Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Trophic flows across ecosystem boundaries : an examination of the strength and consequences of linkages… Marczak, Laurie Beth 2007

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-ubc_2007-267530.pdf [ 7.8MB ]
Metadata
JSON: 831-1.0302171.json
JSON-LD: 831-1.0302171-ld.json
RDF/XML (Pretty): 831-1.0302171-rdf.xml
RDF/JSON: 831-1.0302171-rdf.json
Turtle: 831-1.0302171-turtle.txt
N-Triples: 831-1.0302171-rdf-ntriples.txt
Original Record: 831-1.0302171-source.json
Full Text
831-1.0302171-fulltext.txt
Citation
831-1.0302171.ris

Full Text

T R O P H I C F L O W S A C R O S S E C O S Y S T E M B O U N D A R I E S : A N E X A M I N A T I O N O F T H E S T R E N G T H A N D C O N S E Q U E N C E S O F L I N K A G E S B E T W E E N S T R E A M A N D F O R E S T F O O D W E B S by L A U R I E B E T H M A R C Z A K B . S c , University of Victoria, 1995 M . E . S . , York University, 1997 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R OF P H I L O S O P H Y in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Forestry) T H E U N I V E R S I T Y OF B R I T I S H C O L U M B I A February, 2007 © Laurie Beth Marczak, 2007 Abstract While empirical examples have demonstrated the openness of ecosystems to resource flows, we still have a limited ability to make general predictions about the magnitude of, or controls on, the effects of subsidies. I studied the impacts of spatial resource subsidies, and the controls on those effects, using several different consumer species in stream and riparian habitats. I quantified the variation in effect size between habitats, recipient consumers, trophic level of the consumer, and productivity of recipient and donor habitats and tested the magnitude of the effect of a subsidy in a system where theory predicted small magnitude or little impact based on productivity contrasts. I tested the consequences of subsidies on the individual fitness of a riparian spider when delivery of a subsidy is variable in time. I demonstrated that the quantity of a subsidy arriving in a recipient habitat may be altered by consumers (waterstriders) feeding at the interface between habitats and that this control varies with habitat type. Overall, three key themes emerge: 1) the identity of consumers and habitats matters, subsidies do not act the same way in all circumstances, 2) the quantity and timing of a subsidy can interact with the specific ecological requirements or life history o f an organism such that a subsidy received at the wrong time in an organism's development may actually have negative consequences for relative adult fitness, and 3) the species composition of both boundaries and recipient habitats may alter the consequences of subsidies. This thesis supports the developing consensus that understanding the factors that determine the response of consumers in a recipient community to a resource subsidy is essential to the development of landscape level approaches to open systems. n Table of Contents Abstract ii Table of Contents iii List of Tables vii List of Figures viii Acknowledgements x Dedication xi Co-authorship statement xii Chapter 1 General introduction 1 T H E S I S O V E R V I E W 5 W O R K S C I T E D 1 0 Chapter 2. A meta-analysis of the role of trophic position, habitat type and habitat productivity in determining the food web effects of resource subsidies 13 I N T R O D U C T I O N 13 M E T H O D S ". 1 5 Study selection and data criteria 15 Data Analysis 1 6 R E S U L T S 18 Overview 18 Habitat and recipient taxa effects 1 9 Productivity effects 1 9 D I S C U S S I O N 2 0 Effects of subsidies vary with habitat 2 1 Donor productivity does not determine the magnitude of consumer responses 2 1 The relation of subsidies to in-situ resources modifies subsidy effects 2 3 Trophic levels are influenced differentially by subsidies 2 4 Effects of subsidies are highly variable across and within taxonomic and functional groups 2 4 Conclusions , 2 5 A C K N O W L E D G E M E N T S 2 6 W O R K S C I T E D .". 31 Chapter 3. Spiders and subsidies: results from the riparian zone of a coastal temperate rainforest 34 I N T R O D U C T I O N 3 4 M E T H O D S 3 6 Study Site 3 6 Field experiment 3 7 i i i Study Site 3 8 Field experiment 3 9 Statistical Analyses 4 0 R E S U L T S 4 1 Flying aquatic and terrestrial insect abundance 4 1 Spider density 4 2 D I S C U S S I O N 4 3 S U M M A R Y 4 6 Acknowledgements 4 7 W O R K S C I T E D 5 3 Chapter 4. Timing of resource subsidy alters growth and fitness of the spider Tetragnatha versicolor (Araneae.Tetragnathidae) 56 I N T R O D U C T I O N 5 6 M E T H O D S 5 8 Study organism 5 8 Experiment 1. Food limitation effects on growth and mortality risk 6 0 Experiment 2 . Compensatory growth 6 1 Experiment 3 . Timing of subsidy resource and juvenile development 6 2 R E S U L T S 6 4 Experiment 1: Food limitation effects on growth and mortality risk 6 4 Experiment 2 . Compensatory growth 6 5 Experiment 3 . Timing of subsidy resource in juvenile development 6 5 Growth 6 5 Reproductive stage and condition 6 6 D I S C U S S I O N 6 7 Rapid growth and growth compensation 6 7 Subsidy timing and spider fitness A C K N O W L E D G E M E N T S W O R K S C I T E D Chapter 5. Trophic interception: how a boundary-foraging organism influences cross-ecosystem fluxes I N T R O D U C T I O N M E T H O D S Study location Interception in natural habitats Statistical analysis Controlled arena experiment Experimental methods Statistical Analysis Magnitude of waterstrider interception in natural stream systems R E S U L T S Natural habitat interception experiment Controlled arena experiment . Magnitude of waterstrider interception in natural stream systems D I S C U S S I O N Magnitude of waterstrider interception in natural stream systems 98 Effects of waterstriders on trout and trout on waterstriders 99 Interactions between consumers alter terrestrial subsidy pathways into streams 101 A C K N O W L E D G E M E N T S 104 W O R K S C I T E D I l l CHAPTER 6: Untangling the web: complex effects of resource subsidies 115 C O M M E N T S O N F U T U R E R E S E A R C H 118 W O R K S C I T E D 122 Appendix A 124 Appendix B 130 v List of Tables Table 3.1. Comparison of physical and hydrological characteristics of East Creek and Spring Creek. Means ± 1 standard error 48 Table 3.2. Repeated measures A N O V A s for mean of abundance and biomass per trap (n = 8) of flying aquatic and terrestrial insects. Contributions of random effects (stream, stream*treatment) to the model were assessed using a %2 test (df = 1) of the difference in residual log likelihood for the full and reduced model. Significant p-values are highlighted in bold text 47 Table 3.3. List o f families and species of vegetation-dwelling spiders collected at Spring Creek and East Creek in the Malco lm Knapp Research Forest indicating relative percentages (%) of abundance 48 Table 3.4. Results of separate repeated measures A N O V A s for mean abundance of adult spiders in five families each representing greater than 5% of total spider abundance. Contributions of random effects (stream and stream by treatment) to the model were assessed using a %2 test (df = 1) of the difference in residual log likelihood between the full and reduced model. Significant p-values are highlighted in bold text. A l l p values were adjusted using a sequential Bonferroni procedure 49 Table 4.1. Sample sizes, number and mass of prey offered and treatments used to assess growth rate plasticity and mortality risk of rapid growth 72 Table 4.2. Actual direction of relationships between treatment means (all differences significant) compared with the predicted relationships between treatments i f the experiment had continued until all animals reached maturity (using the same food ration from day 64 of the experiment until maturity). Treatments are C = constant, L H = low to high switch, H L = high to low switch 73 Table 5.1. Physical characteristics of three surface habitat types in Spring Creek. Values are means (± standard error), or proportions 105 Table 5.2. Average lengths and live mass ± standard error for organisms used in experiments. 106 v i List of Figures Figure 1.1 Diagrammatic representation of the subsidy pathways explored in this thesis. In Chapter 2 I attempt a broad synthesis of subsidy effects, predominantly across aquatic:terrestrial boundaries. The effect of the emergence of aquatic invertebrates on the riparian orb-weaving spider community is investigated in Chapter 3. Chapter 4 examines the effects of temporal variation in this subsidy for the spider Tetragnatha versicolor. Chapter 5 focuses on the accidental movement of terrestrial invertebrates to stream habitats and the consequences of competition for this resource on populations of surface-dwelling invertebrates, in-stream fish (trout) and benthic invertebrates 9 Figure 2.1. Mean effect size differed between recipient habitat types. Means with 95% bootstrapped confidence intervals are shown. Numbers below lines indicate sample sizes, different letters indicate significant differences between means; the zero line indicates no effect 28 Figure 2.2. Mean effect size differed with the trophic level of the subsidy consumer. Means with 95% bootstrapped confidence intervals are shown. Numbers below lines indicate sample sizes, different letters indicate significant differences between means. The zero line indicates no effect 29 Figure 2.3. A. Mean effect size differences between levels of further classification within the group of studies that reported results for birds. The overall mean for all studies was significant. Although means did not differ significantly from each other, only aerial insectivores show a mean value significantly different from zero. Numbers below lines indicate sample sizes. Means with 95% bootstrapped confidence intervals are shown. B. Mean effect size differences between levels of further classification within the group of studies that reported results for spiders. The overall mean for all studies was significant. Although means did not differ significantly from each other, only horizontal orb-weavers show a mean value significantly different from zero. Numbers below lines indicate sample sizes. Means with 95% bootstrapped confidence intervals are shown. 30 Figure 2.4. A. Relationship between ratio of donor to recipient habitat productivity and the response ratio (In x). R 2 = 0.025, p = 0.69. B. Relationship between the response ratio (In x) and recipient habitat net primary productivity. R 2 = 0.003, p = 0.52. C . Significant positive relationship between the ratio of resource subsidies to equivalent in situ or ambient resources and the response ratio (In x). R 2 = 0.172, p O.0001 31 Figure 3.1. A. Upstream exclosure along 50 m of East Creek, Malco lm Knapp Research Forest (photo credit Y . Zhang) B. Diagrammatic representation of exclosure placement and experimental design on a single stream 50 Figure 3.2. Effect of exclosure treatments on the (A) overall mean biomass and density of aquatic invertebrates (B) monthly mean biomass of aquatic invertebrates at control (filled bars) and exclosure (open bars) reaches (C) overall mean biomass and density of terrestrial invertebrates and, (D) monthly mean biomass of terrestrial invertebrates at control (filled bars) and exclosure (open bars) reaches. Values are least squares means ± 1 SE 51 Figure 3.3. Effect o f aquatic insect exclosure on the abundance of spiders in five families. Stars indicate significant differences between control and exclosed reaches using differences of treatment means from the mixed model. Values are least squares means ± 1 SE 52 Vll Figure 4.1. Pattern of aquatic insect emergence over one year for three streams (East, Mike and Spring creeks) in the Malco lm Knapp Research Forest during 1999. Data are total individuals per m 2 of stream 74 Figure 4.2. Seasonal timing of cohorts of the spider Tetragnatha versicolor during 2005. The width of each bar represents the proportion of individuals caught within a given size class (based on 1 mm bins of the tibia-patella length of the first pair o f walking legs); widths are a proportional fraction of the labeled legend box representing 100%. Numbers above columns represent sample sizes. Lines below months indicate presence through time of adult males and females 75 Figure 4.3. Cumulative number of prey offered in two treatments, one constant rate of food supply (control) and one switching treatment (switch) for Experiment 2 76 Figure 4.4. Cumulative number of prey offered in three treatments, one constant rate of food supply (constant) and two switching treatments: initial high food followed by low food (High:Low) and initial low food followed by high food (Low:High) for Experiment 3. 77 Figure 4.5. Mean total growth rates (percent change in mg/day) (± 1 SE) for spiders in five treatments ranging from low (I) to high (V) levels of food availability. Different letters above a bar indicate significant differences for pairwise contrasts using a Bonferonni correction (Experiment 1) 78 Figure 4.6. Mean mass of spiders (± 1 SE) over a 28 day experiment fed either a constant food supply (control) or a switching treatment consisting of 7 days of low food followed by 7 days of the same amount of food as received by individuals in the control treatment (switch). Overall growth rates do not differ while growth rates between the start of the experiment and day 7 and between day 7 and the end of the experiment, differ between treatments (Experiment 2) 79 Figure 4.7. Second order polynomials fit to the mean mass (± 1 SE) (mg) of spiders in three treatments over time. Treatments were a constant level o f food supply (C) and two switching treatments: initial high food followed by low food (HL) and initial low food followed by high food (LH) . The leftmost vertical line indicates the time of switching for the H L treatment and the rightmost vertical line indicates the time of treatment switching for the L H treatment (Experiment 3) 80 Figure 4.8. (A) Final mean mass (± 1 SE) of spiders (adjusted for initial mass) fed a constant (C), high to low food (HL) or low to high food (LH) diet. Different letters above a bar indicate significant differences using a Bonferroni correction. (B) Mean mass-specific lipid content per individual (± 1 SE) for spiders fed three diets: C , H L or L H (see Figure 5 or methods for description of treatments). Different letters above a bar indicate significant differences in pairwise contrasts following a Bonferroni correction (Experiment 3) 81 Figure 5.1. Mean proportion of prey (± 1 S.E.) intercepted by waterstriders. A. between different stream habitats and B. for different sizes of terrestrial prey. Different letters above the bars indicate significant differences. Sizes of experimental prey represented by: S = small, M = medium, L = large, A - adult, refer to Table 5.2 for details 107 Figure 5.2. A. Mean proportion of experimental terrestrial inputs (± 1 S.E.) in all prey size categories that were consumed by waterstriders for treatments containing waterstriders. B. Mean proportion of experimental terrestrial inputs (± 1 S.E.) in all prey size v i i i categories that were consumed by trout for treatments containing trout. A n * indicates a significant difference between treatments 108 Figure 5.3. Mean proportion of experimental terrestrial inputs (+ 1 S.E.) in all prey size categories that were assigned to the category "benthos" for each of four treatments. Different letters indicate significant differences... 109 Figure 5.4. Magnitudes of the pathways which terrestrial prey subsidies follow into stream ecosystems when (A) waterstriders are present and (B) waterstriders are absent in the three habitat types (riffles - left, connected pools - center, isolated pools - right). Numbers are proportion of experimental prey diverted to each category or consumer (WS = waterstriders, C T = cutthroat trout, benthos = available to benthic detritivores and scavengers). Proportions do not sum to one because of experimental prey that either drifted out of the study reach or successfully escaped to stream banks or sides of experimental arenas. Asterisks (*) mark those pathways not directly quantified in this study; values are estimates. Dashed lines indicate those pathways that link terrestrial prey to a consumer group found in only low densities in the recipient habitat (e.g., there were very low densities of cutthroat trout in riffle habitats) 110 ix Acknowledgements Consigning yourself to the pursuit of a PhD is something that can only be accomplished with the support and aid of those around you. Time spent in the field and lab was always improved by the company o f one o f the many members o f the STaRR lab. Chief amongst them, my friend and supervisor John Richardson who has supported me throughout this process. Members of my supervisory committee (Robb Bennett, Jon Shurin, Scott Hinch) provided advice, expertise and assistance in numerous ways and I am grateful for their generous gift o f time. Ross Thompson served in many ways as an honorary member of my supervisory committee - without his encouragement my own enthusiasm would have flagged. Extra appreciation is due to Trent Hoover who has been both a source of intellectual challenge and a friend with whom I could share the delight of a sunny afternoon spent watching critters. I couldn't get away without thanking fellow students who came to a Galiano Getaway Write-a-thon, read a draft or made tea during those rainy weekends. Without the seemingly bottomless support of Emi ly Gonzales I would long ago have been a crumpled mess - thanks for the blood sugar restoration, running dates and closed door sessions. Playing music each week was essential for my sanity - Brad Fedy, Trevor Lantz, Rebecca Best, Joe Bennett, Peter Arcese, and Sarah Gergel, thanks for playing with me. Rebecca needs a double mention for her endless patience with my equally endless statistics questions and tolerance for my goofy nicknames. A s always, my family has been a constant source of support, I 'm grateful for the love and caring they have showered on me. A n d o f course, Conan Phelan - who got the worst of the late night sessions, the crankiness, the inattention to personal hygiene - and is undoubtedly the only person more pleased than I am that I have finally finished this degree. Thanks for being in my corner. x For my grandfather, Walter Marczak 1911-2005 Co-authorship statement Chapter 2 was co-authored with Dr. Ross Thompson and Dr. John Richardson. I identified, designed and conducted the research. Ross Thompson and J. Richardson assisted with manuscript preparation and revision. Chapters 3 and 4 were co-authored with Dr. John Richardson. I identified, designed and conducted the research. I am solely responsible for all data analyses, J. Richardson assisted with manuscript preparation and revision. Chapter 5 was co-authored with Trent Hoover and Dr. John Richardson. I identified and designed the research program. T. Hoover assisted with experimental design and we conducted the field work for the "natural habitats" experiments together. I am solely responsible for all data analyses and was assisted with manuscript preparation and revision by J. Richardson. x i i Chapter 1 General introduction Ecologists have defined the term community in many different ways - usually containing some element of "a group of populations that occur together" as in an "oak community" or the "spider community", while other definitions focus more on the interactions between associated populations (Ricklefs 1990). In either case, communities are most difficult to define when the interactions among populations extend beyond arbitrary or even ecologically evident spatial boundaries. The migration of birds between temperate and tropical regions, the anadromous life history of many salmon species, the amphibious development of salamanders and the ontogenetic habitat shifts of aquatic insects to terrestrial habitats, tie together otherwise disparate systems. The influence of a population can extend to ecologically distant portions of a community or across ecosystem boundaries through movements of predators, prey and competitors. Such movement may not be purposive and can include flows of material or inadvertent movement of organisms. Although community ecologists have known that flows of energy, materials or organisms from one habitat to another could influence the structure or composition of food webs since the beginnings of the discipline (see for example Summerhayes and Elton 1923) research over the past century has focused on experimental studies of the effects of species interactions or physical factors within single habitats (largely for logistical reasons). Habitat boundaries and fluxes have been of interest within ecosystem ecology and landscape ecology but at a broad scale where the resolution of population dynamics or species interactions within food webs have not been explicitly considered (Huxel et al. 2004). The relevance of resource subsidies to community ecology was championed by Polis, Anderson and Holt (1997) in their influential review which galvanised and generalised thinking 1 around the ideas of spatial subsidies. According to Polis et al. (1997) subsidies are " . . . a donor-controlled resource (prey, detritus, nutrients) from one habitat to a recipient (plant or consumer) from a second habitat which increases population productivity o f the recipient, potentially altering consumer-resource dynamics in the recipient system" (p. 190). Other definitions proffered in the literature are simple restatements of this basic definition. Witman et al. (2004) define subsidies as donor controlled-resources (prey, nutrients, detritus) moving from one habitat to another that increase the productivity of the recipient habitat. While Henschel et al. (2004) state that subsidies are donor-controlled flows o f resources from external systems or habitats that augment population densities of consumers in recipient habitats, potentially allowing them to exert higher effect strength on their alternative in situ resources. The common elements in these definitions are the ideas of donor-controlled resources arriving in a recipient habitat that increase the abundance, distribution or productivity of a consumer. Resources subsidies can take several forms. Species produced in one habitat frequently move and become food elsewhere, such movement may be accidental or a product of life history (e.g. migration, ontogenetic habitat switches). For instance, researchers have begun to quantify the importance of nutrients from salmon carcasses in riparian forest growth (Wipfli et al. 1999), detrital supplements to consumers (Bastow et al. 2002), and subsidies of emergent aquatic prey to terrestrial predators (Sabo and Power 2002). The extent to which movements of material, energy and prey across habitat boundaries alters the distribution or abundance of recipient consumers has now been demonstrated across a wide range of habitats, subsidy types and focal consumers. Forest and stream food webs are widely viewed as energetically coupled, particularly with respect to the contributions of riparian forests to stream ecosystems. A number of recent 2 studies have focused on specifically aquatic-terrestrial linkages, between ocean and terrestrial settings (Polis and Hurd 1995, 1996, Wipfl i et al. 1999) and streams and riparian areas (Jackson and Fisher 1986, Sabo and Power 2002, Sanzone et al. 2003). Inputs of particulate organic matter from riparian forests represent an important energy source of stream production (Wallace et al. 1997) while accidental inputs of terrestrial invertebrates are a major prey category directly available for stream consumers such as fish (Nakano et al. 1999, Wipfl i 2005). Conversely, the importance of cross-boundary subsidies from streams to riparian forest communities is being increasingly emphasized (e.g., Collier et al. 2002, Sabo and Power 2002, Kato et al. 2003, Helfield and Naiman 2006). Whether or not allochthonous materials or prey have a strong impact in a recipient community is thought to depend heavily on its resource status - additional food in a highly productive environment should prove less likely to be a critical resource and may not even be detectable by researchers against measurement error and other variation. Previous authors have speculated that energy flows between habitats w i l l tend to be larger, and thus produce greater effects when they flow from more to less productive systems (Polis and Hurd 1996, Polis et al. 1997, Power et al. 2004). Additionally, it has been suggested that effects of subsidies should be greatest where the contrast in productivity between habitats is large (Polis and Hurd 1996). Physical forces such as wind, water or gravity can alter the quantity of subsidy moving between habitats (Weathers et al. 2001, Cadenasso et al. 2004). The assertion that boundary conditions regulate fluxes between ecosystems remains a largely untested hypothesis (Cadenasso et al. 2004). Understanding the factors that control the permeability of boundaries to subsidies is an essential step towards developing a landscape level approach to open systems. Many terrestrial ecosystems are characterized by intermittent production of abundant resources for consumers; examples include mast seeding, pulses of heavy primary production following rains, or periodic emergence or eruption of insects. Productivity may vary temporally at scales from hours (diel cycles of photosynthesis) to months (summer growing seasons) to years (wet and dry years, large scale climate patterns). Patterns in productivity range from relative constancy (as in the equatorial tropics) to sharp bursts o f production separated by periods with little or no net primary production (e.g. deserts or other extreme environments) (Southwood 1977, Sher et al. 2004). This seasonal variation in productivity can be an important factor in the dynamics of energy flux between habitats (Polis and Strong 1996, Nakano and Murakami 2001, Kato et al. 2003). Determining the effects of pulsed or seasonal resources within food webs is a major challenge for community ecology. Pulsed or strongly seasonal movements o f organisms or materials between ecosystems can contribute to the tendency of plants and animals to grow and establish reserves during good times which are then used to maintain the population or individual during intervening lean periods. Such pulsed production and storage of reserves permitting persistence from good times to bad occurs to some degree in all seasonal habitats from boreal forests to temperate grasslands to temperate streams and lakes (Polis and Strong 1996).This process can significantly influence the dynamics and structure o f communities and food webs. During less productive periods, the use o f reserves (or dependence on extrinsic productivity or spatial subsidies) maintains higher populations of consumers than possible from local productivity alone. Seasonal peaks in trophic exchange between habitats should reinforce the tendency of organisms to adjust aspects of their ecology (movements, foraging decisions, life history parameters) to synchronize with episodes of cross-habitat movements of energy, materials or prey. Understanding the strength of linkages 4 between habitats is necessary for understanding food web dynamics at larger scales (Jackson and Fisher 1986) and so the effects of these cross-habitat movements of resources and prey on recipient communities have become a focus o f recent work in community and food web ecology. Thesis overview The literature contains many examples of empirical studies which demonstrate the existence of resource subsidies and document some consumer response to that subsidy (e.g. Szepanski et al. 1999, Bouchard and Bjorndal 2000, Sanchez-Pinero and Polis 2000, Gende and Wil lson 2001, Nakano and Murakami 2001, Lynch et al. 2002, Eggert and Wallace 2003, Kato et al. 2003, Mendelssohn and Kuhn 2003, Stapp and Polis 2003). Less has been done to determine the relative importance o f subsidies in different habitats or for different consumers (but see Nakano and Murakami 2001, Callaway and Hastings 2002). A t present, only one study (Sabo and Power 2002) that I am aware of explicitly investigates the effect of resource subsidies on the fitness of a recipient consumer. Very little work has investigated how subsidy effects might fluctuate as a consequence of physical and temporal heterogeneity in subsidy delivery (but see Polis et al. 1998, Ostfeld and Keesing 2000, Witman et al. 2004). Studies of the effects of resource subsidies have almost exclusively been conducted in systems where theory predicts the largest effects of subsidies (strongest contrast between donor system net primary productivity and recipient habitat net primary productivity). A t present there is no evidence in the published literature of any investigations at the other end of the spectrum that attempt to test this prediction, or investigate cases that might specifically shed light on the conditions under which subsidies have large effects in ecological communities. 5 This thesis presents a study of the factors enhancing or restricting the impact of spatial resource subsidies on several different consumers in stream and riparian habitats (Figure 1.1). Theory, models and empirical examples all suggest that resource subsidies are ubiquitous features of ecosystems - reinforcing the re-examination of conceptions about habitat boundaries and the relevance o f system openness in the structuring o f communities. A substantial quantity o f empirical and theoretical work now exists that documents or supports the openness of ecosystems - very little work currently exists that either evaluates the magnitude of that effect, generates predictions about what might control that effect size, or empirically tests theories about controls of effect size of resource subsidies. Understanding some of the factors that may determine the response of consumers in a recipient community to a resource subsidy is essential to the development of landscape level approaches to open systems (or to theory in community ecology, or food web ecology). I use a large array of consumers and habitats drawn from the literature, together with the community of riparian orb-weaving spiders, the tetragnathid spider Tetragnatha versicolor, resident stream fish (Onchorhynchus clarki) and the surface-foraging waterstrider Aquarius remigis found at the interface of streams and forests in the Malco lm Knapp Research Forest, Maple Ridge, British Columbia to examine several key questions. What factors determine the magnitude of subsidy effects in recipient food webs or determine their effect on recipient consumers? H o w are subsidy effects enhanced or dampened by the interactions between temporal or spatial variability and the life history traits or ecological requirements o f recipient consumers? Specific objectives for this thesis include, to: 1. Develop a synthesis of our current understanding of the magnitude of the effect of subsidies across different habitats, resource types and consumers using empirical data from the published literature. 