UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of in-stream structure placements on a macroinvertebrate community : testing of reach-scale… Negishi, Junjiro N. 2001

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

Item Metadata

Download

Media
831-ubc_2001-0480.pdf [ 7.52MB ]
Metadata
JSON: 831-1.0090138.json
JSON-LD: 831-1.0090138-ld.json
RDF/XML (Pretty): 831-1.0090138-rdf.xml
RDF/JSON: 831-1.0090138-rdf.json
Turtle: 831-1.0090138-turtle.txt
N-Triples: 831-1.0090138-rdf-ntriples.txt
Original Record: 831-1.0090138-source.json
Full Text
831-1.0090138-fulltext.txt
Citation
831-1.0090138.ris

Full Text

THE EFFECTS OF IN-STREAM STRUCTURE P L A C E M E N T S ON A M A C R O L N V E R T E B R A T E C O M M U N I T Y : TESTING OF R E A C H - S C A L E RESPONSES A N D MECHANISMS WITH R E G A R D TO FLOW REFUGIA by JUNJIRO N . NEGISHI Bachelor of Agriculture, Hokkaido University, 1997 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES FACULTY OF FORESTRY (Department of Forest Sciences) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A J U L Y 2001 © Junjiro N . Negishi, 2001 w r i U Spec ia Uo e c . i o n s - nes is A u x n o r i s a ; i o n -orm r*age l o" I In presenting t h i s thes is in p a r t i a l fu l f i lment of the requirements for an advanced degree at the Univers i ty of B r i t i s h Columbia, I agree that the L ib ra ry sha l l make i t f ree ly ava i l ab le for reference and study. I further agree that permission for extensive copying of t h i s thes i s for scho la r ly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or pub l i ca t ion of t h i s thes is for f i nanc i a l gain sha l l not be allowed without my wr i t t en permission. Department The Unive r s i ty of B r i t i s h Columbia Vancouver, Canada http:/ /www. I ib ra ry . ubc. c a / s p c o l l / t h e s a u t h . html 7/27/01 A B S T R A C T The effects of placements of in-stream structures on a macroinvertebrate community were examined in Spring Creek, a second-order stream of British Columbia, Canada. I tested the responses of organic matter and macroinvertebrates to structural changes at two spatial scales: reach-scale responses to the placements of boulder clusters; and the mechanisms through which in-stream structure affects macroinvertebrates as flow refugia at a microhabitat scale. The reach-scale study was conducted in three sections (length 40-56 m) within a 300-m reach with at least 20 m buffers between them. Two upstream reference sections were in relatively natural condition with channel meanders and abundant woody debris whereas the downstream treatment section had a relatively straight stream channel with less woody debris. In the treatment section, 6 boulder clusters were installed to manipulate habitat heterogeneity. These three sections were studied for three months prior to and 1.2 year following the placements of boulder clusters. Mean (140 %) and coefficient of variation (115 %) of velocity increased in the treatment section whereas the reference section remained relatively unchanged after the treatment. Through the increased organic matter storage (550 %), total macroinvertebrate abundance, which was dominated by detritus feeders, increased (280 %) in the treatment section to become similar in level to that in the reference sections almost one year after that treatment. However, the effect of placements of boulder clusters on taxonomic richness was negligible. Effects of food resource values on macroinvertebrate colonization of flow refugia during floods were examined in the downstream 180-m reach during three floods in 1999 and 2000. Substrate cages were used as an experimental unit, and were assigned to combinations of three types of treatments: food (natural or artificial leaves); flow refugia (reduced velocity or exposed); and flood (retrieval before or after a flood). Low and stable antecedent flow conditions resulted in significant responses of animals to the flood whereas high and variable antecedent flow resulted in little response. During the flood with the highest peak discharge observed among the three trials, two detritivorous taxa Paraleptophlebia and Despaxia accumulated in refugia and their colonization was disproportionally higher in the cage provided with high food resource value (natural leaves). Nevertheless, the accumulation of macroinvertebrates in flow refugia was species-specific, and was not consistent across the community. It was concluded that in-stream structures (boulder clusters) could be used as a remedy to restore macroinvertebrate community productivity. T A B L E O F C O N T E N T S A B S T R A C T i i T A B L E O F C O N T E N T S iv LIST O F T A B L E S vi LIST O F F I G U R E S vii A C K N O W L E D G E M E N T S viii Chapter 1. General introduction 1 Chapter 2 . Reach-scale responses of benthic organic matter biomass and macroinvertebrates to placements of boulder clusters 3 2.1 Literature review 3 2.2 Objectives and hypotheses 6 2.3 Methods 6 2.3.1 Study site 6 2.3.2 Placements of boulder clusters 8 2.3.3 Benthic samples and measurements of physical variables 9 2.3.4 Habitat characteristics 11 2.3.5 Analyses 11 2.4 Results 14 2.4.1 Habitat characteristics 14 2.4.2 Total organic matter biomass 16 2.4.3 Total abundance of macroinvertebrates 16 2.4.4 Relation between total abundance and total organic matter biomass 16 2.4.5 Number of taxa 17 2.4.6 Community structures 17 2.5 Discussion 19 Chapter 3 . Effects of food resource value on macroinvertebrate colonization in flow refugia during high flows 42 3.1 Literature review 42 3.2 Objectives and hypotheses 45 3.3 Methods 45 3.3.1 Study site 45 3.3.2 Colonization cages 45 3.3.3 Leaf treatment 46 3.3.4 Refugium treatment 46 3.3.5 Flood treatment 47 3.3.6 Cage installations 47 3.3.7 Velocity measurements and cage retrievals 48 3.3.8 Statistical analyses 50 3.4 Results 51 3.4.1 Floods 51 i v 3.4.2 Velocity responses 52 3.4.3 Organic matter responses 52 3.4.4 Macroinvertebrate responses 53 3.5 Discussion 55 Chapter 4 . General conclusions and management implications 74 Literature cited 76 v LIST OF T A B L E S Chapter 2 Table 2.1 General characteristics of each section 26 Table 2.2 Summary of two-way A N O V A s to test effects of occasion, section, and their interactions on physical variables (i.e. substrate coarseness, current velocity, and depth). 27 Table 2.3 Coefficient of variation of physical variables in each section before and after the placements of boulder clusters 28 Table 2.4 Summary of three-way A N O V A s to test effects of year, section, sampling occasion, and their interactions on occasion on total organic matter biomass (> 500 pm) and total abundance of macroinvertebrates 29 Table 2.5 Mean rarefied number of taxa estimated for 1000 individuals in each section on each sampling occasion in 1999 30 Table 2.6 Pearson correlation coefficients between principle components (PC) 1 and 2 and the abundances of the numerically dominant taxa 31 Table 2.7 Pearson correlation coefficients between principle components (PC) and habitat variables 32 Chapter 3 Table 3.1 Summary of three-way A N O V A s to examine the effects of leaf, refugium, flood and their interactions on flow velocity measured at 0.4-depth 63 Table 3.2 Summary of three-way A N O V A s to examine the effects of leaf, refugium, flood and their interactions on total organic matter biomass (> 500 pm) 64 Table 3.3 Summary of three-way ANOVAs to examine the effects of leaf, refugium, flood and their interactions on total macroinvertebrate abundance 65 Table 3.4 Relative abundance of the numerically dominant taxa for each trial 66 Table 3.5 Summary of three-way ANOVAs to examine effects of leaf, refugium, flood and their interactions on the abundances of the numerically dominant taxa 67 vi LIST OF F I G U R E S Chapter 2 Figure 2.1 Top view of the study sections 33 Figure 2.2 Examples of boulder clusters 34 Figure 2.3 Estimated daily mean discharge of Spring Creek during the study period 35 Figure 2.4 Mean (+SE) physical variables in each section before and after the placements of boulder clusters 36 Figure 2.5 Mean (+SE) total organic matter biomass (g AFDM/m 2 ) from 1999 and 2000: all materials greater than 500 pm were included 37 Figure 2.6 Mean (+SE) of total macroinvertebrate abundances (number of individuals/m2) from 1999 and 2000 38 Figure 2.7 Relations between total organic matter biomass and total macroinvertebrate abundance for sampling occasion 39 Figure 2.8 Differences of the mean of rarefied number of taxa for each section on each sampling occasion between 1999 and 2000 40 Figure 2.9 Biplots of the principle components (PC) 1 and 2 based on the abundance of the taxa that comprised greater than 1 % of total abundance based on all samples collected in this study 41 Chapter 3 Figure 3.1 Estimated daily mean discharge of Spring Creek during the study period 68 Figure 3.2 Mean (+1SE) of velocity (m/s) measured at 0.4-depth beside the cages 69 Figure 3.3 Mean (+1SE) of total organic matter biomass (g AFDM/cage) excluding alder leaves used for natural leaf treatment 70 Figure 3.4 Mean (+1SE) of total abundance of macroinvertebrates including individuals found in the leaf bags 71 Figure 3.5 Mean (+1SE) of total abundance of the numerically dominant taxa including the individuals found in the leaf bags 72 vii A C K N O W L E D G E M E N T S First of all, I would like to thank my supervisory committee: Drs. Sue Glenn, Dan Moore, and John Richardson for helpful discussion and continuous patience. Their presence and guidance through the program made it possible for me to reach the finish line after years. I also greatly thank Dr. Hans Schreier for being the external examiner after long hours of flight, and Dr. Diane Srivastava for kindly chairing the defense. I am very grateful to Dr. John Richardson, Chris Allaway, Takashi Gomi, Brent Matsuda for whole-day backbreaking work in helping maneuvering "boulder net" to carry boulders in Spring Creek. I am greatly thankful to Jennifer Louie for helping me to process hundreds of sample jars over two summers. Without her patience, accurate, focused and independent work, I would not have been able to finish this amount of work today. Numbers of people helped me to collect samples, and I am very appreciative of their encouraging help especially in freezing winter: Takashi Gomi, Kaori Kochi, Miho N A K A M U R A (i.e. NEGISHI later on), Yuho Okada, Yasushi Shoji, and Jennifer Wild. Staff at the Malcolm Knapp Research Forest of University of British Columbia provided comfortable research environment including delivering boulders to Spring Creek. Dr. Michael Feller kindly provided flow discharge data. My colleagues from the Richardson lab are to be appreciated for everything that I learned through my program. Among them, I am very thankful to Jennifer Bull, Charlotte Gjer0v, Diane Klimuk, Renata Kolodziezcyk, Brent Matsuda, Heidy Peterson, and Yixin Zhang. I have been so encouraged and inspired by them through friendship, humor, and attitude towards jobs, and research. Given the significance of heavy rainfall in my study, I have to deeply appreciate Vancouver's wet winter, which provided me fairly lucky opportunity to collect valuable samples for this study. viii At last but not at least, my wife Miho and my family have supported me through my project. Their presence and supports in various aspects were indispensable to the completion of this study. ix Chapter 1. General introduction Natural systems are rarely at an equilibrium state and their dynamic characteristic is strongly affected by natural disturbance (Sousa, 1984; White and Pickett, 1985). According to Yount and Niemi (1990), disturbance can be defined as "any relatively discrete event that disrupts ecosystem, community or population structure and changes resources, substrate availability, or the physical environment". I emphasize the expansion of the definition of disturbance in this thesis to include event that causes persistent environmental changes, which result in gradual community changes and thus different community structure from what existed before the disturbance (see Bender et al, 1984; Lake, 2000). Disturbance creates patterns of spatial heterogeneity in physical environments and community structure, and thus affects community diversity as well as persistence against subsequent disturbances (Sousa, 1984; Hildrew and Giller, 1992). Examples of physical disturbance include high flow events and debris torrents in stream ecosystems (Fisher et al, 1982; Lamberti etal., 1991; Lake, 2000), or hurricanes (Myers and Lear, 1998) and forest fires (Weir et al, 2000) in forest ecosystems. The effects of a disturbance may vary depending on its spatial and temporal magnitude and the affected communities. For example a hurricane may create a landscape-scale disturbance affecting the distribution and composition of whole forest communities over a relatively long period, whereas a single fallen tree may create a small gap that affects the microorganism communities greatly but its effects may last only for a relatively short period. Human-induced activities are agents of disturbance that may produce long-lasting effects on a community and its environment, disrupting ecological processes of natural systems. Stream ecosystems have been degraded by human-induced disturbances such as timber harvesting, land-use conversion, physical modification of waterways, and water quality degradation because of sedimentation and discharge of chemicals (Yount and Niemi, 1990; Wallace, 1990; Carignan and Steedman, 2000). Anthropogenic disturbance often negatively affects the structure and functions of ecosystems, and ecological restoration is one of the ways to recover degraded ecosystems. Ecological restoration generally refers to the act of returning anthropogenically-damaged structures and functions of an ecosystem close to the pre-disturbance level (Bradshaw, 1996). Solid knowledge of the processes causing degradation of natural systems and the application of appropriate remedies to the problems are fundamental in ecological restoration. Stream restorations at a relatively small spatial scale occasionally involve the use of in-stream structures to restore habitat heterogeneity. The response of macroinvertebrates to in-stream structures is still poorly understood despite the importance of macroinvertebrates as food resource for higher trophic levels such as fish. This thesis deals with macroinvertebrate responses to artificially-provided structural elements within a stream channel. In-stream structures may affect macroinvertebrates through increased habitat heterogeneity as a part of the long-term recovery from anthropogenic ecosystem disturbance (i.e. loss of habitat heterogeneity). Also structural diversity provided by in-stream structures should be important for community persistence as spatial refugium for macroinvertebrates when the system faces natural disturbance (i.e. high flow events). I examine the responses of organic matter and macroinvertebrates to artificially-provided structures at two spatial and temporal scales. Chapter 2 examines the responses of habitat heterogeneity, organic matter biomass, and macroinvertebrates to placements of in-stream structures at the stream reach scale at a relatively large temporal scale (monthly and yearly). In chapter 3, the functions of in-stream structures as flow refugia and the mechanisms determining their use are examined at a microhabitat scale at a relatively small temporal scale (during high flow events; few days). It is intended that chapter 3 examines the part of mechanisms of community responses that are observed in chapter 2, particularly with regard to flow refugia. 2 Chapter 2 . Reach-scale responses of benthic organic matter biomass and macroinvertebrates to placements of boulder clusters 2.1 Literature review Various biotic and abiotic factors control local abundance and diversity of communities and thus influence their distribution at larger spatial scales. For stream macroinvertebrate communities, experimental manipulations of the environment have shown that factors such as velocity, substrate characteristics and amount of organic matter control abundance and diversity at a microhabitat scale (Egglishaw, 1964; Williams and Mundie, 1978; Culp etal., 1983; O'Conner, 1991; Williams and Smith, 1996; Downes etal., 1998; Angradi, 1999), although these effects can be modified by season (see Peckarsky, 1980). In natural systems, many studies have found that macroinvertebrate communities vary predictably across habitat types (Mackay and Kalff, 1969; Benke etal, 1984, 1985; Huryn and Wallace, 1987; Rhodes and Hubert, 1991; Wohl etal., 1995; Angradi, 1996). These patterns result from species-specific habitat preferences based on food resource and physical habitat requirements (Corkum et al., 1977; Resh, 1977, Gore and Judy, 1981; Statzner et ai, 1988). Community responses to high flows (Lancaster and Hildrew, 1993a; Palmer et ai, 1996) and seasons (Mackay and Kalff, 1969; Benke et ai, 1984) often differ among habitat types, making dynamics of community structure more complex. A stream is a mosaic of habitat patches that have varying physical properties (Hynes, 1970; Pringle et ai, 1988; Hildrew and Giller, 1992), where the composition, proportion, and geometric arrangements of patch types largely determine community structure at larger spatial scales, such as at the stream reach (e.g. Grubaugh et al, 1997). Thus, macroinvertebrate habitat and community structure are strongly controlled by geomorphic channel form, which characterize patch characteristics within the stream channel (Brussock etal., 1985; Huryn and Wallace, 1987; Wohl etal., 1995). 