UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Riparian harvesting and resource limitation in small B.C. interior streams Melody, Katherine Jill 2000

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

Item Metadata

Download

Media
831-ubc_2000-0487.pdf [ 7.56MB ]
Metadata
JSON: 831-1.0089578.json
JSON-LD: 831-1.0089578-ld.json
RDF/XML (Pretty): 831-1.0089578-rdf.xml
RDF/JSON: 831-1.0089578-rdf.json
Turtle: 831-1.0089578-turtle.txt
N-Triples: 831-1.0089578-rdf-ntriples.txt
Original Record: 831-1.0089578-source.json
Full Text
831-1.0089578-fulltext.txt
Citation
831-1.0089578.ris

Full Text

RIPARIAN HARVESTING AND RESOURCE LIMITATION IN SMALL B.C. INTERIOR STREAMS by KATHERINE JILL MELODY B.Sc, University of British Columbia, 1995 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF T H E REQUIREMENTS FOR T H E DEGREE OF MASTER OF SCIENCE in THE F A C U L T Y OF GRADUATE STUDIES Faculty of Forestry Department of Forest Sciences  We accept this thesis as conforming to the required standard  T H E UNIVERSITY OF BRITISH COLUMBIA July 2000 © Katherine Jill Melody, 2000  In  presenting this  degree  at the  thesis in  University of  partial  fulfilment  of  of  department  this thesis for or  by  his  or  scholarly purposes may be her  representatives.  permission.  of  fh ,^+ A  \Sp/eJ^fifi&  The University of British Columbia Vancouver, Canada  DE-6 (2/88)  for  an advanced  Library shall make it  agree that permission for extensive  It  publication of this thesis for financial gain shall not  Department  requirements  British Columbia, I agree that the  freely available for reference and study. I further copying  the  is  granted  by the  understood  that  head  of  copying  my or  be allowed without my written  ABSTRACT  Leaf litter inputs and shading of small streams are altered by forest harvesting. Many of the effects of riparian modification are known for streams in coastal British Columbia, but little is known of how streams in the drier, continental areas of B C respond to logging. The main objectives of this study were to determine how the harvest of the riparian forest affects stream food resources (leaf litter and algae) and macroinvertebrate composition in small, continental streams of BC. I conducted a paired stream-reach study of one harvested and two upstream reaches in each of five streams and measured periphyton, detritus, macroinvertebrate abundance and biomass and many physical measures in both July and October of 1997. In general, harvested stream sections tended to be wider and contained more riffle areas than the upstream forested sections. Leaf litter and algae were higher in some streams and lower in other streams in the harvested stream sections but significantly greater quantities of the food resources were found in the fall season in all streams. Greater macroinvertebrate density and biomass were found in the fall and the shredder and gatherer functional feeding groups were largely responsible for this season increase. The magnitude and direction of differences seen between sites, streams or seasons were stream dependent. To test the hypothesis that changes in light and litter inputs would affect the benthic community in these small streams, as occurred following riparian forest harvesting, I studied the impacts on periphyton and macroinvertebrate abundance and biomass in artificial stream channels. In this experiment I used a 2x2 factorial design with light (shaded or full light) and leaf litter inputs (forested input rate or one quarter that rate to represent the harvested input rate) as factors. A qualitative difference was found in the periphyton, as longer algal filaments occurred in the high light treatments, but there were no quantitative algal responses in chlorophyll a or ash-free dry mass, suggesting that light is not limiting in this system. The leaf litter treatments resulted in higher densities of two macroinvertebrate shredders, Limnephilus sp. and Podmosta sp., in the high leaf litter treatments. This suggests that these shredders may be food limited. This experiment provides evidence that changes to shading and leaf inputs to small streams affect the benthos and may limit secondary production.  