If these attributes could be identified, then they could be included in future tests of container insects and size preferences. More generally, these species may offer insight into the hypothesis that there is a trade-off between competitive ability and stress tolerance, and also demonstrate how a within-trophic level mechanism of coexistence (spatial niche partitioning) functions within the context of multiple trophic levels. Unique insights into the study communities: Boreal forest understory plants The study of boreal forest understory plants (Chapter 3) presented a new experimental framework for testing the effects of extant species on limiting diversity, and more specifically, provided important insights for species coexistence in the Kluane area. Different coexistence mechanisms give distinct predictions about the role of individual species in preventing species from invading a community, and thereby limiting the alpha diversity of that community. Many of these predictions rely on different assumptions about the role of diversity (many species) versus the role of specific, competitively dominant species in limiting invaders.  86  Testing for diversity and species-specific effects in communities has traditionally been difficult, as creating communities through random selection of species can lead to species mixes never seen in nature, and thus not applicable to natural settings (Wardle 2001). For example, a single, competitively dominant species is unlikely to appear in a 'low diversity' community through a random draw, but very likely to occur in a 'high diversity' community by a random draw. However, natural communities are not drawn randomly, and a competitively dominant species may naturally occur in both low and high diversity communities on a landscape. The exclusion of a competitively dominant species from many 'low diversity' treatments may therefore create an experimental bias in the results, and risks artificially confounding a species richness effect with the effect of one species (Wardle 2001, Fridley et al. 2007). On the other extreme, natural communities are difficult to use in a traditional approach of testing the effects of individual species, as different species invariably contribute different amounts of biomass to the extant community (McGill et al. 2007). The design developed in Chapter 3 used a naturallyoccurring community but also decoupled the effects of biomass from that of species identity, thus avoiding many of the problems faced in natural and artificial communities. In addition, linking the dominance of specific species to predictions from differing coexistence mechanisms lead to a strong test of the importance of these mechanisms in preventing the invasion of new species. Results from the study did offer important insights for the establishment of new species, and also highlighted some shortcomings of my design that could be overcome in subsequent tests. In particular, while my design identified competitive mechanisms that limit species diversity, it was only able to reject one facilitation hypothesis. Removals of larger amounts of biomass would be needed to fully explore facilitation effects (below). The results reported in Chapter 3 showed that the role of species in limiting or promoting new seedlings, and thus controlling diversity, depends critically on the degree of disturbance. When all species were completely removed, new seedlings showed a near-complete mortality, but removing only small amounts (7%) of the community biomass produced either no mortality effect or lowered seedling mortality. This non-linear effect of removals suggests that fundamentally different processes may operate during early succession (no species present) compared to late succession (small disturbances). For small disturbances to the community, the strong effect of the herbaceous dominant species (Epilobium angustifolium), compared to both the woody dominant species  87  (Arctostaphylos uva-ursi) and numerous low-abundance species, suggests that there is a competitive hierarchy among extant species. Similarly, the consistent patterns of seedling establishment among treatments demonstrated a strong effect of invader species identity on success in this community, but no species-specific preference for a specific removal. These results suggest that spatial or resource niche partitioning does not underlie the patterns of diversity in this ecosystem, and likewise that temporal niche partitioning, if present, does not limit diversity over the short term (3 years). Despite these results, the patterns of establishment were inconsistent with a competition-colonization trade-off, which is one of the most cited examples of a competitive hierarchy. The results instead suggest that future studies need to focus on the role of dominant species in limiting invasion, and also the mechanisms that restrict