3 Fluvial processes and local hydraulic forces, in addition to the quantitative availability and types of bed materials; interact in a complex manner to form channel morphology (Beschta and Platts, 1986; Massong and Montgomery, 2000). In forested streams, large woody debris (LWD) exerts a profound influence on channel morphology (Kellar and Swanson, 1979; Mosley, 1981; Bisson etal, 1987; Gurnell etal, 1995), particularly in low-order streams where stream size is small relative to the size of riparian trees. L W D controls the transport and storage of sediment and influencing stream flow as well as local morphology of a stream channel, either as debris jams or single pieces (Nakamura and Swanson, 1993). L W D provides important fish habitat by creating pools and providing cover against predators and high flows (Bryant, 1983; Bisson et al, 1987; Andrus et al, 1988; Fausch and Northcote, 1992; Richmond and Fausch, 1995; Crook and Robertson, 1999; Flebbe, 1999). Furthermore, debris dams function as important organic matter storage sites (Bilby and Likens, 1980), and have strong associations with macroinvertebrate production (Smock et al, 1989). The loss of L W D and direct alterations of channel structure often result in degradation of stream communities. Habitat destabilization through increased bedload movement (Bilby, 1984; Heede, 1985; Smith etal, 1993a; Diez et al, 2000), and decreased storage of sediment (Beschta, 1979; Smith et al, 1993&) and organic detritus (Bilby and Likens, 1980; Bilby, 1981; Angermeier and Karr, 1984) have been reported after experimental removal of L W D . Channel morphology often changes through reduction of pool habitats with decreased depth and increased velocity (Angermeier and Karr, 1984; Elliott, 1986; Trotter, 1990). Physical changes of habitat by experimental removal of L W D have resulted in lower productivity of fish and macroinvertebrate communities (Angermeier and Karr, 1984; Elliott, 1986; Fausch and Northcote, 1992). Channelization simplifies channel structure resulting in reduced retention capability of organic detritus (Erman and 4 Lamberti, 1992; Haapala and Muotka, 1998) and in lower abundance and diversity of stream communities (Huggines and Moss, 1974; Moyle, 1976). Through the loss of habitat heterogeneity, community persistence against high flow events may be lowered (Gorman and Karr, 1978; Fausch and Bramblett, 1991; Pearsons et al, 1992; Lancaster et al, 1996) because of loss of spatial refugia (sensu Sedell et al, 1990). Anthropogenic activities such as forest harvesting and urban development have resulted in the degradation of stream ecosystems. Efforts have been made to restore the impaired structure and function of stream ecosystems by altering their physical environment. The placements of artificial in-stream structures such as woody structures, boulders, and their complexes, is a useful technique to restore habitat stability and heterogeneity at a relatively small spatial scale (Swales and O'Hara, 1980; Reeves etal., 1991; House, 1996; Slaney et al, 1997; Ward, 1997; Slaney and Martin, 1997) when a recovery of a natural system is not expected to occur for a long time (e.g. Murphy and Koski, 1989). Numerous failures of in-stream structures have been reported largely due to inappropriate application of restoration techniques without thorough consideration of the natural hydraulic and geomorphic processes of stream at large scales (Frissel and Nawa, 1992; Roper et al, 1997; Roper et al, 1998). The efficacies of in-stream structures need to be monitored in a biological context in ecological restoration (Bryant, 1995; Richardson and Hinch, 1998; Kondolf, 2000). There have been an increasing number of studies on the effects of in-stream structures on fish (Angermeier and Karr, 1984; House and Boehne, 1986; Crispin etal, 1993; Riley and Fausch, 1995; House, 1996; Cederholm et al, 1997), macroinvertebrates (Angermeier and Karr, 1984; Smock et al, 1989; Dobson and Hildrew, 1992; Wallace et al, 1995a; Hilderbrand et al, 1997; Lemly and Hilderbrand, 2000), and organic detritus retention (Smock et al, 1989; Trotter, 1990). These studies have shown that in-stream structures, primarily woody structures, can increase productivity of stream communities through 5 creation of favorable habitat (e.g. pools for fish) as well as increased retention capability of detritus and nutrients. However, few studies have examined both habitat changes due to placements of in-stream structures and its consequences in terms of ecosystem-level processes at the stream-reach scale (see Wallace et al, 1995a; Hilderbrand et al, 1997). Furthermore, the effects of boulder clusters on stream communities have not been extensively studied. 2.2 Objectives and hypotheses The objectives of this chapter were: 1) to examine if boulder clusters affect habitat heterogeneity as well as benthic organic matter biomass; and 2) to examine if abundance and taxonomic diversity of the macroinvertebrate community change at the stream-reach scale due to the placement of boulder clusters. I tested the hypothesis that the placement of boulder clusters would increase habitat heterogeneity through the creation of both depositional and erosional habitats, and thereby increase organic matter biomass, and abundance and diversity of macroinvertebrate community. 2.3 Methods 2.3.1 Study site This study was carried out from January 1999 to June 2000 within a 300-m reach of Spring Creek, a second order stream that flows through University of British Columbia's Malcolm Knapp Research Forest. The Research Forest is located in the Coast Range Mountains approximately 60 km east of Vancouver (122°34'W, 49°16'N). The watershed of Spring Creek lies in the Coastal Western Hemlock biogeoclimatic zone, and the conifers Tsuga heterophylla (western hemlock), Thuja plicata (western red cedar), and Pseudotsuga 6 menziesii (Douglas fir) are the dominant forest species. The riparian vegetation consisted predominantly of Alnus rubra (red alder), Acer circinatum (vine maple), and Rubus spectabilis (salmonberry). Water chemistry and hydrologic features of this creek have been described elsewhere (Feller and Kimmins, 1979; Reece and Richardson, 2000). This creek is characterized by low and stable discharge in summer, and higher and more variable discharge in winter. The discharge of Spring Creek was calibrated from the discharge data of East Creek, a tributary of Spring Creek that enters at the downstream end of the study reach, and which was continuously recorded by a V-notch weir. A discharge-water level relation was independently established for Spring Creek by measurements at a range of water levels and corresponding discharges, and this was used to estimate mean daily discharge of Spring Creek. Spring Creek had an average gradient of 2.1 % and variable wetted width of 2-4 m and bankful width of 4-8 m within the study reach. Minimum and maximum water temperatures of Spring Creek during the study period were 1.5 and 13.7 °C, respectively. The stream channel is characterized by pool-riffle sequences with substrate materials being dominated by cobble with occasional patches of sand and accumulation of organic detritus along the wetted channel margins as well as behind woody debris. Visual inspection revealed that one section of Spring Creek had a relatively straight channel and homogeneous habitat structure with fewer pieces of woody debris compared to the adjacent sections partly due to the presence of a bridge at the downstream end of the section. This section lacked longitudinal and cross-sectional morphological variations and thus depositional habitats where accumulation of food resource (organic detritus) occurs. Low habitat heterogeneity and absence of food resource storage led to the expectation that the abundance and taxonomic diversity of the macroinvertebrate community were lower in this section relative to than that of the adjacent sections. Furthermore, it was predicted that the manipulation of habitat heterogeneity by the placement of boulder clusters would increase macroinvertebrate abundance and taxonomic diversity in this section (treatment section). In addition to the treatment section, two upstream reference sections were selected within the study reach, separated by at least 20 m (Table 2.1 and Figure 2.1). Reference sections were chosen as representatives of stream channel of relatively natural condition, and had visually greater habitat heterogeneity with channel meanders, more abundant woody debris structures, and longitudinal variation of channel morphology relative to the treatment section. Reference sections 1 and 2 were different from each other to some extent in terms of habitat structure. Pools with deep and slow flow dominated reference section 1 with abundant organic matter accumulations along the channel bends. Reference section 2 was shallower than reference section 1 and characterized by a wider channel profile and greater proportion of riffle habitats with shallow and fast flow. In January of 1999, transects were set over each section with longitudinal intervals of 2 m and marked using polyester threads. These transects were perpendicularly marked at 1 m intervals to fix the points at which benthic samples and physical habitat measurements were taken during the study period. These points were referred to as sampling grid points. The number of sampling grid points available varied depending on discharge and wetted channel width. 2.3.2 Placements of boulder clusters It was intended that boulder clusters first alter flow environment and that channel morphology will be subsequently diversified thorough local scours and fills of substrate materials. Boulder clusters were placed as deflectors so that slow flow (i.e. depositional) habitats downstream of the structures and fast flow (i.e. erosional) habitats on the narrowed side of stream channel will be formed. The longitudinal intervals of boulder clusters mimicked the interval of occurrence of woody debris in adjacent natural stream sections, which was found to be approximately 6 m. Round boulders with the longest axes of 35-45 8 cm were collected from a quarry nearby in the research forest as materials for the boulder clusters. On 7 and 14 April 1999, a total of 6 boulder clusters were placed in the treatment section. To minimize disturbance of stream banks and stream habitat, boulders were carried manually in a net. Each of the clusters consisted of 3-7 boulders depending on the bankful channel width at the individual site. They were placed within the channel such that several boulders form a line in an alternating manner on either side of the stream; the upstream end was in contact with the stream bank, and the line of boulder radiated out from the bank at approximately 45° in the downstream direction (Fig. 2.1 and Fig. 2.2). The boulders were merely placed on the streambed without extra excavation activities, and thus boulders were not buried or wedged tightly against each other. Thus substantial flow still existed downstream of the clusters at the time of the placements. It was expected that inorganic and organic materials washed from upstream would eventually fill the spaces between the boulders making them deflect stream flow more efficiently, and channel morphology would adjust itself to the boulder clusters. 2.3.3 Benthic samples and measurements of physical variables Benthic macroinvertebrates and organic matter were collected using a Surber sampler (mesh size 300 pm; area 0.095 m ) in each section on six occasions from 28 February 1999 until 3 June 2000 (Fig. 2.3). In 1999, benthic samples were collected on 28 February, 28 March, and 28 May. Sampling was repeated almost exactly a year later, on 26 February, 27 March, and 3 June 2000, respectively. These three sampling occasions are referred to as February, March and May, respectively, for brevity. Benthic samples were taken with at least 1-month intervals to allow a sufficient colonization period for subsequent samplings and thus these samples were assumed to be independent of one another. Samples were collected based on a stratified random layout. On each sampling occasion, 10 transects were randomly 9 selected in each section and one sampling grid point on each chosen transect was then randomly sampled. When the same sampling grid point was chosen for consecutive sampling occasions, an adjacent patch similar to the sampled patch was substituted; this only occurred twice during the study. Physical variables were measured along with collections of benthic samples. At the location adjacent to where each benthic sample was collected, current velocity was measured at 0.4-depth from streambed along with measurements of depth and substrate coarseness. Substrate coarseness was evaluated using a metal grid frame (25 x 25 cm) that consisted of 9 sub-grids (10 x 10 cm) placed on the streambed. The size of the dominant substrate material within each sub-grid was visually estimated according to the following categories: 1-2 mm = sand; 2-64 mm = cobble; 64-264 mm = pebble; > 264 mm = boulder and bedrock (Wentworth, 1922). These categories were then coded as follows: sand = 1; cobble = 2; pebble = 3; boulder = 4; bedrock = 5. These 9 values within sub-grids were averaged to represent substrate coarseness of each sampling location. Velocity and depth were measured once at the central point of this grid with propeller-based flow meter (propeller diameter of 3.5 cm; model 2100, Swoffer Instruments). Samples were always collected starting from the downstream section and sampling grid points within each section to avoid downstream effects of sampling. Samples were preserved in 4 % formalin in the field and taken back to the laboratory for processing. A l l materials retained on a 500-pm sieve were processed further. Macroinvertebrates were sorted, counted and identified to the genus or species level when possible based on Stewart and Stark (1993), Merritt and Cummins (1996), and Wiggins (1996). Organic materials were separated from macroinvertebrates and dried for at least 24 hours at 60 °C and subsequently ashed for at least 2 hours at 550 °C to obtain ash-free dry mass (AFDM). 10 2.3.4 Habitat characteristics To examine differences in habitat characteristics between sections as well as the changes of habitat structure due to the placement of boulder clusters, physical variables (i.e. current velocity, depth, and substrate coarseness) were measured independently of benthic samples before and after the treatment. These variables were opportunistically measured across all the sampling grid points available in each section at different discharges as an attempt of obtaining variable measures at similar discharges before and after the treatment. In order to minimize disturbance to the benthic community, these measurements were carried out in January 1999, within 5 days after the February and March sampling occasions, and also between June 1999 and January 2000. Consequently, substrate coarseness measurements were done on 25 January 1999 (Q = 0.22 m3/s) and 10 January 2000 (Q = 0.23 m3/s) while depth and velocity measurements were concurrently conducted on 24 January 1999 (Q = 0.35 m3/s) and 18 January 2000 (Q = 0.35 m3/s). 2.3.5 Analyses Ten benthic samples collected from each section were not true, spatial replicates; there were no replicates of section (see Hurlbert, 1984). However, the presence of reference and treatment sites before as well as after the treatment was expected to allow inference of treatment effects. That is, the differences that were observed in connection with the placements of boulder clusters in the treatment section but not in the reference sections were attributable to the effects of treatment. To validate this argument, the following predictions needed to be examined: 1) habitat structure would change only in the treatment section due to the placements of boulder clusters while the change of habitat structure in the reference sections would be negligible during the study period; and 2) biological components of the 11 study, such as organic matter biomass and abundance of macroinvertebrates, would remain constant between years in the reference sections. To examine the differences in habitat characteristics among the sections as well as the effects of placement of boulder clusters on them, two-way analysis of variance (ANOVA) was used for physical variables (i.e. substrate coarseness, current velocity, and depth) with section, occasion (i.e. before or after the treatment), and their interaction as factors. A significant section x occasion interaction would suggest that the response variable changed differently among sections over the treatment. If the significant section x occasion interaction was not detected, the means were compared among sections. If detected, the means were compared among sections before the treatment, and the effect of occasion on the means was also examined for each section. Tukey's test was used for the multiple comparisons (Zar, 1999). Coefficients of variation of physical variables were calculated as indices of habitat heterogeneity for each section before and after the treatment. Total organic matter biomass and total macroinvertebrate abundance were examined using three-way A N O V A with year, section, sampling occasion, and their interactions as factors. To test for statistical assumptions of A N O V A , the data were first fitted to the complete general linear model for unbalanced A N O V A (SAS Version 8, see SAS, 1994) and residuals were compared for deviation from a normal distribution using the Shapiro-Wilks test (Zar, 1999). Logio(n+l)-transformations were performed when needed to improve normality of residual distributions prior to the analyses. Although total abundance of macroinvertebrates did not meet the assumption of normality even after being transformed and improved, the transformed values were used in the analysis. Two types of preplanned orthogonal contrasts were examined (Zar, 1999): 1) the mean of the response variable was higher in 1999 than in 2000 when compared between years on each sampling occasion; and 12 2) the mean of the response variable was higher in the reference sections than in the treatment section when compared for each year on each sampling occasion. In order to find relations between total organic matter biomass and total abundance of macroinvertebrates in three occasions based on each Surber sample, I evaluated the significance of all possible difference between pairs of regression slopes among all three simple regression lines. Before performing multiple slope comparisons, I conducted analysis of covariance (ANCOVA) to examine the homogeneity of the regression lines with total macroinvertebrate abundance as the dependent variable, logio-transformed total organic matter biomass as covariate, and sampling occasion as factor, respectively. If the A N C O V A rejected the hypothesis of homogeneity of the regression slopes, multiple comparisons of the regression lines were conducted for all the combinations of sampling occasions. In these multiple comparisons, comparison-wise significance level was adjusted using Bonferroni method (Zar, 1999). The rarefaction technique (Krebs, 1998, p 412-419) was used to estimate expected total number of taxa for 1000 individuals to exclude the effects of sample size on taxonomic diversity. Rarefaction was conducted by E C O L O G I C A L M E T H O D O L O G Y (Exeter Software) based on pooled total number of taxa and number of individuals found in each section on each occasion (i.e. 10 samples). The means of rarefied number of taxa were compared among sections as well as between years. Correlation-based principal component analysis (PCA) was used to examine 1) the differences in community structure among sections on each sampling occasion, 2) the differences in community structure between years on each sampling occasion, and 3) the effects of placements of boulder clusters on community structure in the treatment section. Logio(n+l)-transformed abundances of the taxa that comprised greater than 1 % of total number of individuals sampled in this study were included. Principal components 1 and 2 13 were then correlated with these taxa as well as habitat variables (i.e. depth, velocity, substrate coarseness, and total organic matter biomass) to examine gradients of taxonomic composition for each axis. Graphical presentations of the ordination results were made separately for three sampling occasions. A l l statistical analyses were performed using SAS (SAS Version 8, see SAS, 1994), unless mentioned otherwise, with the significance level a set at 0.05. 