ii  T A B L E OF CONTENTS  ABSTRACT  ii  LIST OF T A B L E S  v  LIST OF FIGURES  vi  ACKNOWLEDGEMENTS  viii  G E N E R A L INTRODUCTION  1  C H A P T E R O N E : Responses of algae and invertebrates of small streams to experimental manipulation of leaf litter inputs and shading  2  Introduction  2  Materials and Methods  5  Study Site  5  Experimental Design And Sampling Methods  5  Analytical Methods Nutrients and Dissolved Organic Carbon Periphyton Macro invertebrates and Detritus  9 9 9 10  Data Analysis  11  Results  12 Water Chemistry and Temperature  12  Resource Treatments  13  Invertebrate Responses To Treatments Benthos  13 13  Drifting Invertebrates  19  Discussion  21  Invertebrate Responses to Treatments Detritivores (Shredders and Collectors) Other Functional Groups Community Response iii  21 21 23 24  Effectiveness of Shade Treatments  25  Management Implications  28  C H A P T E R TWO: Responses of macroinvertebrates to changes in stream food resources following clearcut logging on small streams in the interior of British Columbia 29 Introduction  29  Materials and Methods Study Sites and Study Design Field Sampling Methods Analytical Methods Periphyton Macroinvertebrates and Detritus Nutrients Statistical Analysis  32 32 35 36 36 36 37 37  Results  39 39 41 50 50 53 55  Physical Data Periphyton and Detritus Macroinvertebrates Functional Feeding Groups Individual Taxa Macroinvertebrate Communities Discussion The Physical System Periphyton Detritus Macroinvertebrates Functional Feeding Groups Management Implications  60 60 61 63 65 66 70  GENERAL CONCLUSION  72  L I T E R A T U R E CITED  74  A P P E N D I X I: Weighted Euclidean distance calculations  83  iv  LIST OF TABLES  Table  Page  1. Characteristics of the five study streams  34  2. Physical measurements from the five study streams  40  V  LIST OF FIGURES  Figure  Page  Chapter One 1. Schematic representation of the mesocosm design  8  2. Pre- and post-treatment organic matter biomass  14  3. Pre-and post-treatment periphyton parameters  15  4. Pre-and post-treatment macroinvertebrate densities and biomass  17  5. Correlation biplot based on the redundancy analysis of the species data from the treatment troughs  18  6. Net and per capita emigration rates  20  Chapter Two 1. Photosynthetically active radiation at the five streams in both harvested and forested stream sections  42  2. Stream discharge for the five streams in summer and fall  42  3. Periphyton parameters from the five streams for all seasons and sites  43  4. Periphyton parameters for each stream across seasons  45  5. Seasonal detrital biomass  46  6. Moss biomass found in the five streams across sites  47  7. Stream*site interactions for needle and fruit & flower biomass in each season  48  8. Stream*site interactions for wood and total detrital biomass in each season  49  9. Functional feeding group density and biomass across streams and seasons.. 51 10. Predator density across seasons and sites  vi  52  11. Individual taxa density and biomass across seasons  54  12. Individual taxa density and biomass across streams  56  13. Correlation biplot based on redundancy analysis of the species data from the Five streams in each season 57 14. Ratio of the weighted Euclidean distance between sites  vii  59  ACKNOWLEDGEMENTS  I would sincerely like to thank my supervisor Dr. John Richardson for his patience and support shown throughout my academic experience. I would also like to thank my committee, Dr. Scott Hinch, Dr. Judy Myers, and Dr. Peter Kiffney for their constructive input and valuable comments. A l l of my comrades in the Richardson lab were always encouraging and helpful and I'm grateful to all. Special thanks to my field assistants, Dierdre Ramsay and Dan O'Donahue, who kept a sense of humour and helped me maintain my sanity. I would also like to thank my family (including my dog Shadow) and especially my husband, Dave, for his support and understanding of this challenging endeavor. Finally, I would like to thank Forest Renewal of British Columbia for funding this project.  viii  GENERAL INTRODUCTION  A major determinant of the structure and function of small stream communities is the relative importance of energy from allochthonous (terrestrial organic matter inputs) and autochthonous (instream primary productivity) energy sources (Cummins 1974, Anderson and Sedell 1979). For small streams, forested riparian vegetation is believed to reduce autotrophic production (within-stream production) by shading resulting in a largely heterotrophic system (outside energy inputs to a stream) dominated by leaf litter inputs (Cummins 1974; Anderson and Sedell 1979; Vannote et al. 1980; Gregory et al. 1991). Harvesting of the riparian forest can reverse the dominant energy source from leaf litter inputs to primary production, which is believed to occur due to an increase in light reaching the stream (Hansmann and Phinney 1973, Wallace and Gurtz 1986, Bilby and Bisson 1992).  