the herbaceous dominant, in particular, from eventually limiting all other species. From a more applied viewpoint, this study suggests that monitoring programs based on a few dominant species, rather than all species, would be better suited to determining the integrity of an ecosystem in terms of its invasibility. The effect of large disturbances was not explained by any competition-based coexistence mechanism, but was instead consistent with predictions of density-dependent facilitation (Callaway and Walker 1997). This finding suggests that the processes that limit or promote diversity following large disturbances are fundamentally different than those processes that occur once a community becomes more dense. The design used in Chapter 3 was sufficient for rejecting a diversity-invasibility hypothesis by showing that removing 70-80% of the species present did not cause net facilitative effects. Nonetheless, it could not discriminate between general facilitation (i.e. any neighbour is a good neighbour) and species-specific effects of one or both dominant species. A future approach to this question could involve a range of removals of the dominant species alone, as this approach could identify the level of disturbance at which interspecific dynamics becomes positive. However, this approach may be inconsistent with the processes that are important in secondary succession; the patterns of natural colonization in cleared plots, and observation of recently burned areas in the region, indicate that the composition of early-successional communities following large disturbance of an area is very different from the composition of the community studied in Chapter 3. Because these communities are distinct, they would require a general test, such as that developed in Chapter 3, along with a specific focus on density-level effects.  88  Unique insights into the study communities: Computer simulations of an idealized community Chapter 4, which explored a theoretical model using computer simulations, showed how dispersal amongst patches with different environmental conditions could create source-sink communities which were in some cases stable and in other cases caused a regional collapse of diversity. The degree of stability depended on both the degree of niche overlap among species and the level of dispersal in the community, with increases in either of these factors increasing sink population sizes and destabilizing the community. However, while increasing both niche overlap and dispersal increased local diversity, they had opposite effects on ecosystem function, with increased dispersal decreasing function. In addition, the presence of sink communities in general shows that the mechanisms that promote coexistence regionally (in this case habitat heterogeneity) may not always be acting locally. This mismatch between local and regional effects of diversity deserve more attention, as almost all experimental studies of diversity test local processes (as in Chapter 2), but most concerns about species loss are at the regional level (Srivastava and Vellend 2005). While Chapter 4 is a theoretical model, and therefore cannot offer insights into any specific ecosystem, it does highlight issues that should be addressed in studies of species coexistence. First, it shows that dispersal processes and the degree of 'neutrality' of a community can have both opposing and synergistic effects on diversity and its consequences. Although dispersal has been actively researched for some time, recent reviews have proposed that the degree of neutrality in communities also deserves further study (Leibold and McPeek 2006, Adler et al. 2007), a view supported by the model results. For example, the strong influence of seed abundance in more neutral communities shown in Chapter 4 suggests that contemporary concerns about invasive species may be partially addressed by assessing their niche overlap with extant species. Second, in terms of experimental methodologies, the scale-depedence of the relationship between biodiversity and environmental heterogeneity (no relationship at small scales, strong relationship at regional scales) suggests that coexistence studies should be performed across scales to test for the robustness of relationships (Shea and Chesson 2002, Davies et al. 2005). While such an approach is fairly straightforward in observational studies (Chapter 2), it can be logistically difficult in experimental studies (Chapter 3).  