2.4 Results 2.4.1 Habitat characteristics Channel morphology did not change immediately after the placements of the boulder clusters. Stream discharge during the spring and summer of 1999 was relatively stable and low due to dry weather and there were few high flow events (Fig. 2.3). These high flow events did not result in detectable changes in channel morphology owing to insufficient hydraulic forces to cause movements of bed materials. The flood that occurred on 14 December 1999 with the largest peak discharge during the study period caused detectable changes in channel morphology and habitat characteristics in the treatment section. Inorganic and organic materials filled the spaces between the boulders, causing boulders to become imbricated, and thereby creating fast flow habitats on the opposite side of the channel by scouring out the streambed, as well as creating a slow flow habitat behind the clusters. During this event, the 2nd boulder cluster from the upstream (see Fig. 2.1) failed since a log (diameter approximately 50 cm), against which the upstream boulder of the structure was resting, was moved downstream approximately 5 m, causing displacement of this boulder. The other boulders in the structure remained in position. After the flood on 14 December, channel morphology did not change to a detectable extent and remained stable. The other five clusters have remained intact until present. Since the change of channel morphology was almost negligible on the May sampling occasion in 1999, 1 month after the boulder installation, the effects of boulder clusters on macroinvertebrates were presumably not yet substantial. The measurements of substrate coarseness on 25 January 1999 and 10 January 2000, and depth and velocity on 24 January 1999 and 18 January 2000 were presumed representative of habitat structure before and after the treatment (placements of boulder clusters), as the high flow event that mostly altered habitat structure occurred between these dates (i.e. 14 December 1999). The physical variables differed among sections before the treatment due to the different channel morphology. Mean substrate coarseness and mean velocity were the highest in reference section 2, followed by the treatment section and reference section 1 (Fig. 2.4), reflecting a high relative proportion of riffle habitats within the section. Reference section 1 and the treatment section showed similar mean velocity but thecoefficient of variation was higher in reference section 1 (Table 2.3 and Fig. 2.4). Mean depth was substantially higher in reference section 1 than the other sections since the former was dominated by slow and deep pool habitats. Overall, the treatment section had the lowest variation in all the three variables before the treatment (Table 2.3). The means of substrate coarseness and depth remained at a similar level in all sections after the treatment. However, velocity changed differently among the sections after the treatment as shown by a significant reach x occasion interaction, and the mean velocity significantly increased in the treatment section (Table 2.2 and Fig. 2.4). The increase of the coefficient of variation of substrate coarseness and velocity after the treatment was the highest in the treatment section relative to the reference sections. Nevertheless, the coefficient of variation of substrate coarseness and depth in the treatment section was still the lowest among the three sections after the treatment. 15 2.4.2 Total organic matter biomass Total organic matter biomass significantly differed among sections (Table 2.4 and Fig. 2.5). Preplanned contrasts showed that organic matter biomass was significantly higher in 2000 than in 1999 only in the treatment section on the May sampling occasion. Furthermore, the organic matter biomass was significantly lower in the treatment section than that in both of the reference sections only on the May sampling occasion in 1999. These findings showed that organic matter biomass in the treatment section was as high as that in the reference section on the May sampling occasion in 2000, after having increased from May 1999 when it was significantly lower than that in the reference sections. 2.4.3 Total abundance of macroinvertebrates Total abundance of macroinvertebrates differed between years and among occasions (Table 2.4 and Fig. 2.6). Preplanned contrasts showed that total macroinvertebrate abundance was significantly higher in 2000 than in 1999 only in the treatment section on the May sampling occasion. Furthermore, macroinvertebrate abundance was significantly lower in the treatment section than that in the reference sections only on the May sampling occasion in 1999. These findings showed that macroinvertebrate abundance in the treatment section was as high as that in the reference section on the May sampling occasion in 2000, after having increased from May 1999 when it was significantly lower than that in the reference sections. 2.4.4 Relation between total abundance and total organic matter biomass Total macroinvertebrate abundance was significantly related to total organic matter biomass for all three sampling occasions (Fig. 2.7). A N C O V A revealed that the slopes of the regression lines were significantly different among sampling occasions (F2 > 174= 6.84, p < 0.001). The slope of the regression line for the May sampling occasion was the greatest, 16 followed by that for the March and February sampling occasions. These findings show that habitat patches with higher organic matter loading are predicted to have higher macroinvertebrate abundance, and that a higher number of macroinvertebrates can be predicted in a given amount of organic matter biomass particularly on the May sampling occasion relative to earlier sampling occasions. 2.4.5 Number of taxa The mean rarefied number of taxa was the lowest in the treatment section only on February sampling occasion before the treatment (Table 2.5). Furthermore, the differences in the mean rarefied number of taxa between years were not consistent among sections (Fig. 2.8). There is high year-to-year variation in the mean rarefied number of taxa in the reference sections and these variations were greater than that in the treatment section. 2.4.6 Community structures Principle components (PC) 1 and 2 together accounted for 53 % of total variance in the data (Table 2.6). Seventy taxa were found in this study and nineteen taxa were included in the PCA as the numerically dominant taxa. Principle component 1 was positively correlated with the abundance of the numerically dominant taxa such as Chironominae, Orthocladiinae, Tanypodinae and Paraleptophlebia, and thus this axis was highly correlated with the total abundance of macroinvertebrates (r = 0.84; p < 0.0001) while PC 2 was positively correlated with the abundance of the taxa such as Baetis, Epeorus and Rhyacophila banksi but not correlated with the total abundance (r = 0.04; p = 0.52). Habitat variables were significantly correlated with PC scores (Table 2.6). Principle component 1 was negatively correlated with velocity and substrate coarseness and positively with organic matter biomass and thus PC 1 can be interpreted as a gradient of the taxa that were found to be abundant in depositional habitat (depositional taxa). Depositional taxa except 17 Tanypodinae and Ceratopogonidae were considered as detritus feeders (Merritt and Cummins, 1996). Principle component 2 was negatively correlated with depth and organic matter biomass and positively with velocity and thus PC 2 was interpreted as a gradient of the taxa that primarily resided in erosional habitat (erosional taxa). Ordination plots of PCA revealed dissimilarities in terms of community structure among sections as well as between years, particularly on the May sampling occasion (Fig. 2.9). On the February and March sampling occasions, there was a high degree of overlap within the variation of both PC 1 and 2 scores between years in each section, showing that the community structures were similar between years on the February and March sampling occasions. On the May sampling occasion, however, the treatment section showed dissimilarity between years without an overlap of the variation in either PC 1 or 2 scores, whereas reference section 2 was dissimilar between years in terms of PC 2 scores. On all three sampling occasions in 1999, the orders of the mean PC scores were consistent among sections. That is, the order of the sections in terms of the mean PC 1 score was reference 1, reference 2, and the treatment section from high to low score; the order of the sections in terms of the mean PC 2 score was the treatment, reference 2, and reference 1 section from high to low score. However, the order of the mean PC 1 score changed to reference section 1, treatment, and reference 2 section on the March and May sampling occasions in 2000. Particularly on the May sampling occasion in 2000, the mean score of PC 1 in the treatment section became fairly similar to that in reference section 1. These findings show that in 1999 there was a gradient of community structure among sections. Reference section 1 was characterized by the highest relative abundance of depositional taxa and lowest relative abundance of erosional taxa; the treatment section was characterized by highest relative abundance of erosional taxa and lowest relative abundance of depositional taxa; and reference section 2 lay between two other sections. Furthermore, 18 the annual shift of community structure in the treatment section, particularly found on the May sampling occasion, involved the increase of the abundance of depositional taxa, which were numerically dominant and highly correlated with total macroinvertebrate abundance. Although the shift of community structure in treatment section and reference section 2 on the May sampling occasion in 2000 involved the increased abundance of erosional taxa, this had little influence on total macroinvertebrate abundance because of lower abundance of erosional taxa relative to depositional taxa. 2.5 Discussion My findings showed that the placements of the boulder clusters increased heterogeneity of habitat at least in terms of velocity and substrate coarseness, organic matter biomass, and macroinvertebrate abundance, supporting the original predictions. However, the effects of the boulder clusters on number of taxa were insignificant. During the study period, habitat structures in the reference sections remained almost unchanged and the pattern of total organic matter biomass and total macroinvertebrate abundance in the reference sections were consistent between the years. Therefore, the changes in the macroinvertebrate community and organic matter biomass observed in the treatment section were inferred to be caused by the placements of the boulder clusters. Most of the change in channel morphology in the treatment section occurred during the flood on December 14, 1999 when hydraulic forces were strong enough to cause movement of substrate materials. However, overall change in channel morphology due to the placements of boulder clusters was slight as shown by similar values of depth and substrate coarseness before and after the treatment. Although the increased variation of substrate coarseness and velocity after the placements of boulder clusters indicate increased habitat heterogeneity in the treatment section, the heterogeneity of substrate coarseness and depth 19 were still lower than that in the reference sections. The velocity environment within the channel was most affected by the boulder clusters. Increased mean velocity in the treatment section was due to the proportional changes of habitat within the channel. That is, proportionally greater area with fast flow was created than slow flow habitats after the treatment. Patches of slow flow habitats were formed behind the clusters. The formation of habitats with contrasting flow environment contributed to the increased heterogeneity of velocity and substrate coarseness. Most of the studies on the effects of in-stream structures (e.g. woody debris and pieces of logs) have shown the increased proportion of pool area (House and Boehne, 1986; Crispen et al, 1993; Riley and Fausch, 1995; Wallace et al, 1995a; Cederholm et al, 1997; Schmetterling and Pierce, 1999; Lemly and Hilderbrand, 2000) by deflecting flow and scouring the stream bed, and thus decreased mean velocity in some cases (e.g. Riley and Fausch, 1995; Wallace et al, 1995a). In my study, the boulder clusters rather acted as flow deflectors, narrowing the stream flow. The extent of the effects of in-stream structures on channel morphology varies depending on local substrate conditions and hydraulic characteristics such as local shear stress. In the study of Hilderbrand et al. (1997) in Virginia streams, installations of log pieces in a high-gradient channel (3-6 %) showed little effect on habitat structure whereas in low-gradient streams (1 %) installations resulted in substantial pool formation. This variability in channel response was attributed to the differences in controlling factors in channel morphology among the sites. Substrate materials tend to consist of coarser materials such as cobbles and boulders in high-gradient streams whereas finer substrate materials predominate in low-gradient streams (Leopold et al, 1964). Therefore, high-gradient streams tend to be more stable and less responsive against in-stream structures. In Spring Creek, substrate materials mainly consisted of cobble and pebble materials. This is probably the reason that boulder clusters did not drastically change the channel morphology. 20 Despite the seemingly slight effects of the boulder clusters on channel morphology, macroinvertebrate habitat was significantly changed. In the treatment section, the channel originally lacked depositional habitat, and thus little organic detritus was stored prior to the placement of boulder clusters, particularly on May occasion. Total benthic organic matter biomass significantly increased in the treatment section on the May occasion from 1999 to 2000 to become similar in level to that in the reference sections. This indicates that retentiveness and storage of organic matter in the treatment section increased due to the boulder clusters. Various obstructions such as woody debris structures, boulders, and shallow riffles trap and store organic materials (Speaker et al, 1984; Ehrman and Lamberti, 1992; Webster et al, 1994; Maridet et al, 1995). Artificially-installed in-stream structures can also increase the storage and biomass of benthic detritus by directly trapping (Bilby, 1981; Bilby and Likens, 1980; Angermeier and Karr, 1984; Dobson et al, 1995; Dobson and Cariss, 2000) or indirectly creating depositional habitats behind the structures (Trotter, 1990; Wallace et al, 1995a; Lemly and Hilderbrand, 2000). In my study, it was observed that accumulations of fine organic matter occurred behind the boulder clusters and also that organic matter was trapped on the upstream side of the clusters (personal observation). The boulder clusters also might have increased substrate stability (Bilby, 1984; Heede, 1985), increasing the trapping efficiency of organic matter within substrate material across the stream channel. Therefore, increased total organic matter biomass in the treatment section was probably a consequence of combinations of these processes. In the treatment section, the increase of organic matter biomass resulted in the increased numerically dominant detritivores, and thus total macroinvertebrate abundance on the May occasion. In forested streams, organic matter derived from the riparian zone serves as a primary energy source for stream communities (Fisher and Likens, 1973; Cummins et al, 1989). Organic matter loadings can influence the local density as well as taxonomic 21 diversity of a macroinvertebrate community (Egglishaw, 1964; Fahy, 1975; Flecker, 1984; Richardson, 1991), and thus physical characteristics of stream channel and types of riparian vegetation affect longitudinal distribution and composition of macroinvertebrate community along river continuum (Vannote et al, 1980; Hawkins and Sedell, 1981; Wallace et al, 1999). For the allochthonous energy to be efficiently utilized by the macroinvertebrate community, however, organic matter needs to be stored within the channel and processed. Therefore, channel structure and retention efficiency have strong associations with the macroinvertebrate community within the channel (Rounick and Winterbourn, 1983; Cummins et al, 1989; Prochazka et al, 1991; Chergui et al, 1993; Haapala and Muotka, 1998), particularly abundance and diversity of macroinvertebrates. Experimental studies have further shown that increased retention can result in community changes (Angermeir and Karr, 1984; Dobson etal, 1995; Dobson and Carris, 2000; Lemly and Hilderbrand, 2000). These studies found a rapidly increased abundance of detritivores and thus shift in community structure to more detrital-based through the increase organic matter biomass. The importance of retention capability may be pronounced in food-limited systems (Richardson, 1991; Dobson and Hildrew, 1992; Wallace etal, 1999). In my study, increased abundance of detritus feeding taxa was observed with the increased level of organic matter biomass. Therefore it is likely that the strong effects of the boulder clusters on total macroinvertebrate abundance were through the provision of a greater amount of food resources available within the channel through increased retention capability. These results suggest that the placement of boulder clusters can create storage of organic matter and can be used as a remedy to restore productivity of macroinvertebrates. Effects of in-stream structures on a macroinvertebrate community can vary among different types of systems. In the study of Hilderbrand et al. (1997) and Lemly and Hilderbrand (2000), macroinvertebrate abundance did not increase at the stream-reach scale 22 after the log placements although net biomass of benthic detritus increased within the channel. In their study, the total abundance of macroinvertebrates did not differ between riffles and pools despite pools having a higher amount of benthic detritus, and hence the proportional change of habitat structure within channel did not result in a net increase of macroinvertebrate abundance at the stream reach scale. Wallace et al. (1995a) found that macroinvertebrate abundance was higher in the depositional habitat that was created by the installations of log pieces than the pre-treatment level. Dobson et al. (1995) also reported that artificially-created leafpacks supported high numbers of detritivorous macroinvertebrates. Community structure in artificially-created habitats was found to be detritivore-domiriant in all of these studies, and thus community structure shifted towards detrital-base after the treatment at the stream-reach scale. In my study, habitat patches with high total organic matter tended to support high total abundance of macroinvertebrates, and this relationship was pronounced on the May occasion. Therefore, creation of organic-matter rich habitat patches by boulder clusters resulted in higher total macroinvertebrate abundance at the stream-reach scale at least on the May occasion. However, the way that in-stream structures are placed and the type of in-stream structure can result in different outcome. For example, Pretty and Dobson (2000) reported that single pieces of logs