6 2. Test for the consequences of spatial and temporal heterogeneity of subsidy delivery to recipient consumers. 3. Examine the influence of biotic controls on the quantity of subsidy transported between habitats. In this thesis I take a multi-level approach towards addressing my main research questions: what factors determine responses to subsidies, how are these responses enhanced or dampened by variability between habitats, and what are the measurable consequences of subsidies for consumers in both streams and riparian forests. I begin by synthesizing a large body of empirical studies of the effects of resource subsidies using meta-analysis (Chapter 2). I use these data to examine the effects of consumer trophic level, habitat type and productivity contrast on the magnitude of subsidy effects. This allows me to assess the strength of the existing evidence about which factors most strongly influence the effects of resource subsidies. In Chapter 3,1 consider the response of consumers (the riparian orb-weaving spider community) in a habitat with high levels of in situ net primary production. This allows me to test the idea that subsidies wi l l have the strongest effects where local production is weak. I compare the response of the spider community to similar published accounts in contrasting habitats. In Chapter 4.1 examine how temporal variability in a subsidy interacts with the developmental timing of a riparian consumer (the spider Tetragnatha versicolor) to produce variable fitness outcomes at the individual level. In Chapter 5 I examine how an organism (the waterstrider Aquarius remigis) foraging at the interface between streams and riparian forests alters the rate of subsidy movement between these two habitats and explore some of the potential consequences for alternate in-stream consumers. Chapter 6 is a general conclusion in which I provide a synthetic overview 7 relating the separate thesis chapters to each other, my initial questions and the overall field of study. Voucher specimens for all species identified in this thesis have been placed in the collection of the Spencer Entomological Museum, University of British Columbia. 8 Figure 1.1 Diagrammatic representation o f the subsidy pathways explored in this thesis. In Chapter 2 I attempt a broad synthesis of subsidy effects, predominantly across aquatic terrestrial boundaries. The effect of the emergence o f aquatic invertebrates on the riparian orb-weaving spider community is investigated in Chapter 3 . Chapter 4 examines the effects of temporal variation in this subsidy for the spider Tetragnatha versicolor. Chapter 5 focuses on the accidental movement of terrestrial invertebrates to stream habitats and the consequences o f competition for this resource on populations o f surface-dwelling invertebrates, in-stream fish (trout) and benthic invertebrates. Works Cited Bastow, J. L . , J. L . Sabo, J. C. Finlay, and M . E . Power. 2002. A basal aquatic-terrestrial trophic link in rivers: algal subsidies via shore-dwelling grasshoppers. Oecologia 131:261 -268. Bouchard, S. S., and K . A . Bjorndal. 2000. Sea turtles as biological transporters of nutrients and energy from marine to terrestrial ecosystems. Ecology 81:2305-2313. Cadenasso, M . L . , S. T. A . Pickett, and K . C. Weathers. 2004. Effects of landscape boundaries on the flux of nutrients, detritus and organisms. Pages 154-168 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Callaway, D . S., and A . Hastings. 2002. Consumer movement through differentially subsidized habitats creates a spatial food web with unexpected results. Ecology Letters 5:329-332. Collier, K . J., S. Bury, and M . Gibbs. 2002. A stable isotope study of linkages between stream and terrestrial food webs through spider predation. Freshwater Biology 47:1651-1659. Eggert, S., L . , and J. B . Wallace. 2003. Reduced detrital resources limit Pycnopsyche gentilis (Tricoptera: Limnephilidae) production and growth. Journal of the North American Benthological Society 22:388-400. Gende, S. M . , and M . F. Willson. 2001. Passerine densities in riparian forests of southeast Alaska: potential effects of anadromous spawning salmon. The Condor 103:624-629. Helfield, J. M . , and R. J. Naiman. 2006. Keystone interactions: salmon and bear in riparian forests of Alaska. Ecosystems 9:167-180. Henschel, J. R. 2004. Subsidized predation along river shores affects terrestrial herbivore and plant success. Pages 189-199 in G . A . Polis, M . E . Power, and G . R. Huxel , editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Huxel, G . R., G . A . Polis, and R. D . Holt. 2004. A t the frontier of the integration of food web ecology and landscape ecology. Pages 434-451 in G . A . Polis, M . E . Power, and G . R. Huxel , editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Jackson, J. K . , and S. G . Fisher. 1986. Secondary production, emergence and export of aquatic insects of a Sonoran desert stream. Ecology 67:629-638. Kato, C , T. Iwata, S. Nakano, and D . Kish i . 2003. Dynamics of aquatic insect flux affects distribution of riparian web-building spiders. Oikos 103:113-120. Lynch, R. J., S. E . Bunn, and C. P. Catterall. 2002. Adult aquatic insects: potential contributors to riparian food webs in Australia's wet-dry tropics. Austral Ecology 27:515-526. Mendelssohn, I. A . , and N . L . Kuhn. 2003. Sediment subsidy: effects on soil-plant responses in a rapidly submerging coastal salt marsh. Ecological Engineering 21:115-128. 1 0 Nakano, S., H . Miyasaka, and N . Kuhara. 1999. Terrestrial-aquatic linkages: riparian arthropod inputs alter trophic cascades in a stream food web. Ecology 80:2435-2441. Nakano, S., and M . Murakami. 2001. Reciprocal subsidies: dynamic interdependence between terrestrial and aquatic food webs. Proceedings of the National Academy of Sciences -U S A 98:166-170. Ostfeld, R. S., and F. Keesing. 2000. Pulsed resources and community dynamics of consumers in terrestrial ecosystems. Trends in Ecology and Evolution 15:232-237. Polis, G . A . , W. B . Anderson, and R. D. Holt. 1997. Toward an integration of landscape and food web ecology: the dynamics of spatially subsidized food webs. Annual Review of Ecological Systematics 28:289-316. Polis, G . A . , and S. D . Hurd. 1995. Extraordinarily high spider densities on islands: flow of energy from the marine to terrestrial food webs and the absence of predation. Proceedings of the National Academy of Sciences - U S A 92:4382-4386. Polis, G . A . , and S. D . Hurd. 1996. Linking marine and terrestrial food webs: allochthonous input from the ocean supports high secondary productivity on small islands and coastal land communities. The American Naturalist 147:396-423. Polis, G . A . , S. D. Hurd, T. C. Jackson, and F. Sanchez-Pinero. 1998. Multifactor population limitation: variable spatial and temporal control of spiders on G u l f of California islands. Ecology 79:490-502. Polis, G . A . , and D . R. Strong. 1996. Food web complexity and community dynamics. The American Naturalist 147:813-846. Power, M . E . , W . E . Rainey, M . S. Parker, J. L . Sabo, A . Smyth, S. Khandwala, J. C. Finlay, F. C. McNeely, K . Marsee, and C. Anderson. 2004. River-to-watershed subsidies in an old-growth conifer forest. Pages 217-240 in G. A . Polis, M . E . Power, and G. R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Ricklefs, R. E . 1990. Ecology, 3rd edition. W . H . Freeman and Company, New York. Sabo, J. L . , and M . E . Power. 2002. River-watershed exchange: effects of riverine subsidies on riparian lizards and their terrestrial prey. Ecology 83:1860-1869. Sanchez-Pinero, F. , and G. A . Polis. 2000. Bottom-up dynamics of allochthonous input: direct and indirect effects of seabirds on islands. Ecology 81:3117-3132. Sanzone, D . M . , J. L . Meyer, E . Marti , E . P. Gardiner, J. L . Tank, and N . B . Grimm. 2003. Carbon and nitrogen transfer from a desert stream to riparian predators. Oecologia. 134"238-250 Sher, A . A . , D . E . Goldberg, and A . Novoplansky. 2004. The effect of mean and variance in resource supply on survival on annuals from Mediterranean and desert environments. Oecologia 141:353-362. 11 Southwood, T. R. E . 1977. Habitat, the template for ecological strategies? The Journal of Animal Ecology 46:336-365. Stapp, P., and G. A . Polis. 2003. Marine resources subsidize insular rodent populations in the G u l f of California, Mexico. Oecologia 134:496-504. Summerhayes, V . S., and C. S. Elton. 1923. Contribution to the ecology of Spitsbergen and Bear Island. Journal of Ecology 11:214-286. Szepanski, M . M . , M . Ben-David, and V . Van Ballenberghe. 1999. Assessment of anadromous salmon resources in the diet of the Alexander Archipelago wol f using stable isotope analysis. Oecologia 120:327-335. Wallace, J. B . , S. L . Eggert, J. L . Meyer, and J. R. Webster. 1997. Multiple trophic levels of a forest stream linked to terrestrial litter inputs. Science 277:102-104. Weathers, K . C , M . L . Cadenasso, and S. T. A . Pickett. 2001. Forest edges as nutrient and pollution concentrators: potential synergisms between fragmentation, forest canopies, and the atmosphere. Conservation Biology 15:1506-1514. Wipf l i , M . S. 2005. Trophic linkages between headwater forests and downstream fish habitats: implications for forest and fish management. Landscape and Urban Planning 72:205-213. Wipf l i , M . S., J. P. Hudson, D . T. Chaloner, and J. P. Caouette. 1999. Influence of salmon spawner densities on stream productivity in Southeast Alaska. Canadian Journal of Fisheries and Aquatic Sciences 56:1600-1611. Witman, J. D . , J. C. El l is , and W. B . Anderson. 2004. The influence of physical processes, organisms, and permeability on cross-ecosystem fluxes. Pages 335-358 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. 12 Chapter 2. A meta-analysis of the role of trophic position, habitat type and habitat productivity in determining the food web effects of resource subsidies^ Introduction A s early as the 1920s ecologists recognized that ecosystem boundaries are frequently crossed by materials, energy and organisms (e.g. Summerhayes and Elton 1923 in Witman et al. 2004). Nonetheless, until the past decade the majority of food web studies were limited to local spatial scales and did not take into account these connections between ecosystems. Recent empirical studies in a range of environments have revealed the magnitude of resource subsidies by demonstrating changes in distribution, abundance or growth of consumers in recipient habitats across a range of vertebrate and invertebrate taxa at multiple trophic levels (e.g. Polis et al. 1998, Anderson and Polis 1999, Nakano et al. 1999, Barrett et al. 2005). Sufficient studies now exist for ecologists to begin testing general predictions about the effects of subsidies across ecosystem types, for different taxonomic and functional groups, and along gradients of productivity. While subsidies have been described in a variety of systems, it is as yet unclear whether they should have similar impacts in all habitat types (Vanni et al. 2004). For example, several authors have hypothesized that habitats with large perimeter to area ratios (e.g. streams, riparian forests, small islands) wi l l have higher relative exposure to resource subsidies, and therefore have proportionately larger responses to subsidies (Polis and Hurd 1996, Polis et al. 1997, Cadenasso et al. 2004, Witman et al. 2004). Empirical studies of resource subsidies have been complemented by theoretical studies that illustrate how cross system subsidies may influence local food web dynamics (Polis et al. 1997), food web stability (Holt 2002, Huxel et al. 2002) and species coexistence (Sears et al. * Published as: Marczak, L . B . R . M . Thompson and J.S. Richardson (2007). Meta-analysis: trophic level, habitat and productivity shape the food web effects of resource subsidies. EcologyM: 140-148. 13 2004). These studies have shown that the trophic position at which subsidies enter a food web has implications for stability (Huxel et al. 2002), and the magnitude of a species response to subsidies (Polis et al. 1997, Anderson and Polis 2004). For example, species feeding on basal resources (e.g. plant tissue and detritus) are likely to respond strongly to subsidies entering at the base of a food web. In contrast, taxa at higher trophic levels may consume prey from multiple food web compartments (Anderson and Polis 2004) and show a reduced response to subsidies affecting only one compartment. We predict that increased dietary generalization moving up food chains w i l l mean that consumer responses to resource subsidies w i l l be less at higher trophic levels. There has been considerable speculation about the effect of recipient habitat productivity on the magnitude of subsidy effects. This has included the idea that effects of subsidies w i l l be strongest where the difference in productivity between donor and recipient habitats is large (Persson et al. 1996, Polis et al. 1997) or where recipient habitats have low productivity (Polis and Hurd 1996, Stapp and Polis 2003). Comparisons of productivity have generally been represented in terms of net primary productivity. However, in cases where the subsidy is of some form other than primary production, this ratio may not be the relevant measure of inter-system differences. Recent narrative reviews have concluded that resource subsidies are ubiquitous across habitats, have generally positive effects on broad taxonomic groups and are strongest where recipient habitat productivity is weak (Fausch et al. 2002, Power and Dietrich 2002, Cadenasso et al. 2003). Our objective was to quantitatively test these hypotheses about resource subsidies across a range o f studies in a meta-analytic review. We defined resource subsidies in the broad sense of any movement of energy across ecotones. Explicitly we tested the following: (1) What is the overall magnitude of the effect of resource subsidies on consumers in recipient habitats? 14 (2) Does the magnitude of this effect vary with the type of recipient habitat? (3) Does the magnitude of the effect vary with the trophic level or taxonomic identity of recipient taxa? (4) H o w does the effect of a subsidy vary with productivity of donor and recipient ecosystems? Methods Study selection and data criteria We restricted our literature search to papers reporting change in consumer biomass (g/m ) or density (number/m2) in recipient habitats in response to a resource subsidy. We searched the databases Biological Abstracts and Web of Science (1965 to present) and the last 10 years of the table of contents of three journals (Ecology, Oecologia, Oikos) for relevant keywords (resource subsidy, allochthonous, ecotone, cross-habitat). Papers were excluded i f they did not report a measure of consumer response or reported consumer responses in units other than density or biomass. We selected studies that manipulated either the amount of a subsidy or used a natural gradient in a subsidy as a proxy for a manipulation. For each study, we recorded the sample size, mean and standard error of the response variable for high and low subsidy contrasts (transect end points or natural experiment contrasts) and for manipulated subsidy and control treatments. We considered multiple results within a single paper as independent studies when they involved different species, habitats, subsidy type or season (Wolf 1986, Gurevitch et al. 1992, Bolnick and Preisser 2005). The selected papers included both experimental and descriptive studies (n = 35 and n = 80 respectively; Appendix A ) . Density or biomass of subsidy resources and the magnitude of the consumer response was extracted for each study. We standardized the results of all studies so that a positive effect value indicates an increase in consumer abundance in response to a resource subsidy. Descriptive studies were o f two types: 1) comparative assessments of areas with and without subsidies (natural experiments), and 2) transect studies that described the effect - 15 of a subsidy as a function of distance from the subsidy input. In order to allow comparisons between subsidized and non-subsidized systems, we characterized sites within studies as high or low subsidy environments. In comparative studies, sites with low subsidies were treated as analogous to manipulated subsidy reductions from experimental studies. In transect studies, we contrasted the transect position closest to the source of the subsidy (highest subsidy input or control) with the farthest transect position. When results were reported as a time series we used the end points in order to capture the time of maximal seasonal difference of the subsidy in our analysis. Study duration was investigated as a possible continuous predictor and level of taxonomic identification as a possible categorical influence in the meta-analytic model. Net primary productivities of recipient and donor habitats were taken from published estimates provided by authors or from supporting materials or publications. In order to contrast resource subsidies with the amount of equivalent resources in the recipient habitat, each study was reviewed in detail. A resource in the recipient environment was considered equivalent to the subsidy input i f it was on the same trophic level and represented a food resource to the same taxonomic and functional groups as the subsidy. For example, flying terrestrial insects were considered equivalent to the subsidy of emerging aquatic insects to terrestrial insectivores. C i Data Analysis The effect size of resource subsidies was measured as the natural log of the response ratio of recipient density or biomass under high and low subsidy scenarios (Gurevitch and Hedges 1999, Rosenberg et al. 2000, Shurin et al. 2002). The response ratio (R) is the ratio of the mean outcome in the experimental group to that in the control group (Rosenberg et al. 2000). This measure o f effect size was used because it is easily interpretable in a biological context and has a sampling distribution that is approximately normal (Hedges et al. 1999, Lajeunesse and Forbes 16 2003). Although numerous studies were either unreplicated or did not report estimates of variability, excluding these studies would increase Type I error (Shurin et al. 2002, Lajeunesse and Forbes 2003). Consequently, we did not weight values by their precision (as in Shurin et al. 2002). Although this increases the rate of Type II errors, it avoids underestimates of effect size (Gurevitch and Hedges 1999, Shurin et al. 2002). Tests for homogeneity of effect sizes were based on the statistic Q with larger values indicating greater heterogeneity in effect sizes among comparisons (Rosenberg et al. 2000). Total heterogeneity for a group of comparisons, Q T , can be partitioned into within-class heterogeneity, Qw, and between-class heterogeneity, Q B , which is analogous to the partitioning of variation in an A N O V A (Rosenberg et al. 2000). We assessed the importance of publication bias using a rank correlation test (Spearman's rho). A significant correlation between effect size and sample size would indicate bias toward the publication of larger effects. We determined that effect sizes were normally distributed based on visual assessment of a normal quantile plot. To test our a priori hypotheses with regard to differences among categories of studies, we followed a hierarchical schema (Gurevitch et al. (1992). We first tested the null hypothesis that all effect sizes were equal, and i f this was rejected, we examined a series of categorical variables (experimental vs. manipulative studies, recipient habitat type, trophic level of the recipient) subdividing by testing the within-class fit. I f the within-class fit to a single effect size is rejected, categories can be further subdivided. We also examined a series of methodological categories to determine their effect on response magnitude including transect length, study duration, sample size, and taxonomic resolution. We regressed the productivity contrast between habitats (dononrecipient productivity), the productivity of the recipient habitat, the treatment ratio (magnitude of the difference between treatments) and the ratio of subsidy to ambient resources against effect size by treating these as continuous variables in the analysis. We transformed ratio 17 data as In (x + 1) to improve normality. A l l data were analyzed using MetaWin software version 2 . 1 . 4 (Rosenberg et al. 2 0 0 0 ) . Results Overview We analyzed 1 1 5 data sets from 3 2 papers detailing studies in streams, island habitats, marine settings, coastal deserts and forests, with the majority of these studies focusing on land-water interfaces (Appendix B) . The overall mean effect size for all studies indicated significant heterogeneity (mean = 0 . 8 9 , Q T = 1 8 3 . 6 , d f = 1 0 2 , p < 0 . 0 0 1 ) indicating that further data structure existed. There was no difference in mean effect size between experimental studies and studies that were descriptive ( Q B = 4 1 . 9 , df = 1, p = 0 . 4 0 8 ) justifying combining them in a single meta-analysis. The data used in this study came from publications on a wide diversity of organisms; however the vast majority of studies focused on comparatively few taxa (Appendix A ) . Effect size varied according to level of taxonomic resolution employed in a study (between groups Q B = 1 1 . 5 , p = 0 . 0 2 ) . Largest mean effect sizes occurred at coarsest levels of taxonomic resolution. Effect sizes were smallest when responses of individual species were considered. Studies ranged from 1 to 3 6 months duration (median 3 months). Study duration was not significantly related to effect size (F = 1 .11 , p = 0 . 2 9 , R z = 0 . 0 1 ) . Number o f replicates ranged from 1 (no replication of treatments) to 2 4 (median 6 ) . High treatment ratios were not related to mean effect size (F = 0 . 2 4 , p = 0 . 6 2 6 , R ' = 0 . 0 0 3 ) . There was a trend towards increased effect size as transect length increased (F .= 4 . 1 , p = 0 . 0 4 8 , R 2 = 0 . 0 7 ) . Spearman's rank correlation test detected no evidence of publication bias (Rs - 0 . 0 1 5 , p = 0 . 8 9 ) 18 Habitat and recipient taxa effects Subsidies significantly affected biomass or density of focal consumers in all recipient habitats except salt marsh and interior forest habitats (Figure 2.1) with largest effect sizes in cobble bars beside streams and coastal habitats. For trophic groupings of recipient organisms, detritivores had the largest significant mean effect size (2.00) compared to omnivores (0.98), all predators (0.77) or just insectivores (0.51). Detritivore response was significantly greater than that of predators and insectivores (between groups Q B = 28.7, p < 0:001; Figure 2.2). Only two taxonomic groups, birds and spiders, contained sufficient sample sizes for further analysis. A total of 36 studies investigated birds as recipient consumers. The overall response for birds across all studies was significant (Figure 2.3 A ) but this was driven by studies containing aerial insectivores; birds in other trophic guilds did not demonstrate a significant mean response. Similarly, 18 studies that examined the response of spiders showed an overall significant effect of subsidies (Figure 2.3B), but only horizontal orb weavers (in the family Tetragnathidae) individually demonstrated a significant response. For both birds and spiders, studies that did not distinguish functional groups or species - i.e. the analysis was done at the level of Class (Aves) or Order (Araneae) - showed no significant effect o f subsidies (Figure 2.3 A , B ) . Productivity effects There was no relationship between effect size and the ratio of donor to recipient net primary productivity (p = 0.69, R 2 = 0.025; Figure 2.4A), or recipient habitat net primary productivity (p = 0.52, R = 0.003; Figure 2.4B). However, there was a significant positive relationship between effect size and the ratio of subsidy resources to trophically equivalent ambient resources (p <0.001, R 2 = 0.172; Figure 2.4C). 19 Discussion Ecologists have made several important predictions about the conditions under which resource subsidies should have large effects in recipient food webs. Several authors have predicted that the effects of subsidies w i l l be strongest where there is a large contrast in net primary productivity between donor and recipient habitats (Persson et al. 1996, Polis et al. 1997). Other authors have suggested the trophic position at which a subsidy enters a food web (Anderson and Polis 2004) or the physical character of the habitat (Witman et al. 2004) as important determinants of how strongly a subsidy affects recipient dynamics. Our meta-analysis confirms that trophic linkages across habitats are detectable and significant, even when systems differ greatly in productivity, species composition and physical structure. In this study we were able for the first time to explore broad scale patterns in the role of subsidies in different habitats, for different trophic and taxonomic groups, and under scenarios of different productivities. These results provide insights into the situations in which subsidies may be important, although it is important to note that the relatively small number of studies currently published and problems inherent in meta-analyses mean that our conclusions must be viewed with some caution. There are a number of possible sources of bias in meta-analyses. First, the researcher, location of a study, habitat and focal consumers are unlikely to be independent. Several authors remain dominant in the field and have produced a disproportionate number of available datasets, across a limited geographic area. Partly as a consequence our data set is strongly biased towards aquatic - terrestrial contrasts. Additionally, published estimates of N P P are rare and even more rarely reported in the context of subsidy studies. The generality of our results are thus limited by the data available. A further limitation of current studies is that they have been frequently conducted precisely where ecologists have anticipated the strongest effects of subsidies (across abrupt boundaries, in high contrast productivity situations). We believe that the absence of a significant publication bias in the studies used for this meta-analysis may be partly a 20 consequence of our inclusion of data for different species from single studies. Non-significant results are thus included in our analysis that might not have been published independently from significantly responding taxa, potentially obscuring a real "file drawer" effect or bias in the habitats, consumers or subsidy situations commonly studied. Effects of subsidies vary with habitat The magnitude of responses to subsidies differed depending on recipient habitats. Previous studies have independently described strong effects of subsidies in environments that are relatively open to neighboring ecosystems or have large perimeter to area ratios such as streams, and beaches (Polis and Strong 1996, Cadenasso et al. 2004, Witman et al. 2004). Our meta-analysis finds the first quantitative support for those patterns. The strongest effects were seen in stream cobble bar environments and coastal habitats, which are both typified by high perimetenarea ratios. We hypothesize that the effects of subsidies w i l l depend on the relative rates of supply (as determined by perimeter to area ratios and the nature of boundaries). However retention of subsidies may also be important. Differences in retention have been documented for a number of systems where subsidies have been shown to be important including beaches (Orr et al. 2005) and streams (Wallace et al. 1995). We hypothesize that large and rapid responses to subsidies may be evident where subsidies are large, and retention is high. Donor productivity does not determine the magnitude of consumer responses Various authors have suggested that large contrasts between the donor and recipient habitat productivity produce the largest effects (e.g., Polis et al. 1997, Kato et al. 2003, Witman et al. 2004). This prediction is based on the untested assumption that the amount of subsidy produced and transported between habitats is proportional to the donor system's productivity. 21 Our analyses revealed no relationship between the consumer response to a subsidy and the ratio of donor to recipient productivity, or the productivity of the recipient habitat. There are several possible explanations for this. Most obviously, the amount of material being transported across ecosystem boundaries as a subsidy does not depend solely on the amount available for transport. Subsidy quantity and rate of delivery may be mediated by physical and biotic features of landscapes (Witman et al. 2004), the nature of the boundaries between habitats (Cadenasso et al. 2004), and the characteristics of the subsidy materials themselves. Physical transport of subsidy materials can be dictated by the dynamics of wind and water. For example, there is considerable evidence that flying insects may be concentrated by winds interacting with topography and local features such as forest edges (Burt and Pedgley 1997). Sea wrack accumulation on beaches may be altered by nearshore hydrodynamics, the buoyancy characteristics of the wrack and beach type (Orr et al. 2005), regardless of the rate of production in nearshore environments. The specific nature of a boundary between donor and recipient habitats may also mediate the flux of material (Cadenasso et al. 2004). Retention and permeability of boundaries may be modified by both physical and biotic factors. For example, differences in forest architecture can influence rates of nutrient subsidies (Cadenasso et al. 2004) and subsidies in the form of dispersing seeds (Cadenasso and Pickett 2001). Biotic vectors can be critical in determining the amount of a subsidy that arrives in a recipient habitat. Some animals may be both vectors and recipients of subsidies as has been shown with migratory geese (Jefferies et al. 2004), while others may act to restrict the transport of material between habitats (e.g. waterstriders or other surface-dwelling organisms that remove some fraction of insects emerging from streams as well as a fraction of insects falling into streams, Marczak et al. unpublished data). Some resources may be more easily mobilized or transported across habitat boundaries (Witman et al. 2004). The large amount of variability in the effects of subsidies seen in this study is likely due in part to 22 interactions between resource subsidy productivity and the biological factors that moderate its movement. Fluctuations in the relative productivity of donor and recipient habitats and the way in which this influences consumer effects have not been addressed by many studies (but see Nakano and Murakami 2001, Kato et al. 2003). The relative productivity of neighboring ecosystems can fluctuate through time, particularly due to seasonal cycles. The overall productivity contrast between two habitats might be equivalent across a year, but seasonally disjunct at the level of specific resources (e.g. aquatic and terrestrial insects) rendering gross contrasts of basal productivity irrelevant. The relation of subsidies to in-situ resources modifies subsidy effects One of the most important findings in our study is a positive relationship between mean effect size and the ratio of subsidy to ambient resources of comparable types. The effects of a subsidy are determined not only by the amount of energy as a whole, but specifically by the way in which this energy is packaged. Subsidy effects appear to be largest when they subsidize a system with low levels of comparable resources. The ratio of subsidy to equivalent ambient resources may shift through time due to episodic or seasonal changes. Community responses to resources which occur variably in time should be functions of the degree of specialization on the subsidy, the rate at which a consumer population can respond to a subsidy and the mobility of the consumer (Ostfeld and Keesing 2000). These patterns are likely to be very important and have been little explored. Future studies of subsidies need to study the specific nature of subsidies relative to the nature of available resources in the recipient habitat, and to incorporate measures of temporal dynamics. 