89  General insights: Integrating the results from three approaches The three distinct approaches to studying coexistence that were used in this thesis draw attention to their strengths and shortcomings, and also demonstrate how each approach can inform the other. One of the clearest examples of these differences lies in the scale at which inferences can be made. For example, Chapter 2 demonstrates the strength of observational studies for exploring understudied communities, and also shows how considering multiple communities at different scales (within and among communities) can lead to strong conclusions about species co-occurrence. Although the patterns are robust in rejecting a neutral hypothesis and also in their consistency with spatial niche partitioning, they nonetheless lead to several questions that would best be tested experimentally: are Wyeomyia melanopus better adapted to surviving desiccation events, and if so, how? Do Culex rejector have predator avoidance techniques that other species do not share? In contrast, Chapter 3, which tests local-scale dynamics, demonstrates that even though the community studied had a typical rank-abundance curve (McGill et al. 2007), the numerous mechanisms of coexistence that could describe such a pattern are all insufficient to do so. These results ultimately demand further exploration of the effects of specific species, but also beg the question of how the processes operating within stands would scale up to coexistence patterns among different communities and across larger time scales. Would the same experiment, performed over several years, show similar patterns? Do species distributions in adjacent communities, such as nearby shrubby grasslands, suggest that source-sink dynamics might be responsible for the patterns of diversity shown here? In other words, the experimental results lead to further questions, some of which require observational studies. This interdependence among methods is probably best illustrated by the theoretical work presented in Chapter 4. Although the mechanistic model of source-sink dynamics is plausible, is there any evidence of either of the underlying assumptions and the outcomes in natural systems? To answer this question, both observational studies of local and meso-scale patterns, and experimental studies of the local impacts of dispersal on competition dynamics, are essential. As a result, I dedicated a significant portion of Chapter 4 to explaining how the proposed model could be tested. Just as the three approaches used provide insights into their strengths and interdependencies, they also offer insights into the range of mechanisms that can underlie species diversity and the impacts of these mechanisms on ecosystem function and resistance to invasion. For example, the mosquito community in Costa Rica showed patterns of niche partitioning that  90  are consistent with classic theories of species coexistence (MacArthur and Levins 1967a, Chesson 2000). These results suggest that, all else being equal, regions with a diversity of mosquito species will be better at limiting the invasion and using available resources than depauperate communities. In contrast, the competitive hierarchy found in the boreal forest understory indicates that disturbances that affect dominant species are more important to the establishment of both native and exotic species than small disturbances that reduce diversity. Although this study did not show a similar trend with resource use, it suggests that monitoring specific species, rather than the full range of native species, would be best for recording disturbances that are most likely to affect community invasibility. However, the results from the model in Chapter 4 provide a caveat for this result: results from the local-scale study in the boreal understory may not persist at the regional scale. Thus, while the importance of the dominant species is shown for aspen stands (Chapter 3), when considered in the larger context of adjacent habitat types, the species that are rare in aspen stands may play different roles in limiting invasion. These contrasting results are far from surprising given the large variation in species' roles that is commonly found even among communities of similar organisms (Lawton 1999, Knapp et al. 2004). Rather, they emphasize the need for further studies in communities that are poorly understood, such as those considered in this thesis. It is only with this base level of knowledge that decisions on protecting species and regenerating degraded systems can be successful.  91  Literature cited Adler, P. B., J. HilleRisLambers, and J. M. Levine. 2007. A niche for neutrality. Ecology Letters 10:95-104. Bradshaw, W. E., and C. M. Holzapfel. 1983. Predator-mediated, non-equilibrium coexistence of tree-hole mosquitoes in southeastern North America. Oecologia 57:239-256. Callaway, R. M., and L. R. Walker. 1997. Competition and facilitation: a synthetic approach to interactions in plant communities. Ecology 78:1958-1965. Chesson, P. 2000. Mechanisms of maintenance of species diversity. Annual Review of Ecology and Systematics 31:343-366. Davies, K. F., P. Chesson, S. Harrison, B. D. Inouye, B. A. Melbourne, and K. J. Rice. 2005. Spatial heterogeneity explains the scale dependence of the native-exotic diversity relationship. Ecology 86:1602-1610. Fridley, J. D., J. J. Stachowicz, S. Naeem, D. F. Sax, E. W. Seabloom, M. D. Smith, T. J. Stohlgren, D. Tilman, and B. Von Holle. 