perpendicularly placed in the stream channel did not result in substantial increase in biomass of organic detritus. This was attributed to the low effectiveness of log structures for trapping of organic matter. Furthermore, Hilderbrand et al. (1997) found that human-judged placements of log pieces were more effective than random placements in forming pool habitats. Even before the treatment (i.e. in 1999), neither organic matter biomass nor macroinvertebrate abundance was lower in the treatment section than the reference sections on early sampling occasions such as February and March. However, on the May occasion, the organic matter biomass was significantly lower in the treatment section than the reference 23 sections and this was reflected in significantly lower total abundance in the treatment section. I found that there is positive relationship between organic biomass and total macroinvertebrate abundance and that highest number of animals can be predicted to be found in a given organic matter biomass on the May occasion. Therefore, the level of organic matter biomass present within the stream channel had stronger effects on the productivity of macroinvertebrates particularly on later sampling occasion. This suggests that low levels of organic matter in later season can strongly limit the productivity of macroinvertebrates. Contrary to the prediction, the placements of boulder clusters had no significant impact on taxonomic richness in the treatment section. The year-to-year variation of number of taxa was high even in the reference sections. Furthermore, even before the treatment, the number of taxa was not necessarily lowest in the treatment section. Past studies have shown there is positive relation between habitat heterogeneity and taxonomic diversity in different types of ecosystems (Abele, 1974; Tonn and Magnuson, 1982; O'Conner, 1991; Downes et al, 1998). The source of colonizers might be limited at the spatial scale examined in my study. The three sections were relatively close to each other, being separated by less than 50 m. Macroinvertebrates continuously disperse through drifting for a long distance (Waters, 1972; Townsend and Hildrew, 1976) and thus these sections might have already shared most of the species before the treatment. The three study sections differed in channel morphology. Therefore, the composition and proportion of habitat patches in terms of physical and biological characteristics differed among sections. The different habitat structure can result in different responses of macroinvertebrates among seasons and years (Benke, 1984; Rhodes and Hubert, 1991). For example, the abundance of the taxa that preferred erosional habitat (erosional taxa) showed higher year-to-year variation relative to depositional taxa particularly on the May occasion as 24 shown in the results of PCA. This pattern of a high year-to-year variation in terms of erosional taxa wasn't observed in reference section 1 probably because habitat structure was different from two other sections being dominated by depositional habitats. Habitats in reference section 1 might have been less variable compared to that in the other two sections, which were characterized more by erosional habitat. The difference in annual variation between depositional habitat dominated and erosional habitat dominated sections may be due to the differences in responses to high flow events between depositional and erosional habitats. Alterations of habitat structure in the treatment section changed its habitat structure and thus the responses of macroinvertebrates to season. In the study system, habitat patches with high biomass of organic detritus can be considered as primary determinant of community productivity of macroinvertebrates. 25 Table 2.1 General characteristics of each section. Channel length (m) Channel gradient (%) Reference 1 40.0 1.08 Reference 2 45.3 1.89 Treatment 56.2 3.54 26 Table 2.2 Summary of two-way A N O V A s to test effects of occasion, section, and their interactions on physical variables (i.e. substrate coarseness, current velocity, and depth). Effects df MS F P Substrate coarseness Occasion (0) 1 0.11 0.56 0.46 Section (S) 2 0.63 3.12 0.04 0 x S 2 0.03 0.15 0.86 Error 442 Velocity Occasion (0) 1 0.34 4.07 0.04 Section (S) 2 0.78 9.28 0.0001 0 x S 2 0.26 3.07 0.04 Error 603 Depth Occasion (0) 1 55.28 0.44 0.51 Section (S) 2 5583.52 44.15 < 0.0001 0 x s 2 6.28 0.05 0.95 Error 603 27 Table 2.3 Coefficient of variation of physical variables in each section before and after the placements of boulder clusters. Before After % change from 1999 to 2000 Substrate coarseness^ Reference 1 Reference 2 Treatment 0.19 0.25 0.11 0.17 0.26 0.12 -10.5 +4.0 +9.1 Velocity (m/s) tt Reference 1 Reference 2 Treatment 0.94 0.78 0.66 0.99 0.76 0.76 +5.3 -2.6 +15.2 Depth (cm) tt Reference 1 Reference 2 Treatment 0.60 0.55 0.47 0.59 0.55 0.47 -1.6 0 0 Substrate coarseness was measured on 25 January 1999 (0.22 m3/s) and 19 January 2000 (0.23 m7s). Velocity and depth were measured on 24 January 1999 (0.35 m3/s) and 18 January 2000 (0.35 m3/s). tt 28 Table 2.4 Summary of three-way A N O V A s to test effects of year, section, sampling occasion, and their interactions on occasion on total organic matter biomass (> 500 pm) and total abundance of macroinvertebrates. Effects df MS F P Total organic matter biomass Year (Y) 1 0.49 1.95 0.16 Section (S) 2 2.53 10.08 < 0.0001 Occasion (0) 2 0.17 0.68 0.50 Y x S 2 0.30 1.23 0.29 S x 0 4 0.14 0.57 0.69 Y x 0 2 0.71 2.82 0.06 Y x s xO 4 0.17 0.69 0.60 Error 162 Total abundance Year (Y) 1 0.54 4.56 0.03 Section (S) 2 0.15 1.23 0.30 Occasion (0) 2 2.14 18.08 < 0.0001 Y x S 2 0.35 2.97 0.05 S x 0 4 0.12 1.01 0.40 Y x O 2 0.04 0.33 0.72 Y x s x 0 4 0.12 1.01 0.40 Error 162 29 Table 2.5 Mean rarefied number of taxa estimated for 1000 individuals in each section on each sampling occasion in 1999. Reference 1 Reference 2 Treatment February 49.48 44.10 42.29 March 39.66 44.68 40.98 May 34.66 38.48 37.83 Table 2.6 Pearson correlation coefficients between principle components (PC) 1 and 2 and the abundances of the numerically dominant taxa. Only significant correlations with correlation coefficients of greater than 0.60 are shown for clarity. Numbers in the parentheses denote relative abundance of each taxon based on all samples pooled. *** p < 0.001 Taxa PC 1 PC 2 Chironominae (14 %) +0.83*** Orthocladiinae (11 %) +0.87*** Oligochaeta (8 %) Epeorus (8 %) +0.63*** Tanypodinae (7 %) +0.83*** Paraleptophlebia (7 %) +0.87*** Baetis (6 %) +0 79*** Malenka (5 %) Sweltsa (5 %) Zaiteria (4 %) Ameletus (3 %) +0.69*** Lepidostoma (3 %) +0.69*** Cinygmula (2 %) Despaxia (2 %) +0 73*** Zapada cinctipes (1%) RhyacophUa banksi (1 %) +0.68*** Heterlimnius (1 %) Ceratopogonidae (1 %) +0.70*** Glossosoma (1 %) Variance explained (%) 34.8 % 16.2 % 31 Table 2.7 Pearson correlation coefficients between principle components (PC) and habitat variables. ***p< 0.001 PC 1 PC 2 Velocity -0.52*** +0.43*** Depth -0.46*** Substrate coarseness -0.42*** Total organic matter biomass +0.51*** -0.25*** 32 10m Figure 2.1 Top view of the study sections. The order of reference 1, 2, and treatment sections corresponds to the upstream-downstream order. Reference section 2 is separated from reference section 1 by 20 m whereas the treatment section is separated from reference section 2 by 50 m. These maps are drawn based on the condition on 15 May 1999 at a discharge of 0.14 m3/s. Major large woody debris (LWD) refers to woody material with a diameter greater than 10 cm. 33 Figure 2.2 Examples of boulder clusters. These photos were taken looking upstream. A) and B) corresponds to the 3 r d and 5 t h clusters from upstream in Figure 2.1, respectively. Figure 2.3 Estimated daily mean discharge of Spring Creek during the study period. The upward arrows indicate the occasions of benthic sampling. • i " Before (25 January, 1999) « n » After (10 January, 2000) Before (24 January, 1999) After (18 January, 2000) Before (24 January, 1999) i After (18 January, 2000) R 1 R 2 Figure 2.4 Mean (+SE) physical variables in each section before and after the placements of boulder clusters: a) Substrate coarseness, b) current velocity, and c) depth, t Section x occasion interaction was not significant: Tukey's multiple comparison was conducted among sections; different capital letters denoted significant difference, ft Section x occasion interaction was significant: Tukey's multiple comparisons were conducted among sections before the treatment as well as between occasions for each section; significant differences between occasions were shown by asterisk. R 1, R 2, and T refer to reference 1, reference 2, and treatment section, respectively. 36 C\J E 250 -I Q LL 200 -< CO CO ma 150 -o ! Q s_ CD -*—< 100 -03 E o 50 -' c 03 O iS 0 -o !— Reference 1 Reference 2 Treatment 1999 2000 February 1999 2000 March 1999 2000 May Figure 2.5 Mean (+SE) total organic matter biomass (g AFDM/m 2 ) from 1999 and 2000: all materials greater than 500 um were included. Different letters above the bars denote significant differences between years in each section on each sampling occasion as results of orthogonal contrast. Significant differences between the reference sections and the treatment section as results of orthogonal contrasts were denoted by asterisk. 37 •3 6000 •«= 4000 2000 0 I Reference 1 a Reference 2 I Treatment 1999 2000 February 1999 2000 March 1999 2000 May Figure 2.6 Mean (+SE) of total macroinvertebrate abundances (number of individuals/m ) from 1999 and 2000. Different letters above the bars denote significant differences between years in each section on each sampling occasion as results of orthogonal contrast. Significant differences between the reference sections and the treatment section as results of orthogonal contrasts were denoted by asterisk. 38 16000 14000 12000 i 10000 8000 6000 ^ 4000 2000 ^ a) February y = 1887x-903 r 2 _ • (r^ = 0.47; p< 0.0001) y = 2200x-639 (r 2 = 0.46; p< 0.0001) y = 3878X-1827 (r 2 = 0.45; p< 0.0001) 0 1 2 Logio(Total organic matter biomass) (AFDM g/m 2) Figure 2.7 Relations between total organic matter biomass and total macroinvertebrate abundance for sampling occasion: a) February, b) March, and c) May. Regression lines denoted by different letters are significantly different as result of slope homogeneity test with the significance level set at 0.05/3. 39 o c CD 0 8 o o o CTJ X CTj - 2 0 Q) _Q E c T3 "2 CD ^ -4 V . CTj £ -6 •8 i i i—i R 1 R 2 T February R 1 R 2 T March R 1 R 2 T May Figure 2.8 Differences of the mean of rarefied number of taxa for each section on each sampling occasion between 1999 and 2000. Rarefaction estimated the number of taxa for 1000 individuals. R 1, R 2, and T refer to reference 1, reference 2, and treatment section, respectively. 40 4 a) February -2 -2 -4 -2 PC 1 • Reference 1 o Reference 2 • Treatment • Reference 1 O Reference 2 a Treatment : -A I 1 I i • Reference 1 o Reference 2 El Treatment Figure 2.9 Biplots of the principle components (PC) 1 and 2 based on the abundance of the taxa that comprised greater than 1 % of total abundance based on all samples collected in this study. Means (±95 % confidence intervals) are shown for each sampling occasion. The directions of arrows denotes the directional change from 1999 to 2000. 41 Chapter 3 . Effects of food resource value on macroinvertebrate colonization in flow refugia during high flows 3.1 Literature review Disturbance is an important feature of dynamic natural environments and is predicted to influence community characteristics such as species diversity and abundance (Connell, 1978; Sousa, 1984; White and Pickett, 1985). In stream ecosystems, fluctuation of flows is characteristic (Poff and Ward, 1989) and flow affects macroinvertebrate communities in various ways (Stazner et al, 1988; Hart and Finelli, 1999). It has been increasingly recognized that flood disturbances exert influential roles on the structure and distribution of macroinvertebrate communities (Stanford and Ward, 1983; Resh et al, 1988; Townsend 1989). Floods can reduce populations of macroinvertebrates and biomass of algae communities through direct washout and perturbation of substrate materials (Minshall, 1968; Fisher et al, 1982; Scrimgeour and Winterbourn, 1989; Giller et al, 1991; Flecker and Feifarek, 1994; Angradi, 1997). Floods also result in inputs and redistribution of nutrients, organic food resources, and biota across stream habitats, contributing to a patchy distribution of resources and varying disturbance history with a corresponding benthic community (Southwood, 1988; Townsend, 1989; Francoeur et al, 1998; Matthaei etal, 1999, 2000). Experimental studies have shown that disturbance can alter predator-prey interaction (Walde, 1986) and the levels of community diversity at a microhabitat scale (Clifford, 1982; Death and Winterbourn, 1995; Death, 1996; Matthaei et al, 1996a; Townsend et al, 1997). And natural system is observed to conform such disturbance history (Townsend and Scarsbrook, 1997). The spatial heterogeneity of habitats influences the formation of such patchiness by differential susceptibility to the influence of flood, and responses of biota at different spatial 42 scales (Hildrew and Giller, 1992; Lake, 2000). It in turn facilitates community persistence against floods through the provision of spatial refugia (sensu Sedell et al, 1990). Several different types of in-stream spatial refugia have been proposed and examined to explain the rapid recovery of macroinvertebrate populations and communities after flooding. A hyporheic zone (Harvey and Wagner, 2000) is potentially important as refugium although its function and significance for community persistence is not clear or consistent (Palmer et al, 1992; Dole-Olivier et al, 1997). Another kind is in-stream flow refugium (referred to as flow refugium hereafter), which is of interest in this study. Lancaster and Hildrew (1993&) defined a flow refugium as a place where the hydraulic influence of a high discharge event is relatively low so that density-independent loss or mortality of macroinvertebrates is likely to be slight. Lancaster and Hildrew (1993a) subsequently demonstrated that such flow refugium habitat patches had higher macroinvertebrate abundance relative to non-refugium patches after floods, suggesting its function as refugium. Stream channel structure is predicted to influence the nature and availability of flow refugia. Positive relations between the persistence of fish and macroinvertebrate communities and the extent of hydraulic or physical heterogeneity of stream channels have been shown (Fausch and Bramblett, 1991; Pearsons etal, 1992; Borchart, 1993). Furthermore, loss of macroinvertebrates by drift from stream reaches is inversely related to the proportion of hydraulically dead zones within stream reach (Lancaster et al, 1996). Several types of stream habitat have been explicitly examined and shown to have potential as flow refugia by providing a hydraulically benign environment during high discharge events (inundated flood plain: Badri et al, 1987; Rempel et al, 1999; Matthaei and Townsend, 2000; woody debris jam: Palmer etal, 1995, 1996). The mechanisms of refugia have also been a focus of disturbance studies because of the predicted importance of refugia for community persistence. One of the models proposed 43 is that dislodged macroinvertebrates accumulate in flow refugia during floods and possibly contribute to a rapid recovery after floods as recolonizers (Robertson et al, 1995). This phenomenon of macroinvertebrate accumulation in depositional habitats during high flow events has been suggested in earlier studies (see Lancaster et al, 1990). Winterbottom et al. (1997a) tested this model and empirically demonstrated that the higher abundance of macroinvertebrates in flow refugia after floods relative to habitat exposed to flow force was due to the accumulation of individuals. Nevertheless such accumulations were dependent on stream discharge and were taxon-specific phenomena (see also Lancaster, 2000). Whether such accumulation is passive or active has been also examined. Lancaster (1999) provided evidence that some taxa actively move into flow refugia via crawling over the streambed whereas others used body movement to promote reattachment to the streambed after being entrained in the water column. However, further research is required to generalize the mechanisms of refugia use under different flow conditions and in different systems. The depositional flow environment in flow refugia accumulates organic materials as well as animals that are washed downstream through entrainment and deposition. Detritus loading is one of the factors that controls local abundance and diversity of the macroinvertebrate community (Egglishaw, 1964; Peckarsky, 1980; Flecker, 1984). Thus, flow refugia potentially harbor high food resource values for macroinvertebrates, at least for detritus feeders. The importance of food resource value in flow refugia may be pronounced when food resources are depleted from disturbed habitat as a consequence of floods. Therefore, the accumulation of macroinvertebrates in flow refugia during floods may have an association with relatively high food resource value in addition to the depositional and benign flow environment. 44 3.2 Objectives and hypotheses In this chapter, my objectives were to examine 1) if macroinvertebrate communities accumulate in flow refugia during floods, and 2) if the patterns of refugium accumulation are influenced by the food resource value. I tested the following two hypotheses: 1) the abundance of macroinvertebrates in flow refugia increases after a flood whereas non-refugia habitat decreases or maintains the pre-flood levels; and 2) colonization of flow refugia is pronounced when provided with higher food resource value. In particular, the colonization of detritivores in flow refugia was predicted to be higher with a higher food resource value under the second hypothesis. 3.3 Methods 3.3.1 Study site This study was carried out in a 180-m reach of Spring Creek, a second order stream that flows through the University of British Columbia's Malcolm Knapp Research Forest as described in Chapter 2. This reach was located downstream of the reach used in Chapter 2. Spring Creek had a gradient of approximately 1.7% and variable wetted width of 2-4 m and bankful width of 4-8 m within the study reach under low flow conditions. 3.3.2 Colonization cages Colonization cages were used in this study as experimental units. Cylinder-shaped cages were constructed with plastic mesh (mesh size approximately 1.5 cm) and a round plastic base (diameter 21 cm). The round plastic plates constituted the base of the cages whereas the sides and top were made with the plastic mesh such that the cage had a height of 15 cm. Each part was tightly tied together with plastic straps. 40 cages were prepared and 8 combinations of three treatments with two levels within each of them (i.e. a 2 x 2 x 2 45 factorial design) were assigned to the cages with five replicates. The three treatments were leaf (real, artificial), refugium (presence, absence), and flood (before, after). 