To date, most stream research has been concentrated in coastal areas and riparian management guidelines have been based on this research (e.g. Hartmann and Scrivener 1990). Streams in the interior of the province may react differently from those on the coast as a result of differing climate, soil types, forest cover and a snowmelt dominated runoff regime (Richardson 1994, Heise 1997). Therefore, there is a need to determine the effects of riparian forest harvesting on streams in the dry, continental interior. I conducted a field study to determine how riparian forest harvesting affects stream food resources and macroinvertebrates in the Horsefly Forest District in central BC. A n experiment manipulating both leaf litter and light levels was also conducted to isolate a community response from other possible factors that could change community structure. 1  CHAPTER 1: RESPONSES OF A L G A E AND INVERTEBRATES OF S M A L L STREAMS TO EXPERIMENTAL MANIPULATION OF L E A F LITTER INPUTS AND SHADING  INTRODUCTION  The river continuum concept hypothesizes that species and their communities will exploit their environment as efficiently as possible resulting in a composite species assemblage that will optimize energy use (Vannote et al. 1980). This would result in a predictable community dependent on the dominant energy sources, i.e., detritus and periphyton, at certain times of the year. In small, forested streams, leaf litter and other detritus fall into the stream and comprise the dominant energy source for stream food webs, especially macroinvertebrates (Cummins 1974, Anderson and Sedell 1979, Cummins et al. 1989).  Stream algae attached to a substrate is labeled periphyton or Aufwuchs and includes benthic algae, bacteria, associated fine detritus and some species of microscopic invertebrates (Lamberti and Moore 1984). Periphyton can be limited by abiotic resources, i.e., nutrients and light, or can be controlled by consumption by herbivores (Rosemond et al. 1993). Light quantity (photon density) is thought to be one of the main physical factors affecting periphyton communities (Fuller 1986, Raven 1992). Stream sections with open canopies have been shown to support higher algal standing crops and primary production than more shaded sections (Lyford and Gregory 1975, Hawkins et al. 1982, Murphy 1984, Behmer and Hawkins 1986). In experimental tests, light has been shown to be the factor limiting primary productivity (Hill and Harvey 1990, Ulrich et. al 1993, K i m 1999), but in many small, forested streams there is a potential for periphyton to be limited by all three factors (Rosemond et al. 1993, Wellnitz et al. 1996£>).  2  Benthic macroinvertebrates are categorized into functional feeding groups based on feeding adaptations and invertebrate taxa that perform different functions within aquatic ecosystems are distinguished based on their method of processing nutritional resources (Merritt and Cummins 1996a). Each functional group can be disproportionately affected by shifts in energy resource abundance. Macroinvertebrates that feed on coarse particulate organic matter (CPOM, detritus particles >lmm in size) are labelled shredders. Collectors are those macroinvertebrates that feed on fine particulate organic matter (FPOM, detrital particles < 1 mm and > 0.45 um in size). Scrapers feed on periphyton and predators feed on prey, i.e. other invertebrates. Within the collector functional group, there exist two sub-groups: the filterers, which feed by filtering particles from the water column, and the gatherers, which feed by gathering particles from the stream bottom (Merritt and Cummins 1996a).  Many benthic macroinvertebrates have high reproductive rates, short generation times, long periods of recruitment and a strong probability that productivity will track the food supply (Richardson 1993). Any factor that can cause mortality or reduce the per capita rate of increase will limit populations (Sinclair and Pech 1996), assuming compensation does not occur. The above observations lead to the hypothesis that productivity of many stream invertebrates is limited by food supply (Richardson 1993). Using artificial, in-situ stream channels, Richardson (1991) experimentally showed that some shredder invertebrates are seasonally food limited in terms of food quantity and that food supplementation led to an increase in net colonization of the supplemented channels and a release from food limitation. Grazers may also be food limited as it has been found that sun exposed stream sections can have significantly higher invertebrate biomass relative to shaded stream sections, and that invertebrate production can be higher in the open stream section (Behmer and Hawkins 1986, Hill et al. 1995). If grazer biomass increases 3  with increases in algal biomass, this would suggest that grazer and algal growth may be tightly coupled and that grazer growth is food limited (Lamberti et al. 1995). Within the overall limits set by stream temperature, the growth rate and to some extent survivorship, of aquatic macroinvertebrates are controlled by food quality and quantity (Cummins 1974).  