2007. The invasion paradox: Reconciling pattern and process in species invasions. Ecology 88:3-17. Knapp, A. K., M. D. Smith, S. L. Collins, N. Zambatis, M. Peel, S. Emery, J. Wojdak, M. C. Horner-Devine, H. Biggs, J. Kruger, and S. J. Andelman. 2004. Generality in ecology: testing North American grassland rules in South African savannas. Frontiers in Ecology and the Environment 2:483-491. Lawton, J. H. 1999. Are there general laws in ecology? Oikos 84:177-192. Leibold, M. A., and M. A. McPeek. 2006. Coexistence of the niche and neutral perspectives in community ecology. Ecology 87:1399-1410. MacArthur, R., and R. Levins. 1967. The limiting similarity, convergence, and divergence of coexisting species. American Naturalist 101:377-385. McGill, B. J., R. S. Etienne, J. S. Gray, D. Alonso, M. J. Anderson, H. K. Benecha, M. Dornelas, B. J. Enquist, J. L. Green, F. He, A. H. Hurlbert, A. E. Magurran, P. A. Marquet, B. A. Maurer, A. Ostling, C. U. Soykan, K. I. Ugland, and E. P. White. 2007. Species abundance distributions: moving beyond single prediction theories to integration within an ecological framework. Ecology Letters 10:995-1015.  92  SCBD. 2006. Global biodiversity outlook 2. Secretariat of the Convention on Biological Diversity, Montreal, Canada. Shea, K., and P. Chesson. 2002. Community ecology theory as aframeworkfor biological invasions. Trends in Ecology and Evolution 17:170-176. Srivastava, D. S., and M. Vellend. 2005. Biodiversity-ecosystem function research: Is it relevant to conservation? Annual Review of Ecology Evolution and Systematics 36:267-294. Wardle, D. A. 2001. Experimental demonstration that plant diversity reduces invasibility evidence of a biological mechanism or a consequence of sampling effect? Oikos 95:161170. Yee, D. A., B. Kesavaraju, and S. A. Juliano. 2004. Larval feeding behaviour of three cooccurring species of container mosquitoes. Journal of Vector Ecology 29:315-322.  93  Appendix A: Competitive interactions amongst mosquito larvae Introduction Several studies have suggested that competitive interactions among mosquito larvae may determine the abundance and distribution of mosquito species (Livdahl and Willey 1991, Edgerly et al. 1998, Schneider et al. 2000, Lounibos et al. 2003). Nonetheless, there are also several examples of other mechanisms that may be important to mosquito distributions, such as sensitivity to predators (Fincke et al. 1997), drought tolerance (Sunahara and Mogi. 2002a) and chemical inhibition (Sunahara and Mogi. 2002b). In addition, priority effects, which may be due to early-arriving individuals being larger, can sometimes influence the outcome of competitive interactions (Sunahara and Mogi. 2002b, Lounibos et al. 2003). This study was initially designed to test all pairwise interactions among mosquito larvae that occur in bromeliads in the Guanacaste region (see Chapter 2). However, due to logistical constraints of collecting sufficient numbers and sizes of larvae of all species, the experiment was reduced to examine the interactions of a focal species, Anopheles nievai, with two other species, Wyeomyia melanopus and Culex rejector (all taxonomic names as in Chapter 2). In addition, two size classes of each species was used to determine the degree to which size, and thus size-based priority effects, alter interactions. The sampling results from Chapter 2 indicated that, in order to test competitive hierarchies amongst species, all pairwise interactions must be considered in a range of bromeliad sizes. This study did not fill that role, but rather was designed to determine if competitive interactions amongst species are indeed occurring, and also if further research in this area would be fruitful. I therefore present this study and its results as an experimental exploration of mosquito interactions.  Methods Six second instar A. neivai were raised alone or with six small (second instar) or large (third instar) individuals from the species A. neivai, W. melanopus or C. rejector. There were therefore 7 treatments in total: a control with no competitors and 6 treatments with competitors of one of  94  three species which were in one of two size classes. Each treatment was housed within a 50 ml transparent centrifuge tube that was partly submerged in the well of a bromeliad leaf. A full block (7 tubes) of the experiment was housed in each large bromeliad, for a total of four bromeliads (blocks). Each of these bromeliads had all detritus and water removed prior to the experiment, and was refilled with spring water. The tubes had a small (approx. 