3.3.3 Leaf treatment To test the effect of food resource values on macroinvertebrate colonization during a flood, the cages were provided with either natural or artificial leaves. The natural leaves were red alder (Alnus rubra) leaves that were collected from the nearby riparian forest floor in early September 1999 and air-dried in the laboratory for at least two weeks. Only leaves that were recently (within a few days) fallen to the ground were collected and used for the experiment. Red alder is a common riparian species that naturally provides a large amount of leaf litter into the stream in this area. As the artificial leaf, heavy polyester cloth was cut into one of four different representative sizes of red alder leaves and used for the experiment. The edges of the artificial leaves were passed through a flame to seal the edges and keep them from weathering. Each natural leaf pack was made by grouping 10 (±0.25) g of alder (dry weight) leaves. The artificial leaf packs consisted of a randomly chosen mixture of different sized artificial leaves that provided a similar surface area to that of 10 g of alder leaves. Thus artificial leaves had no food resource value but provided a similar microhabitat for macroinvertebrates. These leaf packs were separately stored in Ziplock bags until the day of the cage installation. 3.3.4 Refugium treatment To examine the effects of a sheltered flow environment (flow refugium) on macroinvertebrate colonization during a flood, the cages were assigned to either a refugium or exposed treatment. Refugium cages were each provided with a single concrete block (40 x 19 x 19 cm) placed upstream of the cage in a way that the distance between the cage and block was approximately 7 cm. Concrete blocks were placed such that the longest axis lay 46 perpendicular to the flow direction and the largest aspect faced the flow direction so that refugium cages were sheltered from flow behind the blocks. Each block had two holes (12 x 12x19 cm) on the second largest aspect; these holes were filled with substrate materials after steel bars were driven through to the streambed by 30 cm to support the block. The bottom part of the concrete blocks was buried by approximately 3 cm to keep upwelling flow from coming through underneath the blocks. Exposed cages were not provided with concrete blocks so that they were directly exposed to flow. 3.3.5 Flood treatment To examine the effects of both food resource value and refugium on macroinvertebrate colonization during a flood, the cages were collected either before or after a flood. Pre-flood samples were collected within 2 days before the onset of flood (i.e. increase of discharge) whereas post-flood sampling was carried out as soon as possible after the peak discharge of flood occurred and flow conditions allowed sampling. 3.3.6 Cage installations The 180-m study segment was divided into five sections to control for differences in the macroinvertebrate community and benthic organic matter biomass owing to any longitudinal differences in stream conditions such as local gradient and characteristics of riparian vegetation. These sections were treated as a block effect. Thus, all 8 combinations of treatments were arranged within each of five sections. The colonization cages were filled with similar substrate materials (diameter 3-10 cm) to control for the differences in substrate characteristics among the cages. The substrate materials were collected in situ from Spring Creek and rinsed to remove all macroinvertebrates but not biofilm on the substrate surface. These cages were then embedded in the streambed along the wetted margin of the stream channel (average depth of 10 cm) in glide habitats (transition areas between pools and riffles) 47 such that substrates within the cages were level with the surrounding materials. Cages was separated by at least 1.5 m from each other, and refugium and exposed cages were arranged in an alternate order to avoid confounding treatments by affecting flow conditions. Leaf and flood treatments were randomly assigned within this constraint. The cages were fixed with metal pins (15 cm long) driven into the streambed at the upstream end of the cages. Leaf bags were prepared at the site after soaking leaves in stream water within the Ziplock bags for 2 hours and subsequently placing them into 1 cm mesh plastic bags (10 x 10 cm), and placed over the substrate materials within the cages such that leaf bags were submerged in water. These installations were carried out under low flow conditions to prevent the cages from being out of water during the colonization periods. After installation of the cages, at least a two-week colonization period was set to allow conditioning of leaf materials as a food resource and macroinvertebrate colonization. 3.3.7 Velocity measurements and cage retrievals Current velocity measured at 0.4-depth from the streambed was used as a hydraulic measurement on which the manipulation of flow refugium was based. At both pre- and post-flood samplings, current velocity was measured at both sides and upstream side of the cages at the point of four tenths of water depth using a propeller-based flow meter (propeller diameter 3.5 cm; model 2100, Swoffer Instruments). These velocities were assumed to represent the flow environment within the colonization cages, and these three values were averaged to obtain a 0.4-depth velocity for each cage. A mesh net (mesh size 250 pm) large enough to hold a cage was held immediately downstream of each cage, and the cages were quickly lifted out of water. The contents were transferred to a bucket where leaf bags were first collected and organic detritus and macroinvertebrates were then elutriated and separated from inorganic sediment from the contents of the cages. These samples were taken back to 48 the laboratory and all macroinvertebrates were washed out of the leaf pack materials within 6 hours. Alder leaves that originated from the leaf bags were separated and air-dried for two weeks to be reweighed. Other organic material and macroinvertebrates were preserved in 4 % formalin until further processing. Macroinvertebrates and organic material that were retained on a 500-pm sieve were sorted, identified and counted. In this process, samples from the leaf bags and cages were pooled. Macroinvertebrates were identified to the genus or species level where possible based on Stewart and Stark (1993), Merritt and Cummins (1996), and Wiggins (1996). Organic material was oven-dried for at least 24 hours at 60 °C and subsequently ashed for 2 hours at 550 °C to obtain ash-free dry mass (AFDM). This experiment was repeated three times over the course of a year, and these are hereafter referred to as trial 1, 2, and 3, respectively. Trial 1, 2, and 3 were set up on 30 September 1999, 23 January 2000, and 12 September 2000, respectively. The discharge of Spring Creek was calibrated from discharge data, which was continuously monitored by a data logger in East Creek, which flows into Spring Creek at the upstream end of the study segment. A discharge-water level relation was independently established for Spring Creek by measurements at a range of water levels and corresponding discharges, and used to estimate daily mean discharge of Spring Creek. Various materials that wedged on the upstream side of exposed cages and concrete blocks were regularly cleared to assure the refugium treatment effect: every 3 days for trial 1 and 3; every 5-7 days for trial 2. Concrete blocks and cages were taken out of the stream at the end of each trial, cleaned well, and reinstalled for the following trials and thus each trial was assumed to be independent. 49 3.3.8 Statistical analyses Three-way analyses of variance (ANOVAs) were used to examine the effect of food resource value, flow refugium, flood, and their interactions. The response measures included velocity, total organic matter biomass, total abundance of macroinvertebrates, and abundance of numerically dominant taxa. These analyses were carried out separately for each of the three trials. Abundance was calculated as number of individuals within each cage. The total organic matter biomass referred to the biomass of organic matter that accumulated within the cages and thus excluded the biomass of alder leaves used for natural leaf treatment. To test for statistical assumptions of A N O V A , the data were first fitted to the complete general linear model for unbalanced A N O V A (SAS Version 8, see SAS, 1994) and residuals were compared for deviation from a normal distribution using the Shapiro-Wilks test. Several variables were logio(n+l)-transformed to improve normality of residual distributions wherever appropriate. For trial 3, transformation did not improve the normality of velocity, total organic matter biomass, and total macroinvertebrates and thus raw data were used for these variables although the raw data did not meet the assumptions of normality. Statistical significance level a was set at 0.05. The following were the statistical hypotheses in the A N O V A s that were of particular interest. For velocity, the refugium x flood interaction was tested to examine if refugium cages provided sheltered flow environments relative to the exposed cages during flood; the velocity for exposed cages was predicted to be higher at post-flood sampling than pre-flood occasion for the exposed cages, while that for the refugium cages was expected to remain at low level for both occasions. For macroinvertebrates and total organic matter biomass, refugium x flood interaction was tested to examine if organic matter and macroinvertebrates accumulated in refugium cages during a flood relative to the exposed cages; macroinvertebrate abundance and organic matter biomass was predicted to increase in 50 refugium cages whereas in exposed cages these should decrease or maintain pre-flood levels at the post-flood sampling. Furthermore, for macroinvertebrates, the three-way interaction was tested to examine if macroinvertebrates preferentially accumulated into refugium cages provided with higher food resource value during a flood; macroinvertebrate abundance in refugium cages was predicted to be higher, particularly for refugium cages provided with the higher food resource value on the post-flood sampling occasion. 3.4 Results 3.4.1 Floods The three sampled floods varied in the magnitude of peak discharge (Fig. 3.1). The peak discharge of the flood in trial 1 was the highest among the three floods and the other two floods had similar peak discharges, which were less than half that of the first flood. The flood in trial 1 was of a magnitude that was less common to occur than the other two. The antecedent flow condition and the length of the colonization period were different among the three trials (Fig. 3.1). The floods sampled in trials 1 and 3 occurred after a relatively stable and low antecedent flow condition. Particularly, there had been no precipitation events or detectable flow fluctuations during the colonization period in trial 3. In contrast, the flood sampled in trial 2 occurred after a period of continuous precipitation events and high and variable flow discharge. It is important to note that a flood of a similar magnitude to that of the sampled flood occurred 12 days before the sampling in trial 2. The length of colonization period (i.e. the period between the installation until pre-flood sampling occasion) differed among the three trials since samples were opportunistically collected for relatively intense floods; colonization periods were 27, 52, and 14 days for the first, second and third floods, respectively (Fig. 3.1). The first flood was unexpectedly intense and several post-flood 51 samples were lost owing to washout. Exposed cages were severely damaged (7 out of 10 exposed cages were lost: four natural leaf treatment and three artificial leaf treatment) and several refugium cages were also destroyed owing to the rollover of concrete blocks (three natural leaf treatments were lost). In subsequent trials, the anchoring for the concrete blocks and colonization cages were strengthened to prevent further losses. In trial 1, one cage with natural leaf, refugium and pre-flood treatment was excluded from analyses owing to dewatering of the cage before the flood. 3.4.2 Velocity responses The responses of current velocity beside the colonization cages to the floods differed between refugium and exposed treatment in all three trials, as shown by a significant refugium x flood interactions (Table 3.1 and Fig. 3.2). This indicates that the refugium cages were highly sheltered from flow forces at both pre- and post-sampling occasions whereas the exposed cages experienced higher flow velocity particularly at post-flood occasion. Leaf effects were not significant, showing that the refugium effect was independent of leaf treatment (Table 3.1). Therefore, the velocity manipulation was achieved to test our hypotheses regarding flow refugium. The difference in flow velocity between refugium and exposed cages at the post-flood sampling was the greatest for trial 1 (Fig. 3.2). This was largely due to the difference in the peak discharges among the three floods. 3.4.3 Organic matter responses The responses of total organic matter biomass varied among the trials. Total organic matter biomass responded to the floods in trials 1 and 3 as shown by a significant flood effect (Table 3.2 and Fig. 3.3). Block effect was significant in trial 1. In trial 1, total organic matter biomass increased in the refugium cages after the flood whereas that in the exposed 52 cages remained similar to the pre-flood level although this pattern was not shown as a significant refugium x flood interaction (Table 3.2). In trial 3, total organic matter biomass was higher after the flood independently of the leaf and refugium effect. In trial 2, on the other hand, total organic matter biomass did not show any significant response to flood, refugium, or leaf treatment as shown by insignificant main effects (Table 3.2). In all three trials, no significant effect of leaf on total organic matter biomass was detected (Table 3.2), suggesting that there was no difference in the accumulation of organic matter biomass between natural and artificial leaf treatment over the floods. 3.4.4 Macroinvertebrate responses Responses to the flood of total macroinvertebrate abundance varied among the trials. In all three trials, the total abundance of macroinvertebrates was higher in the cages provided with natural leaf packs as shown by the significant leaf effects (Table 3.3 and Fig. 3.4). Total macroinvertebrate abundance showed a significant response to the flood only in trial 3 (Table 3.3) ; the total abundance increased after the flood, as did the total organic matter biomass. In neither trial 1 nor 2 did macroinvertebrates show significant response to the floods. In trial 2, higher macroinvertebrate abundance in exposed cages particularly provided with natural leaves over the refugium cages was consistent over the flood (Fig. 3.4). These results suggest that in none of the three trials was there the pattern of macroinvertebrate accumulation in flow refugia as a whole community. Compositions and relative abundance of numerically dominant taxa differed among the three trials due to seasonal effects and varying lengths of colonization periods (Table 3.4) . To compare the macroinvertebrate responses for numerically dominant taxa within each trial and among the three trials, the taxa that comprised at least 3 % of the total abundance for at least one of the three trials, and greater than 1 % in all the three trials were 53 selected. For example, Capnia was excluded from further analyses owing to low relative abundance of less than 1 % in trial 2 although it comprised greater than 10 % of the total abundance in trial 3. Although Zapada cinctipes was not dominant in trial 2, an exception was made and it was included together with Malenka in trial 2 as Nemouridae, since these two taxa share similar morphology and food basis as detritivores. Consequently eight taxa were further examined as the numerically dominant taxa, comprising relative abundances of 73, 84, and 73 % of total abundance in trials 1,2, and 3, respectively. Examinations of the numerically dominant taxa revealed taxon-level patterns, particularly in trial 1. A significant three-way interaction and refugium x flood interactions were observed for the detritivorous mayfly Paraleptophlebia and the stonefly Despaxia in trial 1 (Table 3.5 and Fig. 3.5), suggesting that these taxa disproportionately colonized the refugium cages provided with natural leaf packs during the flood. In trial 2 and 3, on the other hand, the responses of the numerically dominant taxa were similar to that at the community level. In trial 3, six taxa showed increased abundance during the flood independent of refugium treatments, as shown by significant flood effects (Table 3.5). Although two taxa, Tanypodinae and Lepidostoma, showed significant three-way interactions in trial 3, these were not the consequence of the pattern that I wished to test (Fig. 3.5). It is important to note that colonization rate during floods differed among taxa. For example, colonization rates of Paraleptophlebia and Despaxia during the flood were particularly high relative to other taxa in trial 3 (Fig. 3.5). In trial 2, no dominant taxa showed significant response to the flood except for Chironominae, which decreased during the flood (Table 3.5 and Fig. 3.5). Four taxa were more abundant in the exposed cages than the refugium cages in trial 2, being similar to the pattern found for total macroinvertebrate abundance. The responses to the three trials varied among the numerically dominant taxa. The numerically most dominant taxon, Orthocladiinae, responded little to the flood in all three 54 trials, only showing strong disproportionate colonization to natural leaf packs, and the different colonization rate between natural and artificial leaf treatments was pronounced in the refugium cages in trial 2 (Table 3.5 and Fig. 3.6). Paraleptophlebia and Despaxia showed disproportionate accumulation in natural leaf treated refugium cages only during the flood in trial 1. None of the Tanypodinae, Chironominae, or Lepidostoma showed significant responses to the flood in trial 1 owing to high variability, but these three taxa showed a pattern of increased abundance during the flood in trial 3. Baetis spp. was the least responsive taxon to any of the treatments. Nemouridae only showed a response to the flood in trial 3, and this increase of abundance was higher in the natural leaf-treated cages. Orthocladiinae, Paraleptophlebia and Nemouridae always showed a higher abundance in natural leaf treatment in all three trials. 