Leaf litter inputs and shading of small streams are altered by forest harvesting. The general effects of riparian modification are known for streams in coastal British Columbia, but little is known of how streams in the drier continental ("interior") areas of B C respond. In many coastal studies, it has been shown that an increase in light reaching a stream following harvesting can result in greater primary production (Hansmann and Phinney 1973, Murphy and Hall 1981, Murphy et al. 1981, Webster et al. 1983, Noel et al. 1986, Wallace and Gurtz 1986, Gregory et al. 1987, Bilby and Bisson 1992, Davies and Nelson 1994, Kim 1999). I predicted that a decrease in direct leaf litter input following harvesting and an increase in periphyton would affect the benthic community in these small streams.  To determine how stream macroinvertebrates respond to changes in light and litter inputs, and if they are food limited in B C interior streams, I conducted a food manipulation experiment using artificial streams, or mesocosms. Although a mesocosm is a simplification of the natural stream, it allows the mechanism of a community response to be isolated from the other possible factors that could cause a change in community structure, allows for replication and reduces variation (Mundie et al. 1991). The artificial streams give experimental units where the factors of interest, periphyton and detrital levels, can be controlled and replicated.  Detritus and light levels were manipulated in a completely randomized, two-by two factorial design. The two levels of detritus were selected to represent the expected input rates of forested 4  and harvested streams. Because the amount of light reaching a stream dramatically increases when it is harvested, light was selected as the dominant factor that may limit periphyton levels. Shade treatments were meant to represent conditions found in either forested stream sections (i.e. shaded) or harvested stream sections (i.e. in direct sunlight). The major objectives of this experiment were to determine if stream macroinvertebrates are food limited and the relative importance of different food resources to macroinvertebrate biomass, relative composition and density.  MATERIALS AND METHODS  Study Site The experiment was conducted in the interior of British Columbia in the Sub-boreal Spruce biogeoclimatic zone in the Cariboo Forest Region (Meidinger and Pojar 1991). Sixteen troughs, or mesocosms, were set up in parallel on the stream bank of Moffat Creek, which drains into the Horsefly River located in the Horsefly Forest District. Moffat Creek is a fourth order stream, 900m above sea level with a 1% slope. Discharge is dominated by snowmelt with the highest and most variable flows occurring in the spring and lower stable flows occurring in the winter (Rothacher 1970). Throughout the summer the stream is characterized by flashy discharge events due to short, but heavy rainfall during afternoon thunderstorms. This site was in full sunlight as the surrounding area consists mostly of rangeland.  Experimental Design and Sampling Methods The mesocosm consisted of sixteen troughs. Each trough was 7.5m long and 0.2m wide, and filled with gravel (2-3cm-diameter) and sand to a depth of 10 cm. The area of the bottom of each trough was 1.5 m and the average slope of the channels was 1%. A sill was placed at the 2  5  downstream end of each trough to maintain a water depth of at least 5cm. Water from the stream was diverted through 6 " P V C sewer pipe from 250m upstream of the channels into 2 settling boxes to prevent sediment from entering the streams. The water flowed from the settling boxes through small outflow valves into each stream where the velocity of the water was maintained at 0.5L/s. The water intake pipe had three aluminum grates over the opening to allow invertebrates in, but to keep out large debris and to prevent the access of fish. Ten benthic baskets, to sample macroinvertebrates, and ten unglazed ceramic tiles, to sample algae, were randomly placed in each trough. The benthic baskets consisted of perforated plastic 500ml tubs with a wire handle, which were embedded in the substrate of the troughs. The perforations allowed invertebrates to move between the substrate in the basket and the substrate in the trough. Stream macroinvertebrates would colonize the troughs by drifting through the pipes. To augment the troughs with invertebrates, three Surber samples taken from within Moffat Creek on 12 September 1997 were placed in each trough. The distances from the water inflow of each trough to the individual baskets and tiles were measured to determine if there was a location effect along the length of the trough.  