2 cm diameter) hole drilled in the base, which was covered with 102 wm Nitex mesh (Sefar America Inc., Depew, NY) to allow free passage of water from the surrounding bromeliad while preventing predators and other insects from entering. Prior to adding the mosquito larvae, each tube was filled with 0.4 g of fine detritus (1mm < diameter < 5mm) that had been collected from other bromeliads and oven dried for 4 hours at approximately 65° C to kill any insect eggs. Mesh was also used to cover the top of the tube, allowing entry by rainwater but preventing insect oviposition. Initial conditions in the experimental tubes were set so as to mimic natural levels of mosquito density (12 individuals per 50 ml) and resource availability (0.4 g detritus) based on previous survey results (Dr. D. Srivastava, unpub.). Mosquitoes used in the experiment were collected from a separate bromeliad-sampling project, with individuals of the appropriate species and size randomized prior to placement in a treatment. Per capita growth and survival rates were measured as the total change in length or survival between a given census and the previous census, with the time to the first census treated as a 'transplant shock' and not considered further. The experiment was monitored every 9 days for 27 days by removing all detritus and mosquito pupae and larvae from a tube, measuring growth and survival, and replacing the mosquitoes and detritus. Pupae counts represent a minimum number of pupae as some individuals appear to have emerged and died between censuses, or to have escaped through small tears in the mesh. In the large A. neivai versus young A. neivai treatment cohorts were separated in the analysis by assuming that no individual decreased in size between censuses. In the young A. neivai versus young A. neivai treatment, growth and survival rates were averaged over all individuals.  Analysis Mosquito growth data from the competition experiment were first analysed with a blocked ANOVA, with all treatments considered. Following a statistically significant outcome, the data were reanalysed with the control treatment removed using a mixed model to test for species- and  95  size-specific effects on A neivai growth at the 2n and 3r census. The total number of competitors in a treatment, which varied among experimental units due to competitor mortality and differential emergence times, was included as a covariate in the mixed model analysis; including this covariate did not change the patterns observed. The mixed model included block and number of competitors as random effects, and census time as a repeated measure. The covariance structures were modelled with variance components (block, number of competitors) and compound symmetry (census time), which were determined by comparing the Akaike's information criterion for many models. A Bonferroni correction was used for these and all other post-hoc comparisons. Survival data from the competition experiment were binary (alive vs. dead), and therefore analysed with a general linear model (GLM) with a complementary log-log link and a binomial error distribution following Allison (1995), with the number of competitors included as a covariate. Survival analyses were first done with the control (no competitors) included. For both growth and survival analyses each tube was considered as an independent observation.  Results Competitor species and size affected growth of A. neivai, with growth being reduced by 63% in the presence of large A. neivai compared to large C. rejector (Fig. A.l). Growth results showed a size by species interaction when the effect of the number of competitors was removed as a covariate (mixed model F2,i4= 35.48, p<0.0001). Specifically, small A neivai had a smaller impact than larger conspecifics, but small C. rejector had a greater impact than large C. rejector. W. melanopus showed no significant difference between instar sizes. When averaged within species, A. neivai and W. melanopus did not have significantly different effects, but both had significantly larger effects than C. rejector (both posthoc p<0.01).  96  (0 •D  > c  4O .«B  f*  E c  I  1 Large A. nei.  Small A. nei  Small C. rej.  Large W. met.  Small W. mel.  Figure A.l: The effect of competitors on the growth (mm) of 2nd instar Anopheles neivai. A) Mean growth (+ standard error) after competing against six large (3rd instar) or small (2nd instar) individuals of the same species, Wyeomyia melanopus, or Culex rejector. Values are least-squared means with number of competitors removed statistically. Different letters are statistically different (Bonferroni corrected a=0.05). The numbers of competitors in each treatment differed, mainly due to differences in pupation times; treatments with large competitors had significantly fewer individuals by the third census than treatments with small competitors (ANOVA Fi>i5=6.4, p<0.03). No A. neivai were observed pupating during the experiment, and approximately 38% of individuals from other species were observed pupating. Consequently, by the third census, treatments with just A. neivai had more individuals than those with W. melanopus and C. rejector (ANOVA F2,i5=5.4, p<0.02). When the number of competitors was not controlled statistically, the effects of A. neivai increased relative to the other species, but overall trends did not differ. Survival did not differ significantly among treatments (GLM %2= 10.83, 6 d.f, p>0.05).  