3.5 Discussion The hydraulically benign flow environment similar to what was provided by concrete blocks in this study can occur naturally in places such as behind boulders and woody debris during high flow events. Therefore, this study examined the potential of such habitats to serve as flow refuges in Spring Creek. These findings did not support our first hypothesis that macroinvertebrates accumulate in flow refugia during floods as a whole community. However, the second hypothesis was partly supported in that two detritivorous taxa Paraleptophlebia and Despaxia showed disproportionately accumulation in flow refugia provided with higher food resource value during a flood. However, the occurrence of such accumulations was likely dependent on the characteristics of the flood. Despite varying colonization periods of the cages and leaf packs, the presence of natural leaves influenced the rate of macroinvertebrate colonization in the cages in all three trials. In all three trials, total macroinvertebrate abundance was higher in the natural leaf 55 treatment than in the artificial leaf treatment. It has been shown that macroinvertebrate abundance is positively correlated with the biomass of organic detritus (Egglishaw, 1964; Mackay and Kalff, 1968; Flecker, 1984). In forested streams, leaf input has strong influence on macroinvertebrate abundance (Richardson, 1991; Wallace etal., 1999). Richardson (1992a) demonstrated that leaf materials primarily served as a food resource for macroinvertebrates and thus resulted in high rates of colonization of individuals (see also Dobson et al., 1992; Dudgeon and Wu, 1999). In my study, high total abundance in the natural leaf treatment was particularly due to the high abundance of the numerically dominant taxa such as Orthocladiinae, Paraleptophlebia and Nemouridae. These taxa are considered detritivores (Merritt and Cummins, 1996) and thus the high abundance of macroinvertebrates observed in the natural leaf treatment was due to higher food resource value of natural leaves. Richardson (1992a) revealed that the different colonization rates of non-shredder animals between artificial and natural leaf packs were due to different accumulation of fine organic matter between the leaf types. In my study, however, the effect of leaf type on accumulation of organic matter biomass was not significant whereas leaf types on total abundance was highly significant, suggesting the high abundance was largely due to direct food value of leaf treatment. The continuous redistribution of individuals through various colonization pathways is an important process in maintaining the populations of macroinvertebrate communities in stream ecosystems where unidirectional flow can result in downstream loss of individuals (Hynes, 1970; Waters, 1972). Among the different kinds of colonization pathways, drifting has been considered of primary importance to macroinvertebrates (Townsend and Hildrew, 1976; Matthaei et al, 19966). Change in flow discharge can result in different rates of macroinvertebrate colonization (Lancaster et al, 1990; Winterbottom et al., 19976), and this has been attributed to increased drift rates of individuals (Townsend and Hildrew, 1976; 56 Perry and Perry, 1986; Allan and Feifarek, 1989; Poff and Ward, 1991). However, if flow increases significantly and consequently disturbs the streambed, local populations of macroinvertebrates could decrease (Fisher et al, 1982; Scrimgeour and Winterbourn, 1989; Giller et al, 1991; Flecker and Feifarek, 1994; Angradi, 1997). Population loss and mortality through flood disturbance can be moderated through the presence of spatial refugia where the hydraulic environment is benign relative to other places exposed to high flow forces. In such cases, colonization can be hindered in exposed habitats while refugium habitats may act as sinks for dislodged individuals (see Robertson et al, 1995). Therefore, the function of refugia as sinks is strongly related to flood characteristics and condition of the biological community at the time of the event. In this study, three trials resulted in different responses of macroinvertebrates to the floods. Responses of macroinvertebrates and organic matter were minimal in trial 2 while the responses that occurred in trials 1 and 3 were different from each other to some extent. Interpretations of these results need to be made carefully with consideration of magnitude and antecedent conditions of the sampled floods since these trials were carried out under different conditions. Movements of organic matter can be assumed as passive and subject to stream flow. Thus the absence of organic matter response in trial 2 was probably the consequence of minimal effects of the flood across the stream. The flood in trial 2 was preceded by variable discharge and occurrence of floods with a similar intensity. In February and March, furthermore, organic matter input from allochthonous sources was fairly low except the input of needles (see Richardson, 1992£>), and organic matter stored within the stream channel might have been depleted from the system through previous high flow events during winter (Webster et al, 1987; Cuffney and Wallace, 1989; Kiffney et al, 2000). Therefore, it is likely that in trial 2, easily transportable organic materials and macroinvertebrates were already depleted by variable flows in winter and a flood of similar 57 magnitude did not result in further movement. The fact that the flood in trial 3 with a similar peak discharge did result in responses of organic materials and macroinvertebrates further supports the argument that antecedent flow condition influenced response patterns. In contrast, trials 1 and 3 coincided with the timing of high allochthonous material input to the stream (Richardson, 19926) and these floods were both relatively large floods that occurred after relatively dry season. Low stable flow conditions as well as high input of organic materials resulted in high abundance of easily transportable materials and presumably formed a macroinvertebrate community susceptible to floods. A fairly high number of macroinvertebrates and abundant leaf materials were trapped and deposited on riffle areas and channel margins at low flow but were washed out during high flow conditions (personal observation). Previous studies have shown that retentiveness of stream channels is inversely related to discharge and thus the standing crop of organic matter within a channel is related to the timing of leaf input (King et al, 1987; Snaddon et al, 1992). Other studies have showed that export of organic materials is highly associated with timing of flood and input of allochthonous organic matter (Malmqvist et al, 1978; Cuffney and Wallace, 1989; Wallace et al, 19956). It has also been shown that subsequent floods can result in lower drift and thus less clear responses of macroinvertebrates, probably due to depletion of the benthos (Irvine, 1985). Therefore, different antecedent conditions probably resulted in different macroinvertebrate response to a flood and colonization of refugia in this study. Besides the antecedent flow conditions, the magnitude of the flood (peak discharge of flood) was important in determining macroinvertebrate responses. Trials 1 and 3 were conducted at a similar time of year and with similar antecedent flow characteristics (i.e. low and stable flow). However, these trials were carried out for floods of different magnitude and resulted in different responses of macroinvertebrates. In trial 3, macroinvertebrate abundance and organic matter biomass increased in both exposed and refugium cages whereas in trial 1 58 the increases were mainly shown in the refugium cages. As shown by the fact that several cages and concrete blocks were washed out in trial 1 during the flood, the stream was more disturbed in trial 1 than trial 3, and the influence of exposure to the floods in the "exposed" treatment were different. Although velocity increased in the exposed cages relative to the refugium cages in trial 3, this probably was not strong enough to adversely affect colonization of macroinvertebrates and the accumulation of organic matter. These colonization cages were located along channel margins and the location of the cages might have resulted in this pattern. It has been shown that stream flow tends to accumulate animals along margins (see Ciborowski, 1983). In trial 3, the flood might have deposited animals near the places where the cages were embedded. These findings suggest that colonization into refuges might have happened, but not on the spatial scale that we expected. In trial 1, owing to the influence of high flow velocity during the flood, colonization was impeded in the exposed habitat, which therefore did not act as a sink for macroinvertebrates. Hence, organic matter biomass and macroinvertebrate abundance did not increase in the exposed cages. The response to the flood differed among the numerically dominant taxa, and detritivorous taxa Paraleptophlebia and Despaxia showed a pattern of increased colonization of the refugium cages provided with natural leaves. The accumulation of organic material did not differ between natural and artificial leaf treatments, suggesting that the higher colonization rate was due to higher food resource value of natural leaves. In previous studies on flow refugia, the influence of organic matter biomass on the abundance or accumulation of macroinvertebrates into refugia has been refuted (Lancaster and Hildrew, 1993a; Winterbottom et ai, 1997a). They argued that food resources are not limited in their study sites and also that taxa responding were not necessarily detritivores. The primary energy source in forested stream comes from allochthonous input (Fishers and Likens, 1973). Furthermore, macroinvertebrates strongly depend on such energy sources from riparian input 59 (Cummins et al, 1989). In our system, the detrital food resource has been shown to be limiting for the macroinvertebrate community (Richardson, 1991). Thus the effect of organic matter might have been pronounced owing to its higher importance as a trophic base in our system. These results suggest that food resource quality of flow refugia can influence patterns of colonization during high flow events. There was probably an artifact of using colonization cages in this study, and the macroinvertebrate responses in the exposed treatment were particularly underestimated in two senses. Firstly, the colonization cages did not allow the substrate materials within them to be freely influenced by high flow events as much as were the surrounding materials. It is thus likely that substrate materials were less disturbed by the floods. Secondly, leafpacks were fixed within the cages. Thus leafpacks remained in position even in the exposed habitat during high flow events, which was very unlikely to occur in the natural situation. Leaf packs could have been washed away if they were not fixed within the cages. These facts could have resulted in the following consequences. Substrate within each cage was more stable relative to surrounding materials and thus could have provided refugium (see Matthaei et al, 1999). Furthermore, leaf packs could have provided micro-scale refugia for macroinvertebrates if they could hold on to leaf pack materials or hide inside during floods. Particularly in trial 1, exposed cages were heavily disturbed and some of the cages were lost by the force of flow. In such a condition, fewer animals could have been left in a real situation. Furthermore, in trial 3, the exposed habitat might have accumulated materials and animals during the flood for this reason. Although these results need to be carefully interpreted, these facts may further emphasize the importance of colonization patterns found in trial 1 and 3 where flow refugia acted as sink for individuals. The fixed nature of the cages and leafpacks yield interesting insights into the differences in the responses to floods among the dominant taxa. In trial 1, some taxa such as Orthocladiinae and Nemouridae did 60 not respond to the flood in the exposed treatment whereas other taxa such as Paraleptophlebia and Despaxia decreased in the exposed cages and increased in the refugium cages. The former taxa might have hidden within leaves or stayed within substrate during the flood, whereas Paraleptophlebia and Despaxia left the exposed treatment. It is not clear whether the act of leaving was active or passive. Nevertheless, the fact that the high colonization rate of these two taxa during the flood in trial 3 suggests that these two taxa may be sensitive to flow changes and actively seek other microhabitats. These are probably related to taxon-specific colonization rates and responses to flow fluctuations (Winterbottom etal., \991b; Lancaster, 1999; Holomuzki and Biggs, 2000). Mechanisms of colonization (i.e. active or passive) remain unclear. That is, how did Paraleptophlebia and Despaxia preferentially accumulate in flow refugia provided with high food value? In order to identify the food resource value of flow refugia, these individuals had to colonize in the first place. The probability of reaching flow refugia is assumed to be independent of leaf types. Therefore, it is likely that individuals made a decision about staying or leaving based on the habitat quality available in the flow refugia. These two taxa probably left flow refugia to other places even during high flow events if food resource value was not adequate, suggesting that colonization occurs quickly. For these taxa high flow events that can adversely affect other taxa may be merely a way of redistribution. My findings suggest that the function of flow refugia is dependent on flood characteristics. Geomorphic elements such as boulders and woody debris generally provide important habitat for macroinvertebrates, supporting high abundance of animals because of the presence of abundant organic matter. These geomorphic elements can provide hydraulically benign habitats during high flow events and may also be important as flow refugia for some taxa. If food resource value is low in such potential flow refugia, persistence and local populations of certain taxa could be lower through reduced colonization 61 of refugia during high flow events. Therefore, preservation of allochthonous input of organic detritus as well as geomorphic features that provide hydraulically sheltered environment may be important to maintain local populations and persistence of the community. Only one kind of refugium was tested for a limited set of flow conditions in this study. There are various types of refugia existing at different spatial and temporal scales. Furthermore, the functions of refugia can vary in different systems. Given the influential role of refugia in community persistence and dynamics of macroinvertebrates, further research needs to be done on the function of different refugia in different systems. 62 Table 3.1 Summary of three-way A N O V A s to examine the effects of leaf, refugium, flood and their interactions on flow velocity measured at 0.4-depth. Effects df MS F P Trial 1 Leaf(L) ! <0.01 0.04 0.8458 Refugium (R) 1 1.46 101.61 <0.0001 Flood (F) 1 1.90 132.34 <0.0001 Block 0.01 0.92 0.4675 L x R 1 <0.01 0.01 0.9418 L x F 1 0.01 0.99 0.3284 R x F 1 0.99 69.07 <0.0001 L x R x F 1 <0.01 0.23 0.6341 Error 25 Trial 2 Leaf(L) 1 <0.01 0.01 0.9064 Refugium (R) 1 0.17 179.48 <0.0001 Flood (F) 1 0.06 64.95 <0.0001 Block <0.01 0.63 0.6438 L x R 1 <0.01 1.00 0.3251 L x F 1 <0.01 0.11 0.3251 R x F 1 0.06 64.16 <0.0001 L x R x F 1 <0.01 2.53 0.1226 Error 28 Trial 3 Leaf(L) 1 <0.01 3.60 0.0680 Refugium (R) 1 0.15 537.53 <0.0001 Flood (F) 1 0.06 215.29 <0.0001 Block <0.01 2.18 0.0973 L x R 1 <0.01 1.06 0.3129 L x F 1 <0.01 0.05 0.8315 R x F 1 0.04 156.86 <0.0001 L x R x F 1 <0.01 0.15 0.7002 Error 28 63 Table 3.2 Summary of three-way A N O V A s to examine the effects of leaf, refugium, flood and their interactions on total organic matter biomass (> 500 urn). Effects df MS F P Trial 1 Leaf(L) ! 26.38 1.52 0.2340 Refugium (R) 1 72.70 4.20 0.0562 Flood (F) 1 244.02 14.09 0.0016 Block 114.38 6.60 0.0021 L x R 1 34.34 1.98 0.1771 L x F 1 30.68 1.77 0.2007 R x F 1 21.51 1.24 0.2806 L x R x F 1 1.96 0.11 0.7406 Error 17 Trial 2 Leaf(L) 1 0.02 0.30 0.5874 Refugium (R) 1 0.03 0.35 0.5572 Flood (F) 1 <0.01 0.01 0.9333 Block 0.09 1.12 0.3656 L x R 1 <0.01 0.02 0.8861 L x F 1 0.03 0.42 0.5244 R x F 1 0.15 1.83 0.1870 L x R x F 1 <0.01 0.08 0.7850 Error 28 Trial 3 Leaf(L) 1 0.01 0.39 0.5360 Refugium (R) 1 0.07 1.75 0.1963 Flood (F) 1 3.18 83.83 <0.0001 Block 0.09 2.60 0.0574 L x R 1 0.09 2.25 0.1447 L x F 1 <0.01 0.13 0.7195 R x F 1 0.03 0.75 0.3948 L x R x F 1 0.13 3.35 0.0778 Error 28 64 Table 3.3 Summary of three-way A N O V A s to examine the effects of leaf, refugium, flood and their interactions on total macroinvertebrate abundance. Effects df MS F P Trial 1 Leaf(L) ! 556895.42 38.57 <0.0001 Refugium (R) 1 916.27 0.06 0.8041 Flood (F) 1 22442.13 1.55 0.2294 Block 42368.64 2.93 0.0516 L x R 1 2816.27 0.20 0.6643 L x F 1 25533.62 1.77 0.2011 R x F 1 16808.47 1.16 0.2957 L x R x F 1 45170.46 3.13 0.0949 Error 17 Trial 2 Leaf(L) 1 1058200.90 30.24 <0.0001 Refugium (R) 1 609102.40 17.41 0.0003 Flood (F) 1 48441.60 1.38 0.2492 Block 85948.46 2.46 0.0688 L x R 1 95062.50 2.72 0.1105 L x F 1 36120.10 1.03 0.3183 R x F 1 3763.60 0.11 0.7454 L x R x F 1 136.90 0.00 0.9506 Error 28 Trial 3 Leaf(L) 1 1.68 73.18 <0.0001 Refugium (R) 1 <0.01 0.12 0.7308 Flood (F) 1 0.28 12.04 0.0017 Block 0.02 0.82 0.5234 L x R 1 0.11 4.96 0.0341 L x F 1 <0.01 0.01 0.9150 R x F 1 <0.01 0.21 0.6487 L x R x F 1 0.02 1.06 0.3118 Error 28 65 Table 3.4 Relative abundance of the numerically dominant taxa for each trial. Taxon Relative abundance (%) Trial 1 Orthocladiinae 35.31 Paraleptophlebia 9.40 Chironominae 7.76 Baetis spp. 6.44 Zapada cinctipes 5.87 Tanypodinae 3.64 Lepidostoma 3.54 Ameletus 2.75 Capnia 2.12 Sweltsa 2.02 Oligochaeta 1.72 Ironodes 1.71 Despaxia 1.62 Trial 2 Orthocladiinae 37.94 Chironominae 12.66 Paraleptophlebia 9.54 Malenka 9.16 Baetis spp. 6.16 Lepidostoma 5.05 Swefrsa 2.94 Tanypodinae 2.47 Cinygmula 1.82 Oligochaeta 1.64 Ameletus 163 Despaxia 1.37 Simuliidae 1.04 Trial 3 Orthocladiinae 35.31 Paraleptophlebia 12.99 Capm'a 10.06 Zapada cinctipes 8.61 Baetis spp. 5.56 Despaxia 3.84 Chironominae 2.81 Oligochaeta 2.54 Chironomidae pupae 2.45 Tanypodinae 2.21 Ameletus 2.07 Sweltsa 1-81 Lepidostoma 1 -29 1) cd p o £ 3 t<5 <^ QJ c o c o cd C o >> e CO f—i S 3 G JS "o o O cd •c * * * * * * r--* * * # * * * * * * * * * * * * * * * ca 3 i-i CN m 3 '5b a oi ' - ' ( N O T3 O O ^ C N cn M o o fe P3 ~ * C N cn Di X - H C N cn fe X — I C N cn fe X Di '—i C N cn | fe X Di X CO o o CD > 1.0 -j 0.8 -0.6 -0.4 -0.2 0.0 1.0 0.8 0.6 0.4 0.2 0.0 1.0 0.8 -0.6 -0.4 0.2 0.0 Trial 1 Trial 2 Trial 3 R Pre _ I L H R Post Natural I I Artificial Figure 3.2 Mean (+1SE) of velocity (m/s) measured at 0.4-depth beside the cages. E and R denote "exposed" and "refugium" treatment, respectively. Natural and artificial denotes the types of leaf treatment. "Pre" and "post" refer to pre- and post-sampling occasions, respectively. 6 9 0 CO O Q LL < C7) CO CO E o l a CD •i—i •*—> CO E o 'c CO CD 40 n 30 20 J 0 40 30 20 10 0 40 _ 30 CO o 20 10 Trial 1 • Natural • Artificial 10 JJI Trial 2 Mm Trial 3 R R Pre Post Figure 3.3 Mean (+1SE) of total organic matter biomass (g AFDM/cage) excluding alder leaves used for natural leaf treatment. A l l materials greater than 500um were included. E and R denote "exposed" and "refugium" treatment, respectively. Natural and artificial denotes the types of leaf treatment. "Pre" and "post" refer to pre- and post-sampling occasions, respectively. 