23 Trophic levels are influenced differentially by subsidies Studies modeling the effects of resource subsidies have shown dynamic implications for the stability of food webs depending on which trophic level is subsidized (e.g. Huxel et al. 2002). Our meta-analysis showed that the response of lower trophic levels (herbivores and detritivores) to subsidies was roughly twice that of the effect on predators. Some of this difference is likely due to "source omnivory" (Anderson and Polis 2004) whereby higher-level consumers have access to resources derived from multiple productivity compartments. Thus a subsidy to a single productivity compartment may have its effect muted by a lack of change in other compartments. A significant factor in the nature of predator responses may be having the life history and mobility traits to respond quickly to an increase in a productivity compartment (Power and Rainey 2000). Predators tend to be longer lived than lower trophic levels and may take longer to respond numerically to a subsidy. Studies of the interaction between species traits and responses to subsidies would be an interesting next step. It is likely also that there is an attenuation of the effect of basal subsidy fluctuations as they cascade up through food chains. Effects of subsidies are highly variable across and within taxonomic and functional groups Because of a lack of data on other groups, we were forced to limit an analysis of differing responses at the level of functional group to studies of birds and spiders - both of these groups were responding to subsidies of emerging aquatic insects. Within the ranks of generalist predators such as spiders and birds, the specialization represented by different capture modes results in differing responses to resource subsidies. We observed significant effects across broad taxonomic groups that were driven only by strong effects on small groups of taxa in specialized feeding guilds. For example, of the three spider families most commonly examined, only horizontal orb-weavers (Tetragnathidae) respond significantly to prey subsidies. Studies that simply examined "spiders" without further differentiation often found effects and frequently 24 interpreted these effects as influencing all spider taxa. In birds, only studies that separated the responses of aerial insectivores (consumers that might be expected to exhibit some degree of specialization on emerging aquatic insects) from those of other insectivores showed effects of subsidies. Although our analysis is on a relatively limited subset of the data, we believe that the effects of functional grouping on the ability to detect subsidies may be an important issue in all studies of this type, and should be taken into account when designing and executing these studies. The subtleties of small group or individual species' responses may be overlooked when researchers lump organisms too broadly. Conclusions Unti l recently, there were insufficient independent studies to develop a quantitative assessment of the magnitude of the effects of spatial subsidies. Our analysis is the first significant meta-analysis in this area and showed clear effects of subsidies across a range of taxa. Subsidy effects varied between habitats and differentially influenced taxa both trophically and functionally. Effects were strongest in habitats such as cobble bars and coastlines where perimetenarea ratios were high, and where the recipient consumer was a detritivore. When predators were the recipient species, subsidies had the strongest effect on insectivores. Importantly, we found that strongest effects of subsidies did not occur in systems of contrasting primary productivity (Persson et al. 1996, Polis et al. 1997, Kato et al. 2003, Sears et al. 2004), unless productivity was carefully defined in terms of relevant resources. The amount of a subsidy arriving in a recipient habitat is likely to be modified by the type of subsidy (detritus, prey, nutrients), agent of transport (passive or active, physical or biotic vector) and the physical (topography, vegetative architecture) and biotic characteristics of the interface between donor and recipient habitats. For these reasons, the ratio of subsidy to equivalent resources was a more useful tool for predicting the possible effect of a subsidy than coarser contrasts of basal productivity. Our study shows the 25 trophic importance of subsidies across habitat boundaries, but also indicates that the biological effects of subsidies w i l l depend upon the nature of the recipient habitat and biota. Future studies have the opportunity to explore how recipient habitat characteristics and species traits determine the magnitude of responses to subsidies. Acknowledgements We acknowledge the help of J. Gurevitch and participants in the 2005 meta-analysis workshop in Kananaskis, Alberta. Valuable feedback on early drafts was provided by T . M . Hoover, J. Shurin and the Stream and Riparian Research lab at U B C . The manuscript was substantially improved by comments from D.S. Gruner and several anonymous reviewers. Funding was provided by the University of British Columbia and the Natural Sciences and Engineering Research Council (Canada). 26 5 33 46 17 marine: marine: freshwater: terrestrial: marine: marine terrestrial terrestrial freshwater freshwater B ab b i 2 I 19 bed cd 39 o 14 cd I 16 Figure 2.1. Mean effect size differed between recipient habitat types. Means with 95% bootstrapped confidence intervals are shown. Numbers below lines indicate sample sizes, different letters indicate significant differences between means; the zero line indicates no effect. 27 5 4 H ab N •6 i c (0 e 2H 1 H -1 H -2 20 6* ab b 32 V I 36 c <» 3 Figure 2.2. Mean effect size differed with the trophic level of the subsidy consumer. Means with 95% bootstrapped confidence intervals are shown. Numbers below lines indicate sample sizes, different letters indicate significant differences between means. The zero line indicates no effect. 28 5 A 36 -3 NI '55 d) c as a) E - l H -2 H B o 1R < 6 > c 3 Figure 2.3. A . Mean effect size differences between levels of further classification within the group of studies that reported results for birds. The overall mean for all studies was significant. Although means did not differ significantly from each other, only aerial insectivores show a mean value significantly different from zero. Numbers below lines indicate sample sizes. Means with 95% bootstrapped confidence intervals are shown. B. Mean effect size differences between levels of further classification within the group of studies that reported results for spiders. The overall mean for all studies was significant. Although means did not differ significantly from each other, only horizontal orb-weavers show a mean value significantly different from zero. Numbers below lines indicate sample sizes. Means with 95% bootstrapped confidence intervals are shown. 29 -1 H 1 1 1 1 T . -6 -4 -2 0 2 4 6 In ratio of subsidy to ambient resources Figure 2.4. A. Relationship between ratio o f donor to recipient habitat productivity and the response ratio (In x). R 2 = 0.025, p = 0.69. B. Relationship between the response ratio (In x) and recipient habitat net primary productivity. R 2 = 0.003, p = 0.52. C. Significant positive relationship between the ratio o f resource subsidies to equivalent in situ or ambient resources and the response ratio (In x). R 2 = 0.172, p <0.0001. 30 Works cited Anderson, W . B . , and G . A . Polis. 1999. Nutrient fluxes from water to land: seabirds affect plant nutrient status on G u l f of California islands. Oecologia 118:324-332. Anderson, W . B . , and G . A . Polis. 2004. Allochthonous nutrient and food inputs: consequences for temporal stability. Pages 82-95 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Barrett, K . , W . B . Anderson, A . Wait, L . L . Grismer, G . A . Polis, and M . D . Rose. 2005. Marine subsidies alter the diet and abundance of insular and coastal lizard populations. Oikos 109:145-153. Bolnick, D . I., and E . L . Preisser. 2005. Resource competition modifies the strength of trait-mediated predator-prey interactions: a meta-analysis. Ecology 86:2771-2779. Burt, P. J. A . , and D . E . Pedgley. 1997. Nocturnal insect migrations: effects of local winds. Advances in Ecological Research 27:61-92. Cadenasso, M . L . , and S. T. A . Pickett. 2001. Effects of edge structure on the flux of species into forest interiors. Conservation Biology 15:91-97. Cadenasso, M . L . , S. T. A . Pickett, and K . C. Weathers. 2004. Effects of landscape boundaries on the flux of nutrients, detritus and organisms. Pages 154-168 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Cadenasso, M . L . , S. T. A . Pickett, K . C. Weathers, and C. G . Jones. 2003. A framework for a theory of ecological boundaries. Bioscience 53:750-758. Fausch, K . D . , M . E . Power, and M . Murakami. 2002. Linkages between stream and forest food webs: Shigeru Nakano's legacy for ecology in Japan. Trends in Ecology and Evolution 17:429-434. Gurevitch, J., and L . V . Hedges. 1999. Statistical issues in ecological meta-analyses. Ecology 80:1142-1149. Gurevitch, J., L . L . Morrow, A . Wallace, and J. S. Walsh. 1992. A meta-analysis of competition in field experiments. The American Naturalist 140:539-572. Hedges, L . V . , J. Gurevitch, and P. S. Curtis. 1999. The meta-analysis of response ratios in experimental ecology. Ecology 80:1150-1156. Holt, R. D . 2002. Food webs in space: on the interplay of dynamic instability and spatial processes. Ecological Research 17:261-273. Huxel, G . R., K . McCann, and G . A . Polis. 2002. Effects of partitioning allocthonous and autocthonous resources on food web stability. Ecological Research 17:419-432. 31 Jefferies, R. L . , H . A . L . Henry, and K . F. Abraham. 2004. Agricultural nutrient subsidies to migratory geese and change in Arctic coastal habitats. Pages 268-283 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Kato, C , T. Iwata, S. Nakano, and D . Kish i . 2003. Dynamics of aquatic insect flux affects distribution of riparian web-building spiders. Oikos 103:113-120. Lajeunesse, M . J., and M . R. Forbes. 2003. Variable reporting and quantitative reviews: a comparison of three meta-analytic techniques. Ecology Letters 6:448-454. Nakano, S., Y . Kawaguchi, Y . Taniguchi, H . Miyasaka, Y . Shibata, H . Urabe, and N . Kuhara. 1999. Selective foraging on terrestrial invertebrates by rainbow trout in a forested headwater stream in northern Japan. Ecological Research 14:351-360. Nakano, S., and M . Murakami. 2001. Reciprocal subsidies: dynamic interdependence between terrestrial and aquatic food webs. Proceedings of the National Academy of Sciences -U S A 98:166-170. Orr, M . , M . Zimmer, D . E . Jelinski, and M . Mews. 2005. Wrack deposition on different beach types: spatial and temporal variation in the pattern of subsidy. Ecology 86:1496-1507. Ostfeld, R. S., and F. Keesing. 2000. Pulsed resources and community dynamics of consumers in terrestrial ecosystems. Trends in Ecology and Evolution 15:232-237. Persson, L . , J. Bengtsson, B . A . Menge, and M . E . Power. 1996. Productivity and consumer regulation - concepts, patterns and mechanisms. Pages 396-434 in G . A . Polis and K . O. Winemiller, editors. Food webs: integration of patterns and dynamics. Chapman & Hal l , N e w York. Polis, G . A . , W . B . Anderson, and R. D . Holt. 1997. Toward an integration of landscape and food web ecology: the dynamics of spatially subsidized food webs. Annual Review of Ecological Systematics 28:289-316. Polis, G . A . , and S. D . Hurd. 1996. Linking marine and terrestrial food webs: allocthonous input from the ocean supports high secondary productivity on small islands and coastal land communities. The American Naturalist 147:396-423. Polis, G . A . , S. D . Hurd, T. C. Jackson, and F. Sanchez-Pinero. 1998. Multifactor population limitation: variable spatial and temporal control of spiders on G u l f of California islands. Ecology 79:490-502. Polis, G . A . , and D . R. Strong. 1996. Food web complexity and community dynamics. The American Naturalist 147:813-846. Power, M . E . and W . E . Rainey. 2000. Food webs and resource sheds: Towards spatially delimiting trophic interactions, pp. 291-314- in M . J . Hutchings, E.A.John and A . J . A . Stewart (eds.) Ecological Consequences of Habitat Heterogeneity. Blackwell Scientific, Oxford, U K . 32 Power, M . E . , and W . E . Dietrich. 2002. Food webs in river networks. Ecological Research 17:451-471. Rosenberg, M . S., D . C . Adams, and J. Gurevitch. 2000. Meta W i n statistical software for meta-analysis version 2. Sinauer Associates, Inc., Sunderland, Massachusetts. Sears, A . L . W. , R. D . Holt, and G . A . Polis. 2004. Feast and famine in food webs: the effects of pulsed productivity. Pages 359-386 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Shurin, J. B . , E . T. Borer, E . W. Seabloom, K . Anderson, C . A . Blanchett, B . Broitman, S. D . Cooper, and B . S. Halpern. 2002. A cross-ecosystem comparison of the strength of trophic cascades. Ecology Letters 5:785-791. Stapp, P., and G . A . Polis. 2003. Marine resources subsidize insular rodent populations in the G u l f of California, Mexico. Oecologia 134:496-504. Vanni , M . J., D . L . DeAngelis, D . E . Schindler, and G . R. Huxel. 2004. Overview: Cross-habitat flux of nutrients and detritus. Pages 3-11 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Wallace, J. B . , M . R. Whiles, S. Eggert, T. F. Cuffhey, G . H . Lugthart, and K . Chung. 1995. Long-term dynamics of coarse particulate organic matter in three Appalachian mountain streams. Journal of the North American Benthological Society 14:217-232. Witman, J. D . , J. C. El l is , and W. B . Anderson. 2004. The influence of physical processes, organisms, and permeability on cross-ecosystem fluxes. Pages 335-358 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Wolf, F . M . 1986. Meta-analysis: quantitative methods for research synthesis. Sage Publications, London. 33 Chapter 3. Spiders and subsidies: results from the riparian zone of a coastal temperate rainforest.^ Introduction Resource flows between habitats (hereafter referred to as subsidies) can have important implications for food web dynamics in recipient environments (Bastow et al, 2002; Collier, Bury & Gibbs, 2002; Sabo & Power, 2002b). The movement of energy, nutrients or prey between systems is ubiquitous, although it is not yet clear how strong the effects of these subsidies are in all systems (Polis, Anderson & Holt, 1997). In this context it is critical to understand how different characteristics of subsidies (e.g. trophic level of a subsidy, subsidy type, variability in space or time), habitats (e.g. permeability, seasonality of in situ production) or consumers (e.g. trophic level, functional feeding group) might determine the relative importance of cross-habitat subsidies. Explici t in the definition of resource subsidies has been the idea of recipient benefit. We follow Polis et al. (1997) in defining subsidies as any movement of energy and materials across a habitat boundary that provides benefit to a recipient consumer. Although not implicit or required by this definition, the vast majority of studies have supposed that cross-habitat subsidies move predominantly into less productive systems, and there strongly influence abundance and distribution of consumers (Paetzold, Bernet & Tockner, 2006; Polis & Hurd, 1996; Sears, Holt & Polis, 2004). This premise is based on the assertion that passive dispersal or movement w i l l occur by diffusion from areas of higher to lower density, and that these flows wi l l have strong effects in lower productivity recipient habitats, and by extension negligible effects in productive recipient habitats (Sears et al, 2004). Indeed, examples of strong effects of resource subsidies in low productivity environments are well known, particularly in deserts, dry coastlines and arctic systems (Polis et al, 1996). In Polis and Hurd's (1996) island food-webs, marine algae and * Marczak, L . B . and J.S. Richardson (2007). Spiders and subsidies: results from the riparian zone of a coastal temperate rainforest. Journal of Animal Ecology.volume to be determined. 34 carrion beached on the shores of small, very dry islands supported extremely large populations of detritivorous arthropods which formed greater than 90% of the food available to web-building spider populations. Spiders were 1 -2 orders of magnitude greater in abundance than in similar areas without detrital inputs. Marine-derived material has similarly been found to subsidize coyotes along desert coastlines in Baja, California (Rose & Polis, 1998). Similarly, emerging aquatic insects have been found to affect the growth and fitness of lizards on dry cobble bars in California (Sabo et al, 2002b) and to alter the distribution of ground-dwelling arthropods on low productivity gravel bars in Italy (Collier et al, 2002; Paetzold et al, 2006). H o w do consumers respond to subsidies in highly productive recipient habitats? It has been supposed that the effect of subsidies arriving in high productivity habitats w i l l be small in magnitude. However, empirical studies in such habitats are scarce - researchers have understandably focused on systems with strong edges and where the ability to detect a response to subsidies is correspondingly large. The emphasis on studying systems where strong responses to resource subsidies are anticipated has limited our ability to ascertain the variables that determine the strength of responses, or to appreciate the ubiquity of subsidy effects. Aquatic-terrestrial contrasts have been popular in studies of subsidy effects, at least in part because of the distinct boundary between these habitats. In particular, the study of forest contributions to headwater streams was fruitful long before it was framed in the context of resource subsidies (e.g. Jackson & Fisher, 1986; Mason & Macdonald, 1982; Richardson, 1991). More recent investigations have highlighted the potential contribution of streams to forest habitats in the form of salmon (Wipfl i , 2005), algal mats (Bastow et al, 2002), and emerging aquatic invertebrates (Sanzone et al, 2003). Recipient consumers for these subsidies include riparian vegetation (nutrient addition from salmon carcasses), invertebrates (decomposer fauna for carcasses, algae and detritus) and spiders, birds, bats and similar organisms (emerging aquatic invertebrates). In keeping with the suppositions above, such work has focused on systems with 35 large contrasts in donor to recipient primary productivity and strongly oligotrophic recipient habitats. In this study, we conducted a manipulative field experiment to test whether the abundance of emerging aquatic insects affected the distribution and abundance of riparian web-building spiders in a highly productive rainforest of coastal British Columbia, Canada. Given the high net primary productivity of the recipient habitat at our study site, we predicted that the response of web-building spiders to aquatic insect subsidy exclusion would be small and limited to those spiders with a strong dependence or specialization on adult aquatic insects. Methods S t u d y Site The field experiment was conducted from May through July of 2004 in two similar headwater streams (Spring Creek and East Creek; Table 1) within the Malcolm Knapp Research Forest ( M K R F ; 49°18 '40"N, 122°32'40"W). The forest is located in the Pacific coastal rainforest of southwestern British Columbia. Average mean air temperature ranges from a low of 2°C in January to a high of 16°C in July. Precipitation is approximately 2500 mm annually. More than 70% of this precipitation falls between October and March. The coastal temperate rainforests of British Columbia are substantially more productive (above ground N P P estimated at 1050 - 1300 g C m-2 yr-1 from a model calibrated for the M K R F , Forest Ecosystem Modell ing, 2006) than the headwater streams that flow through them (production of algal biomass estimated at 3.64 g C m-2 yr-1; data adapted from Kiffney, Richardson & Feller, 2000). The dominant vegetation surrounding these streams included red alder (Alnus rubra) and vine maple (Acer circinatum) with a canopy composed largely of western hemlock (Tsuga heterophylla), western redcedar (Thuja plicata) and Douglas-fir (Pseudotsuga menziesii). Understorey vegetation immediately adjacent to the streams consisted of 1-2 m tall shrubs, particularly salmonberry (Rubus spectabUis) and huckleberry (Vaccinium ovatum).The immediate riparian cover of East Creek is 36 more strongly deciduous than Spring Creek, with large numbers of red alder dominating the stream banks. F i e l d e x p e r i m e n t Between May 16 t h and May 19 t h, 2004 two greenhouse-type exclosures were constructed on each of the two streams in the M K R F (Figure 1 A ) . Exclosures were constructed from transparent plastic sheeting supported by semi-circular P V C frames that were anchored to the banks. The edges of the plastic sheeting were partly dug into the stream bank to form a complete seal over the streambed. Each exclosure covered a 50 m long stream reach, and was separated by 50 m long stream reaches that served as controls. Controls and exclosures alternated from the upstream direction (Figure IB) . We sampled orbweaving spider abundance by using timed vegetation shake samples (Costello & Daane, 1997). We used a 0.3 m tray with steep sides and shook vegetation directly over this tray for 20 seconds. Individual spiders were removed from the tray using an aspirator and stored in 75% ethanol until identified. Orbweaving spiders were collected in this fashion in the week prior to construction of the exclosures and every week for 10 weeks following construction of the exclosures with four subsamples taken in the middle 15 m of each reach within 2 m o f the stream wetted edge, two on each side of the stream (Figure IB) . Precise collecting locations within a reach varied between weeks. The abundances of flying aquatic and terrestrial insects were estimated each month by sticky trap sampling. Traps were composed of Tanglefoot (The Tanglefoot Company, Grand Rapids, MI) thinly spread on one side of an acetate sheet (each sheet represents a 600 c m 2 surface area) and suspended between two garden stakes approximately 1.5 to 2 m above the ground facing the stream in roughly the same locations as spider shake samples. Sticky trap samples were set for 7 days in the middle of each month (May, June and July), and collected samples were frozen until sorted. 37 Adult spiders were sorted and identified to species. Insects were identified to Order, or to Family in the case of Diptera, and assigned to either an aquatic or terrestrial group based on published life history details (Merritt & Cummins, 1996). Length and width measurements of insects were determined with an optical micrometer to the nearest 0.01 mm. Biomass estimates of spiders were based on measurements of the length of the tibia-patella of the first pair of walking legs (Higgins, 1992). We used published length-mass regression equations to determine biomass of spiders and aquatic and terrestrial insects (Rogers, Hinds & Buschbom, 1976; Sabo, Bastow & Power, 2002a; Sample et ai, 1993). Statistical Analyses Repeated measures and nested designs involve spatial and temporal autocorrelation that violate assumptions of independence of data points necessary for conventional general-linear modelling (Buckley, Briese & Rees, 2003). We used mixed-effects models that can account for the correlated error structures present in our data. B y combining both repeated measures and spatially nested random effects in a general linear mixed effects ( G L M E ) model we mitigate problems of non-independence and pseudoreplication - the combination of random spatial effects and repeated measures made the use of this technique necessary. The abundance and biomass of flying aquatic and terrestrial insects were each analysed using repeated measures A N O V A with treatment (ambient and reduced insects) and stream (East and Spring Creeks) as main factors, time as the repeated measure (3 sampling periods) and sticky trap samples as replicates. A s multiple samples from each reach were not independent, these data were combined to expand the total area of habitat sampled (4 subsamples in each reach combined into one weekly sample, n = 8 per sampling period) for both insect and spider abundances. We also used repeated measures A N O V A with the abundance of all spiders as the response variable, treatment and stream as factors and date (10 sampling periods) as the repeated 38 measure. Only adult spiders were utilized in statistical analyses since the identification of juveniles is less reliable. Subsequent repeated measures A N O V A s were performed separately for the most abundant spider families (5 families each representing > 5% of total abundance). We applied a sequential Bonferroni procedure to correct for multiple tests. Stream was considered a random factor in all models as this has the advantage of using fewer degrees of freedom. We used the Satterthwaite approximation to estimate denominator degrees of freedom for both fixed and random effects as recommended by Schabenberger and Pierce (2002). The hypothesized time correlation structure used for these models was heterogeneous autoregressive (i.e. correlation between samples is assumed to decrease as separation in time increases). The assumption of a correct covariance model was examined using a likelihood ratio test (with a yr2 distribution) against models with compound symmetry. Heterogeneous autoregressive was the better covariance model for all insect and spider data. We also used likelihood ratio tests to assess the contribution of the two spatial correlation parameters included as random factors (stream and the stream by treatment interaction). For each model, we first checked that the assumptions of normally distributed data and linearly related fixed-effects means were met by examining residual vs. predicted plots and normal probability plots. A l l data required In transformation to meet these assumptions. A l l analyses were conducted using P R O C M I X E D in the statistical package S A S v 9.0 (SAS Institute Inc, Cary, U S A ) . Results Flying aquatic and terrestrial insect abundance The greenhouse cover significantly reduced the abundance and biomass of flying insects of aquatic origin in exclusion reaches relative to control reaches (abundance, F 1,5.23 = 15.07, p < 0.01; biomass F12.1 = 16.35, p = 0.05; Table 2; Figure 2A). Adult aquatic insect biomass in the exclusion reaches was 55.9 % lower than in control reaches when all dates were combined while 39 adult aquatic insect abundance was 62.9 % lower. Flying terrestrial insect abundance and biomass were unaffected by the exclusion treatment (abundance, F15.36 = 0.08, p = 0.79; biomass, F i ^ = 0.02, p = 0.89; Table 2; Figure 2C). Neither time, stream nor the treatment by stream interaction were significant factors for either aquatic or terrestrial insects although there was trend towards overall greater abundances of emerging aquatic invertebrates at East Creek (deciduous canopy). There was no significant change in terrestrial insect abundance or biomass over the sampling period. In control reaches (representing ambient conditions) flying aquatic insects were 6.49 times more abundant than flying terrestrial insects across the entire sampling period and had 4.25 times greater biomass. This pattern did not change over the sampling period; there was no evidence of alternating peaks in abundance between these two prey sources over the three month study period (Figure 2B and 2D). Spider density There was no effect of future exclosure site, stream or the stream by exclosure site interaction on spider abundance in the week prior to construction of experimental exclosures (effect of future exclosure site: F 1 5 = 0.50, p = 0.51). A total of 26 spider species, representing 9 families were collected during the experiment (Table 3). The experimental reduction of aquatic insects depressed the overall abundance of spiders adjacent to exclusion reaches (treatment, F 1,16.1 = \11.59, p < 0.01). There was no significant effect o f time or the interaction of time and treatment. The random effect of stream was not significant across all adult spiders and there was no significant interaction between stream and treatment although there was a trend towards greater overall spider abundance at East Creek. Five families were present in abundances large enough to merit further analyses (Araneidae, Hahniidae, Linyphiidae, Tetragnathidae, Theridiidae, each representing > 5% of total abundance). Repeated measures A N O V A s for mean abundances of the most common families showed that four of the five families included in the analysis were 40 significantly depressed by the exclusion of aquatic insects (Table 4; Figure 3). There was no significant interaction between stream and treatment for any of the spider families. Tetragnathids did significantly decrease in abundance over the sampling period (Fc.,36.2 = 2.22, p = 0.04), while only linyphiids differed in abundance between the two streams (p = 0.04). Hahniids and araneids showed the largest magnitude responses to aquatic insect exclusion, being 89.1 % and 85.7 % lower in exclosure reaches relative to control reaches, while linyphiids and tetragnathids were 50.1 % and 42.9 % lower respectively. Discussion This study demonstrates that predators in highly productive terrestrial habitats can respond strongly