Discussion The competition study showed three trends that were consistent with the predictions of Chapter 2. First, the effect of competitors on the growth rate of A. neivai varied with competitor species and size (Fig. A.l), contrary to the hypothesis of equal competition among species that is necessary for neutral drift (Hubbell 2001). Second, as with the sampling results, I found no support for consistent priority effects among species. Third, if species' occurrence patterns  97  reflect their competitive abilities in a given environment, we would expect C. rejector to have the smallest impact on A. neivai in the experimental tubes, as they simulate a small bromeliad size (50 ml) that A. neivai and W. melanopus occupy much more than C. rejector (Fig. 2.4). This was indeed the case. Despite these findings, which appear consistent with the predictions of Chapter 2, this competition experiment showed a number of trends that suggest that mechanisms other than simple spatial niche partitioning may be important in this community. For example, despite competitive effects on larvae growth rates that may scale up to reproductive success (Livdahl and Willey 1991), species interactions did not differentially influenced, neivai survival. This result suggests that oviposition decisions among adults likely drive larvae distributional patterns, as has been observed in other insects (e.g. Srivastava et al. 2005), a hypothesis that could be tested by future sampling. The survival results also indicate that, even within a given habitat, there is some degree of overlap among species' competitive abilities, especially in the case of A. neivai and W. melanopus. This overlap is also apparent in species' selection of bromeliad size (Fig. 2.4), which indicates that each species has a range of optimal bromeliad sizes, but that there is some overlap even among species with similar spatial niches (C rejector and W. melanopus above 40 ml volume). Thus, despite the important role of spatial niche partitioning in this community (Figs. 2.2 - 2.4), equalizing mechanisms also appeared to be important. This type of niche overlap and intraspecific variability is common in most populations (e.g. Clark 2003), and can lead to stability in communities by slowing competitive exclusion in sink populations (Chesson 2000, Gravel et al. 2006). A second result that should be investigated further is that large mosquitoes sometimes have smaller competitive impacts than smaller mosquitoes (Fig. A.l). Coupled with different maturation rates of the three species studied, this result suggests that life-history traits amongst mosquito species may influence competitive interactions through time. Such differences in lifehistory traits are often associated with some trade-off between competition and colonization ability (Tilman 1994), although other trade-off axes may also be important (Chase and Leibold 2003). This result, coupled with differences in species' exposure to drought and predators (Chapter 2), suggests that more complex interactions may interact with the spatial niches of mosquito larvae to determine their abundances and distributions.  98  Although the experiment presented here is not meant as a definitive test of the sampling results found in Chapter 2, it nonetheless proved useful in confirming some of the expectations of that study and also provided new hypotheses about the roles of species traits in mediating mosquito coexistence.  99  Literature cited Allison, P.D. 1995. Survival analysis using SAS: a practical guide. SAS institute Inc. Cary, NC, USA. Chase, J.M. and M.A. Leibold. 2003. Ecological Niches: Linking Classical and Contemporary Approaches. University of Chicago Press, Chicago. Chesson, P. 2000. Mechanisms of maintenance of species diversity. Annual Review of Ecology and Systematics 31:343-366. Clark, J.S. 2003. Uncertainty in ecological inference and forecasting. Ecology 84: 1349-1350. Edgerly, J.S., M. McFarland, P. Morgan and T. Livdahl. 1998. A seasonal shift in egg-laying behaviour in response to cues of future competition in a treehole mosquito. Journal of Animal Ecology 607: 805-818. Fincke, O.M., S.P. Yanoviak and R.D. Hanschu. 1997. Predation by odonates depresses mosquito abundance in water-filled tree holes in Panama. Oecologia 112: 244-253. Gravel, D., C. D. Canham, M. Beaudet, and C. Messier. 