The mesocosms were left for two weeks to allow for sufficient invertebrate and algal colonization and for background detrital materials to become established. Leaf Utter was collected from the forest floor from the dominant tree species found at the field study sites (see Chapter 2) which were trembling aspen (Populus tremuloides), black cottonwood {Populus trichocarpa), mountain alder (Alnus incana ssp. tenuifolid) and hybrid white spruce (Picea glauca x engelmannii). The day prior to the application of treatments, drift nets with a 250um mesh size were placed over the water outflow of each trough to measure the drift rates of invertebrates over a 24-h period. On 29 September 1997, four benthic baskets and 4 clay tiles were randomly removed from each trough to constitute the pre-treatment samples to be used for 6  comparison to the post-treatment samples. The clay tiles were placed in labelled plastic bags and placed in a freezer for later analysis. The benthic basket samples were preserved with buffered formaldehyde and stored.  The experiment began on 29 September 1997 and consisted of two factors giving four treatments: high detritus/shade, high detritus/no shade, low detritus/shade, low detritus/no shade. Each treatment was replicated four times (Figure 1). To shade the troughs, a 90% shade cloth (American Horticultural Supplies) was placed over 8 randomly selected channels. Detrital levels were manipulated by placing 12.5g/m of air-dried leaf litter in the 8 randomly selected low 2  detrital treatment troughs and 50 g/m in the high detrital treatment troughs. Detrital levels were 2  based on published rates of litterfall for the Sub-boreal Spruce biogeoclimatic zone for September/October which were 50g/m of dried leaf litter (286 gm" yr ) (Prescott et al. 1989). 2  2  _1  Thus, the high detrital treatments represent the natural leaf litter input rate of 50g/m and the low 2  detrital treatments were one fourth of the natural input in attempts to mimic a harvested stream reach. The proportion of each leafneedle type was based on the area covered by each tree species found at the field sites which resulted in 50% spruce needles, 22.5% aspen leaves, 22.5% alder leaves and 5% cottonwood leaves.  Four temperature loggers were used, one in each of the 2 settling tanks, one in a shaded trough and one in an open trough. The experiment was conducted for 30 days, with 24-hour drift samples taken once each week from each trough to measure invertebrate emigration. Invertebrate immigration was also measured by placing a drift net over the water inflow of an individual trough for 24 hours. This was repeated for all 16 troughs on different days throughout the experiment. After 30 days, on 28 October 1997, the shade cloth was removed and lengths of the algal filaments on the clay tiles were measured. Water samples were 7  I I I I I I I I  W H H H  Water outflow  0.2m  Figure 1. Schematic representation of the 2x2 factorial design of the mesocosm experiment. Eight channels were randomly chosen for the high detrital treatment represented by a leaf and eight channels were chosen for the shade treatment represented by the shaded troughs. Circles represent the benthic invertebrate basket samples and the squares represent the clay tiles used to sample periphyton.  8  taken from the outflow of each trough for determination of dissolved organic carbon (DOC). Water samples were also taken from Moffat Creek, adjacent to the experimental apparatus and frozen to determine ambient nutrient levels. The remaining 6 benthic baskets and 6 clay tiles from each trough were then removed. The clay tiles were placed in labelled plastic bags and frozen. The benthic samples were preserved in formaldehyde and stored.  Analytical Methods  Nutrients and Dissolved Organic Carbon Nitrogen and phosphorus concentrations were measured to the nearest 0.35 ug N 0 -N/L and 2.4 3  ug PO4-P/L respectively according to the methods of Wood et al (1967) for nitrogen and Hager et al. (1968) for phosphorus. The analysis was done on previously frozen water samples taken from Moffat Creek using a Technicon AutoAnalyzer II. To determine DOC concentrations, water samples were immediately filtered in the field through pre-ashed 47mm glass fiber filters (Whatman GF/F filters). Two methods were used: combustion-infrared method on Shimadzu Carbon Analyzer (Model TOC-500) and gas chromatography (Carlo Erba Elemental Analyzer Model 1106).  