to trophic subsidies. Most previous studies have focused on the case where local production is largely absent and consumers are obligately dependent on allochthonous contributions. For example, Paetzold et al. (2006) noted that large differences in productivity, such as occurred at their river cobble bar study site, should result in greater transfers of energy from donor to recipient habitats (via incorporation of subsidies by recipient habitat dwelling consumers). They found that ground-dwelling arthropods showed a substantial numerical response to the exclusion or addition of drifting and emerging aquatic invertebrate subsidies. Polis and Hurd (1995) found that, in most years allochthonous marine inputs controlled the dynamics of web-building spiders on dry G u l f of California islands - allowing large populations of consumers to persist despite terrestrial primary productivity fluctuations. These and other empirical results have led to an assumption that the effects of subsidies w i l l only be significant when recipient primary productivity is lower than that of the donor system. Our results are the first to test that assumption in a highly productive terrestrial setting. Although yearly aquatic N P P was substantially lower than terrestrial N P P , the average emerging aquatic insect abundance was 5.9 times higher than terrestrial insects. This indicates that assumptions about subsidy 41 movements based purely on ratios of N P P may be misleading. Large subsidies (relative to terrestrial production) of emerging aquatic insects occurred despite low stream N P P . This is perhaps not entirely surprising given the well-known relationship between secondary production in these headwater streams and detrital inputs (from the surrounding forest). In these systems it appears that detrital inputs from the forest to streams is driving growth and development of aquatic invertebrates which feeds energy back into the surrounding forest. Within the stream environment, these detrital inputs may be converted to insect biomass at a substantially higher rate than would occur for the equivalent material on the forest floor (Shurin, Gruner & Hillebrand, 2006). Streams may therefore be important bioreactors for converting relatively recalcitrant forest litter into energy sources that are available to higher trophic levels in terrestrial settings. Results from our system showed that spiders in diverse families, with widely divergent web morphologies and capture techniques, tracked aquatic insect subsidies for some portion of their diet. In their study of the effects of aquatic insect exclusion on riparian spiders in Horonai, Japan, Kato et al. (2003) found that horizontal orb-weavers were noticeably affected while vertical orb-weaving and sheet-web weaving spiders were not. They attributed this result to the different feeding strategies implied by the web morphologies and thus prey preferences of these distinctive families. Sheetweb weavers (Linyphiidae) have a high investment in three-dimensional web structures that make it costly to move in order to track spatially and temporally ephemeral resources. Vertical orb weavers (Araneidae) construct large vertical webs that are structurally more suited to catching larger, faster-flying terrestrial prey (Foelix, 1996; Olive, 1982). In contrast, the large open webs and non sticky silk of horizontal orb weavers are often viewed as adapted to the capture of small or weakly flying insects (Olive, 1982) and represent a lower investment that may promote resource tracking. The insect exclusion in this study generated a response in a larger number of spider groups, including those that are not associated 42 with resource tracking of aquatically derived insects. This unexpected pattern of response by non-specialists on aquatic resources suggests that, in the riparian habitats of the M K R F , adult aquatic insects play a disproportionately large role in determining the distribution of many web-building spiders. This may be particularly true when subsidy resources form a large fraction of total available resources or more broadly when the ratio of subsidy to equivalent local resources is greater than one. The high abundance of aquatic insects evidently supports groups of spiders with a broader range of capture techniques than may be occurring where aquatic insects form a much smaller proportion of the overall prey base. Recent studies have emphasized the importance o f alternating periods of productivity between habitats, creating seasonally reciprocal flows of energy (Kato et al, 2003; Nakano & Murakami, 2001; Takimoto, Iwata & Murakami, 2002). At the interface between streams and forests, the seasonal emergence of aquatic insects may be temporally offset from the time of maximum secondary terrestrial productivity - influencing the distribution and abundance of generalist consumers such as riparian spiders (Kato et al., 2003) and birds (Nakano et al., 2001; Uesugi, 2002). Terrestrial invertebrate production should be greatest during the spring and early summer as deciduous trees leaf out, coniferous trees put on new growth and understorey vegetation is flush. While previous studies have shown strong variability in the relative availability of aquatic and terrestrial insect prey, this pattern was not detectable in the three months (spring through mid-summer) of this experiment. In our study, availability of emerging aquatic insect prey was always greater than terrestrial insect abundance in the riparian forest. It seems probable that, in the highly productive rainforests of the Pacific Northwest, emerging aquatic insects are more abundant than flying terrestrial invertebrates at most times in the year. However this production and export of material to the riparian forest is itself a consequence of terrestrial detrital inputs. The high secondary productivity of headwater streams in this region supports a diverse group of invertebrate predators. This secondary productivity is itself created 43 by the higher trophic efficiency within streams, based on terrestrially-derived materials indicating that for small headwater streams and their adjacent forests, the land-water boundary is particularly porous. Summary In our system, existing theory predicted that responses to subsidy exclusion would be weak, or constrained to family groups known to specialize on aquatic insects as a subsidy resource. Contrary to this expectation, the overall abundance of several spider families decreased with subsidy exclusion, including families not thought to be particularly sensitive trackers of aquatic insect abundance. This suggests that even in highly productive settings, specific subsidy types can have important effects on fauna. Although allochthonous resources may indeed contribute close to 100% of productivity in some habitats (e.g. headwater streams, caves, snowfields, islands etc; Vanni et al., 2004), the importance of subsidies appears nearly as great in habitats with substantial in situ primary productivity, such as that described here. This study provides direct evidence that the local distribution of multiple families of riparian orb-weavers can be controlled by changes in inputs of emerging aquatic insects, particularly when that input to adjacent terrestrial habitats is sufficiently large compared to equivalent terrestrial resources (terrestrial insects). The high relative abundance of aquatic insects in headwater, temperate rainforest streams may impact riparian food web dynamics particularly where terrestrial insect production is limited, even though riparian forest primary productivity is relatively high. Subsidy effects appear to be largest when they subsidize a system with comparable resources that are at low levels - the pool of labile or available carbon is often not equivalent to the overall contrast in donor and recipient habitat primary productivity, particularly when the focal consumer is a predator. The development of predictions about where subsidies are likely to produce the greatest effects in recipient habitats w i l l require more specific 44 studies that examine the nature of the subsidy relative to the nature of available resources in the recipient habitat. Acknowledgements We thank Y i x i n Zhang, Deirdre Leard, Conan Phelan, Tatiana Lee, Trent Hoover and other members o f the Stream and Riparian Research group (StaRR) for assistance in the field and lab. Ross Thompson and Rebecca Best provided statistical guidance. Robb Bennett provided expert assistance with spider species identification. The authors acknowledge funding assistance from the Natural Sciences and Engineering Research Council (Canada) and the Forest Sciences Program (British Columbia). 45 / Table 3.1. Comparison of physical and hydrological characteristics of East Creek and Spring Creek. Means ± 1 standard error. East Creek Spring Creek Gradient (%)f 1.9 1.8 Elevation (m asl) 154 160 Watershed area (ha) 35.0 44.0 Channel wetted width (m) (mean ± SE) f 2.4 i 0.2 3.0 ± 0.6 Mean water depth (m) (mean ± SE) 0.18 ± 0 . 0 1 0.14 ± 0 . 0 2 Mean velocity during experiment (m/s) <0.14 <0.10 Mean annual discharge (L/s) * 32 ± 3 n/a Substrate size (% distribution ± SE): § Sand and silt (< 2mm) 6.3 ± 1.3 70 ± 10.8 Gravel (2mm - 64 mm) 70.3 ± 3 . 0 26.3 ± 9 . 7 Cobble (64 mm -130 mm) 22.3 ± 3 . 7 3.75 ± 1.3 t Kiffney et al 2000 § Boss and Richardson 2002 * J. Caulkin unpublished data 46 Table 3.2. Repeated measures A N O V A s for mean of abundance and biomass per trap (n = 8) of flying aquatic and terrestrial insects. Contributions of random effects (stream, stream*treatment) to the model were assessed using a x 2 test (df = 1) o f the difference in residual log likelihood for the full and reduced model. Significant p-values are highlighted in bold text. Flying aquatic insects Flying terrestrial insects Abundance Biomass Abundance Biomass Effect Test p Test P Test P Test P Treatment Fi, 5 .23= 15.07 0.01 F i , 2 . i = 16.35 0.05 F l , 5.36 = 0.08 0.79 F l , 5.82 = 0.02 0.89 Time F2,9.99 = 3.46 0.07 F2,13.5 = 0.87 0.44 F2,,o.3 = 0.09 0.92 F 2 , io.9 = 0.15 0.87 " Time* treatment F2^ 9.99 = 4.25 0.06 F2,13.5 = 1.00 0.39 F2^ 10.3 = 0.57 x - 0 0.58 F2,10.9 = 0.72 0.51 Stream x 2 = o 1.00 x 2 = o 1.00 1.00 x 2 = o 1.00 Stream*treatment x 2 = o 1.00 X 2 = 0.9 0.34 x 2 = o 1.00 x 2 = o 1.00 Table 3.3. List o f families and species of vegetation-dwelling spiders collected at Spring Creek and East Creek in the Malcolm Knapp Research Forest indicating relative percentages (%) of abundance. Trap type Family Species Percent of total abundance (adults) Vertical orb-webs Araneidae Araneus nordmanni 0.14 Araniella displicata 0.01 Cyclosa conica 0.14 Larinioides sclopetarius 4.8 Active hunters (with Clubionidae Clubiona pacifica 3.5 silk retreat) Dwarf sheet-webs Hahniidae Cryphoeca exlineae 0.9 Dirksia cinctipes 4.1 Sheet-webs Linyphiidae Helophora sp. 3.8 Linyphiid morphospecies 0.3 Microlinyphia mandibulata 5.2 Neriene digna 1.0 Pityohyphantes costatus 3.7 Walckenaeria kochi 0.3 Active hunters (no web) Philodromidae Philodromus rodecki 1.7 Active hunters (no web) Salticidae Saltieid spp. 0.1 Horizontal orb-webs Tetragnathidae Tetragnatha versicolor 1.7 Metellina curtisi 14.2 Tangle-webs Theridiidae Emblyna peragrata 0.3 Enoplognatha ovata 7.0 Pholcomma sp. 0.1 Theridiid morphospecies 0.1 Rugathodes sexpunctatus 43.3 Theridion varians 0.4 Hackled orb-webs Uloboridae Hyptiotes gertschi 3.3 48 Table 3.4. Results of separate repeated measures A N O V A s for mean abundance of adult spiders in five families each representing greater than 5% of total spider abundance. Contributions of random effects (stream and stream by treatment) to the model were assessed using a x 2 test (df = 1) of the difference in residual log likelihood between the full and reduced model. Significant p-values are highlighted in bold text. A l l p values were adjusted using a sequential Bonferroni procedure. Araneidae Hahniidae Linyphiidae Tetragnathidae Theridiidae Test p Test p Test P Test p Test p Treatment Fi, ,8.9 = 9.38 <0.01 F 1,26.9= 17.45 <0.01 F, , ,5.4 = 4.53 0.05 F,, ,9.8 = 4.69 0.04 F i , 17.9 = 1.20 0.29 Time F934.4 = 0.82 0.61 F 9 ,33 .4 = 1.28 0.28 F 9 ,37 .7 =2.04 0.06 F9,36.2 = 2.22 0.04 F 9 , 3 2 . 9 = 1.75 0.12 Time* Treatment F 9 34 4 =0.75 0.75 F 9 ' 3 3.4= 1.00 0.46 F 9 37 7 — 0.82 0.60 F936 .2 = 0.83 0.59 F9,32.9 = 0.5 8 0.81 Stream x2 = o 1.00 X2'=0.7 0.40 X2'=4.1 0.04 x2=i.o 0.32 X 2 =2.2 0.14 Stream* Treatment >C2 =0.1 0.75 X 2 =0.7 0.40 X 2 =3.8 0.08 x2=o.i 0.75 X 2=0.1 0.75 B. Greenhouse-type cover s u b — » 5 0 m 5 0 m 5 0 m 5 0 m Control Exclusion Control Exclusion Figure 3.1. A. Upstream exclosure along 5 0 m of East Creek, Malcolm Knapp Research Forest (photo credit Y . Zhang) B. Diagrammatic representation of exclosure placement and experimental design on a single stream. 5 0 Aquatic Invertebrates control exclosure control exclosure 45-40-- .35-Q. | 30-£ 2 5 -I M H TO TO IO-H 5-1 0 Terrestrial Invertebrates D control exclosure 2 -| 1.8 1.6 1.4 1.2 1 0.8 -| 0.6 0.4 0.2 0 60 5Q-\ a. £ 4 0 -t 01 §20-1 CD •10 control exclosure May June July Figure 3.2. Effect of exclosure treatments on the (A) overall mean biomass and density of aquatic invertebrates (B) monthly mean biomass of aquatic invertebrates at control (filled bars) and exclosure (open bars) reaches (C) overall mean biomass and density of terrestrial invertebrates and, (D) monthly mean biomass of terrestrial invertebrates at control (filled bars) and exclosure (open bars) reaches. Values are least squares means ± 1 SE. 51 • control • exclosure Theridiidae Araneidae Tetragnathidae Liny phi idae Hahniidae Figure 3.3. Effect of aquatic insect exclosure on the abundance of spiders in five families. Stars indicate significant differences between control and exclosed reaches using differences of treatment means from the mixed model. Values are least squares means ± 1 SE. 52 Works Cited Bastow, J. L . , J. L . Sabo, J. C. Finlay, and M . E . Power. 2002. A basal aquatic-terrestrial trophic link in rivers: algal subsidies via shore-dwelling grasshoppers. Oecologia 131:261-268. Buckley, Y . M . , D . T. Briese, and M . Rees. 2003. Demography and management of the invasive plant species Hypericum perforatum. I. Using multi-level mixed-effects models for characterizing growth, survival and fecundity in a long-term data set. Journal of Applied Ecology 40:481-493. Collier, K . J., S. Bury, and M . Gibbs. 2002. A stable isotope study of linkages between stream and terrestrial food webs through spider predation. Freshwater Biology 47:1651-1659. Forest Ecosystem Modelling. 2006. Canadian Research Chair in Forest Ecosystem Modell ing. http://www.forestry.ubc.ca/modelresearch/index.htm. Department of Forest Sciences, Faculty of Forestry, University of British Columbia. Higgins, L . E . 1992. Developmental plasticity and fecundity in the orb-weaving spider Nephila clavipes. The Journal of Arachnology 20:94-106. Jackson, J. K . , and S. G. Fisher. 1986. Secondary production, emergence and export of aquatic insects of a Sonoran desert stream. Ecology 67:629-638. Kato, C , T. Iwata, S. Nakano, and D . Kish i . 2003. Dynamics of aquatic insect flux affects distribution of riparian web-building spiders. Oikos 103:113-120. Kiffney, P. M . , J. S. Richardson, and M . Feller, C. 2000. Fluvial and epilithic organic matter dynamics in headwater streams of southwestern British Columbia, Canada. Archiv fur Hydrobiologie 149:109-129. Mason, C. F. , and S. M . Macdonald. 1982. The input of terrestrial invertebrates from tree canopies to a stream. Freshwater Biology 12:305-311. Merritt, R. W. , and K . W. Cummins. 1996. A n introduction to the aquatic insects of North America. Kendall/Hunt Publishing Company, Dubuque, Iowa. Nakano, S., and M . Murakami. 2001. Reciprocal subsidies: dynamic interdependence between terrestrial and aquatic food webs. Proceedings of the National Academy of Sciences -U S A 98:166-170. Olive, C. W. 1982. Behavioural response of a sit-and-wait predator to spatial variation in foraging gain. Ecology 63:912-920. Paetzold, A . , J. F. Bernet, and K . Tockner. 2006. Consumer-specific responses to riverine subsidy pulses in a riparian arthropod assemblage. Freshwater Biology 51:1103-1115. Polis, G . A . , W. B . Anderson, and R. D. Holt. 1997. Toward an integration of landscape and food web ecology: the dynamics of spatially subsidized food webs. Annual Review of Ecological Systematics 28:289-316. 53 Polis, G . A . , and S. D . Hurd. 1995. Extraordinarily high spider densities on islands: flow of energy from the marine to terrestrial food webs and the absence of predation. Proceedings of the National Academy of Sciences - U S A 92:4382-4386. Polis, G . A . , and S. D . Hurd. 1996. Linking marine and terrestrial food webs: allocthonous input from the ocean supports high secondary productivity on small islands and coastal land communities. The American Naturalist 147:396-423. Richardson, J. S. 1991. Seasonal food limitation of detritivores in a montane stream: an experimental test. Ecology 72:873-887. Rogers, L . E . , W. T. Hinds, and R. L . Buschbom. 1976. A general weight vs. length relationship for insects. Annals of the Entomological Society of America 69:387-389. Rose, M . D . , and G. A . Polis. 1998. The distribution and abundance of coyotes: the effects of allocthonous food subsidies from the sea. Ecology 79:998-1007. Sabo, J. L . , J. L . Bastow, and M . E . Power. 2002. Length-mass relationships for adult aquatic and terrestrial invertebrates in a California watershed. Journal of the North American Benthological Society 21:336-343. Sabo, J. L . , and M . E . Power. 2002. River-watershed exchange: effects of riverine subsidies on riparian lizards and their terrestrial prey. Ecology 83:1860-1869. Sample, B . E . , R. J. Cooper, R. D. Greer, and R. C. Whitmore. 1993. Estimation of insect biomass by length and width. American Midland Naturalist 129:234-240. Sanzone, D . M . , J. L . Meyer, E . Mart i , E . P. Gardiner, J. L . Tank, and N . B . Grimm. 2003. Carbon and nitrogen transfer from a desert stream to rioarian predators. Oecologia 134: 238-250. Schabenberger, O., and F. J. Pierce. 2002. Contemporary Statistical Models for the plant and soil sciences. C R C Press Ltd. Sears, A . L . W., R. D . Holt, and G. A . Polis. 2004. Feast and famine in food webs: the effects of pulsed productivity. Pages 359-386 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Shurin, J. B . , D . S. Gruner, and H . Hillebrand. 2006. A l l wet or dried up? Real differences between aquatic and terrestrial food webs. Proceedings of the Royal Society of London:B 273:1-9. Takimoto, G. , T. Iwata, and M . Murakami. 2002. Seasonal subsidy stabilizes food web dynamics: balance in a heterogeneous landscape. Ecological Research 17:433-439. Uesugi, A . 2002. Expanding effects of stream-driven allocthonous subsidies on watershed bird communities. Master's Thesis. Hokkaido University. Varmi, M . J., D . L . DeAngelis, D . E . Schindler, and G . R. Huxel. 2004. Overview: Cross-habitat flux of nutrients and detritus. Pages 3-11 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. 54 Wipf l i , M . S. 2005. Trophic linkages between headwater forests and downstream fish habitats: implications for forest and fish management. Landscape and Urban Planning 72:205-213. 55 Chapter 4. Timing of resource subsidy alters growth and fitness of the spider Tetragnatha versicolor (AraneaeiTetragnathidae)^ Introduction Many strongly seasonal environments receive short duration pulses of resource subsidies from adjacent habitats. Such temporally and spatially variable pulses of prey or nutrients have garnered significant attention from the ecological community in recent years (e.g. Szepanski et al. 1999, Gende and Wil lson 2001, Mendelssohn and Kuhn 2003, Briers et al. 2005, Orr et al. 2005). However, there have been few studies of the ultimate consequences of this temporal variability in resource subsidies for fitness of recipient organisms (but see Sabo and Power 2002). Resource subsidies, by definition, must represent a net benefit to recipient consumers (Polis et al. 1997). While many of these resource flows are seasonal - as in the summer emergence of aquatic insects - the timing of their arrival within a season has implications for organisms that show growth plasticity or experience constraints and thresholds related to life history phenologies. Rates of food uptake affect an organism's fitness in terms of survivorship and reproductive potential (Olive 1982). Being larger, and being so earlier, often confers fitness advantages. Across many taxa, female fecundity increases with increased body size and decreases with increased age at maturity (Blanckenhorn 1991, Heiling and Herberstein 1999, Danielson-Francois et al. 2002, De Block and Stoks 2005). For a wide variety of organisms, variable nutrition, particularly during early development, can influence adult traits related to fitness (e.g. Higgins 1992, Beck 1997, Amal in et al. 1999, Higgins and Rankin 2001, Metcalfe and Monaghan 2001, Phillips 2004). A n organisms' growth rate affects lifetime reproduction (Fischer et al. 2004), average lifespan (Parsons 2004) and survival to adulthood (Higgins and Rankin 2001). * Submitted as: Marczak, L . B . and J.S. Richardson. Timing of resource subsidy alters growth and fitness of the spider Tetragnatha versicolor (Araneae:Tetragnathidae). Functional Ecology 56 Following periods of food shortage, organisms may show accelerated growth (Metcalfe and Monaghan 2001, Jespersen and Toft 2003) which can exceed that normally observed on the same ration. Periods of rapid growth may be induced by environmental variation in food supply and may carry inherent ecological and physiological risks as well as clear benefits (Fischer et al. 2004). Ecological costs include an increased exposure to predators and parasites during increased foraging (Abrams et al. 1996). Increasingly, researchers have demonstrated that there may also be intrinsic physiological costs associated with these periods of rapid growth including a decreased resistance to starvation and other environmental stresses (Metcalfe and Monaghan 2001, Jespersen and Toft 2003, Fischer et al. 2004, Yearsley et al. 2004). These costs of rapid growth may be traded off against the potential for organisms to catch up to or exceed the size attained by conspecifics that did not experience a resource restriction (Yearsley et al. 2004). Growth compensation may allow an organism to retain the optimal seasonal timing of its life cycle, to mitigate the effects of poor provisioning from the egg stage (maternal effects), poor early growth or late date of birth on final body size. This ability is not universal, it varies by species and has been shown to vary between sexes or to be limited to certain life stages or traits (Metcalfe and Monaghan 2001). Organisms inhabiting highly seasonal environments are likely to be exposed to costs associated with variable rates of growth. Spiders living in riparian areas receive frequent, episodic bursts of aquatic prey (subsidies) in addition to variable terrestrial prey abundance. Many spiders have been shown to have a combination of very low metabolic rates and a capacity to opportunistically ingest large amounts of prey (Foelix 1996). Orb-weaving spiders are generally assumed to be sit-and-wait predators that gorge when prey are abundant (Foelix 1996). This strategy should result in periods of very rapid increases in mass. Spiders living in strongly seasonal environments, where synchronized aquatic insect emergence results in cycles of prey availability and food limitation, might be particularly vulnerable to costs associated with rapid 57 growth or delayed costs of compensatory growth due to variability in prey abundance during different stages of spider development. Although riparian spiders have been the subject of studies of changes in distribution or abundance in response to subsidies (e.g. Polis and Hurd 1995, Collier et al. 2002, Kato et al. 2003, Sanzone et al. 2003, Briers et al. 2005), the impact of aquatic insect subsidies on the reproductive success or fitness of riparian spiders has yet to be examined. Depending on the timing, consistency and extent of aquatic insect movements into riparian habitats, species that obtain a significant proportion of their diet from aquatic insects may depend on this subsidy to complete their life cycles. We tested the ability of the long-jawed orbweaver, Tetragnatha versicolor to exhibit both plastic growth in response to a variable, food environment and its ability to compensate for early nutritional shortages. We then compared the effect of resource subsidies received early in development (early rapid growth) and in the penultimate molt (later compensatory growth). We conducted three laboratory experiments in which juvenile T versicolor spiders were randomly assigned to treatments which varied food availability quantitatively and temporally. We address the following questions: (1) Is there a detectable increase in mortality risk associated with rapid growth? (2) Are individual T versicolor capable of compensatory growth responses? (3) How does the variable timing of resource abundance intersect with tradeoffs in fitness consequences of rapid or compensatory growth? Methods Study organism The orb weaving spider Tetragnatha versicolor Walckenaer 1842 (Araneae: Tetragnathidae) is a broadly distributed species with a consistent association with freshwater habitats, largely feeding on emerging aquatic insects (Williams et al. 1995). Although aquatic insects are emerging throughout the year in the region where this study was conducted, peaks in emergence occur at 58 different times from different streams (J. Richardson unpublished data). In coastal British Columbia, this spider has an adult body length of 6 to 10 mm (tibia-patella length or T P L ) and has a mainly annual life cycle with multiple cohorts of overlapping adults and juveniles during the summer reproductive period. For our experiments we used either wi ld caught juveniles (experiment 1: variable growth and mortality risk) or laboratory reared individuals from wi ld collected females (22 spiderlings from 3 wi ld caught females for compensatory growth experiment, 70 spiderlings from 10 wi ld caught females for timing of subsidy experiment). W i l d collected spiders were hand-picked from webs adjacent to small headwater streams in Malcolm Knapp Research Forest ( M K R F ; 49°18 '40"N, 122°32'40"W), in coastal British Columbia, between May and June 2005 and in June 2006. W i l d collected, visibly gravid female spiders were kept for one week in the lab at an average temperature of 21°C until an egg case was laid. Lab-reared spiders were maintained on an ad libitum diet of local honeybee pollen and Drosophila melanogaster reared in a medium enriched with crushed high protein dogfood (Jespersen and Toft 2003) until they successfully moulted to the second instar (7-10 days). A l l experimental spiders were reared in individual Perspex containers (8 cm height by 4.5 cm diameter) screened with fine mesh on the top and equipped with a 1 cm deep layer of sand which was kept semi-saturated to provide humidity. These microcosms were maintained at 21 °C on a long day cycle (16:8) for the duration of each experiment. A plastic straw was placed vertically in the container to provide web attachment points. A l l spiders in experiments were fed adult Drosophila melanogaster (average individual mass: 1.07 ± 0.02 mg). Experiment 1. Food limitation effects on growth and mortality risk Wild-collected juvenile spiders varying in initial size between 3.0 and 5.0 mm (first walking leg tibia + patella length or T P L ) were used to test the hypotheses that growth rate is plastic and that 59 rapid growth may carry inherent physiological costs for 71 versicolor. Individual spiders were randomly assigned to one of five food levels, calculated as a percentage of the average initial mass of a spider at the beginning of the experiment (Table 1). The experimental treatments ended after 40 days or when each individual moulted. Immediately following each moult, T P L , carapace width (CW) and mass were recorded. Length of time between moults (intermoult period) or time to death were also recorded. The intermoult period for spiders dying during a moult was measured as the time to death or moribund state (hanging or fallen from web with no response to contact by the observer). Following Higgins and Rankin (2001), we calculated the change in mass as a percentage change per unit time: growth = -1 100 Using the maximum mass prior to the moult (t = intermoult duration in days) as the second measure of mass (m 2). Growth rates were ln(x+l) transformed to successfully improve normality of ratio data. We used an A N C O V A to assess the variation in growth rate between treatments with the initial mass (mi) of spiders as the covariate. Since our assumption of homogeneous slopes was met, we removed the interaction between initial mass and growth rate from the model as recommended by Engqvist (2005). Differences in the length of the intermoult period between the five feeding levels were assessed using A N O V A . We used a x2 test and a 3 by 3 contingency table (PROC G E N M O D in S A S system) to determine the effect of feeding rate on total mortality by treatment. The two lowest feeding levels and the two highest feeding levels were combined (treatments = low, medium, high) to minimize difficulties with small cell sizes. Spiders were scored as either alive, dead or dead during the moult. We confirmed that model assumptions were met using visual assessment of residuals and normal probability plots. Following post-hoc comparisons we used a Bonferroni correction to control the experiment-wise error. 60 Experiment 2. Compensatory growth We conducted a laboratory experiment to test the ability of the long-jawed spider T. versicolor to catch up on growth following nutritional deficiency in the initial period of independent life. Once lab reared spiderlings moulted to the second instar, they were randomly assigned to one of two treatments. Treatments were either a constant (C, n = 11) supply of 2 fruit flies per day for 28 days, and a switching treatment (S, n = 11) consisting of 14 days at one quarter of the food in the constant group (1 fruit fly every second day) followed by 14 days at full food (Figure 1). We measured the T P L and mass (mg) of all spiders on the first day of the experiment, on the day of the switch in food rations and at the conclusion of the experiment. To account for the effect of decreasing food amounts relative to body size, we calculated the natural log transformed rate of growth between the starting mass and mass at the time of food ration switch (rate 1) and the natural log transformed rate of growth between the time of food ration switch and the completion of the experiment (rate 2) and standardized both of these rates by the ration provided in each time step to produce the ration specific growth rates for each time period. We then combined. these response variables in a M A N O V A design. We used separate A N O V A s for each time period as post-hoc tests with a Bonferroni correction to control the experiment-wise error. Additionally, we tested whether differences in growth rates produced differences in final size using an A N C O V A with final mass (mg) as the response variable and initial mass as the covariate. Since our assumption of homogeneous slopes was met, we removed the interaction between initial and final mass from the model as recommended by Engqvist (2005). We used visual assessments of residual and normal probability plots to confirm that data met the assumptions of these models. 