2006. Reconciling niche and neutrality: The continuum hypothesis. Ecology Letters 9: 399-409. Hubbell, S.P. 2001. The Unified Neutral Theory of Biodiversity and Biogeography. Princeton University Press, Princeton, N.J. Livdahl, T.P. and W.S. Willey. 1991. Prospects for an invasion: competition between Aedes albopictus and native Aedes triseriatus. Science 253: 189-191. Lounibos, L.P., G.F. O'Meara, N. Nishimura and R.L. Escher. 2003. Interactions with native mosquito larvae regulate the production of Aedes albopictus from bromeliads in Florida. Ecological Entomology 28: 551-558. Schneider, P., W. Takken and P.J. McCall. 2000. Interspecific competition between sibling species larvae of Anopheles arabiensis and An. gambiae. Medical and Veterinary Entomology 14: 165-170. Srivastava, D.S., M.C. Melnychuk and J.T. Ngai. 2005. Landscape variation in the larval density of a bromeliad-dwelling zygopteran, Mecistogaster modesta (Odonata: Pseudostigmatidae). International Journal of Odonatology 8: 67-79. Sunahara, T. and M. Mogi. 2002a. Priority effects of bamboo-stump mosquito larvae: influences of water exchange and leaf litter input. Ecological Entomology 27: 346-354.  100  Sunahara T. and M. Mogi. 2002b. Variability of intra- and interspecific competitions of bamboo stump mosquito larvae over small and large spatial scales. Oikos 97: 87-96. Tilman, D. 1994. Competition and biodiversity in spatially structured habitats. Ecology 75: 2-16.  101  Appendix B: Principal facilitator expectations from a LotkaVolterra model In order to determine the relationship between a principal facilitator and relative abundance, we evaluated invasibility criteria of the Lotka-Volterra competition model. The Lotka-Volterra competition model for species i is given as: = r.«. 1 dt ''  >« K,  Bl  Where r is the intrinsic population growth rate, n is the population size and K is the carrying capacity. The alpha terms determine the strength of interspecific interactions and their directions (facilitative or competitive). For the following invasion criteria, we assumed that one species facilitates one or more species (a.; <0), but that its abundance is depressed by those species (a;. >0). This general case is commonly reported in the facilitation literature (Bertness and Callaway 1994, Callaway and Walker 1997, Bruno et al. 2003, Callaway 2007), and tends to occur when a principal species facilitates other species by reducing stress through mechanisms such as providing shelter or increasing surface soil moisture through hydraulic lift. We also assumed that all species have a positive carrying capacity and growth rate in the absence of other species. In the two species model, two non-trivial equilibria exist in which one species is absent, n;=0 and nj=Kj, and similarly with the subscripts switched. Each species must be able to invade the community when initially absent if both are to coexist in the community. This invasion occurs when the leading eigenvalue of the Jacobian matrix is positive, indicating that the population of the initially absent species increases. The eigenvalues for the two species model when one species is absent are:  a„Kt  A  1  v  V  *.  J  K = ~r2  J  The second eigenvalue is always negative by assumption. The first eigenvalue can be solved for cases in which it is positive, which gives:  K, > aijKj  B2  If species j facilitates species i, a<0 and species i will always invade. However, if species j competes with species i, then the carrying capacity of i must be greater than j by a factor of the competition coefficient. In the extreme case where the facilitated species (j) competes solely for the same limiting resource as the facilitator (i), the carrying capacity of i must be greater than j if they have similar conversion of resources to biomass. The two-species scenario can be easily extended to multiple species so long as the other species form a stable equilibrium. In this case, the invasion criteria is (MacArthur and Levins 1967b):  Kt > X apt *  B3  Where n* is the equilibrium population size in the absence of species i. From this equation, we can see two inclusive conditions that promote the establishment of species i. If species i is negatively impacted by all other species (a;. >0), both low niche overlap with other species and a high carrying capacity allow it to invade. This result is identical to results given for the coexistence of competing species (MacArthur and Levins 1967b). However, if species i also facilitates other species, then the greater the equilibrium population size of species i, the more it will increase the facilitated species can invade. In the study area considered, only the two dominant species have high abundances and are consistently present in the community (Fig. 1), suggesting that if a principal facilitator is present, it is likely to be one of these species.  Literature cited Bertness, M. D., and R. Callaway. 