70 Trial 1 0 O) CO .o tn co "D '> c 0 £ C 0 CJ c CO c _ Q < Natural i i Artificial 1000 800 ] 600 400 ] 200 Trial 3 n i l R R Pre Post Figure 3.4 Mean (+1SE) of total abundance of macroinvertebrates including individuals found in the leaf bags. E and R denote "exposed" and "refugium" treatment, respectively. Natural and artificial denotes the types of leaf treatment. "Pre" and "post" refer to pre- and post-sampling occasions, respectively. 71 (96eo/S|enp!A!pu! jo jeqiunu) eouepunqy 0 3 o . J3 C A *•> W> cn ^ u o •° o SP j5 .2 s .3 is 5 B rt ,0 c 3 rt c 1/1 r-rt "§ c o 3 3 ^ . ^ O --3 — £ u tjf) •»-> <D ^ ft U cfl ft s u -3 c -3 3 rt D cfl >5 3 w s i ° rt C W> 3 - 3 S | 1 o 3 S rt 2 g 1 ^ T3 rt e rt O 3 p 3 T3 I 8 , i3 o . 12 s 3 aj cfl 3 -3 O ^ *S "§ o rt ca £ W R« w £ (Z» "e3 <j T1 "G 3 + •"-> CD ^ 'o* £ rt • r rt 0) CN <U ^ : «* * s CU .. o FL —< u .£f .2 ft Cti h >^ 00 O O O (96eo/s|enp!A!pu! jo jeqiunu) eouepunqv CD o c rt -a so c o "cfl rt u CJ o c 2 rt rt =f i> <D Q O 00 * E * c a i -rt K " * ^ C f l *J e ° 8 S ft o g * ft .s3 CD s 2 8 7 3 0 & 5 o ™ = I i 13 7 3 D H fe "2 « | § | -*—< i 1 S s s s 43 <N cfl 5 rt O-• 2 3 « 0 1 - o u cfl B o c T3 O T - H 4—I ° -c — i . rt + cfl -r; w bo ,a § -O CD «*a rt 2 J J j - s CD 3 C a — •B -g i/> J3 ^ rt s E .2 i -a c rt "rt s-rt Z _>> 13 "3 _> _> o o CD CD ft & cfl cfl CD <D Chapter 4 . General conclusions and management implications The findings of the preceding two chapters suggest that the placements of boulder clusters may influence macroinvertebrates through two mechanisms, which were operating at two different temporal scales. Firstly, boulder clusters increased the abundance of macroinvertebrates through the increased food resources stored within the stream channel, as a relatively long-term recovery of ecosystem-level processes from loss of habitat heterogeneity and retentiveness. Secondly, placements of boulder clusters might be important in maintaining local populations of certain detritivorous taxa and other species, whose persistence against short-term high flow events is associated with the presence of the sheltered habitat with appropriate biomass of organic detritus. Therefore, the increased abundance of macroinvertebrates observed in the treatment section after the placements of boulder clusters may be partly due to the increased availability of flow refugia with high food resource value within the channel. It has been emphasized that process-based elucidation of natural systems, identification of the factors that are responsible for the degradation of systems, and application of appropriate remedies based on them are important for ecological restoration (Frissell, 1997; Angermeier, 1997). In the case of small-scale stream restorations with in-stream structures, it is important to understand the functions of different microhabitats and incorporate such knowledge to predict the outcomes of channel modification (Rabeni and Jacobson, 1993; Gore et al, 1998). Ecological restoration further emphasizes the consideration of ecological processes at large spatial scales (i.e. watershed), and the priority of stopping source-problems whenever possible (Roper et al, 1997; Kauffman et al, 1997; Wissmar and Beschta, 1998). If the process-based restoration at small spatial scales can be undertaken with thorough consideration of local geomorphic and hydraulic conditions (Beschta and Platts, 1986) as well as large-scale processes, the benefits can be substantial. 74 My findings in this thesis emphasize that the maintenance of heterogeneity of stream channels, which is associated with the retention capability of organic detritus and also availability of flow refugia, and appropriate levels of allochthonous organic detritus are crucial in maintaining local populations of macroinvertebrates. Consequently, riparian vegetation needs to be appropriately managed to provide continuous input of large organic debris as well as leaf litter (Naiman et al, 2000). Furthermore, in places where these structures and functions of ecosystems are already lost, the restoration efforts such as planting trees along stream banks to facilitate the foundation of future riparian vegetation, and placing in-stream structures as substitutes of large organic debris can be useful until these planted riparian vegetations become established. Most in-stream structures have been reported to last for a limited period after installation (Frissell and Nawa, 1992; Roper et al, 1998). It is therefore important to monitor effects of practices for guidance of future works (Osborne et al, 1993; Wissmar and Beschta, 1998). In this study, long-term (e.g. decades) effects of in-stream structures are unknown. Furthermore, the different kinds of placement methods of in-stream structure (e.g. types and sizes of structures, locations within channel, and arrangement patterns) need to be further examined to test potentially different effects of practices. The most difficult task may be to make sufficient spatial replication to test the numbers of treatments and to generalize the outcomes. 75 Literature cited Abele, L .G . (1974) Species diversity of decapod crustaceans in marine habitats. Ecology 55: 156-161. Allan, J.D. and B.P. Feifarek (1989) Distances travailed by drifting mayflay nymphs: factors influencing return to the substrate. Journal of the North American Benthological Society 8: 322-330. Andrus, C.W., B .A. Long, and H.A. Froehlich (1988) Woody debris and its contribution to pool formation in a coastal stream 50 years after logging. Canadian Journal of Fisheries and Aquatic Sciences 45: 2080-2086. Angradi, T.R. (1996) Inter-habitat variation in benthic community structure, function, and organic matter storage in 3 Appalachian headwater streams. Journal of the North American Benthological Society 15: 42-63. Angradi, T.R. (1997) Hydrologic context and macroinvertebrate community response to floods in an Appalachian headwater stream. American Midland Naturalist 138: 371-386. Angradi, T.R. (1999) Fine sediment and macroinvertebrate assemblages in Appalachian streams: a field experiment with biomonitoring applications. Journal of the North American Benthological Society 18: 49-66. Angermeier, P.L. (1997) Conceptual roles of biological integrity and diversity. Pp. 49-65 in J.E. Williams, C A . Wood, and M.P. Dombeck, editors. Watershed Restoration: Principles and Practices. American Fisheries Society, Bethesda, Maryland, USA. Angermeier, P.L. and J.R. Karr (1984) Relationships between woody debris and fish habitat in a small warmwater steam. Transactions of the American Fisheries Society 113: 716-726. Badri, A.J. , J. Giudicelli, and G. Prevot (1987) Effects of a flood on the benthic invertebrate community in a Mediterranean river, the Rdat (Morocco). Acta Oecologia 8: 481-500. Bender, E.A., T.J. Case, and M.E. Gilpin (1984) Perturbation experiments in community ecology: theory and practice. Ecology 65: 1-13. Benke, A.C. , R.L. Henry, D . M . Gillespie, and R.J. Hunter (1985) Importance of snag habitat for animal production in southeastern streams. Fisheries 10: 8-13. Benke, A.C. , T.C. Van Arsdall, Jr., D . M . Gillespie, and F.K. Parrish (1984) Invertebrate productivity in a subtropical blackwater river: the importance of habitat and life history. Ecological Monographs 54: 25-63. Beschta, R.L. (1979) Debris removal and its effects on sedimentation in an Oregon coast range system. Northwest Science 53: 71-77. Beschta, R.L. and W.S. Platts (1986) Morphological features of small streams: significance and function. Water Resources Bulletin 22: 369-379. 76 Bilby, R.E. (1981) Role of organic debris dams in regulating the export of dissolved and particulate matter from a forested watershed. Ecology 62: 1234-1243. Bilby, R.E. (1984) Removal of woody debris may affect stream channel stability. Journal of Forestry 82: 609-613. Bilby, R.E. and G.E. Likens (1980) Importance of organic debris dams in the structure and function of stream ecosystems. Ecology 61:1107-1113. Bisson, P.A., R.E. Bilby, M.D. Bryant, C A . Dolloff, G.B. Grette, R.A. House, M . L . Murphy, K . V . Koski, and J.R. Sedell (1987) Large woody debris in forested streams in the Pacific Northwest: past, present, and future. Pp. 143-190 in E.O. Salo and T.W. Cundy, editors. Streamside Management: Forestry and Fishery Interactions. University of Washington, Institute of Forest Resources, Contribution 57, Seattle. Borchardt, D. (1993) Effects of flow and refugia on drift loss of benthic macroinvertebrates: implications for habitat restoration in lowland streams. Freshwater Biology 29: 221-227. Bradshaw, A .D . (1996) Underlying principles of restoration. Canadian Journal of Fisheries and Aquatic Sciences 53 (supplement 1): 3-9. Brussock, P.P., A . V . Brown, and J.C. Dixon (1985) Channel form and stream ecosystem models. Water Resources Bulletin 21: 859-866. Bryant, M.D. (1983) The role and management of woody debris in west coast salmonid nursery streams. North American Journal of Fisheries Management 3: 322-330. Bryant, M.D. (1995) Pulsed monitoring for watershed and stream restoration. Fisheries 20: 6-12. Carignan, R. and R.J. Steedman (2000) Impacts of major watershed perturbations on aquatic ecosystems. Canadian Journal of Fisheries and Aquatic Sciences 57 (Suppl.): 1-4. Cederholm, C.J., R.E. Bilby, P A . Bisson, T.W. Bumstead, B.R. Fransen, W.J. Scarlett, and J.W. Ward (1997) Response of juvenile coho salmon and steelhead to placement of large woody debris in a coastal Washington stream. North American Journal of Fisheries Management 17: 947-963. Chergui, H. , A . Maamri, and E. Pattee (1993) Leaf litter retention in two reaches of a Moroccan mountain stream. Limnologica 23: 29-37. Ciborowski, J.J.H. (1983) Downstream and lateral transport of nymphs of two mayfly species (Epemeroptera). Canadian Journal of Fisheries and Aquatic Sciences 40: 2025-2029. Clifford, H.F. (1982) Effects of periodically disturbing a small area of substratum in a Brown-water stream of Alberta, Canada. Freshwater Invertebrate Biology 1: 39-47. 77 Connel, J.H. (1978) Diversity in tropical rain forests and coral reefs. Science 199: 1302-1310. Corkum, L.D. , P.J. Pointing, and J.J.H. Ciborowski (1977) The influence of current velocity and substrate on the distribution and drift of two species of mayflies (Ephemeroptera). Canadian Journal of Zoology 55: 1970-1977. Crispin, V. , R. House, and D. Roberts (1993) Changes in instream habitat, large woody debris, and salmon habitat after the restructuring of a coastal Oregon stream. North American Journal of Fisheries Management 13: 96-102. Crook, D.A. and A.I. Robertson (1999) Relationships between riverine fish and woody debris: implications for lowland streams. Marine and Freshwater Research 50: 941-953. Cuffney, T.F. and J.B. Wallace (1989) Discharge-export relationships in headwater streams: the influence of invertebrate manipulations and drought. Journal of the North American Benthological Society 8: 331-341. Culp, J .M., S.J. Walde, and R.W. Davies (1983) Relative importance of substrate particle size and detritus to stream benthic macroinvertebrate microdistribution. Canadian Journal of Fisheries and Aquatic Sciences 40: 1568-1574. Cummins, K.W. , M . A . Wilzbach, D . M . Gates, J.B. Perry, and W.B. Taliaferro (1989) Shredders and riparian vegetation. Bioscience 39: 24-30. Death, R.G. (1996) The effect of patch disturbance on stream invertebrate community structure: the influence of disturbance history. Oecologia 108: 567-576. Death, R.G. and M.J . Winterbourn (1995) Diversity patterns in stream benthic invertebrate communities: the influence of habitat stability. Ecology 76: 1446-1460. Diez, J.R., S. Larrafiaga, A . Elosegi, and J. Pozo (2000) Effect of removal of wood on streambed stability and retention of organic matter. Journal of the North American Benthological Society 19: 621-632. Dobson, M . and H . Cariss (2000) Improving detritus retention as management option in streams impacted by forestry. Verhandlungen. Internationale Vereinigung fur Theoretische und Angewandte Limnologie/Proceedings. International. 27: ?-?. Dobson, M . and A . G . Hildrew (1992) A test of resource limitation among shredding detritivores in low order streams in southern England. Journal of Animal Ecology 61: 69-77. Dobson, M . , A . G . Hildrew, S. Orton, and S.J. Ormerod (1995) Increasing litter retention in moorland streams: ecological and management aspects of a field experiment. Freshwater Biology 33: 325-337. Dole-Olivier, M.-J. , P. Marmonier, and J.-L. Beffy (1997) Response of invertebrates to lotic disturbance: is the hyporheic zone a patchy refugium? Freshwater Biology 37: 257-276. 78 Downes, B.J., P.S. Lake, E.S.G. Schreiber, and A. Glaister (1998) Habitat structure and regulation of local species diversity in a stony, upland stream. Ecological Monographs 68: 237-257. Dudgeon, D. and K . K . Y . Wu (1999) Leaf litter in a tropical stream: food or substrate for macroinvertebrates? Archive fur Hydrobiologia 146: 65-82. Egglishaw, H.J. (1964) The distributional relationship between the bottom fauna and plant detritus in streams. Journal of Animal Ecology 33: 463-476. Ehrman, T.P. and G.A. Lamberti (1992) Hydraulic and particulate matter retention in a 3rd-order Indiana stream. Journal of the North American Benthological Society 11: 341-349. Elliott, S.T. (1986) Reduction of a Dolly Varden population and macrobenthos after removal of logging debris. Transactions of the American Fisheries Society 115: 392-400. Fahy, E. (1975) Quantitative aspects of the distribution of invertebrates in the benthos of a small stream system in western Ireland. Freshwater Biology 5: 167-182. Fausch, K .D. and R.G. Bramblett (1991) Disturbance and fish communities in intermittent tributaries of a western Great Plains river. Copeia 3: 659-674. Fausch, K .D. and T.G. Northcote (1992) Large woody debris and salmonid habitat in a small coastal British Columbia stream. Canadian Journal of Fisheries and Aquatic Sciences 49: 682-693. Feller, M.C . and J.P. Kimmins (1979) Chemical characteristics of small streams near Haney in southwestern British Columbia. Water Resources Research 15: 247-258. Fisher, S.G. and G.E. Likens (1973) Energy flow in Bear Brook, New Hampshire: an integrative approach to stream ecosystem metabolism. Ecological Monographs 43: 421-439. Fisher, S.G., L.J . Gray, N.B. Grimm, and D.E. Busch (1982) Temporal succession in a desert stream ecosystem following flash flooding. Ecological Monographs 52: 93-110. Flebbe, P. A . (1999) Trout use of woody debris and habitat in Wine Spring Creek, North Carolina. Forest Ecology and Management 114: 367-376. Flecker, A.S. (1984) The effects of predation and detritus on the structure of a stream insect community: a field test. Oecologia 64: 300-305. Flecker, A.S. and B. Feifarek (1994) Disturbance and the temporal variability of invertebrate assemblages in two Andean streams. Freshwater Biology 31: 131-142. Francoeur, S.N., B.J.F. Biggs, and R.L. Lowe (1998) Microform bed clusters as refugia for periphyton in a flood-prone headwater stream. New Zealand Journal of Marine and Freshwater Research 32: 363-374. 79 Frissell, C A . (1997) Ecological principles. Pp. 96-115 in J.E. Williams, C A . Wood, and M.P. Dombeck, editors. Watershed Restoration: Principles and Practices. American Fisheries Society, Bethesda, Maryland, USA. Frissell, C A . and R.K. Nawa (1992) Incidence and causes of physical failure of artificial habitat structures in streams of western Oregon and Washington. North American Journal of Fisheries Management 12: 182-197. Giller, P.S., N . Sangpradub, and H . Twomey (1991) Catastrophic flooding and macroinvertebrate community structure. Verhandlungen. Internationale Vereinigung fur Theoretische und Angewandte Limnologie/Proceedings. International. 24: 1724-1729. Gore, J.A. and R.D. Judy, Jr. (1981) Predictive models of benthic macroinvertebrate density for use in instream flow studies and regulated flow management. Canadian Journal of Fisheries and Aquatic Sciences 38: 1363-1370. Gore, J.A., D.J. Crawford, and D.S. Addison (1998) An analysis of artificial riffles and enhancement of benthic community diversity by physical habitat simulation (PHABSIM) and direct observation. Regulated Rivers: Research & Management 14: 69-77. Grubaugh, J.W., J.B. Wallace, and E.S. Houston (1997) Production of benthic macroinvertebrate communities along a southern Appalachian river continuum. Freshwater Biology 37: 581-596. Gurnell, A . M . , K.J . Gregory, and G.E. Petts (1995) The role of coarse organic woody debris in forest aquatic habitat: ecological and management aspects of a field experiment. Aquatic Conservation: Marine and Freshwater Ecosystems 5: 143-166. Haapala, A. and T. Muotka. (1998) Seasonal dynamics of detritus and associated macroinvertebrates in a channelized boreal stream. Archive fur Hydrobiologia 142: 171-189. Harvey, J.W. and B.J. Wagner (2000) Quantifying hydrologic interactions between streams and their subsurface hyporheic zones. Pp. 3-44 in J.B. Jones and P.J. Mulholland, editors. Streams and Ground Waters. Academic Press, SanDiego, California, USA. Hart, D.D. and C M . Finelli (1999) Physical-biological coupling in streams: the pervasive effects of flow on benthic organisms. Annual Review of Ecology and Systematics 30: 363-395. Hawkins, C P . and J.R. Sedell (1981) Longitudinal and seasonal changes in functional organization of macroinvertebrate communities in four Oregon streams. Ecology 62: 387-397. Heede, B.H. (1985) Channel adjustments to the removal of log steps: an experiment in a mountain stream. Environmental Management 9: 427-432. Hilderbrand, R.H., A .D . Lemly, C A . Dolloff, and K . L . Harpster (1997) Effects of large woody debris placement on stream channels and benthic macroinvertebrates. Canadian Journal of Fisheries and Aquatic Sciences 54: 931-939. 