61 Experiment 3. Timing of subsidy resource and juvenile development Early resource abundance can produce rapid growth, while later abundance (with early restriction in resources) can produce compensatory growth. Alternately, animals might experience a more constant environment. In order to determine which developmental path wi l l produce the greatest payoff in terms of adult condition and fitness we tested the effect of these different subsidy schedules on measures of spider fitness. Lab-reared juveniles that had successfully moulted to the second instar were randomly assigned to one of three treatments: two variable or switching food levels and a constant "control". Animals assigned to the high:low ( H L , n = 24) treatment received a large pulse of food for the first seven days of the experiment and then a constant low rate of food. Animals in the low:high ( L H , n = 23) treatment received a constant low rate of food until the penultimate week of the experiment when they received a seven day pulse of high food. Animals in the constant (C, n = 23) treatment received the same amount of food each day of the experiment. The length of the experiment (64 days) was selected to ensure that individuals in all 3 treatments received the same total number of prey over the course of the experiment and that only the timing of prey delivery was varied (Figure 2). The experimental treatments ended after 64 days or when each individual died. Spiders that died before the completion of the experiment were eliminated from the growth analysis to avoid biasing overall growth rates. We recorded the length of time between moults (intermoult period) for each spider. Immediately following each moult, T P L (mm) and mass (mg) were measured. We fit second order polynomial equations to the growth data for each spider and used the coefficients from these equations as response variables in a M A N O V A to compare the shape of the growth functions for individual spiders by treatment (Meredith and Stehman 1991). This analysis of linear and quadratic coefficients directly addresses the shape of the response function. Rejection of the hypothesis of no treatment differences between the linear 62 component of the time effect indicates that the trend in linear time (slope) is not the same for all treatments while a similar test for the quadratic time by treatment interaction determines whether different curvature of the response function exists. Differences between pairs of treatments were determined using M A N O V A s for each possible pair of treatments with a Bonferroni correction to control the experiment-wise error. We compared the mean number of moults over the 64 day experiment using a single factor A N O V A and an A N C O V A model to compare the final mass of spiders with treatment as the factor and initial size (TPL) as the covariate. For A N C O V A s , we tested for interactions between the main factor and the covariate and removed these interactions from models when they proved to be non-significant (Engqvisx 2005). We used visual assessments of residual and normal probability plots to confirm that data met the assumptions of these models. We developed a qualitative scale of reproductive development with four categories: no evidence of gonadal tissue (juvenile), minor sclerotization present or minor tissue swellings where gonads wi l l develop (immature), presence of gonads obvious, but not yet complete (subadult) or male palps fully formed, female epigynum sclerotized and ovaries present on dissection (adult). B y the end of our 64 day experiment, only one individual had completed a final molt to maturity, since distinguishing between males and females is not possible for juveniles, and difficult for immature spiders we did not include sex in any of our analyses. The single adult male was excluded from analyses of body size and condition since male size:shape relationships change dramatically at the final molt. We used a x2 test to determine differences in reproductive stage between treatments at the end of the experiment (PROC G E N M O D in S A S version 9.0, S A S Institute Inc, Cary, U S A ) . We assessed final condition using an A N C O V A with final T P L as the covariate, treatment as the main factor and final body size as the response variable (Garcia-Berthou 2001). 63 We used established methods (Bligh 1959, Frings et al. 1972, Al l en 1976, Van Handel 1985) to measure the absolute quantity of glycogen, lipids, protein and sugars present in individual spiders. Results from glycogen and sugar assays were determined by recording the colorimetric change in controls followed by specimens at 625 nm; the optical density of lipid and protein assays were determined at 525 nm (Beckman D U 640 Spectrophotometer, Beckman Coulter Inc., Fullerton, California). We used a M A N C O V A with the amount (pg) of each storage compound as multiple response variables and final body size as a covariate to explore differences in stored reserves between individual spiders in our treatments at the completion of the experiment. We used separate A N C O V A s for each body compound as post-hoc tests to determine which body components were contributing to a significant overall effect in the M A N C O V A design. Final body size was the covariate in these models. We employed a Bonferroni correction to control the experiment-wise error. Homogeneity of the regression lines and other assumptions required for M A N C O V A were tested using standard procedures (Tabachnick and Fidell 2001). Results Experiment 1: Food limitation effects on growth and mortality risk Growth and development in the spiders differed significantly between the five treatments. A t the start of the experiment, there were no differences among the treatment groups in initial size ( A N O V A , F 4 , 4 2 = 1.49, p = 0.22). The rate of growth differed between treatments ( A N C O V A , treatment, F ^ - 8.81, p < 0.001; Figure 3), increasing with increasing food rations (Figure 3). The model was significantly affected by the initial size (covariate) of the individual ( A N C O V A , initial size, F142 = 5.81, p = 0.02). The interaction between initial size and treatment was not significant ( A N C O V A , initial size by treatment, F 4 ) 3 8 = 1.54, p = 0.21) and was removed from the model prior to interpretation. 64 The length of the intermbult period decreased with increasing food levels ( A N O V A , treatment ¥4^2 = 2.47, p = 0.059). There was a significant effect of treatment on survival to the next instar (Pearson x 2 = 23.34, p < 0.001). Spiders in the two,highest feeding regimes were significantly more likely to die during moulting than spiders in lower feeding regimes. Experiment 2. Compensatory growth There were no differences in initial distribution of masses (mg) between treatments ( A N O V A , Fi,2o = 0.80, p = 0.381). Overall growth rates between the two treatments differed significantly across both time periods ( M A N O V A , Wi lks ' exact Lambda F 2 j i 9 = 18.29, p O.0001). Post-hoc comparisons show that spiders in the switching treatment grew significantly slower relative to their ration during the first time period and significantly faster during the second time period than spiders with constant food conditions (Figure 4). There were no differences between the two treatments in the mass of spiders at the conclusion of the experiment ( A N O V A , Fi^o = 0.735, p = 0.402). Experiment 3. Timing of subsidy resource in juvenile development Growth There were no differences among the treatment groups in initial size (F2,67 = 0.05, p = 0.95, n = 70 individuals). The growth curves differed significantly between treatments ( M A N O V A , effect of treatment, Wi lks ' exact Lambda F6,i2o = 51.57, p < 0.001; Figure 5). Both the linear (slope) and quadratic (curvature) terms differed between treatments (linear F 2 ,6 2 = 89.7, p < 0.001, quadratic F 2 i 6 2 = 131.9, p < 0.001). The overall rate of growth for spiders in the L H and C treatments did not differ (no difference in linear slopes, p = 0.13) while spiders in the H L treatment grew at an overall slower rate than spiders in the other two treatments ( L H ^ H L ^ C , all contrasts p <0.001). The curvature of each growth curve was significantly different for all 65 contrasts (all p < 0.05). The final weight achieved by spiders differed between treatments ( A N O V A , ¥2,62 = 19.86, p < 0.0001); spiders in the H L treatment were significantly lighter than spiders in the L H and C groups (Figure 6A). Timing of food abundance had a significant effect on the total number of moults per treatment ( A N O V A , effect of treatment, ¥2,62 = 3.13, p = 0.05) with spiders in the H L completing significantly fewer moults than spiders in either the control or L H treatments. Reproductive stage and condition Our analysis of the body condition of spiders at the completion of the experiment showed a significant effect of treatment and the covariate final T P L , but not their interaction which was removed from the model ( A N C O V A , treatment: F 2, 6o = 5.70, p = 0.005; final T P L : F U 6 o = 81.87, p < 0.001). H L were lighter for their size than animals in the other two treatments. Treatments differed according to the reproductive stage of spiders at the end of the experiment (X2 = 44.9, p < 0.001). Animals receiving the L H switching treatment finished the experiment with more individuals in the penultimate moult before maturity (sub-adults) followed by C and then H L . The relative composition of spider body components differed at the end of the experiment ( M A N C O V A , Wi lks ' Lambda exact F 8,no = 2.89, p = 0.006). Separate A N C O V A s for lipids, proteins, carbohydrates and glycogen showed that only lipids differed by treatment (effect of treatment F2,57 = 4.32, p = 0.018) with spiders in the L H treatment having greater amounts of lipids relative to their body size compared with animals in the H L treatment (mean mass-specific l ipid content for spiders in treatment L H = 197.09 (± 13.22) pg, H L = 118.52 (± 14.22) pg and C = 183.45 (± 13.86) pg). Animals in the constant (C) treatment did not differ from the other treatments with respect to their lipid content once final body size was controlled for (Figure 6B). 66 Discussion Across a variety of taxa, researchers have demonstrated that the early restriction of food followed by a later switch to high food levels produces compensatory growth responses that may exceed the growth of individuals reared on constant high food amounts. This effect has been demonstrated for the larvae of amphibians (Beck 1997), and freshwater (Twombly 1996) and marine invertebrates (Hentschel and Emlet 2000). In our study we found support for both higher mortality rates during the moult with high juvenile growth rates and compensatory growth responses following food restriction. In our final experiment, when juvenile T. versicolor were exposed to large rations of food after a period of initial low food, their overall response was a dramatic increase in body size, developmental stage and lipid storage. When individuals experienced low food following an initial period of food abundance they experienced decreased growth, developmental stage and l ipid storage relative to animals that received either a constant delivery of food or low food followed by a switch to high food. These results suggest that the seasonal timing of the brief availability of very abundant aquatic insects could play a significant role in the fitness and population dynamics of riparian spiders R a p i d growth and growth compensation Although very high rates of growth might be assumed to increase fitness, allowing animals to avoid costs of slow development, or small size, there is a growing body of evidence supporting the existence of both extrinsic ecological and intrinsic physiological costs associated with rapidly increasing mass (Abrams et al. 1996, Higgins and Rankin 2001, Metcalfe and Monaghan 2001). The ecological costs of rapid growth include an increased risk of predation and parasitism associated with the presumed increased foraging necessary to fuel rapid weight gain (Abrams et al. 1996). Physiological costs have been comparatively less well studied but include a decreased 67 resistance to starvation or other environmental stresses (Stockoff 1991) or a decrease in other functions such as development (Arendt 1997). In univoltine populations of Nephila clavipes, Higgins and Rankin (2001) found evidence for physiological costs of rapidly increasing mass. These results are corroborated here for the multivoltine T. versicolor. We found support for an increasing mortality risk for spiders fed very large numbers of prey. A s in the study by Higgins and Rankin (2001) these deaths occurred during moulting. Potential explanations for the death of well-fed T. versicolor juveniles include gut failure following overeating, changes in the functioning or regulation of moulting hormones or nutritional imbalances. In both N. clavipes and T. versicolor the temporal link between the moult cycle and death suggests that very large increases in mass may interfere with the moulting process. Our results further show that Tetragnatha versicolor is able to respond to short-term food deprivation with increased rates of growth relative to individuals kept on a constant ration. Once recovery had begun, growth accelerated and remained high until the weight deficit had been regained. These spiders accomplished this catch-up growth on a lower total ration than provided to the control group. Spiders as a group demonstrate strong resistance to starvation through physical characters such as a heavily reticulated midgut and greatly expandable abdomen (Foelix 1996) which suggest they have a capacity to take advantage of food when it is periodically abundant. The ability for growth compensation may also have evolved from strong selection on spiders to synchronize their reproduction to a specific season. Subsidy timing and spider fitness In this study, individual T. versicolor receiving late subsidies of food following early food restriction did as well or better by all measures of fitness (same final size, same number of moults, same body condition, same quantity of lipid storage, but faster reproductive development) than animals receiving a constant supply of food. These animals did better for all 68 of these measures than individuals that received a large subsidy of food early in their development and were then food restricted for the remainder of the experiment. One possible explanation for how this occurred includes the development of a more efficient metabolism under stress (Parsons 2004). Individuals that receive luxury rations in early development do not need to conserve energy and may use resources less efficiently than animals that initially developed on low food levels. Animals developing on low food rations may experience induced metabolic efficiency and be able to more efficiently use food received later in their development for rapid growth and swift reproductive development (Smith 1976, Parsons 2004). Since this experiment was designed so as to vary only the timing of food provided (while providing equal amounts of food to all individuals over the course of the experiment) it was not possible to follow each individual to maturity. It is probable that spiders in the constant food treatment would have eventually reached maturity at a slightly larger size than those in the low to high switching treatment. Additionally, low to high treatment spiders likely had a high l ipid content (statistically equivalent to the constant group) at the conclusion of the experiment because they had been reserving energy for their moult to reproductive status or for upcoming reproductive effort (De Block and Stoks 2005). It seems probable that the total energy available for reproduction would eventually be higher in the constant food supply group. Given that animals in the constant and low to high group were the same size at the end of the experiment, the longer time to development of the constant group would require continued growth and probable greater size at maturity. This greater size should translate into a greater overall storage capacity for lipids. We would predict that spiders receiving a constant supply of food should eventually have the capacity for larger clutch sizes, or more sequential clutches (Anderson 1990). Spiders receiving late subsidies of food (as in the low to high treatment) have the advantage of earlier reproduction while spiders receiving a constant supply of food may have the advantage of greater total fecundity. 69 Female spiders must meet the energy demands associated with producing eggs with a high energy content while facing a food supply that fluctuates over time. This difficulty is compounded by the fact that, in common with most animals, the mass-specific energy content of spider eggs is frequently greater than that of the females producing them (Anderson 1990). Van Hook (1971) measured the mass specific energy content of adult lycosids and noted that this value varied seasonally by as much as 64%, reaching a peak value at the time of reproduction. Spider eggs (and the eggs of most animals) contain significant amounts of lipids. A s Anderson (1990) and Downer and Mathews (1976) suggest, lipids are the material of choice for energy storage in seeds, eggs and dormant stages since they contain approximately twice the amount of energy per unit mass as proteins and carbohydrates. This compact energy form may be critical for the survival of spiderlings given the ubiquity of aerial dispersal by ballooning for this life stage. The oxidation of l ipid also provides nearly twice as much water as that provided by the oxidation of carbohydrate (Beenakkers et al. 1985) - this water conservation may be particularly important for newly emerged spiderlings. The offspring of species reproducing in the fall generally overwinter within the egg sac and depend entirely on the maternal energy reserves provisioned at the time of reproduction. Spiders that successfully synchronise their reproductive efforts with seasonal increases in the availability of their prey can increase the amount of energy directed to reproduction. A s a laboratory study, these experiments represent a significant abstraction from processes operating in nature. Although spiders are generalist predators known to perform best on a mixed diet (Amalin et al. 1999, Nyffeler 1999, Toft 1999) we used a single species diet in order to minimize sources of variation. To our knowledge, natural feeding rates for T. versicolor have not been determined. Wise (1979) observed median daytime natural feeding rates for two araneid species of similar size to T. versicolor that were equivalent to ~3 fruit-flies (range = 0 to 21) per 24 hour period. Experimental treatments were selected to work with survival parameters 70 expressed on the artificial diet we used in the lab. Feeding rates in the lab were thus likely lower than those potentially experienced in nature. Our results suggest that spiders receiving large subsidies of food later in their development w i l l reach reproductive capacity in the same or better condition, with the same or better stored resources, and get there faster relative to animals receiving either constant supplies of food or early resource pulses. Receiving large amounts of food early in life appears to actually confer disadvantages in terms of later performance - the physiological costs of early rapid growth appear greater in terms of fitness than whatever delayed costs of compensatory growth might be expressed later in life. Subsidies may not be subsidies (of benefit to consumers) i f environmental variability is sufficient to decouple co-evolved synchrony between a subsidy and the reproductive or growth schedule of the consumer. Acknowledgements The authors acknowledge assistance with rearing and feeding spiders from Kyle Bateson and Nancy Hofer; Kel ly Walker assisted with the analysis of spider body compounds. Members of the Stream and Riparian Research lab at the University of British Columbia provided valuable feedback on early versions of the manuscript. This project was funded in part by the Natural Sciences and Engineering Research Council (NSERC) of Canada and the Forest Sciences Program (British Columbia - Forest Investment Account). 71 Table 4.1. Sample sizes, number and mass of prey offered and treatments used to assess growth rate plasticity and mortality risk of rapid growth. Treatment n Number of prey offered (per day) Mass of prey offered (mg/day) Prey percent of average initial spider mass I 9 0.25 0.27 1.9 II 9 0.5 0.54 3.8 III 10 1 1.07 7.6 IV 9 2 2.14 15.2 V 9 4 4.28 30.3 72 Table 4.2. Actual direction of relationships between treatment means (all differences significant) compared with the predicted relationships between treatments i f the experiment had continued until all animals reached maturity (using the same food ration from day 64 of the experiment until maturity). Treatments are C = constant, L H = low to high switch, H L = high to low switch. Response variable Actual relationship Predicted relationship among among treatment treatment means means Final size (corrected for C = L H > H L C > L H > H L initial size) (C spiders continue to grow) Number of moults C = L H > H L C> L H > H L completed (C require an additional moult to maturity) Body condition C = L H > H L C = L H > H L (mm/rag) Reproductive stage C < L H > H L C > L H < H L Age at maturity Mass specific lipid C = L H > H L C = L H > H L content (ng/mg) 73 May Jun Aug ' Sep Dec ' Figure 4.1. P a t t e r n o f a q u a t i c i n s e c t e m e r g e n c e o v e r o n e y e a r f o r t h r e e s t r e a m s ( E a s t , M i k e a n d S p r i n g c r e e k s ) i n t h e M a l c o l m K n a p p R e s e a r c h F o r e s t d u r i n g 1999. D a t a a r e t o t a l i n d i v i d u a l s p e r m 2 o f s t r e a m . 74 87 57 13 I 25 31 aug sep oct - — — males females Figure 4.2. Seasonal timing of cohorts of the spider Tetragnatha versicolor during 2005. The width of each bar represents the proportion of individuals caught within a given size class (based on 1 mm bins of the tibia-patella length of the first pair of walking legs); widths are a proportional fraction of the labeled legend box representing 100%. Numbers above columns represent sample sizes. Lines below months indicate presence through time of adult males and females. 75 2 4 6 8 1 0 1 2 1 4 1 6 1 8 2 0 2 2 2 4 2 6 2 8 Experiment day Figure 4.3. Cumulative number of prey offered in two treatments, one constant rate of food supply (control) and one switching treatment (switch) for Experiment 2. 76 Experiment day Figure 4.4. Cumulative number of prey offered in three treatments, one constant rate of food supply (constant) and two switching treatments: initial high food followed by low food (High:Low) and initial low food followed by high food (Low:High) for Experiment 3. 77 T r e a t m e n t Figure 4.5. Mean total growth rates (percent change in mg/day) (± 1 SE) for spiders in five treatments ranging from low (I) to high (V) levels of food availability. Different letters above a bar indicate significant differences for pairwise contrasts using a Bonferonni correction (Experiment 1). ( 78 10--m- control - # - switch 1 14 28 Experiment day Figure 4.6. Mean mass of spiders (± 1 SE) over a 28 day experiment fed either a constant food supply (control) or a switching treatment consisting of 7 days of low food followed by 7 days of the same amount of food as received by individuals in the control treatment (switch). Overall growth rates do not differ while growth rates between the start of the experiment and day 7 and between day 7 and the end of the experiment, differ between treatments (Experiment 2). 79 12 1 0 J E m 6 E 24 •— • - c ... ..jSjfc. - H L A - L H 14 63 7 0 28 35 42 49 56 Experiment day Figure 4.7. Second order polynomials fit to the mean mass (± 1 SE) (mg) of spiders in three treatments over time. Treatments were a constant level of food supply (C) and two switching treatments: initial high food followed by low food (HL) and initial low food followed by high food (LH) . The leftmost vertical line indicates the time of switching for the H L treatment and the rightmost vertical line indicates the time of treatment switching for the L H treatment (Experiment 3). 80 12-j OS E m | 1 0 -•E 13 c c xa m E 8 6 , 2 v 0 C H L LH B 220 _ _ 2 0 0 1 . 1 8 0 -& 0 160-* 1 4 0 ^ f1Z0 1 1 0 0 I 20 • C HL LH Figure 4.8. (A) Final mean mass (± 1 SE) of spiders (adjusted for initial mass) fed a constant (C), high to low food (HL) or low to high food (LH) diet. Different letters above a bar indicate significant differences using a Bonferroni correction. ( B ) Mean mass-specific l ipid content per individual (± 1 SE) for spiders fed three diets: C , H L or L H (see Figure 5 or methods for description of treatments). Different letters above a bar indicate significant differences in pairwise contrasts following a Bonferroni correction (Experiment 3). 81 Works Cited Abrams, P. A . , O. Leimar, S. Ny l in , and C. Wiklund. 1996". The effect of flexible growth rates on optimal sizes and development times in a seasonal environment. American Naturalist 147:381-395. Al len , W . V . 1976. Biochemical aspects of lipid storage and utilization in animals. American Zoologist 16:631-647. Amal in , D . M . , J. Reiskind, R. McSorley, and J. Pena. 1999. Survival of the hunting spider Hibana velox (Araneae, Anyphaenidae), raised on different artificial diets. The Journal of Arachnology 27:692-696. Anderson, J. F. 1990. The size of spider eggs and estimates of their energy content. Journal of Arachnology 18:73-78. Arendt, J. D . 1997. Adaptive intrinsic growth rates: an integration across taxa. Quarterly Review of Biology 72:149-177. Beck, C. W. 1997. Effect of changes in resource level on age and size at metamorphosis in Hyla squirella. Oecologia 112:187-192. Beenakkers, A . M . T., D . J. Vanderhorst, and W. J. A . Vanmarrewijk. 1985. Insect lipids and lipoproteins and their role in physiological processes. Progress in L ip id Research 24:19-67. Blanckenhorn, W. U . 1991. Foraging in groups of water striders (Gerris remigis): effects of variability in prey arrivals and handling times. Behavioural Ecology and Sociobiology 28:221-226. Bl igh, E . G. 1959. A rapid method of total l ipid extraction and purification. Canadian Journal of Biochemistry and Physiology 37:911. Briers, R. A . , H . M . Cariss, R. Geoghegan, and J. H . R. Gee. 2005. The lateral extent of the subsidy from an upland stream to riparian lycosid spiders. Ecography 28:165-170. Collier, K . J., S. Bury, and M . Gibbs. 2002. A stable isotope study of linkages between stream and terrestrial food webs through spider predation. Freshwater Biology 47:1651-1659. Danielson-Francois, A . , C. A . Fetterer, and P. D . Smallwood. 2002. Body condition and mate choice in Tetragnatha elongata (Araneae:Tetragnathidae). The Journal of Arachnology 30:20-30. De Block, M . , and R. Stoks. 2005. Fitness effects from egg to reproduction: bridging the life history transition. Ecology 86:185-187. Downer, R. G . FL, and J. R. Matthews. 1976. Patterns of lipid distribution and utilisation in insects. American Zoologist 16:733-745. 82 Engqvist, L . 2005. The mistreatment of covariate interaction terms in linear model analyses of behavioural and evolutionary ecology studies. Animal Behaviour 70:967-971. Fischer, K . , I. Zeilstra, S. K . Hetz, and K . Fiedler. 2004. Physiological costs of growing fast: does accelerated growth reduce pay-off in adult fitness? Evolutionary Ecology 18:343-353. Foelix, R. F. 1996. Biology of spiders, 2nd edition. Oxford University Press, New York. Frings, C. S., T. W . Fendley, R. T. Dunn, and C. A . Queen. 1972. Improved determination of total serum lipids by the sulfo-phospho-vanillin reaction. Clinical Chemistry 18:673-674. Garcia-Berthou, E . 2001. On the misuse of residuals in ecology: testing regression residuals vs. the analysis of covariance. Journal of Animal Ecology 70:708-711. Gende, S. M . , and M . F. Willson. 2001. Passerine densities in riparian forests of southeast Alaska: potential effects of anadromous spawning salmon. The Condor 103:624-629. Heiling, A . M . , and M . E . Herberstein. 1999. The importance of being larger: intraspecific competition for prime web sites in orb-web spiders (Araneae:Araneidae). Behaviour 136:669-677. Hentschel, B . T., and R. B . Emlet. 2000. Metamorphosis of barnacle nauplii: effects of food variability and a comparison with amphibian models. Ecology 81:3495-3508. Higgins, L . E . 1992. Developmental plasticity and fecundity in the orb-weaving spider Nephila clavipes. The Journal of Arachnology 20:94-106. Higgins, L . E . , and M . A . Rankin. 2001. Mortality risk in the spider Nephila clavipes. Functional Ecology 15:24-28. Jespersen, L . B . , and S. Toft. 2003. Compensatory growth following early nutritional stress in the W o l f Spider Pardosa prativaga. Functional Ecology 17:737-746. Kato, C , T. Iwata, S. Nakano, and D. Kishi . 2003. Dynamics of aquatic insect flux affects distribution of riparian web-building spiders. Oikos 103:113-120. Mendelssohn, I. A . , and N . L . Kuhn. 2003. Sediment subsidy: effects on soil-plant responses in a rapidly submerging coastal salt marsh. Ecological Engineering 21: IT 5-128. Meredith, M . P., and S. V . Stehman. 1991. Repeated measures in forestry: focus on analysis of response curves. Canadian Journal of Forestry 21:957-965. Metcalfe, N . B . , and P. Monaghan. 2001. Compensation for a bad start: grow now, pay later? Trends in Ecology and Evolution 16:254-260. Nyffeler, M . 1999. Prey selection of spiders in the field. The Journal of Arachnology 27:317-324. Olive, C. W. 1982. Behavioural response of a sit-and-wait predator to spatial variation in foraging gain. Ecology 63:912-920. 83 Orr, M . , M . Zimmer, D. E . Jelinski, and M . Mews. 2005. Wrack deposition on different beach types: spatial and temporal variation in the pattern of subsidy. Ecology 86:1496-1507. Parsons, P. A . 2004. From energy efficiency under stress to rapid development and a long life in natural populations. Biogerontology 5:201-210. Phillips, N . E . 