1994. Positive interactions in communities. Trends in Ecology and Evolution 9:191-193. Bruno, J. F., J. J. Stachowicz, and M. D. Bertness. 2003. Inclusion of facilitation into ecological theory. Trends in Ecology and Evolution 18:119-125. Callaway, R. M. 2007. Positive Interactions and Interdependence in Plant Communities. Springer, Dordrecht, The Netherlands. Callaway, R. M., and L. R. Walker. 1997. Competition and facilitation: a synthetic approach to interactions in plant communities. Ecology 78:1958-1965. MacArthur, R., and R. Levins. 1967. The limiting similarity, convergence, and divergence of coexisting species. American Naturalist 101:377-385.  104  Appendix C: Species in extant community and transplanted into plots Over 32 species occurred in the surveyed plots (Table CI) used in Chapter 3. All species names are from the published flora of the area (Cody 2000).  Table CI: Extant species in sampled aspen stands in Kluane area, Yukon. Species  Average  Allometric  Frequencyft  Relationship  Percent of Rz  Community  d.f.  Form  Biomassf , Achillea millefolium  0.445  0.83  0.87  60  Power  Anemone multifield  0.157  0.33  0.90  24  Power  Antennaria sp. *  >0.001  0.01  Arctostaphylos uva-ursi  83.1  1  0.97  18  Power  Arnica cordifolia  0.009  0.05  0.94  8  Power  Aster sibiricus * *  0.290  0.59  0.94  20  Power  Astragalus alpinus  0.033  0.05  0.92  18  Power  Calamagrostis  0.430  0.71  0.94  38  Power  Carex sp. *  0.001  0.04  Castilleja unalaschcensis  0.002  0.01  0.90  18  Power  Delphinium glaucum  0.283  0.43  0.93  34  Power  Epilobium angustifolium  9.79  1  0.97  38  Power  Festuca altaica  2.06  0.95  0.97  30  Power  Galium boreale  0.098  0.2  0.91  23  Power  Gentiana sp*  0.001  0.01  Grass morphospecies 1 f  1.58  0.80  0.94  38  Power  Grass morphospecies 2*  0.001  0.01  Hedysarum alpinum  0.079  0.12  0.96  13  Power  purpurascens  Linnaea borealis  0.013  0.07  0.95  11  Linear  Mertensia paniculata  0.221  0.41  0.96  28  Power  Moehringia lateriflora*  0.003  0.01  Orthilia secunda  0.338  0.17  0.92  26  Power  Oxytropis deflexa *  0.001  0.01  Penstemon procerus  0.104  0.32  0.98  16  Power  Polemonium pulcherrimum  0.004  0.05  0.97  19  Linear  Populus tremuloides  0.036  0.16  0.85  18  Power  Potentilla diversifolia  0.005  0.01  0.92  13  Power  Pyrola asarifolia  0.027  0.03  0.95  25  Power  Rosa acicularis  0.065  0.08  0.90  18  Power  Salixsp. *  0.002  0.01  Senecio lugens  0.021  0.09  0.95  15  Power  Solidago multiradiata  0.795  0.48  0.96  14  Power  f Calculated as percent biomass per plot averaged across all plots ff Proportion of plots in which the species was present * species that were only encountered once had biomass estimated as the mean biomass of similarly sized individuals outside of the plot ** Name has changed to Eurybia sibirica (L.) G. L. Nesom J At least two broad-leaved grasses that were morphologically similar and flowered infrequently. The two indentified were Festuca richardsonii and multiple subspecies of Elymus trachycaulus. The species used as transplants were selected from native and exotic species that occur in the plots or in the locality (Table C2). Except for T. officinale, all exotic species were grown from seed provided by Richardson Seed Company (Burnaby, Canada) and the native grass seeds were provided by Dr. Manivalde Vaartnou (Richmond, Canada). All other species were grown from seed collected locally.  106  Table C2: Species used as transplants in Chapter 3. Species  Native / Exotic  Type  Present in plots, in stands or locallyf  Agrostis scabra  Native  Grass  Locally  Anemone multifida  Native  Forb  Plots  Elymus trachycaulus  Native  Grass  Plots  Festuca saximontana  Native  Grass  Locally  Phleum alpinum  Native  Grass  Locally  Hedysarum mackenzii  Native  Forb (Legume)  Locally  Medicago sativa  Exotic  Forb (Legume)  Locally*  Poa compressa  Exotic  Grass  Locally  Poa glauca  Native  Grass  Locally  Poa palustris  Native  Grass  Locally  Poa pratensis  Exotic**  Grass  Locally  Exotic  Forb  Stands  ssp. subsecundus  Taraxacum officinale  f In plots refers to plants found in the initial census of the 75 plots. In stands refers to those seen in the same aspen stands, but not found in plots. Those found locally were determined from Cody (2000) beforehand to ensure that they can survive in the area. * Occurrence in Yukon is mainly restricted to areas around Whitehorse, approximately 180 km to the southeast. ** a native subspecies of P. pratensis does occur in the area, but the species used in our experiment was an agricultural subspecies that is considered exotic (Cody 2000).  107  Literature cited Cody, W. J. 2000. Flora of the Yukon Territory. Second Edition. National Research Press, National Research Council of Canada, Ottawa.  108  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0067972/manifest

Comment

Related Items