80 Hildrew, A .G . and P.S. Giller (1992) Patchiness, species interactions and disturbance in the stream benthos. Pp. 21-62 in P.S. Giller, A . G . Hildrew, and D.G. Raffaelli, editors. Aquatic Ecology: Scale, Pattern and Process. Blackwell Science, Cambridge, U K . Holomuzki, J.R. and B.J. Biggs (2000) Taxon-specific responses to high-flow disturbance in streams: implications for population persistence. Journal of the North American Benthological Society 19: 670-679. House, R. (1996) An evaluation of stream restoration structures in a coastal Oregon stream, 1981-1993. North American Journal of Fisheries Management 16: 272-281. House, R.A. and P.L. Boehne (1986) Effects of instream structures on salmonid habitat and populations in Tobe Creek, Oregon. North American Journal of Fisheries Management 6: 38-46. Huggins, D.G. and R.E. Moss (1974) Fish population structure in altered and unaltered areas of a small Kansas stream. Transactions of the Kansas Academy of Science 77: 18-30. Hurlbert, S.H. (1984) Pseudoreplication and the design of ecological field experiments. Ecological Monographs 54: 187-211. Huryn, A.D. and J.B. Wallace (1987) Local geomorphology as a determinant of macrofaunal production in a mountain stream. Ecology 68: 1932-1942. Hynes, H .B.N (1970) The ecology of running waters. University of Toronto Press, Toronto. Irvine, J.R. (1985) Effects of successive flow perturbations on stream invertebrates. Canadian Journal of Fisheries and Aquatic Sciences 42: 1922-1927. Kauffman, J.B., R.L. Beschta, N . Otting, and D. Lytjen (1997) An ecological perspective of riparian and stream restoration in the western United States. Fisheries 22: 12-24. Kellar, E.A. and F.J. Swanson (1979) Effects of large organic material on channel form and fluvial processes. Earth Surface Processes and Landforms 4: 361-380. Kiffney, P .M. , J.S. Richardson, and M.C . Feller (2000) Fluvial and epilithic organic matter dynamics in headwater streams of southwestern British Columbia, Canada. Archive fur Hydrobiologia 149: 109-129. King, J .M., M-P. Henshall-Howard, J.A. Day, and B.R. Davies (1987) Leaf-pack dynamics in a southern African mountain stream. Freshwater Biology 18: 325-340. Kondolf, G . M . (2000) Some suggested guidelines for geomorphic aspects of anadromous salmonid habitat restoration. Restoration Ecology 8: 48-56. Krebs, C.J. (1998) Ecological methodology, Benjamin/Cummings, Menlo Park, CA. 81 Lake, P.S. (2000) Disturbance, patchiness, and diversity in streams. Journal of the North American Benthological Society 19: 573-592. Lamberti, G.A., S.V. Gregory, L.R. Ashkenas, R.C. Wildman, and K.M.S . Moore (1991) Stream ecosystem recovery following a catastrophic debris flow. Canadian Journal of Fisheries and Aquatic Sciences 48: 196-208. Lancaster, J. (1999) Small-scale movements of lotic macroinvertebrates with variations in flow. Freshwater Biology 41: 605-619. Lancaster, J. (2000) Geomorphic scaling of microhabitat patches and their efficacy as refugia during disturbance. Journal of Animal Ecology 69: 442-457. Lancaster, J. and A . G . Hildrew (1993a) Flow refugia and the microdistribution of lotic macroinvertebrates. Journal of the North American Benthological Society 12: 385-393. Lancaster, J. and A . G . Hildrew (1993&) Characterizing in-stream flow refugia. Canadian Journal of Fisheries and Aquatic Sciences 50: 1663-1675. Lancaster, J., A . G . Hildrew, and C. Gjerlov (1996) Invertebrate drift and longitudinal transport processes in streams. Canadian Journal of Fisheries and Aquatic Sciences 53: 572-582. Lancaster, J., A . G . Hildrew, and C R . Townsend (1990) Stream flow and predation effects on the spatial dynamics of benthic invertebrates, Hydrobiologia 203: 177-190. Lemly, A .D . and R.H. Hilderbrand (2000) Influence of large woody debris on stream insect communities and benthic detritus. Hydrobiologia 421: 179-185. Leopold, L .B . , M . G . Wolman, and J.P. Miller (1964) Fluvial processes in geomorphology. W.H. Freeman & Company, San Francisco. Mackay, R.J. and J. Kalff (1969) Seasonal variation in standing crop and species diversity of insect communities in a small Quebec stream. Ecology 50: 101-109. Malmqvist, B. , L . M . Nilsson, and B.S. Svensson (1978) Dynamics of detritus in a small stream in southern Sweden and its influence on the distribution of the bottom animal communities. Oikos 31: 3-16. Maridet, L. , J.G. Wasson, M . Philippe, and C. Moros (1995) Benthic organic matter dynamics in three streams: riparian vegetation or bed morphology control? Archive fur Hydrobiologia 132: 415-425. Massong, T.M. and D.R. Montgomery (2000) Influence of sediment supply, lithology, and wood debris on the distribution of bedrock and alluvial channels. Geological Society of America Bulletin 112: 591-599. 82 Matthaei, C D . and C R . Townsend (2000) Inundated floodplain gravels in a stream with an unstable bed: temporary shelter or true invertebrate refugium? New Zealand Journal of Marine and Freshwater Research 34: 147-156. Matthaei, C D . , C J . Arbuckle, and C R . Townsend (2000) Stable surface stones as refugia for invertebrates during disturbance in a New Zealand stream. Journal of the North American Benthological Society 19: 82-93. Matthaei, C D . , K . A . Peacock, and C R . Townsend (1999) Patchy surface stone movement during disturbance in a New Zealand stream and its potential significance for the fauna. Limnology and Oceanography 44: 1091-1102. Matthaei, C D . , D. Werthmiiller, and A. Frutiger (1996&) Invertebrate recovery from a bed-moving spate: the role of drift versus movements inside or over the substratum. Archive fur Hydrobiologia 140: 221-235. Matthaei, C D . , U . Uehlinger, E.I. Meyer, and A. Frutiger (1996a) Recolonization by benthic invertebrates after experimental disturbance in a Swiss prealpine river. Freshwater Biology 35: 233-248. Merritt, R.W. and K.W. Cummins (1996) An introduction to the aquatic insects of North America. Kendall/Hunt Publishing Company, Dubuque, Iowa. Minshall, G.W. (1968) Community dynamics of the benthic fauna in a woodland springbrook. Hydrobiologia 32: 305-339. Mosley, M.P. (1981) The influence of organic debris on channel morphology and bedload transport in a New Zealand forest stream. Earth Surface Processes and Landforms 6: 571-579. Moyle, P.B. (1976) Some effects of channelization on the fishes and invertebrates of Rush Creek, Modoc County, California. California Fish and Game 62: 179-186. Murphy, M . L . and K . V . Koski (1989) Input and depletion of woody debris in Alaska streams and implications for streamside management. North American Journal of Fisheries Management 9: 427-436. Myers, R.K. and D.H. van Lear (1998) Hurricane-fire interactions in coastal forests of the south: a review and hypothesis. Forest Ecology and Management 103: 265-276. Naiman, J.R., R.E. Bilby, and P.A. Bisson (2000) Riparian ecology and management in the Pacific coastal rain forest. Bioscience 50: 996-1011. Nakamura, F. and F. Swanson (1993) Effects of coarse woody debris on morphology and sediment storage of a mountain stream system in western Oregon. Earth Surface Processes and Landforms 18: 43-61. O'Connor, N .A. (1991) The effects of habitat complexity on the macroinvertebrates colonizing wood substrates in a lowland stream. Oecologia 85: 504-512. 83 Osborne, L .L . , P.B. Bayley, L.W.G. Higler, B. Statzner, F. Triska, and T .M. Iversen (1993) Restoration of lowland streams: an introduction. Freshwater Biology 29: 187-194. Palmer, M.A . , A .E . Bely, and K . E . Berg (1992) Response of invertebrates to lotic disturbance: a test of the hyporheic refuge hypothesis. Oecologia 89: 182-194. Palmer, M.A. , P. Arensburger, A.P. Martin, and D.W. Denman (1996) Disturbance and patch-specific responses: the interactive effects of woody debris and floods on lotic invertebrates. Oecologia 105: 247-257. Palmer, M.A. , P. Arensburger, P.S. Botts, C C . Hakenkamp, and J.W. Reid (1995) Disturbance and the community structure of stream invertebrates: patch-specific effects and the role of refugia. Freshwater Biology 34: 343-356. Pearsons, T.D., H.W. L i , and G.A. Lamberti (1992) Influence of habitat complexity on resistance to flooding and resilience of stream fish assemblages. Transactions of the American Fisheries Society 121: 427-436. Peckarsky, B .L . (1980) Influence of detritus upon colonization of stream invertebrates. Canadian Journal of Fisheries and Aquatic Sciences 37: 957-963. Perry, S.A. and W.B. Perry (1986) Effects of experimental flow regulation on invertebrate drift and stranding in the Flathead and Kootenai Rivers, Montana, USA. Hydrobiologia 134: 171-182. Poff, N .L . and J.V. Ward (1989) Implications of streamflow variability and predictability for lotic community structure: a regional analysis of streamflow patterns. Canadian Journal of Fisheries and Aquatic Sciences 46: 1805-1818. Poff, N .L . and J.V. Ward (1991) Drift responses of benthic invertebrates to experimental streamflow variation in a hydrologically stable stream. Canadian Journal of Fisheries and Aquatic Sciences 48: 1926-1936. Pretty, J.L. and M . Dobson (2000) A possible strategy for enhancing the stream biota in areas of intensive commercial forestry. Verhandlungen. Internationale Vereinigung fur Theoretische und Angewandte Limnologie/Proceedings. International. 27: ???. Pringle, C M . , R.J. Naiman, G. Bretschko, J.R. Karr, M.W. Oswood, J.R. Webster, R.L. Welcomme, and M.J . Winterbourn (1988) Patch dynamics in lotic streams: the stream as a mosaic. Journal of the North American Benthological Society 7: 503-524. Prochazka, K., B .A. Stewart, and B.R. Davies (1991) Leaf litter retention and its implications for shredder distribution in two headwater streams. Archive fur Hydrobioligia 120: 315-325. Rabeni, C F . and R.B. Jacobson (1993) The importance of fluvial hydraulics to fish-habitat restoration in low-gradient alluvial streams. Freshwater Biology 29: 211-220. 84 Reece, P.F. and J.S. Richardson (2000) Benthic macroinvertebrate assemblages of coastal and continental streams and large rivers of southwestern British Columbia, Canada. Hydrobiologia 439: 77-89. Reeves, G.H., J.D. Hall, T.D. Roelofs, T.L. Hickman, and C O . Baker (1991) Rehabilitating and modifying stream habitats. Pp. 519-557 in W. R. Meehan, editor. Influences of Forest and Rangeland Management on Salmonid Fishes and Their Habitats. American Fisheries Publication, Bethesda, Maryland, USA. Rempel, L .L . , J.S. Richardson, and M.C. Healey (1999) Flow refugia for benthic macroinvertebrates during flooding of a large river. Journal of the North American Benthological Society 18: 34-48. Resh, V . H . (1977) Habitat and substrate influences on population and production dynamics of stream caddisfly, Ceraclea ancylus (Leptoceridae). Freshwater Biology 7: 261-277. Resh, V . H . , A . V . Brown, A.P. Coyich, M.E . Gurtz, H.W. L i , G.W. Minshall, S.R. Reice, A . L . Sheldon, J.B. Wallace, and R.C. Wissmar (1988) The role of disturbance in stream ecology. Journal of the North American Benthological Society 7: 433-455. Rhodes, H. and W.A. Hubert (1991) Submerged undercut banks as macroinvertebrate habitat in a subalpine meadow stream. Hydrobiologia 213: 149-153. Richardson, J.S. (1991) Seasonal food limitation of detritivores in a montane stream: an experimental test. Ecology 72: 873-887. Richardson, J.S. (1992a) Food, microhabitat, or both? Macroinvertebrate use of leaf accumulations in a montane stream. Freshwater Biology 27: 169-176. Richardson, J.S. (19926) Coarse particulate detritus dynamics in small, montane streams of southwestern British Columbia. Canadian Journal of Fisheries and Aquatic Sciences 49: 337-346. Richardson, J.S. and S.G. Hinch (1998) Ecological objectives for stream and watershed restoration along the Pacific Coast of North America. Pp. 47-56 in Proceedings of The International Workshop on Environmental Hydrodynamics and Ecological Restoration in Cold Regions. STNTEF, Norway. Richmond, A.D. and K.D. Fausch (1995) Characteristics and function of large woody debris in subalpine Rocky Mountain streams in northern Colorado. Canadian Journal of Fisheries and Aquatic Sciences 52: 1789-1802. Riley, S .C and K.D. Fausch (1995) Trout population response to habitat enhancement in six northern Colorado streams. Canadian Journal of Fisheries and Aquatic Sciences 52: 34-53. Robertson, A . L . , J. Lancaster, and A.G . Hildrew (1995) Stream hydraulics and the distribution of microcrustacea: a role for refugia. Freshwater Biology 33: 469-484. 85 Roper, B.B. , J.J. Dose, and J.E. Williams (1997) Stream restoration: is fisheries biology enough? Fisheries 22: 6-11. Roper, B.B. , D. Konnoff, D. Heller, and K. Wieman (1998) Durability of Pacific Northwest instream structures following floods. North American Journal of Fisheries Management 18: 686-693. Rounick, J.S. and M.J . Winterbourn (1983) Leaf processing in two contrasting beech forest streams: effects of physical and biotic factors on leaf breakdown. Archive fur Hydrobiologia 96: 448-474. SAS (1994) S AS/ST AT user's guide. Release 6.04 edition. SAS institute, Cary, North Carolina. Schmetterling, D.A. and R.W. Pierce (1999) Success of instream habitat structures after a 50-year flood in Gold Creek, Montana. Restoration Ecology 7: 369-375. Scrimgeour, G.J. and M.J . Winterbourn (1989) Effects of floods on epilithon and benthic macroinvertebrate populations in an unstable New Zealand river. Hydrobiologia 171: 33-44. Sedell, J.R., G.H. Reeves, F.R. Hauer, J.A. Stanford, and C P . Hawkins (1990) Role of refugia in recovery from disturbances: modern fragmented and disconnected river systems. Environmental Management 14: 711-724. Slaney, P.A. and A.D. Martin (1997) The watershed restoration program of British Columbia: accelerating natural recovery processes. Water Quality Resources Journal of Canada 32: 325-346. Slaney P.A., R.J. Finnigan, and R.G. Millar (1997) Acceleration the recovery of log-jam habitats: large woody debris-boulder complexes. Chapter 9 in Fish Habitat Rehabilitation Procedures. Ministry of Environment, Lands and Parks and Ministry of Forests, BC, Canada. Smith, R.D., R.C. Sidle, and P.E. Porter (1993a) Effects on bedload transport of experimental removal of woody debris from a forest gravel-bed stream. Earth Surface Processes and Landforms 18: 455-468. Smith, R.D., R.C. Sidle, P.E. Porter, and J.R. Noel (19936) Effects of experimental removal of woody debris on the channel morphology of a forest, gravel-bed stream. Journal of Hydrology 152: 153-178. Smock, L .A. , G . M . Metzler, and J.E. Gladden (1989) Role of debris dams in the structure and functioning of low-gradient headwater streams. Ecology 70: 764-775. Snaddon, C D . , B.A. Stewart, and B.R. Davies (1992) The effect of discharge on leaf retention in two headwater streams. Archive fur Hydrobiologia 125: 109-120. Sousa, W.P. (1984) The role of disturbance in natural communities. Annual Reviews of Ecology and Systematics 15: 353-391. 86 Southwood, T.R.E. (1988) Tactics, strategies and templets. Oikos 52: 3-18. Speaker, R., K. Moore, and S.V. Gregory (1984) Analysis of the process of retention of organic matter in stream ecosystems. Verhandlungen. Internationale Vereinigung fur Theoretische und Angewandte Limnologie/Proceedings. International. 22: 1835-1841. Stanford, J.A. and J.V. Ward (1983) Insect species diversity as a function of environmental variability and disturbance in streams. Pp. 265-278 in J.R. Barnes and G.W. Minshall, editors. Stream Ecology: Application and Testing of General Ecological Theory. Plenum Press, New York, Y N , USA. Statzner, B., J.A. Gore, and V. Resh (1988) Hydraulic stream ecology: observed patterns and potential applications. Journal of the North American Benthological Society 7: 307-360. Stewart, K.W. and B.P. Stark (1993) Nymphs of North America stonefly genera (Plecoptera). University of North Texas Press, Denton, Texas. Swales, S. and K. O'Hara (1980) Instream habitat improvement devices and their use in freshwater fisheries management. Journal of Environmental Management 10: 167-179. Tonn, W.M. and J.J. Magnuson (1982) Patterns in the species composition and richness of fish assemblages in northern Wisconsin lakes. Ecology 63: 1149-1166. Townsend, C.R.(1989) The patch dynamics concept of stream community ecology. Journal of the North American Benthological Society 8: 36-50. Townsend, C.R. and A . G . Hildrew (1976) Field experiments on the drifting, colonization and continuous redistribution of stream benthos. Journal of Animal Ecology 45: 759-772. Townsend, C.R. and M.R. Scarsbrook (1997) The intermediate disturbance hypothesis, refugia, and biodiversity in streams. Limnology and Oceanography 42: 938-949. Townsend, C.R., M.R. Scarsbrook, and S. Doledec (1997) Quantifying disturbance in streams: alternative measures of disturbance in relation to macroinvertebrate species traits and species richness. Journal of the North American Benthological Society 16: 531-544. Trotter, E.H. (1990) Woody debris, forest-stream succession, and catchment geomorphology. Journal of the North American Benthological Society 9: 141-156. Vannote, R.L., G.W. Minshall, K.W. Cummins, J.R. Sedell, and C E . Cushing (1980) The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences 37: 130-137. Walde, S.J. (1986) Effect of an abiotic disturbance on a lotic predator-prey interaction. Oecologia 69: 243-247. Wallace, J.B. (1990) Recovery of lotic macroinvertebrate communities from disturbance. Environmental Management 14: 605-620. 87 Wallace, J.B., J.R. Webster, and J.L. Meyer (1995a) Influence of log additions on physical and biotic characteristics of a mountain stream. Canadian Journal of Fisheries and Aquatic Sciences 52: 2120-2137. Wallace, J.B., S.L. Eggert, J.L. Meyer, and J.R. Webster (1999) Effects of resource limitation on a detrital-based ecosystem. Ecological Monographs 69: 409-442. Wallace, J.B., M.R. Whiles, S. Eggert, T.F. Cuffney, G.J. Lugthart, and K. Chung (19956) Long-term dynamics of coarse particulate organic mater in three Appalachian Mountain streams. Journal of the North American Benthological Society 14: 217-232. Ward, B.R. (1997) Acceleration the recovery of log-jam habitats: large woody debris-boulder complexes. Chapter 10 in Fish Habitat Rehabilitation Procedures. Ministry of Environment, Lands and Parks and Ministry of Forests, BC, Canada. Waters, T.F. (1972) The drift of stream insects. Annual Review of Entomology 17: 253-272. Webster, J.R., A.P. Covich, J.L. Tank, and T V . Crockett (1994) Retention of coarse organic particles in streams in the southern Appalachian Mountains. Journal of the North American Benthological Society 13: 140-150. Webster, J.R., E.F. Benfield, S.W. Golladay, B.H. Hil l , L .E . Hornick, R.F. Kazmierczak, Jr., and W.B Perry. (1987) Experimental studies of physical factors affecting seston transport in streams. Limnology and Oceanography 32: 848-863. Weir, J.M.H., E.A. Johnson, and K. Miyanishi (2000) Fire frequency and the spatial age mosaic of the mixed-wood boreal forest in western Canada. Ecological Applications 10: 1162-1177. Wentworth, C K . (1922) A scale of grade and class terms for clastic sediments. Journal of Geology 377-392. White, P.S. and S.T.A. Pickett (1985) Natural disturbance and patch dynamics: an introduction. Pp. 3-13 in S.T.A. Pickett and P.S. White, editors. The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, London, U K . Wiggins, G.B. (1996) Larvae of the North American caddisfly genera (Trichoptera). University of Toronto Press, Toronto. Williams, D.D. and J.H. Mundie (1978) Substrate size selection by stream invertebrates and the influence of sand. Limnology and Oceanography 23: 1030-1033. Williams, D.D. and M.R. Smith (1996) Colonization dynamics of river benthos in response to local changes in bed characteristics. Freshwater Biology 36: 237-248. Winterbottom, J.H., S.E. Orton, and A . G . Hildrew (19976) Field experiments on the mobility of benthic invertebrates in a southern English stream. Freshwater Biology 38: 37-47. 88 Winterbottom, J.H., S.E. Orton, A .G . Hildrew, and J. Lancaster (1997a) Field experiments on flow refugia in streams. Freshwater Biology 37: 569-580. Wissmar, R.C. and R.L. Beschta (1998) Restoration and management of riparian ecosystems: a catchment perspective. Freshwater Biology 40: 571-585. Wohl, D.L., J.B. Wallace, and J.L. Meyer (1995) Benthic macroinvertebrate community structure, function and production with respect to habitat type, reach and drainage basin in the southern Appalachians (U.S.A.). Freshwater Biology 34: 447-464. Yount, J.D. and G.J. Niemi (1990) Recovery of lotic communities and ecosystems from disturbance-a narrative review of case studies. Environmental Management 14: 547-569. Zar, J.H. (1999) Biostatistical analysis. Fourth edition. Prentice Hall, Upper Saddle River, NJ, USA. 89 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0090138/manifest

Comment

Related Items