2004. Variable timing of larval food has consequences for early juvenile performance in a marine mussel. Ecology 85:2341-2346. Polis, G . A . , W. B . Anderson, and R. D. Holt. 1997. Toward an integration of landscape and food web ecology: the dynamics of spatially subsidized food webs. Annual Review of Ecological Systematics 28:289-316. Polis, G . A . , and S. D . Hurd. 1995. Extraordinarily high spider densities on islands: flow of energy from the marine to terrestrial food webs and the absence of predation. Proceedings of the National Academy of Sciences - U S A 92:4382-4386. Sanzone, D . M . , J. L . Meyer, E . Mart i , E . P. Gardiner, J. L . Tank, and N . B . Grimm. 2003. Carbon and nitrogen transfer from a desert stream to riparian predators. Oecologia. 134:238-250. Smith, C. C. 1976. When and how much to reproduce: the trade-off between power and efficiency. American Zoologist 16:763-774. Stockoff, B . A . 1991. Starvation resistance of gypsy moth, Lymantria dispar (L.) (Lepidoptera: Lymantriidae): trade-offs among growth, body size and survival. Oecologia:422-429. Szepanski, M . M . , M . Ben-David, and V . Van Ballenberghe. 1999. Assessment of anadromous salmon resources in the diet of the Alexander Archipelago wol f using stable isotope analysis. Oecologia 120:327-335. ] Tabachnick, B . G , and L . S. Fidell . 2001. Using Multivariate Statistics, 4th edition. A l l y n and Bacon, Boston. Toft, S. 1999. Prey choice and spider fitness. The Journal of Arachnology 27:301-307. Twombly, S. 1996. Timing of metamorphosis in a freshwater crustacean: comparison with anuran models. Ecology:1855-1866. Van Handel, E . 1985. Rapid determination of total lipids in mosquitoes. Journal of the American Mosquito Control Association 1:302-304. Van Hook, R. I. 1971. Energy and nutrient dynamics of spider and orthopteran populations in a grassland ecosystem. Ecological Monographs 41:1-26. Will iams, D . D. , L . G . Ambrose, and L . N . Browning. 1995. Trophic dynamics of two sympatric species of riparian spider (Araneae: Tetragnathidae). Canadian Journal of Zoology 73:1545-1553. Wise, D . H . 1979. Effects of an experimental increase in prey abundance upon the reproductive rates of two orb-weaving spider species (Araneae: Araneidae). Oecologia 41:289-300. 84 Yearsley, J. M . , I. Kryriazakis, and I. J. Gordon. 2004. Delayed costs of growth and compensatory growth rates. Functional Ecology 18:563-570. Chapter 5. Trophic interception: how a boundary-foraging organism influences cross-ecosystem fluxes1^ Introduction Numerous empirical studies have shown that resource subsidies are ubiquitous (see for example, Bustamente and Branch 1996, Takimoto et al. 2002, Mendelssohn and Kunn 2003, Baxter et al. 2004, Paetzold et al. 2006). This realisation has led to the re-examination of conceptions about habitat boundaries and the degree to which connectivity between systems influences community structure. A t the same time, the assertion that boundary conditions regulate fluxes between ecosystems remains a largely untested hypothesis (Cadenasso et al. 2004). While many studies have documented the open nature of ecosystems, little attention has been directed towards how conditions at the boundaries themselves may influence the between-habitat flows of energy or nutrients. The physical and biological differences between two habitats together with processes acting at the boundary itself can concentrate (Witman et al. 2004) or potentially reduce a subsidy. This later possibility, that organisms living or foraging at the boundaries between habitats may act to restrict the movement of resource subsidies between habitats, has not yet been explicitly investigated. In this paper we introduce the concept of trophic interception, where the flow of resources across a habitat boundary is interrupted or redirected by an organism foraging at that boundary. This concept would likely apply most strongly in instances where resources flow from one habitat to another along a simple, distinct pathway. Trophic interception potentially alters both the amount of inputs that eventually enter the recipient habitat and the specific pathways that these material inputs follow. Although few studies have highlighted the role of consumer species that limit the amount of subsidy materials arriving in a recipient habitat, examples can be found. Although not * Currently in review in Ecology as: Marczak, L . B . , T . M . Hoover and J.S. Richardson. Trophic interception: how a boundary-foraging organism influences cross-ecosystem fluxes. 86 recognized as such, trophic interception may be particularly common where strong physical differences characterise the interface between habitats - illustrated by the effects of beetles, ants and spiders (Paetzold and Tockner 2005), lizards (Sabo and Power 2002) or shorebirds (Dugan et al. 2003) foraging at marine-terrestrial or stream-riparian margins. Phytoplankton in lakes have been shown to regulate the productivity of macrophytes and benthic periphyton by intercepting light (through physical shading effects) and nutrients (Sand-Jensen and Borum 1991, Schindler and Scheuerell 2002). Recently, Maron et al. (2006) showed that introduced foxes in the Aleutian islands interrupt the flow of marine-derived nutrient subsidies to terrestrial plants by reducing seabird populations and therefore the production of guano. Bears have been shown to reduce the flow of marine nutrients to freshwater systems by diverting salmon carcasses to terrestrial habitats (Gende et al. 2002, Helfield and Naiman 2006). A n ecologically important example of trophic interception occurs where multiple predators compete for invertebrates crossing the air-water interface of streams. The suspension of insects in the surface film constitutes a distinct pathway across a permeable boundary that prey or materials must cross in order to move between terrestrial and aquatic habitats. This subsidy occurs as either incidental inputs of terrestrial invertebrates (forest to stream) or during periods of emergence of aquatic invertebrates (stream to forest). Invertebrates crossing the air-water surface in streams constitute an important trophic subsidy to both aquatic (Nakano et al. .1999) and terrestrial consumers (Sanzone et al. 2003). In coastal headwater streams of southwestern British Columbia (Canada) such consumers include trout (Oncorhynchus clarki), benthic omnivores such as crayfish (Pacifastacus leniusculus), and a host of benthic invertebrates that may eventually consume terrestrial invertebrate inputs as detritus (Bondar et al. 2005). In terrestrial habitats, numerous predators have been found to respond to changing densities or abundances of emerging aquatic insects. These include spiders (Sanzone et al. 2003, 87 Marczak and Richardson in prep), bats (Power et al. 2004), lizards (Sabo and Power 2002) and beetles (Paetzold et al. 2006) amongst others. Previously unexamined in this context is a distinctive group of organisms that forage at the interface between land and water. These include fishing spiders (Dolomedes), backswimmers (Notonecta), whirligig beetles (Gyrinidae), and waterstriders (Gerridae). The waterstrider Aquarius remigis (Say) (Hemiptera:Gerridae) is a common inhabitant of the surface of streams and lakes throughout North America. Densities of these insects on streams near the Pacific Coast of North America range from 3 m" 2 to more than 10 m" 2 (Cooper 1984a; present study), and vary between stream habitat types (e.g. riffles and pools). Adult gerrids are found primarily where surface currents range between 0 and 10 cm s"1 and are generally absent when currents exceed 15 cm s"1 (Fairbairn and Brassard 1988; L . Marczak, personal observation). Waterstriders feed almost exclusively on terrestrial insects trapped in the surface fi lm of streams, ponds, and lakes, and on aquatic insects emerging from the water, although they have also been observed to feed upon dead vertebrates and even living fish in rare instances (Spence and Andersen 1994). B y foraging at the air-water interface for fallen terrestrial or emerging aquatic insects, surface-foraging consumers such as waterstriders have the potential to interrupt the flow of nutrients across the terrestrial-aquatic boundary, nutrients that otherwise would be incorporated directly into aquatic or riparian food webs. In the following study, we experimentally quantify the magnitude of waterstrider interception of terrestrial invertebrate subsidies to streams and examine the potential impact of this interception on in-stream consumers (fish, benthic consumers). Since waterstrider foraging efficiency is known to vary with surface velocity (Spence and Andersen 1994), we quantify the variation in the efficiency of waterstrider interception of terrestrial prey across the range of natural stream habitats (riffles, connected pools, and isolated pools) under field conditions, and estimate the potential amount of terrestrial-to-stream resources that are intercepted in a coastal 88 headwater stream in southwestern British Columbia (Canada). Finally, using a series of controlled field experiments in constructed arenas with known densities of waterstriders and cutthroat trout, we quantify how sensitive the pathways which incorporate terrestrial prey inputs into stream food webs are to the presence of interceptors and in-stream consumers. We use these experiments to show how fish alter waterstrider interception of terrestrial inputs, how waterstriders alter the amount of terrestrial subsidies acquired by fish, and how the presence or absence of both of these consumers affect the amount o f this subsidy that ultimately becomes available to benthic detritivores. Methods Study location Field collections and experiments were conducted on Spring Creek, a low-gradient, second-order stream in the Malcolm Knapp Research Forest ( M K R F ; 49°18 '40"N, 122°32'40"W). The M K R F is located in the Pacific coastal rainforest of southwestern British Columbia, and average mean air temperature ranges from a low of 2°C in January to a high of 16°C in July (Kiffney et al. 2004). Precipitation is approximately 2500 mm annually, mainly in the form of rain. More than 70 % of this precipitation falls between October and March. The reach of Spring Creek in which the study was conducted is located approximately 160 m above sea level and has a mean wetted width of 3 m (SE = 0.6 m), a mean water depth of 0.15 m (SE = 0.01 m), and a mean summer velocity < 0.10 m s"1 (Kiffney et al. 2000, Boss and Richardson 2002). The study reach consists of a series of riffles and pools whose beds are dominated by sands and coarse gravels. Due to low summertime precipitation in the M K R F , discharge in Spring Creek is at its lowest point in late summer. At that time, the stream reach consists of deep, slow moving pools that retain surface connections to the main flow of the channel (pools), shallower pools no longer connected by surface flow to the main channel 89 (isolated pools), and shallow, fast flowing segments (riffles) (Table 5.1). The overall density of waterstriders in summer was 2.0 m" (range: 0 m" to 10.6 m"), while densities of trout were approximately 0.7 m"2 (range: 0 m" 2 to 3.0 m"2). The dominant vegetation surrounding the creek included red alder (Alnus rubra) and vine maple (Acer circiriatum) with a canopy composed largely of western hemlock (Tsuga heterophylla), western redcedar (Thuja plicata) and Douglas-fir (Pseudotsuga menziesii). Understorey vegetation immediately adjacent to the stream consisted of 1-2 m tall shrubs, particularly salmonberry (Rubus spectabilis) and huckleberry (Vaccinium ovatum). Interception in natural habitats In order to characterize differences in waterstrider interception rates of terrestrially derived prey in the three types of stream habitat (n = 11 pools, n = 10 isolated pools, and n = 8 riffles), we conducted random tosses of experimental prey to the surface of pools, isolated pools and riffles on a 200 m reach of Spring Creek. We did not manipulate the movement, density, or composition of either waterstriders or fish populations present. To experimentally simulate the input of falling terrestrial invertebrates, we used crickets in four size classes (adults (A) , large (L), medium (M), and small (S); Table 5.2). Individual crickets were dropped onto a haphazardly selected location on the surface of the pool or riffle. We followed each prey item until a fate was determined, before introducing the next cricket. The time to consumption, fate (eaten by waterstrider (WS), eaten by trout (CT), escaped (ES), retained against rocks or submerged for greater than 60 seconds, making them available to the benthos (BE), drifted out of experimental reach (DR)), and total surface distance travelled was determined for each cricket. A cricket was assumed eaten by a fish only when we witnessed a fish rise to the surface and consume the cricket. Waterstrider consumption of a cricket was considered final when one or more waterstriders remained attached to the cricket for greater than 60 seconds. Once a fate had been 90 determined, we marked the locations of each cricket on a scale map of the study reach. At the conclusion of a trial, we measured the depth (cm) and distance from the habitat margin for each individual cricket at the point of capture. We used each habitat unit once, randomized the order of prey presented in each trial, and limited the total number of prey items per trial to 24 (within a trial, each size class was represented by 3 to 8 individuals). Each trial required between 30 and 45 minutes to complete. Statistical analysis We analysed the proportion of prey intercepted by waterstriders using a split plot, mixed effects model. In this model, habitat (fixed) was the whole-plot factor and prey size (fixed) was the split plot factor, trial was a random blocking factor (n = 29 trials) and waterstrider density in each habitat was used as a covariate. We employed the Kenward-Rogers method to approximate denominator degrees of freedom for unbalanced data with multiple variance components (Spilke et al. 2005). Only second order interactions between factors or between factors and the covariate were included in the model (prey size x habitat, and waterstrider density x habitat). The interaction with the covariate was not significant and was removed from the model. We checked model assumptions by inspecting residuals and normal probability plots. Proportions data were arcsine square root transformed to successfully meet assumptions of normality and homogeneity of variances. In order to test whether the presence of fish was constraining the foraging area of waterstriders to the margins of pools (Cooper 1984a) we used Bartlett's test for equal variances to determine whether waterstrider captures of prey in connected versus isolated pools occurred over a greater range of depths (standardized for mean habitat depth) or over a greater range of distances from the pool edge (standardized by the mean radius of the habitat). To determine i f fish presence was lengthening the response time of waterstriders to a prey cue, we assessed 91 differences in surface drifting times of prey prior to capture by waterstriders on connected (fish present) versus isolated pools (fish absent) using an A N O V A . - A l l analyses were conducted using the M I X E D procedure or the G L M procedure in the statistical package S A S v 9.0 (SAS Institute Inc, Cary, U S A ) . Controlled arena experiment To quantify the direct effect of waterstrider interception on the amount of subsidies acquired by fish, the reciprocal effect of fish presence on waterstrider interception, and the effect of both of these predators on the transfer of terrestrial subsidies to the stream benthos we conducted a series of controlled trials of experimental prey inputs to in-stream arenas containing known combinations of waterstriders and fish. In early July, 2006 we established three circular arenas in Spring Creek, each with a water surface area of approximately 2 m 2 . Arenas were constructed from 0.5 cm wire screen mesh, with 15-20 cm of this material buried in the stream bed, and 30-40 cm of material projecting above the water surface. We lined the above-water portions of the fence with mosquito-netting to prevent waterstriders from escaping through the mesh. We ensured that each arena contained at least two areas with a depth of ~ 40 cm and large overhanging rocks to provide a refuge for fish. The arena perimeter was modified to contain 8-10 rocks that protruded above the water surface in a region with a depth of ~8 cm as a refuge for waterstriders. Following construction we removed any waterstriders present by handnetting and ensured arenas contained no fish by placing three baited minnow traps in each arena for three consecutive 24 hour periods. Experimental methods Over ten days from July 12 to July 22 we conducted six independent trials of four treatments. We randomized the treatment and the arena in which each trial was conducted. The treatments 92 consisted of combinations of trout and waterstrider presence: no waterstriders + 3 trout (FISH), 20 waterstriders + no trout (WS), 20 waterstriders + 3 trout (BOTH) , and no waterstriders + no trout (NONE) . We used adult waterstriders caught by handnetting in nearby pools and trout between 7 - 10 cm forklength caught with baited minnow traps in pools downstream of our constructed arenas (Table 5.2). A l l animals were caught the day before their use in an experimental trial and kept in separate, fully enclosed mesh containers embedded in the stream channel. Individual animals were used in only one trial. A t the start of each trial, animals were transferred to the arena and allowed to acclimate for two hours. In each trial, we tossed six cricket prey of each size class (A, L , M , S; Table 5.2) into the arena, randomising the order in which individual prey of various size classes were introduced. The fate of each prey item was recorded before adding a new prey item following the same protocols as outlined for the first experiment (natural habitats). We calculated the proportion of prey in each of four possible fates (ES, B E , W S , C T as in the natural habitat experiment) within each trial, by prey size. Statistical Analysis We were interested in (1) the effects of waterstriders presence on the amount of experimental subsidy acquired by fish, (2) the effect of fish presence on the amount of experimental subsidy acquired by waterstriders and, (3) the effect of the presence or absence of both consumers on the amount o f experimental subsidy that is directed to the benthos. Each of these questions uses a different response variable (proportion of prey consumed by fish, proportion of prey consumed by waterstriders, proportion of prey directed to the benthos), some of which are not represented in certain treatment combinations. Including all treatment combinations in our models would lead to the inclusion of treatments containing only zeros and create problems with underdispersion. To address the effects of unnecessary zeros on our ability to meet model 93 assumptions, we subdivided our dataset into preplanned groupings of treatments for each response variable. For example, to address the question of how waterstrider presence altered the amount of experimental terrestrial subsidy acquired by fish we excluded treatment combinations that did not contain fish (by definition - the response variable w i l l always equal zero in these treatments). Each of these groupings was analysed using separate split-plot mixed models. The effect of trout presence and absence on the proportion of experimental terrestrial inputs captured by waterstriders was analysed using a mixed model with treatment (WS and B O T H ) as the whole plot fixed factor and prey size as the split plot fixed factor. We used trial as a random blocking factor (n = 12 trials). We used an identical approach to analyse the effect of waterstrider presence and absence on the proportion of experimental terrestrial inputs captured by cutthroat trout (treatments F I S H and B O T H , n = 12 trials). Finally, we assessed the impact of stream consumer presence and identity on the proportion of experimental terrestrial inputs that were incorporated into the benthos (all treatments: W S , F ISH, B O T H , N O N E , n = 24 trials). We checked key model assumptions by inspecting residual and normal probability plots. Proportions data were arcsine square root transformed to successfully meet these assumptions. A l l analyses were conducted using the M I X E D procedure with Satterthwaite's degrees of freedom method for balanced data in the statistical package S A S v 9.0 (SAS Institute Inc, Cary, U S A ) . Magnitude of waterstrider interception in natural stream systems In an attempt to understand the potential significance of the interception of terrestrial prey to fish, we calculated crude estimates of the amount of material waterstriders could be intercepting over the course of the summer (roughly June through July) and the subsequent number of juvenile salmonids this material could support (Wipfl i and Gregovich 2002). First, we calculated the average surface area of connected pools, isolated pools and riffle habitats over a standardized 94 100 m reach of Spring Creek based on initial habitat surveys conducted during early August 2005 (Table 5.1). We estimated the range of terrestrial prey inputs during summer low flows from datasets provided by colleagues (A. Reiss unpublished data, L . Marczak and T. Hoover unpublished data). We made the assumptions that waterstrider density was equal to that observed in this study, that the relative distribution of habitat units was constant throughout the summer, that the dominant size class of invertebrate prey falling on the stream was approximated by the smallest size class of experimental prey used in our study (~5 mm) and that waterstrider interception of these prey was consistent with that observed in our natural habitat experiment (41.4% in connected pools, 97.5% in isolated pools and 3.0 % in riffles). We combined these values to produce an estimate of the potential magnitude of waterstrider interception during summer flows on Spring Creek. Results Natural habitat interception experiment The ability of waterstriders to intercept terrestrial prey differed between the three habitat types (F2,27.7 = 53.2, p < 0.0001; Figure 5.1A). Waterstriders intercepted an average of 71.8%) of prey items placed on the surface of isolated pools, but only 21.5% in connected pools and < 1% in riffles. There was a significant interaction between prey size and habitat type (F6,89.2 = 7.35, p < 0.0001), however post hoc comparisons indicated that the source of this interaction was the large number of zero values in riffle habitats; interpretation of main effects between connected and isolated pools is thus unaffected. Waterstriders were more effective at intercepting smaller prey in all habitat types (effect of prey size, F 3, 89.2 = 28.05, p < 0.0001; Figure 5. IB) . Across all habitat units, waterstriders intercepted 45% of the smallest size classes, with that effectiveness dropping to less than 10% for the largest prey. 95 Prey items captured by waterstriders on connected pools had significantly longer surface residence times before consumption than prey items captured by waterstriders on isolated pools (Fi,5i = 3.55, p < 0.001). Waterstriders consistently captured prey over a greater range of depths in isolated pools than in connected pools (x2 = 23.44, df = 1, p < 0.0001) and over a greater range of distances from the edges of isolated pools than on connected pools (x2 = 8.867, df = 1, p = 0.003). Controlled arena experiment The presence of cutthroat trout significantly decreased waterstrider interception of falling terrestrial invertebrates (comparison of treatments WS versus B O T H : Fi jo = 50.02, p < 0.0001). Waterstrider interception consistently varied with the size of the prey (F3 ;3o= 9.27, p = 0.0002) with consistently greater success with smaller prey items (Tukey-Kramers correction all p < 0.02) whether fish were present or not. In the presence of fish, waterstrider interception decreased across all prey sizes by approximately 60.6 % (mean proportion with fish = 0.26 (0.02 S.E.), mean proportion without fish = 0.66 (0.05 S.E.); Figure 5.2A) or 39.4 % of the amount waterstriders captured when alone. Waterstriders decreased the percentage of falling terrestrial invertebrates that were captured by trout (Fyo = 10.30, p = 0.009; Figure 5.2B). This pattern did not differ by prey size (WS versus F ISH, F3.30 = 1.49, p = 0.237). When waterstriders were present, fish consumption of terrestrial inputs decreased across all prey sizes by approximately 45 % (i.e., 55 % of the terrestrial inputs they acquired when alone, mean proportion with waterstriders = 0.22 (0.07 S.E.), mean proportion without waterstriders = 0.40 (0.05 S.E.)). The amount of terrestrial inputs that went to the benthos varied significantly depending on the presence or absence of both fish and waterstriders ^3,20 = 4.34, p = 0.017). Post-hoc tests show that inputs to the benthos when both waterstriders and fish are present are significantly 96 greater than for all other combinations (Tukey-Kramers correction all p < 0.03, Figure 5.3). Prey size was also a significant factor (F3 j 6o = 3.12, p = 0.032 with the smallest prey items entering the benthos more often than the other three size categories (Tukey-Kramers correction all p < 0.02). N o interaction terms were significant in any of these models. Magnitude of waterstrider interception in natural stream systems We estimated the range of terrestrial prey inputs during sumrher low flows in Spring Creek to be between 11.5 and 50.2 mg m" 2 day"1 (A. Reiss unpublished data; Marczak and Hoover unpublished data). Given the areal proportions of pools, isolated pools and riffles on Spring Creek during the summer (Table 5.2), we determined that every 100 m reach of Spring Creek receives 8.8 to 38.4 g 100 m"1 day"1 o f terrestrial invertebrates or 0.8 to 3.5 kg 100 m ' 1 over the course of a 90 day summer period (June through August) and waterstrider interception of these inputs is equal to 0.3 to 1.2 kg 100 m"1 over the same summer period. At a maximum sustained consumption rate of 20 mg prey day"1 for 2-3 g young-of-the-year ( Y O Y ) salmonids at 10 °C (Dunbrack 1988, Wipf l i and Gregovich 2002), the material intercepted by waterstriders over a 100 m reach of stream during summer could theoretically feed between 13.2 and 57.6 fish (or a density of 0.1 to 0.6 fish m"2). Discussion A growing body of theoretical and empirical work has highlighted the role of consumers in linking adjacent habitats or increasing the diffusion of subsidies into a recipient habitat. In this study we have demonstrated experimentally that some consumers can also act as interceptors that limit the penetration or uptake of subsidies in a recipient habitat by interrupting or redirecting subsidies at the air-water boundary. A s such, stream-surface dwelling invertebrates have the 97 potential to exert partial control over the quantity and size distribution of terrestrial invertebrate subsidies moving from forests to streams. In our study, waterstriders monopolized small and medium prey falling on the stream surface, thus intercepting a substantial proportion of experimental prey subsidies before they could enter the drift, be captured by in-stream consumers, or enter the detrital pool. The magnitude of this trophic interception was modified by the distribution and area of stream habitat types and the presence or absence of in-stream consumers. When waterstriders were present they diverted substantial quantities of material from the stream and unexpectedly, in combination with cutthroat trout, contributed to an increased diversion of terrestrial invertebrate inputs to the stream benthos (Figure 5.4). Magnitude of waterstrider interception in natural stream systems We know very little about how different geomorphic features of the channel or spatial distributions of stream habitat units affect the transfer of subsidies between land and water (but see Iwata et al. 2003). In this study, waterstrider interception of falling terrestrial invertebrates varied between surface habitat types of the stream ecosystem. Interception of all prey items was greatest in isolated stream segments where velocity was low and (in this study) where fish were absent. Since waterstrider (or other surface predator) interception of these inputs effectively reduces the amount of subsidy actually entering the stream, the seasonal shift in the area, relative abundance, and physical characteristics of stream subunits may be critical determinants of the response of in-stream consumers to subsidies. Several studies have demonstrated that populations of drift-feeding fish in many streams and rivers cannot be maintained by benthic production alone (Kawaguchi and Nakano 2001, Allan et al. 2003, Wipfli 2005). In these systems, large quantities of terrestrial invertebrate inputs allow fish to reach their daily caloric requirements. We estimated waterstrider interception of 98 terrestrial inputs to be 0.3 to 1.2 kg 100 m"1 during the period of summer low flows. This material could theoretically feed between 13 and 58 fish (or a density of 0.1 to 0.6 fish m~2). This estimate is likely conservative since crickets, our choice of experimental prey, were strong surface swimmers - unlike the majority of prey items that fall onto the surface of streams (e.g. adult Diptera) - and thus waterstriders may have been consistently less successful at subduing these prey than would be possible with natural prey inputs. I f salmonid populations along the west coast of North America are indeed food limited (Chapman 1966, Boss and Richardson 2002) then this scale of restriction in available allochthonous inputs to headwater streams during the peak of summer could have substantial impacts on fish populations. This effect might be felt in either small headwater streams themselves or transferred to lower velocity downstream reaches (Boss and Richardson 2002, Wipf l i 2005). Effects of waterstriders on trout and trout on waterstriders Coexisting predator species may interact in a number of ways; they may compete for a common resource, prey on one another, or alter one another's behaviour. In small coastal streams, trout and waterstriders share a common food supply (terrestrial insects trapped in the surface film) and directly compete for food resources. These interactions directly affected the foraging ability of both predator species. In our arena experiments, where velocity was held constant, the amount of terrestrial invertebrate subsidy intercepted by waterstriders was significantly affected by the presence of fish. Waterstrider capture rates of prey in the presence of fish were only 40.2% of the capture rates observed when fish were absent. Fish both competed directly with waterstriders for terrestrial prey subsidies and presented a predation threat which altered the foraging behaviour of waterstriders present. Trout harass waterstriders, which results in increased energy expenditure for these insects and may reduce foraging efficiency (Cooper 1984a). Cooper (1984a) demonstrated that the presence of trout 99 influenced the surface distribution of waterstriders within stream pools. In pools with fish, waterstriders were restricted to the margins of pools where they can more easily avoid attack by trout. This effect was also noted in this study; when trout were present, waterstriders caught their prey closer to the edges of pools and over shallower water than in the absence of trout, and the time they took to respond to a prey item was also increased. The presence of cutthroat trout limits the foraging surface available to waterstriders, effectively altering the amount of terrestrial inputs acquired by waterstriders. In arena trials where trout were present, waterstriders were concentrated near pool margins, and thus likely experienced stronger intraspecific interactions (e.g., competition, cannibalism). Waterstriders feed more effectively in relatively low water velocities; during periods of high discharge (and thus high channel velocity) waterstriders are likely restricted to stream margins as high mid-channel water velocities would limit their ability to maintain their position (Fairbairn and Brassard 1988). A s such, interception rates would decrease and subsidies to stream food webs potentially increase during periods of higher streamflow. Waterstriders also reduced the rate at which fish captured experimental terrestrial prey. When waterstriders were present, fish caught roughly half of the terrestrial inputs they acquired when alone. Although there are reports of waterstriders successfully capturing small fish (see references in Spence and Andersen 1994), this is unlikely in Spring Creek due to the relatively large size and aggressive foraging behaviours of juvenile cutthroat trout. Rather, this effect is due to the high rates of prey capture by waterstriders at pool margins; i f fallen prey are captured at stream margins before they drift out into the main channel, they are not available to the fish predators that inhabit deeper pool areas. Trout seldom forage in shallow areas along stream margins (Cooper 1984b), and thus prey floating on the surface at the periphery of pools are relatively inaccessible. This type of spatial partitioning of terrestrial prey subsidies may also occur in other types of freshwater systems. Invertebrate prey trapped on the water surface of 100 lakes or rivers may become available to fish inhabiting deeper areas when wind or storm events increase the distance that falling insects travel before hitting the water, or when prey not captured by other predators drift out of shallower areas; these inputs, however, w i l l likely be smaller than amounts available near shore. Interactions between consumers alter terrestrial subsidy pathways into streams In addition to the direct effects that the interception of terrestrial prey has on fish populations, the presence of waterstriders on the stream surface may have a multitude of indirect effects in stream ecosystems. Decreases in fish populations via food limitation (Boss and Richardson 2002) could produce cascading positive effects on the biomass of benthic invertebrates which are also utilised as prey by trout. Alternatively, it has been shown that when terrestrial invertebrates in the drift are scarce, fish can switch to feeding on benthic prey (Kawaguchi et al. 2003, Zhang et al. 2003). B y reducing the terrestrial prey subsidies available, waterstriders could increase the degree of benthic feeding by fish, producing indirect negative effects on benthic prey populations. Not all terrestrial invertebrate subsidies falling onto the stream surface are consumed directly by surface dwelling organisms or large vertebrate predators such as fish. While these consumers may account for the majority of prey subsidies that enter the type of stream system described in this study, a fraction of the subsidy wi l l enter the food web via other pathways. Some terrestrially derived prey wi l l be lost to the system entirely, either by drifting out of the stream into higher-order reaches downstream or by being consumed by terrestrial consumers inhabiting the riparian zone. However, some terrestrial prey that fall into streams wi l l eventually become waterlogged or drawn under the water surface and trapped on the streambed by turbulent flows. When this happens, these terrestrial subsidies become available to a diverse assemblage of benthic invertebrate detritivores. In our arenas experiment we tracked the proportion of experimental inputs that became trapped on the streambed or on debris. Interestingly, more prey 101 became available to benthic detritivores (i.e., were trapped on the streambed or debris) when there were multiple consumers present (waterstriders and trout) than when only one consumer or no consumers were present (waterstriders or trout or none; Figure 5.4) and significantly more of this material was in the smallest experimental prey size category. We observed that waterstriders frequently abandoned smaller prey items when the initial contact occurred close to the deeper portions of our experimental arenas. Such prey items often went unnoticed by fish since they were small and no longer moving. These prey items often then became waterlogged and trapped against a protruding stone or sank to the streambed, and were rarely subsequently captured by either waterstriders or trout. Waterstriders may abandon smaller prey in favour of larger prey items as group feeding on larger prey has been shown to reduce the risk of predation by fish (Di l l and Ydenberg 1987). Individual waterstriders are known to monopolize smaller prey while groups of waterstriders may more effectively subdue larger prey items (Erlandsson 1988), a phenomenon observed in the waterstriders in this study. Smaller items (which while easier to subdue may be riskier to feed on) may therefore be more likely to be attacked and abandoned, whereas larger food items are "safer" once subdued. Waterstriders may also be diverting material directly to the benthic detritivore pool when they abandon the carcasses of their prey after their feeding is complete. Gerrids are predator-scavengers with the piercing and sucking mouthparts typical of predatory Hemiptera, and do not engulf the exoskeleton of their prey. Jamieson and Scudder (1977) predicted that when prey are relatively scarce, waterstrider utilization of prey contents is likely to be relatively complete. However, when prey are abundant or during storm- or wind-induced mass inputs of terrestrial invertebrates to the stream surface, waterstriders may not completely ingest the soft parts of any given prey. Carcasses of partially digested prey are typically abandoned once a waterstrider achieves satiation or maximum extraction of prey contents. A s these carcasses have been wetted for some time, they may then become easily waterlogged or trapped against rocks and sink to the 102 benthos where they enter the detrital pool. Although the nutritional quality of these prey has been reduced, these carcasses still represent a type of input to the benthic foodweb that would not occur i f waterstriders were absent. Waterstriders thus not only directly intercept a substantial portion of material that might otherwise enter the stream food web, but also potentially redirect material to the stream benthos through competitive interactions with fish and via the discarded carcasses of their prey. In this study we focused on the role that water-surface foraging predators play in controlling the amount of terrestrial prey subsidies to streams and the pathways by which these subsidies are incorporated into stream food webs. However, surface-dwelling predators may also affect the distribution and abundance of terrestrial consumers by reducing the abundance of emerging aquatic invertebrates. Waterstriders on Californian streams have been shown to effectively reduce the quantity of emerging aquatic insects by up to 46% (Wiseman and Cooper 1988). Sweeney and Vannote (1982) similarly noted that predation by water surface-dwelling gyrinid beetles on emerging mayfly larvae reduced the abundance of adult mayflies in riparian areas. More recently, Paetzold and Tockner (2005) experimentally demonstrated that riparian arthropods (ground-dwelling beetles, spiders and ants) consume nearly half of newly emerged aquatic insect biomass. The potential consequences of trophic interception of aquatic-terrestrial subsidies for terrestrial and riparian consumers remains largely unstudied. We have shown that organisms foraging at habitat boundaries can intercept substantial quantities of material, altering rates of resource transfer between habitats. This process, trophic interception, alters the quantity or quality of subsidies moving between habitats and has the potential to alter the magnitude of the response of recipient food webs to cross-habitat subsidies. The ways in which both physical and organismal agents may either concentrate or limit the magnitude of a subsidy in a recipient habitat is a critical piece in a broader attempt to understand where and why subsidies have large effects in some habitats, but not others. Attempts to 103 understand the factors that control the permeability of boundaries to subsidies are an essential step towards developing a landscape level approach to open systems. Acknowledgements We thank Babita Bains and other members of the Stream and Riparian Research ( S T A R R ) Laboratory at the University of British Columbia for assistance in the field and lab, and A y a Reiss for sharing some of her unpublished data. Locke Rowe provided expert confirmation of the waterstride species identification. The authors acknowledge funding from the Natural Sciences and Engineering Research Council (Canada) and the Forest Sciences Program (BC). 104 Table 5.1. Physical characteristics of three surface habitat types in Spring Creek. Values are means (± standard error), or proportions. Isolated pools Connected Riffles pools Mean depth (m) 0.16(0.02) 0.40(0.04) 0.09(0.001) Mean velocity (cm s"1) 1(0.02) 6(0.9) 19(2.4) Mean surface area (m 2) 8.78 (0.66) 11.3 (3.56) 10.64 (1.49) Proportion of total surface area (over 100 m) 0.075 0.478 0.449 during summer low flows 105 V Table 5.2. Average lengths and live mass ± standard error for organisms used in experiments. Length (cm) Mass (mg) Crickets (prey) Adult (A) 2.0 ± 0.02 430.38 ± 22.08 Large (L) 1.55 ± 0.55 212.25 ± 16.46 Medium (M) 0.95 ± 0.03 54.49 ± 5.50 Small (S) 0.57 ± 0.03 16.26 ± 2 . 2 1 Waterstriders 1.12 ± 0 . 0 5 22.82 ± 2.56 Cutthroat trout 7.81 ± 0 . 1 7 6580 ± 420 106 i s o l a t e d p o o l s pools Figure 5.1. Mean proportion of prey (± 1 S.E.) intercepted by waterstriders. A . between different stream habitats and B. for different sizes of terrestrial prey. Different letters above the bars indicate significant differences. Sizes of experimental prey represented by: S = small, M = medium, L = large, A = adult, refer to Table 5.2 for details. 107 fish both Figure 5.2. A . Mean proportion of experimental terrestrial inputs (± 1 S .E.) in all prey size categories that were consumed by waterstriders for treatments containing waterstriders. B. Mean proportion of experimental terrestrial inputs (± 1 S .E.) in all prey size categories that were consumed by trout for treatments containing trout. A n * indicates a significant difference between treatments. 108 0.3-c "ti o 2 . ft 0,25-O JO 0.2-1 0.15-<£d O 0.1-ndi »— 0,05-U -Figure 5.3. Mean proportion of experimental terrestrial inputs (+ 1 S.E.) in all prey size categories that were assigned to the category "benthos"- for each of four treatments. Different letters indicate significant differences. 109 Riffles Connected Pools Isolated Pools Figure 5.4. Magnitudes of the pathways which terrestrial prey subsidies follow into stream ecosystems when (A) waterstriders are present and ( B ) waterstriders are absent in the three habitat types (riffles - left, connected pools - center, isolated pools - right). Numbers are proportion of experimental prey diverted to each category or consumer (WS = waterstriders, C T = cutthroat trout, benthos = available to benthic detritivores and scavengers). Proportions do not sum to one because of experimental prey that either drifted out of the study reach or successfully escaped to stream banks or sides of experimental arenas. Asterisks (*) mark those pathways not directly quantified in this study; values are estimates. Dashed lines indicate those pathways that link terrestrial prey to a consumer group found in only low densities in the recipient habitat (e.g., there were very low densities of cutthroat trout in riffle habitats). 110 Works Cited Allan , J. D. , M . S. Wipf l i , J. P. Caouette, A . Prussian, and J. Rodgers. 2003. Influence of streamside vegetation on inputs of terrestrial invertebrates to salmonid food webs. Canadian Journal of Fisheries and Aquatic Sciences 60:309-320. Baxter, C. V . , K . D . Fausch, M . Murakami, and P. L . Chapman. 2004. Fish invasion restructures stream and forest food webs by interrupting reciprocal prey subsidies. Ecology 85:2656-2663. Bondar, C. A . , B . K . , K . Zeron, and J. S. Richardson. 2005. Does trophic position of the omnivorous signal crayfish (Pasifastacus leniusculus) in a stream food web vary with life history stage or density? Canadian Journal of Fisheries and Aquatic Sciences 62:2632-2639. Boss, S. M . , and J. S. Richardson. 2002. Effects of food and cover on the growth, survival and movement of cutthroat trout (Oncorhynchus clarki) in coastal streams. Canadian Journal of Fisheries and Aquatic Sciences 59:1044-1053. Bustamente, R. H . , and G. M . Branch. 1996. The dependence of intertidal consumers on kelp-derived organic matter on the west coast of South Africa. Journal of Experimental Marine Biology and Ecology 196:1-28. Cadenasso, M . L . , S. T. A . Pickett, and K . C. Weathers. 2004. Effects of landscape boundaries on the flux of nutrients, detritus and organisms. Pages 154-168 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Chapman, D . W. 1966. Food and space as regulators of salmonid populations in streams. American Naturalist 100:345-357. Cooper, S. D . 1984a. he effects of trout on water striders in stream pools. Oecologia 63:376-379. Cooper, S. D . 1984b. Prey preferences and interactions of predators from stream pools. Verh. Internat. Verein. L imnol . 22:1853-1857. D i l l , L . M . , and R. C. Ydenberg. 1987. The group size-flight distance relationships in waterstriders (Gerris remigis). Canadian Journal of Zoology 65:223-226. Dugan, J. E . , D . M . Hubbard, M . D. McCrary, and M . O. Pierson. 2003. The response of macrofauna communities and shorebirds to macrophyte wrack subsidies on exposed sandy beaches of southern California. Estuarine, Coastal and Shelf Science 58ST33-148. Dunbrack, W. F. 1988. Feeding of juvenile coho salmon (On'chorhynchus kisutch): maximum appetite, sustained feeding rate, appetite return, and body size. Canadian Journal of Fisheries and Aquatic Sciences 45:1191-1196. Erlandsson, A . 1988. Food-sharing vs. monopolising prey: a form of kleptoparasitism in Velia caprai (Heteroptera). Oikos 53:203-206. I l l Fairbairn, D . J., and J. Brassard. 1988. Dispersion and spatial orientation of Gerris remigis in response to water current: a comparison of pre- and post-diapause adults. Physiological Entomology 13:153-164. Gende, S. M . , R. T. Edwards, M . F. Willson, and M . S. Wipf l i . 2002. Pacific salmon in aquatic and terrestrial ecosystems. Bioscience 52:917-928. Helfield, J. M . , and R. J. Naiman. 2006. Keystone interactions: salmon and bear in riparian forests of Alaska. Ecosystems 9:167-180. Iwata, T., S. Nakano, and M . Murakami. 2003. Stream meanders increase insectivorous bird abundance in riparian deciduous forests. Ecography 26:325-337. Jamieson, G . S., and G . G . E . Scudder. 1977. Food consumption in Gerris (Hemiptera). Oecologia 30:23-41. Kawaguchi, Y . , and S. Nakano. 2001. Contribution of terrestrial invertebrates to the annual resource budget for salmonids in forest and grassland reaches of a headwater stream. Freshwater Biology 46:303-316. Kawaguchi, Y . , Y . Taniguchi, and S. Nakano. 2003. Terrestrial invertebrate inputs determine the local abundance of stream fishes in a forested stream. Ecology 84:701-708. Kiffney, P. M . , J. S. Richardson, and J. P. B u l l . 2004. Establishing light as a causal mechanism structuring stream communities in response to experimental manipulation of riparian buffer width. Journal of the North American Benthological Society 23:542-555. Kiffney, P. M . , J. S. Richardson, and M . Feller, C. 2000. Fluvial and epilithic organic matter dynamics in headwater streams of southwestern British Columbia, Canada. Archiv fur Hydrobiologie 149:109-129. Marczak, L . B . , and J. S. Richardson, in prep. Spiders and subsidies: results from the riparian zone of a coastal temperate rainforest. Maron, J., J. A . Estes, D . A . Crol l , E . M . Danner, S. C. Elmendorf, and S. L . Buckelew. 2006. A n introduced predator alters Aleutian Island plant communities by thwarting nutrient subsidies. Ecological Monographs 76:3-24. Mendelssohn, I. A . , and N . L . Kuhn. 2003. Sediment subsidy: effects on soil-plant responses in a rapidly submerging coastal salt marsh. Ecological Engineering 21:115-128. Nakano, S., H . Miyasaka, and N . Kuhara. 1999. Terrestrial-aquatic linkages: riparian arthropod inputs alter trophic cascades in a stream food web. Ecology 80:2435-2441. Paetzold, A . , J. F. Bernet, and K . Tockner. 2006. Consumer-specific responses to riverine subsidy pulses in a riparian arthropod assemblage. Freshwater Biology 51:1103-1115. Paetzold, A . , and K . Tockner. 2005. Effects of riparian arthropod predation on the biomass and abundance of aquatic insect emergence. Journal of the North American Benthological Society 24:395-402. 112 Power, M . E . , W. E . Rainey, M . S. Parker, J. L . Sabo, A . Smyth, S. Khandwala, J. C. Finlay, F. C. McNeely, K . Marsee, and C. Anderson. 2004. River-to-watershed subsidies in an old-growth conifer forest. Pages 217-240 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Sabo, J. L . , and M . E . Power. 2002. River-watershed exchange: effects of riverine subsidies on riparian lizards and their terrestrial prey. Ecology 83:1860-1869. Sand-Jensen, K . , and J. Borum. 1991. Interactions among phytoplankton, periphyton and macrophytes in temperate freshwaters and estuaries. Aquatic Botany 41:137-175. Sanzone, D . M . , J. L . Meyer, E . Mart i , E . P. Gardiner, J. L . Tank, and N . B . Grimm. 2003. Carbon and nitrogen transfer from a desert stream to riparian predators. Oecologia. 134:238-250. Schindler, D . E . , and M . D. Scheuerell. 2002. Habitat coupling in lake ecosystems. Oikos 98:177-189. Spence, J. R., andN. M . Andersen. 1994. Biology of waterstriders: interactions between systematics and ecology. Annual Review of Entomology 39:101-128. Spilke, J., H . P. Piepho, and X . Hu. 2005. Analysis of unbalanced data by mixed linear models using the M I X E D Procedure of the S A S system. Journal of Agronomy and Crop Science 191:47-54. Sweeney, B . W. , and R. L . Vannote. 1982. Population synchrony in mayflies: a predator satiation hypothesis. Evolution 36:810-821. Takimoto, G , T. Iwata, and M . Murakami. 2002. Seasonal subsidy stabilizes food web dynamics: balance in a heterogeneous landscape. Ecological Research 17:433-439. Wipf l i , M . S. 2005. Trophic linkages between headwater forests and downstream fish habitats: implications for forest and fish management. Landscape and Urban Planning 72:205-213. Wipf l i , M . S., and D . P. Gregovich. 2002. Export of invertebrates and detritus from Ashless headwater streams in southeastern Alaska: implications for downstream salmonid production. Freshwater Biology 47:957-969. Wiseman, S. W. , and S. D. Cooper. 1988. The predatory effects of water striders on emerging insects. Verh. Internat. Verein. Limnol . 23:2141-2144. Witman, J. D . , J. C. El l is , and W. B . Anderson. 2004. The influence of physical processes, organisms, and permeability on cross-ecosystem fluxes. Pages 335-358 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Zhang, Y . , J. N . Negishi, J. S. Richardson, and R. Kolodziejczyk. 2003. Impacts of marine-derived nutrients on stream ecosystem functioning. Proceedings of the Royal Society of London 270:2117-2123. 113 C H A P T E R 6: Untangling the web: complex effects of resource subsidies The movement of energy, nutrients and organisms between habitats is not a new topic for ecologists. A n increasing number of authors have considered the importance of the openness of systems (e.g. Holt 2002, Huxel et al. 2004, McCann et al. 2005) and the movement across system boundaries of materials (e.g. Wallace et al. 1997, Rose and Polis 1998, Anderson and Polis 1999, Mendelssohn and Kuhn 2003, Stapp and Polis 2003) and prey (e.g. Henschel et al. 2001, Nakano and Murakami 2001, Lynch et al. 2002, Sabo and Power 2002, Baxter et al. 2004). Both theoretical and empirical works have demonstrated the importance of allochthonous inputs of resources for population dynamics and food web structure. A number of researchers (e.g. Holt 2002, Huxel et al. 2002, McCann et al. 2005) have modelled the importance of resource subsidies as a factor determining the stability of recipient ecosystems. Witman et al. (2004) have discussed the ways in which physical processes and landscape variables might affect cross-ecosystem fluxes. Gary Polis, together with various co-authors (e.g. Polis and Hurd 1996, Polis et al. 1997, Polis et al. 1998, Rose and Polis 1998, Stapp and Polis 2003, Anderson and Polis 2004), considered the effects of temporal variability on the response to resource subsidies as well as outlining the importance of various geometric relationships (such as perimeter to area relationships). Despite these advances, theory has lagged behind empirical demonstrations of resource subsidy impacts. In particular, we lack a strong framework for predicting or understanding the conditions under which flows of material become subsidies and where we might expect these flows to occur. The research in this thesis has focused on our ability to make predictions about which habitats or consumers might be most influenced by system openness. It has been hypothesized (e.g., Polis et al. 1997, Kato et al. 2003, Witman et al. 2004) that primary productivity gradients 114 determine the overall importance of subsidies in a recipient habitat, with large contrasts between the donor and recipient habitat productivity producing the largest effects. This prediction is based on the reasonable assumption that the amount of subsidy produced and transported between habitats should be proportional to the overall productivity of the donor habitat in addition to the unreasonable assumption that such fluxes follow a diffusion model of transport from more to less productive habitats. In Chapter 2,1 demonstrated that strongly linear habitats such as coastlines or riparian forests had greater magnitude responses to resource subsidies, that lower trophic levels (detritivores, herbivores) respond more strongly than higher trophic levels (predators), and that these responses do not depend on gross contrasts of net primary productivity between donor and recipient habitats. Instead, I proposed that strong effects of subsidies could be anticipated where the abundance of a subsidy resource is high relative to the equivalent resource in the recipient habitat. Although there remain significant constraints and difficulties inherent to meta-analyses (e.g. limited data sets, non-independence of studies, publication bias), this analysis represents the first attempt at a broad synthesis of the many empirical studies present in the literature. I further explored the relationship between donor and recipient habitat productivity in an empirical test of the effect of a subsidy of emerging aquatic insects on the distribution and abundance of riparian orb-weaving spiders in a productive riparian rainforest (Chapter 3). I used a replicated, large scale habitat manipulation to empirically demonstrate that strong responses to subsidies can still occur where donor productivity (small headwater streams) is less than that of the recipient habitat (riparian forest). Not only was the distribution of riparian orb-weaving spiders significantly reduced beside stream reaches with aquatic insect exclusion, but this reduction occurred for taxonomic groupings not previously demonstrated to strongly track the abundance of this subsidy. Many authors have noted the effects of temporal variability in resource availability as a determining factor in the distribution, abundance and fitness of organisms. I used a riparian orb-115 weaving spider (Tetragnatha.versicolor) to investigate the consequences of variability in the timing of resource subsidies for the fitness of a recipient consumer (Chapter 4). I demonstrated that the way in which a subsidy intersects with the life history of an organism may critically alter whether that pulse of allocthonous food is received as a subsidy. In a series of controlled laboratory trials I showed that individuals of T. versicolor that received an initial high pulse of food (mimicking a short emergence of aquatic insects) followed by constricted food rations, performed worse on all measures of fitness (growth rate, final size, development time, body condition, l ipid storage) than spiders that either experienced initially low levels of food availability followed by a late pulse of food (mimicking a late emergence of aquatic insects) or a constant supply of food. Receiving large subsidies of food early in life appears to actually confer disadvantages for this species relative to a constant food supply or a subsidy late in development - illustrating that the physiological costs of rapid growth are likely greater than the later costs associated with compensatory growth (Yearsley et al. 2004, Johnsson and Bohlin 2005, Emlet and Sadro 2006). The timing of exposure to a resource subsidy can thus determine whether a subsidy produces benefits for an organism or produces the paradoxical effect of negative consequences (i.e. is by definition no longer a subsidy). I further demonstrated that the rate or amount of a subsidy reaching a consumer may be controlled by animals feeding at the interface between two ecosystems and that this control varies in response to the physical character of the habitat and the other species present (Chapter 5). B y foraging at the air-water interface for fallen terrestrial or emerging aquatic insects, surface-foraging consumers such as waterstriders have the potential to interrupt the flow of nutrients across the terrestrial-aquatic boundary. This biotic control on the quantity of subsidy transiting between two habitats has the potential to disconnect responses to subsidies from the relative productivities of donor and recipient habitats. Nutrients intercepted by boundary-foraging organisms may be redirected to unexpected habitat compartments or prevented from 116 reaching the usual recipient consumer in amounts sufficient to alter the overall strength of the i response of that consumer. This demonstration of trophic interception - or the ways in which physical or organismal agents may either concentrate or limit the magnitude of a subsidy -represents a critical piece in a broader attempt to understand jwhere and why subsidies have large \ effects in some habitats but not in others. i i Allochthonous nutrient inputs can be a key component that contribute greatly to system-wide productivity and affect community dynamics in recipient habitats. However it is critically important to determine when and how this general principle is systematically modified by habitat or species characteristics. Knowledge of the commonalities and differences in life history, productivity or other ecological characteristics that modify the response of consumers to resource subsidies (such as the preliminary work developed in this thesis) w i l l allow us to more efficiently predict those effects. Comments on future research There are a number of extensions or additional questions based on the research presented in this thesis that would prove fruitful to explore in greater detail. In particular, we lack further robust examples that illustrate the relationship between subsidy responses and productivity contrasts. Further field studies that are designed explicitly to address this question are necessary before we can fully reject a connection between donor recipient habitat productivity and the magnitude of consumer responses, or confidently modify this principle to reflect consumer-specific definitions of productivity. Although I have demonstrated a particular case where trophic interception occurs - the consequences of this interception for alternate consumers (in this case, fish and benthic consumers) has not been experimentally demonstrated. Determining both the commonness of this effect (other habitats, other consumers) and the magnitude of the direct and indirect 117 ( i I ) 1 consequences of interception are areas with substantial scope <for both theoretical and empirical work. In the case of the particular waterstrider-cutthroat trout system described in this thesis, it would be valuable to measure the magnitude of waterstrider trophic interception for resources moving from the stream to the riparian forest (interception of emerging aquatic invertebrates). A s I noted in chapter 5, estimates of the impact of generalist predators on prey populations exist in the literature (e.g. Sweeney and Vannote 1982, Cooper and Walde 1990, Paetzold and Tockner 2005), including measures of the effectiveness of waterstriders in reducing the abundance of emerging aquatic invertebrates (Wiseman and Cooper 1988). The potential consequences of interception of these stream-to-forest subsidies for terrestrial consumers has not yet been addressed and could be pursued within the context of the riparian spider community studied here, providing a more complete picture of the way in which mobile prey connect the abundance, distribution and potential fitness of in-stream consumers, surface-foraging organisms and terrestrial predators. I have begun work on two analytical models: (1) a model that assesses the role of temporal variability in terrestrial subsidy inputs to streams on the effectiveness of waterstriders interception and subsequent effects on fish populations and benthic consumers, and (2) a more general model that describes the effect of temporal variability and productivity contrasts in donor and recipient habitats for the magnitude of consumer responses to resource subsidies in both recipient and donor habitats. Researchers interested in cross-habitat fluxes have so far been exclusively concerned with the effects of spatial subsidies within recipient habitats. Investigators have looked at the movements of detritus from high productivity ocean environments to dry islands (Polis and Hurd 1996, Polis et al. 1997, Takimoto et al. 2002), from streams to riparian areas in deserts (Sanzone et al. 2003) and subsidies to lizards on the dry rocky shores of California rivers (Sabo and Power 2002). While several recent authors have examined the importance of reciprocal subsidies, the focus of these works has remained on effects in recipient systems (Nakano et al. 1999, Power 118 2001, Baxter et al. 2005). A s of yet, rio-dhe has investigated the potential effects, or potentially unique consumer-resource dynamics engendered by the loss of subsidies within a donor system. I believe this has occurred in part because of assumptions built into the implicit definition of what constitutes a spatial subsidy and a focus on the case most likely to produce strong results. Although internally consistent, the definition of a resource subsidy as one having a benefit to a consumer in the recipient habitat (Polis et al. 1997) has directed research towards simple demonstrations of this fact while broader considerations of system openness have languished. While this definition may be an accurate description of a resource subsidy it does not address the case of resource flows between systems that have no benefit or even a negative effect for a recipient (as demonstrated in Chapter 4). Defining subsidies only in terms of recipient benefits deflects our attention from the holistic consequences of resource movements between systems onto the specific effects in only one habitat type (the recipient habitat). The result has been that the full panoply of potential effects of resource subsidies has yet to be considered. Holt (2004) noted that spatial fluxes may exhibit strong directionality due to the action of prevailing winds, currents or gravity. There is little reason for these abiotic forces to work solely from productive to unproductive habitats. There is no logical reason, however, to suppose that the effects of spatial subsidies are important in recipient systems only. A s Loreau et al. (2003) comments in passing, a subsidy entering one ecosystem must necessarily be drawn from another. Subsidies in one place are losses in another, and as such should have an impact on both the donor and recipient habitats. Examining the effects of these involuntary losses to such source habitats has been neglected. The neglect of consequences of subsidies in donor systems is symptomatic of the absence of a consistent framework for understanding the potential consequences of resource subsidies. Since some resource depauperate systems receive subsidies from "wealthy" systems, it is logical to suppose the existence of donor systems which are themselves resource poor. L o w productivity 119 donor habitats, in this analogy, are the working poor of ecological systems. Producing just enough resources to support resident species, they may i l l afford any losses - whether these are to systems with a high contrast in productivity or not. I suggest that the focus on a single direction of subsidy movement with respect to system productivity, and with the effects of this movement in recipient ecosystems only, does not allow for a full understanding of the importance of openness for ecosystem dynamics. Recently Baxter et al., (2005) suggested that direct and indirect effects, the potential stabilizing effect of subsidies and the importance of human disturbance in reciprocal subsidies are important directions for new research in this field. To this list I would add: (l)What is the magnitude of flow required to have significant ecological effect between habitats? and (2) What are the effects of subsidy movement in donor systems? The flow of resources between habitats and across landscapes is now widely recognized as an important process that has the potential to both alter recipient community dynamics and our understanding of the linkages between food webs and landscapes. The receipt of allochthonous material (and potentially the donation of this material) can cause significant changes in the composition and structure of communities. However, the strength of the effect of subsidies is modified by the ability of a given consumer to use the subsidy and the permeability of a habitat boundary to a subsidy (Huxel et al. 2004). The conservation of some species or habitats may critically depend on the conservation of their adjacent or linked habitats. Beyond theoretical interest in the complexities of food webs - the renewed interest in the openness of ecosystems highlights the need to address system complexity in conservation and management. 120 Works Cited Anderson, W . B . , and G . A . Polis. 1999. Nutrient fluxes from water to land: seabirds affect plant nutrient status on G u l f of California islands. Oecologia 118:324-332. Anderson, W. B . , and G . A . Polis. 2004. Allochthonous nutrient and food inputs: consequences for temporal stability. Pages 82-95 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Baxter, C. V . , K . D . Fausch, M . Murakami, and P. L . Chapman. 2004. Fish invasion restructures stream and forest food webs by interrupting reciprocal prey subsidies. Ecology 85:2656-2663. Baxter, C. V . , K . D . Fausch, and C. W . Saunders. 2005. Tangled webs: reciprocal flows of invertebrate prey link streams and riparian zones. Freshwater Biology 50:201-220. Cooper, S. D. , and S. J. Walde. 1990. Prey exchange rates and the impact of predators on prey populations in streams. Ecology 71:1503-1514. Emlet, R. B . , and S. S. Sadro. 2006. Linking stages of life history: how larval quality translates into juvenile performance for an intertidal barnacle (Balanus glandula). Integrative and Comparative Biology 46:334-346. Henschel, J. R., D . Mahsberg, and H . Stumpf. 2001. Allocthonous aquatic insects increase predation and decrease herbivory in river shore food webs. Oikos 93:429-438. Holt, R. D . 2002. Food webs in space: on the interplay of dynamic instability and spatial processes. Ecological Research 17:261-273. Holt, R. D . 2004. Implications of system openness for local community structure and ecosystem function. Pages 96-114 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. University of Chicago Press, Chicago. Huxel, G . R., K . McCann, and G . A . Polis. 2002. Effects of partitioning allocthonous and autocthonous resources on food web stability. Ecological Research 17:419-432. Huxel, G . R., G . A . Polis, and R. D . Holt. 2004. A t the frontier of the integration of food web ecology and landscape ecology. Pages 434-451 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Johnsson, J. I., and T. Bohlin. 2005. Compensatory growth for free? a field experiment on brown trout, Salmo trutta. Oikos 111:31-38. Kato, C , T. Iwata, S. Nakano, and D . Kish i . 2003. Dynamics of aquatic insect flux affects distribution of riparian web-building spiders. Oikos 103:113-120. Loreau, M . , N . Mouquet, and R. D . Holt. 2003. Meta-ecosystems: a theoretical framework for a spatial ecosystem ecology. Ecology Letters 6:673-679. 121 Lynch, R. J., S. E . Bunn, and C. P. Catterall. 2002. Adult aquatic insects: potential contributors to riparian food webs in Australia's wet-dry tropics. Austral Ecology 27:515-526. McCann, K . S., J. B . Rasmussen, and J. Umbanhowar. 2005. The dynamics of spatially coupled food webs. Ecology Letters 8:513-523. Mendelssohn, I. A . , and N . L . Kuhn. 2003. Sediment subsidy: effects on soil-plant responses in a rapidly submerging coastal salt marsh. Ecological Engineering 21:115-128. Nakano, S., H . Miyasaka, and N . Kuhara. 1999. Terrestrial-aquatic linkages: riparian arthropod inputs alter trophic cascades in a stream food web. Ecology 80:2435-2441. Nakano, S., and M . Murakami. 2001. Reciprocal subsidies: dynamic interdependence between terrestrial and aquatic food webs. Proceedings of the National Academy of Sciences -U S A 98:166-170. Paetzold, A . , and K . Tockner. 2005. Effects of riparian arthropod predation on the biomass and abundance of aquatic insect emergence. Journal of the North American Benthological Society 24:395-402. Polis, G . A . , W. B . Anderson, and R. D. Holt. 1997. Toward an integration of landscape and food web ecology: the dynamics of spatially subsidized food webs. Annual Review of Ecological Systematics 28:289-316. Polis, G . A . , and S. D . Hurd. 1996. Linking marine and terrestrial food webs: allocthonous input from the ocean supports high secondary productivity on small islands and coastal land communities. The American Naturalist 147:396-423. Polis, G . A . , S. D . Hurd, T. C. Jackson, and F. Sanchez-Pinero. 1998. Multifactor population limitation: variable spatial and temporal control of spiders on G u l f of California islands. Ecology 79:490-502. Power, M . E . 2001. Prey exchange between a stream and its forested watershed elevates predator densities in both habitats. Proceedings of the National Academy of Sciences - U S A 98:14-15. Rose, M . D . , and G. A . Polis. 1998. The distribution and abundance of coyotes: the effects of allocthonous food subsidies from the sea. Ecology 79:998-1007. Sabo, J. L . , and M . E . Power. 2002. River-watershed exchange: effects of riverine subsidies on riparian lizards and their terrestrial prey. Ecology 83:1860-1869. Sanzone, D . M . , J. L . Meyer, E . Marti , E . P. Gardiner, J. L . Tank, and N . B . Grimm. 2003. Carbon and nitrogen transfer from a desert stream to riparian predators. Oecologia. 134: 238-250 Stapp, P., and G. A . Polis. 2003. Marine resources subsidize insular rodent populations in the G u l f of California, Mexico. Oecologia 134:496-504. 122 Sweeney, B . W. , and R. L . Vannote. 1982. Population synchrony in mayflies: a predator satiation hypothesis. Evolution 36:810-821. Takimoto, G . , T. Iwata, and M . Murakami. 2002. Seasonal subsidy stabilizes food web dynamics: balance in a heterogeneous landscape. Ecological Research 17:433-439. Wallace, J. B . , S. L . Eggert, J. L . Meyer, and J. R. Webster. 1997. Multiple trophic levels of a forest stream linked to terrestrial litter inputs. Science 277:102-104. Wiseman, S. W., and S. D. Cooper. 1988. The predatory effects of water striders on emerging insects. Verh. Internat. Verein. L imnol . 23:2141-2144. Witman, J. D . , J. C. El l is , and W. B . Anderson. 2004. The influence of physical processes, organisms, and permeability on cross-ecosystem fluxes. Pages 335-358 in G . A . Polis, M . E . Power, and G . R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago. Yearsley, J. M . , I. Kryriazakis, and I. J. Gordon. 2004. Delayed costs of growth and compensatory growth rates. Functional Ecology 18:563-570. 123 Appendix A . Studies included in meta-analysis with treatment and control means and standard deviation. author-date code subsidy type recipient taxa trophic level study type n treatment mean units TRT sd control mean units control sd response variable Barrett et al.2005 marine debris Uta stansburiana predator descriptive 2 9 # sighted/hr 1 25 # sighted/hr 4 abundance Barrett et al.2005 marine bird guano Uta stansburiana predator descriptive 9 26 # sighted/hr 8 78 # sighted/hr 20 abundance Bastow et al. 2002 algal mats tetrigidae detritivore descriptive 3 0.0707 #/m2 0.1141 0.2173 #/m2 0.2082 density Bastow et al. 2002 algal mats tetrigidae detritivore descriptive 3 0.0294 #/m2 0.0336 1.075 #/m2 0.6041 density Bastow et al. 2002 algal mats tetrigidae detritivore descriptive 3 0.0191 #/m2 0.0336 1.154 #/m2 0.8693 density Bastow et al. 2002 algal mats tetrigidae detritivore manipulative 6 1.12! #/m2 0.8355 35.96 #/m2 24.4046 density Bastow et al. 2002 algal mats acrididae detritivore manipulative 3 0.02 #/m2 0.0346 0.14 #/m2 0.1386 density Bastow et al. 2002 algal mats gelastocoridae detritivore manipulative 3 0.05 #/m2 0.052 4.1 #/m2 0.1559 density Baxter et al. 2004 aquatic insects • spiders predator manipulative 4 0.0252 #/m2 0.0271 0.1452 #/m2 0.04 density Bomkamp et al. 2004 marine debris echinoderms predator manipulative 3 0.5 #/m2 0.4 2.5 #/m2 1 density Bomkamp et al. 2004 marine debris mollusks detritivore manipulative 3 0.1 #/m2 0.2 0.45 #/m2 0.2 density Bouchard and Bjorndal 2000 marine turtle eggs racoons omnivore descriptive 2 1 percent 28 percent percent of total Bouchard and Bjorndal 2000 marine turtle eggs ghost crabs omnivore descriptive 2 1 percent 2.5 percent percent of total Bouchard and Bjorndal 2000 marine turtle eggs plants producer descriptive 2 1 percent 23 percent percent of total Eggert and Wallace 2003 leaf litter Pycnopsyche gentilis detritivore manipulative 4 2 #/m2 175 #/m2 abundance Gende and Willson 2001 salmon birds insectivore descriptive 7 4.21 #/station/ day 0.7 5.14 #/station/ day 0.7408 density Gende and Willson 2001 salmon Pacific slope flycatcher insectivore descriptive 7 0.29 #/station/ day 0.7 #/station/ day density Gende and Willson 2001 salmon townsends warbler insectivore descriptive 7 0.45 #/station/ day 0.6 #/station/ day density Gende and Willson 2001 salmon golden crowned kinglet insectivore descriptive 7 0.43 #/station/ day 0.54 #/station/ day density Gende and Willson 2001 salmon chestnut backed chickadee insectivore descriptive 7 0.36 #/station/ day 0.42 #/station/ day density Gende and Willson 2001 salmon dark-eyed junco insectivore descriptive 7 0 #/station/ day 0.03 #/station/ day density author-date code subsidy type recipient taxa trophic level study type n treatment mean units TRTsd control mean units control sd response variable Gende and Willson 2001 salmon pine siskin insectivore descriptive 7 0 #/station/ day 0.03 ^/station/ day density Gende and Willson 2001 salmon tree swallow insectivore descriptive 7 0 #/station/ day 0.01 #/station/ day density Gende and Willson 2001 salmon western tanager insectivore descriptive 7 0 #/station/ day 0.01 #/station/ day density Gende and Willson 2001 salmon varied thrush insectivore descriptive 7 0.39 #/station/ day 0.4 #/station/ day density Gende and Willson 2001 salmon american robin insectivore descriptive 7 0.11 #/station/ day 0.1 #/station/ day density Gende and Willson 2001 salmon fox sparrow insectivore descriptive 7 0.04 #/station/ day 0 #/station/ day density Gende and Willson 2001 salmon winter wren insectivore descriptive 7 0.63 #/station/ day 0.58 A/station/ day density Gende and Willson 2001 salmon ruby crowned kinglet insectivore descriptive 7 0.17 #/station/ day 0.13 #/station/ day density Gende and Willson 2001 salmon hermit thrush insectivore descriptive 7 0.3 #/station/ day 0.25 #/station/ day density Gende and Willson 2001 salmon swainson's thrush insectivore descriptive 7 0.07 #/station/ day 0.01 #/station/ day density Gende and Willson 2001 salmon orange-crowned warbler insectivore descriptive 7 0.12 #/station/ day 0.04 #/station/ day density Gende and Willson 2001 salmon yellow warbler insectivore descriptive 7 0.08 ft/station/ day 0 #/station/ day density Gende and Willson 2001 salmon wilson's warbler insectivore descriptive 7 0.28 #/station/ day 0.05 #/station/ day density Henschel et al. 2001 aquatic insects horizontal orb weavers predator descriptive 9 9.2 #/m2 7.1 21.9 #/m2 21.1 density Henschel et al. 2001 aquatic insects horizontal orb weavers predator descriptive 9 10.5 #/m2 8.1 21.9 #/m2 21.1 .density Iwataet al. 2003 aquatic insects flycatchers insectivore descriptive 26 0.44 #ind 41.519 8 U ind density Iwataet al. 2003 aquatic insects gleaners insectivore descriptive 26 11.31 #ind 219.90 78 # ind density Iwata et al. 2003 aquatic insects bark probers insectivore descriptive 26 2.43 # ind 17.893 4 # ind density Johnson et al. 2003 leaf litter stoneflies predator manipulative 2 0.9 percent mortality 0.3 0.55 percent mortality 0.2 mortality Johnson et al. 2003 leaf litter stoneflies predator manipulative 2 0.9 percent mortality 0.3 0.55 percent mortality 0.2 mortality Johnson et al. 2003 leaf litter stoneflies predator manipulative 2 0.9 percent mortality 0.3 0.55 percent mortality 0.2 mortality author-date code subsidy type recipient taxa trophic level study type n treatment mean units TRT sd control mean units control sd response variable Johnson et al. 2003 leaf litter stoneflies predator manipulative 2 0.9 percent mortality 0.3 0.55 percent mortality 0.2 mortality Kato et al. 2003 aquatic insects horizontal orb weavers predator manipulative 6 0.25 #/m2 0.098 0.65 #/m2 0.2205 density Kato et al. 2003 aquatic insects horizontal orb weavers predator manipulative 6 0.09 #/m2 0.0735 0.34 #/m2 0.1225 density Kato et al. 2003 aquatic insects horizontal orb weavers predator manipulative 6 0.03 #/m2 0.049 0.1 #/m2 0.1225 density Kato et al. 2003 aquatic insects vertical orb weavers predator manipulative 6 0.31 #/m2 0.0735 0.24 #/m2 0.098 density Katoet al.2003 aquatic insects vertical orb weavers predator manipulative 6 0.24 #/m2 0.0735 0.24 #/m2 0.0735 density Kato et al. 2003 aquatic insects vertical orb weavers predator manipulative 6 0.06 #/m2 0.049 0.07 #/m2 0.049 density Kato et al. 2003 aquatic insects sheet weavers predator manipulative 6 0.41 #/m2 0.1715 0.52 #/m2 0.2449 density Kato et al. 2003 aquatic insects sheet weavers predator manipulative 6 0.24 #/m2 0.1225 0.33 #/m2 0.147 density Kato et al. 2003 aquatic insects sheet weavers predator manipulative 6 0.11 #/m2 0.098 0.17 #/m2 0.098 density Kawaguchi and Nakano 2001 terrestrial invertebrates salmonids predator descriptive 2 1273 mass/100 mg 325.127 7 2821 mass/100 mg 379.2921 biomass Kawaguchi and Nakano2001 terrestrial invertebrates salmonids predator descriptive 2 1692 mass/100 mg 531.037 2 4052 mass/100 mg 433.4565 biomass Kawaguchi and Nakano 2001 terrestrial invertebrates salmonids predator descriptive 2 2171 mass/100 mg 498.510 3 5972 mass/100 mg 444.3459 biomass Kawaguchi and Nakano 2001 terrestrial invertebrates salmonids predator descriptive 2 1456 mass/100 mg 270.963 3 2099 mass/100 mg 359.4931 biomass Kawaguchi et-al 2003 terrestrial invertebrates salmonids predator manipulative 4 41.7 mass/100 mg 1.5 55.2 mass/100 mg 2.9 biomass Kawaguchi et al 2003 terrestrial invertebrates salmonids predator manipulative 4 212.5 mass/100 mg 96 412.5 mass/100 mg 35.18 biomass Marczak and Richardson 2006 aquatic insects vertical orb weavers predator descriptive 2 6.5 # ind/m2 0.7 13.5 tt ind/m2 2.12 abundance Marczak and Richardson 2006 aquatic insects sheet weavers predator descriptive 2 19 tt ind/m2 5.65 20 # ind/m2 12.02 abundance Marczak and Richardson 2006 aquatic insects horizontal orb weavers predator descriptive 2 7.5 tt ind/m2 0.7 15 # ind/m2 9.89 abundance Marczak and Richardson 2006 aquatic insects tangle web weavers predator descriptive 2 11 tt ind/m2 15.5 10 # ind/m2 11.3 abundance Marczak unpublished data aquatic insects vertical orb weavers predator manipulative 16 0.00997 #ind 0.032 0.0523 tt ind 0.032 density Marczak unpublished data aquatic insects horizontal orb weavers predator manipulative 16 0.07 tt ind 0.0532 0.1228 tt ind 0.0532 density author-date code subsidy type recipient taxa trophic level study type n treatment mean units TRT sd control mean units control sd response variable Marczak unpublished data aquatic insects sheet weavers predator manipulative 16 0.0589 # ind 0.0576 0.109 # ind ' ' 0.0576 density Marczak unpublished data aquatic insects dwarf sheet weavers predator manipulative 16 0.0059 # ind 0.028 0.0444 # ind 0.028 density Medelssohn and Kuhn 2003 marine debris plants producer manipulative 5 47 % cover 15.65 44 % cover 6.71 percent cover Medelssohn and Kuhn 2003 marine debris plants producer manipulative 5 47 % cover 15.65 76 % cover 8.94 percent cover Murakami and Nakano 2002 aquatic insects Great tit insectivore descriptive 3 0.588 visitation rate/hour 0.6495 4.671 visitation rate/hour 1.0685 density Murakami and Nakano 2002 aquatic insects Marsh tit insectivore descriptive 3 2.078 visitation rate/hour 1.5084 1.588 visitation rate/hour 1.0477 density Murakami and Nakano 2002 aquatic insects Crowned willow warbler insectivore descriptive 3 0.1199 visitation rate/hour 0.1152 0.6583 visitation rate/hour 0.3038 density Nakano etal. 1999 terrestrial invertebrates dolly varden charr predator manipulative 8 130 mm 5.6569 132 mm 5.6569 biomass Polis and Hurd 1995 marine debris orb weavers predator descriptive 7 0.025 #/m2 0.017 0.155 #/m2 0.04 density Polis and Hurd 1995 marine bird guano orb weavers predator descriptive 7 6.15 #/m3 3.92 25.1 #/m3 2.11 density Polis and Hurd 1995 marine bird guano orb weavers predator descriptive 7 1.8 #/m3 1.5 14.3 #/m3 1.5 density Polis and Hurd 1995 marine debris insects detritivore descriptive 7 1.6 mm/trap/h r 2.5 175.8 mm/trap/ hr 2 biomass Polis and Hurd 1995 marine bird guano insects detritivore descriptive 7 5.46 mm/trap/h r 1.45 12.2 mm/trap/ hr 1.65 density Polis and Hurd 1995 marine bird guano insects detritivore descriptive 7 4.5 #/day 1.8 12.3 #/day 1 density Rose and Polis 1998 marine debris coyote omnivore descriptive 4 2.799 tracks/tran sect/night 1.4242 14.2 tracks/tra nsect/nig ht 2.216 density Sabo and Power 2002 aquatic insects Sceloporus sp. predator manipulative 4 0.0317 g/day 0.5084 0.2097 g/day 0.22 growth rate Sabo and Power 2002 aquatic insects Sceloporus sp. predator manipulative 4 2.12 ind/plot 0.12 3.37 ind/plot 1.06 abundance Sabo and Power 2002a aquatic insects Sceloporus sp. predator manipulative 4 0.036 g/day 0.0671 0.0219 g/day 0.0178 growth rate Sanchez-Pinero and Polis 2000 marine carrion beetles detritivore descriptive 4 3 #/trap 4 12.5 #/trap 8 abundance Sanchez-Pinero and Polis 2000 marine bird guano beetles detritivore descriptive 17 19 #/trap 10 73 #/trap 12 abundance Stapp and Polis 2003 marine debris Peromyscus omnivore descriptive 3 23.25 proportion of captures 15.7963 49.73 proportio n of captures 8.3485 density trophic treatment control author-date code subsidy type recipient taxa level study type n mean units TRTsd mean units control sd response variable proportion of proportio n of density Stapp and Polis 2003 marine debris Peromyscus omnivore descriptive 3 68.45 captures 13.9257 8.87 captures 8.0194 proportion of proportio n of Stapp and Polis 2003a marine debris Peromyscus omnivore descriptive 6 22.47 captures 16.0931 57.26 captures 29.6878 density birds/ha/ birds/ha/h Uesugi 2002 aquatic insects riparian birds insectivore descriptive 12 0.1 hr 1.0392 3.7 r 4.1569 abundance birds/ha/ birds/ha/h Uesugi 2002 aquatic insects upland birds insectivore descriptive 12 21.6 hr 8.6603 21.5 r 9.3531 abundance black faced birds/ha/ birds/ha/h Uesugi 2002 aquatic insects bunting insectivore descriptive 12 2 hr 0.6928 2 r 0.7967 abundance e. crowned willow birds/ha/ birds/ha/h Uesugi 2002 aquatic insects warbler insectivore descriptive 12 3.1 hr 0.6928 2.9 r 0.6928 abundance birds/ha/ birds/ha/h Uesugi 2002 aquatic insects great tit insectivore descriptive 12 2.4 hr 0.6928 2.6 r 1.0392 abundance birds/ha/ birds/ha/h Uesugi 2002 aquatic insects nuthatch insectivore descriptive 12 0.6 hr 1.2124 1.2 r 1.0392 abundance Japanese grey birds/ha/ birds/ha/h Uesugi 2002 aquatic insects thrush insectivore descriptive 12 2 hr 1.7321 2.7 r 1.7321 abundance birds/ha/ birds/ha/h Uesugi 2002 aquatic insects marsh tit insectivore descriptive 12 0.6 hr 1.7321 0.5 r 1.7321 abundance birds/ha birds/ha/h Uesugi 2002 aquatic insects treecreeper insectivore descriptive 12 0.9 /hr 2.0785 0.9 r 1.7321 abundance short-tailed birds/ha/ birds/ha/h Uesugi 2002 aquatic insects bush warbler insectivore descriptive 12 0.9 hr 1.3856 1 r 1.3856 abundance birds/ha/ birds/ha/h Uesugi 2002 aquatic insects coal tit insectivore descriptive 12 1.55 hr 3.4641 1.56 r 3.4641 abundance great spotted birds/ha/ birds/ha/h Uesugi 2002 aquatic insects woodpecker insectivore descriptive 12 0.3 hr 0.6928 0.6 r 2.0785 abundance birds/ha/ birds/ha/h Uesugi 2002 aquatic insects long tailed tit insectivore descriptive 12 0.5 hr 1.3856 0.6 r 1.0392 abundance Japanese pygmy birds/ha/ birds/ha/h Uesugi 2002 aquatic insects woodpecker insectivore descriptive 12 0.9 hr 1.7321 1.4 r 2.0785 abundance birds/ha/ birds/ha/h Uesugi 2002 aquatic insects varied tit insectivore descriptive 12 0.3 hr 1.3856 0.5 r 1.3856 abundance narcissus birds/ha/ birds/ha/h Uesugi 2002 aquatic insects flycatcher insectivore descriptive 12 3.1 hr 3.1177 2.9 r 3.1177 abundance brown birds/ha/ birds/ha/h Uesugi 2002 aquatic insects flycatcher insectivore descriptive 12 0.1 hr 0.6928 1.3 r 2.7713 abundance ON author-date code subsidy type recipient taxa trophic level study type n treatment mean units TRT sd control mean units control sd response variable Uesugi 2002 aquatic insects Blue, white flycatcher insectivore descriptive 12 0 birds/ha/ hr 0 0.4 birds/ha/h r 0.866 abundance Uesugi 2002 aquatic insects pale legged willow warbler insectivore descriptive 12 0 birds/ha/ hr 0 1.2 birds/ha/h r 2.4249 abundance Uesugi 2002 aquatic insects olive backed tree pipit insectivore descriptive 12 0 birds/ha/ hr 0 0.65 birds/ha/h r 2.2517 abundance Wallace etal. 1997 detritus Mytilus spp detritivore descriptive 6 23 g/m2 61.2372 478 g/m2 244.949 biomass Wallace etal. 1997 leaf litter scraper detritivore manipulative 2 367 #/m2 314 562 #/m2 469 abundance Wallace etal. 1997 leaf litter shredders detritivore manipulative 2 554 #/m2 431 954 #/m2 588 abundance Wallace etal. 199/ leaf litter gatherers detritivore manipulctive 2 18019 #/m2 9874 30940 #/m2 14431 abundance Wallace etal. 1997 leaf litter filterers detritivore manipulative 2 188 #/m2 233 323 #/m2 546 abundance Wallace etal. 1997 leaf litter consumers consumer manipulative 2 19128 #/m2 10237 32779 #/m2 14927 abundance Wallace etal. 1997 leaf litter predators predator manipulative 2 2883 #/m2 2121 4892 #/m2 2486 abundance Wallace etal. 1997 leaf litter salamander predator manipulative 2 1 #/m2 2 4 #/m2 6 abundance Wallace etal. 1999 leaf litter shredders detritivore manipulative 2 554 #/m2 431 954 #/m2 588 density Wallace etal. 1999 leaf litter gatherers detritivore manipulative 2 18019 #/m2 9874 30940 #/m2 14431 density Wipfli etal. 1999 salmon invertebrates omnivore manipulative 6 18.53 #/100cm2 2.23 64.84 #/100cm2 11.51 abundance Wipfli etal. 1999 salmon invertebrates omnivore manipulative 6 140.53 #/100cm2 32.65 402.9 AV100cm2 119 abundance Appendix B . Ful l citations of all papers used in the meta-analysis. Author-Date code F u l l Citat ion Barrett et al. 2005 Bastow et al. 2002 Baxter et al. 2004 Bomkamp et al. 2004 Bouchard and Bjorndal 2000 Eggert and Wallace 2003 Gende and Wil lson 2001 Henschel et al. 2001 Iwata et al. 2003 Johnson et al. 2003 Kato et al. 2003 Barrett, K . , Anderson, W. B . , Wait, A . , Grismer, L . L . , Polis, G . A . and Rose, M . D . 2005. Marine subsidies alter the diet and abundance of insular and coastal lizard populations. Oikos 109: 145-153. Bastow, J. L . , Sabo, J. L . , Finlay, J. C. and Power, M . E . 2002. A basal aquatic-terrestrial trophic link in rivers: algal subsidies via shore-dwelling grasshoppers. Oecologia 131: 261-268. Baxter, C. V . , Fausch, K . D. , Murakami, M . and Chapman, P. L . 2004. Fish invasion restructures stream and forest food webs by interrupting reciprocal prey subsidies. Ecology 85:2656-2663. Bomkamp, R. E . , Page, H . M . and Dugan, J. E . 2004. Role of food subsidies and habitat structure in influencing benthic communities of shell mounds at sites of existing and former offshore oil platforms. Marine Biology 146: 201-211. Bouchard, S. S. and Bjorndal, K . A . 2000. Sea turtles as biological transporters of nutrients and energy from marine to terrestrial ecosystems. Ecology 81: 2305-2313. Eggert, S., L . and Wallace, J. B . 2003. Reduced detrital resources limit Pychnopsyche gentilis (Tricoptera: Limnephilidae) production and growth. Journal of the North American Benthological Society 22: 388-400. Gende, S. M . and Willson, M . F. 2001. Passerine densities in riparian forests of southeast Alaska: potential effects of anadromous spawning salmon. The Condor 103: 624-629. Henschel, J. R., Mahsberg, D . and Stumpf, H . 2001. Allocthonous aquatic insects increase predation and decrease herbivory in river shore food webs. Oikos 93: 429-438. Iwata, T., Nakano, S. and Murakami, M . 2003. Stream meanders increase insectivorous bird abundance in riparian deciduous forests. Ecography 26: 325-337. Johnson, B . R., Cross, W. F. and Wallace, J. B . 2003. Long-term resource limitation reduces insect detritivore growth in a headwater stream. Journal of the North American Benthological Society. Kato, C , Iwata, T., Nakano, S. and Kishi , D . 2003. Dynamics of aquatic insect flux affects distribution of riparian web-building spiders. Oikos 103: 113-120 Author-Date code F u l l Citat ion Kawaguchi and Nakano 2001. Kawaguchi, Y . and Nakano, S. 2001. Contribution of terrestrial invertebrates to the annual resource budget for salmonids in forest and grassland reaches of a headwater stream. Freshwater Biology 46: 303-. Kawaguchi et al. 2003 Kawaguchi, Y . , Taniguchi, Y . and Nakano, S. 2003. Terrestrial invertebrate inputs determine the local abundance of stream fishes in a forested stream. Ecology 84: 701-708. Marczak and Richardson 2006 Marczak, L . B . and J.S. Richardson, (in review). Resource subsidies to productive habitats: aquatic insect emergence and the abundance of riparian web building spiders in a coastal temperate rainforest. Marczak unpublished Marczak, L . B . unpublished data Mendelssohn and Kuhn 2003 Mendelssohn, I. A . and Kuhn, N . L . 2003. Sediment subsidy: effects on soil-plant responses in a rapidly submerging coastal salt marsh. Ecological Engineering 21: 115-128. Murakami and Nakano 2002 Murakami, M . and Nakano, S. 2002. Indirect effect of aquatic insect emergence on a terrestrial insect population through birds predation. Ecology Letters 5: 333-227. Nakano et al. 1999 Nakano, S., Miyasaka, H . and Kuhara, N . 1999. Terrestrial-aquatic linkages: riparian arthropod inputs alter trophic cascades in a stream food web. Ecology 80: 2435-2441. Polis and Hurd 1995 Polis, G . A . and Hurd, S. D . 1995. Extraordinarily high spider densities on islands: flow of energy from the marine to terrestrial food webs and the absence of predation. Proceedings of the National Academy of Sciences - U S A 92: 4382-4386. Polis et al. 1998 Polis, G . A . , Hurd, S. D. , Jackson, T. C. and Sanchez-Pinero, F . 1998. Multifactor population limitation: variable spatial and temporal control of spiders on G u l f of California islands. Ecology 79: 490-502. Rose and Polis 1998 Rose, M . D. and Polis, G . A . 1998. The distribution and abundance of coyotes: the effects of allocthonous food subsidies from the sea. Ecology 79: 998-1007. Sabo and Power 2002 Sabo, J. L . and Power, M . E . 2002. River-watershed exchange: effects of riverine subsidies on riparian lizards and their terrestrial prey. Ecology 83: 1860-1869. Sabo and Power 2002a Sabo, J. L . and Power, M . E . 2002. Numerical response of riparian lizards to aquatic insects and short-term consequences for alternate terrestrial prey. Ecology. Sanchez-Pinero and Polis 2000 Sanchez-Pinero, F. and Polis, G . A . 2000. Bottom-up dynamics of allochthonous input: direct and indirect effects of seabirds on islands. Ecology 81: 3117-3132. 

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0302171/manifest

Comment

Related Items