Periphyton The periphyton was scraped off the unglazed ceramic tiles with a razorblade and toothbrush into a bucket resulting in a slurry. The slurry was then thoroughly mixed and split into two equal parts and filtered onto pre-ashed 47mm glass fiber filters (Whatman GF/F filters), one for periphytic biomass analysis and the other for chlorophyll a analysis. The filters destined for periphytic biomass were oven dried (60°C for at least 24 h), weighed, ashed (500°C for at least 1 h) and reweighed to determine ash-free dry mass by weight loss on ignition. The chlorophyll a 9  samples were soaked in 90% acetone for 24 h and processed according to the extraction procedure outlined in Strickland and Parsons (1972) using a Turner fluorometer (model 10-005 R). The surface areas of all the tiles were measured to determine the periphytic measurements on a per area basis.  Macroinvertebrates and Detritus The benthic basket samples were washed through 3 nested sieves (4mm, 1mm and 500pm). Using a dissecting microscope at 6 and 10-x magnification, all of the animals were removed from each sieve fraction and identified, counted and digitized for animal biomass. The digitizer consisted of a drawing tube attached to the dissecting microscope and a digitizing tablet with a computer mouse, which was connected to a computer. The length of the animal was measured and converted to biomass using a computer program, Zoobbiom (Hopcroft, R. 1995. Zoobenthos Biomass Digitizing Program). A l l animals were assigned to functional feeding groups (scrapers, shredders, filterers, gatherers or predators) according to Merritt and Cummins (1996). The organic matter was retained from the benthic basket samples and used to determine organic matter standing crop. The detritus from the 4mm sieve was sorted into leaf or needle categories. The detritus from the 1mm sieve was classified as coarse particulate organic matter and the detritus from the 500pm sieve was called fine particulate organic matter. Organic matter was oven dried (60°C for at least 24 h), weighed, ashed (500 °C for at least 1 h) and reweighed to determine ash free dry mass (AFDM). The emigration and immigration samples were washed through the 500pm sieve and all of the animals were removed, identified, counted and digitized for animal biomass.  10  Data Analysis A l l analyses were done using the General Linear Models procedure for analysis of variance (ANOVA) designs in SAS version 6.12 (SAS 1996). Each data point was calculated by subtracting the pre-treatment value from the post-treatment value for each trough. To determine if there was a relationship between the response variable and the location of the sample within the trough an analysis of covariance (ANCOVA) was performed, with distance from the head of the trough as the covariate. As there was a relationship found between the response variable and the location within the trough a repeated-measures A N O V A was then used on the same data to account for the within-trough (i.e. within-subject) variation, using distance as the repeated measure. In all analyses, the Type III sums of squares were used and compared with an F statistic with 1,12 df. Although there were multiple statistical tests performed, there were no statistical adjustments made. Not only do Bonferroni adjustments decrease type I errors, they also increase the probability of type II errors, which are equally problematic (Perneger 1998).  The invertebrate immigration data were grouped into four time periods and averaged, with each time period corresponding to one of the four sampling dates for invertebrate emigration. Net emigration was calculated by subtracting the mean immigration from the first time period from the mean emigration measured in week 1. This was also done for weeks 2, 3 and 4. Per capita immigration and emigration were also determined by dividing the number of invertebrates immigrating, or emigrating, by the average number of invertebrates found in the total area of the troughs. This gives a better understanding of the influence immigration and emigration have on the standing stock of macroinvertebrates. A l l of the invertebrate drift measures were analyzed using repeated-measures A N O V A , with sampling date as the repeated measure. Again, the Type III sums of squares were used and compared with an F statistic with 1,12 df.  11  The algal filament lengths and the dissolved organic carbon (DOC) were not measured prior to the experiment. Therefore, this analysis included only the post-treatment data using repeated measures A N O V A for the algal filament lengths and a two-way A N O V A for the DOC. The assumption of normality was tested on the residuals of all of the data sets using the conservative Shapiro-Wilk statistic and where necessary, data were logio (x +1) transformed before analysis (Zar 1984). In all tests the critical level of significance was a = 0.05.  To determine how the taxonomic composition differed among the treatments, a direct multivariate approach was used. A detrended correspondence analysis (DCA) was conducted to determine if the data were linear or unimodal (ter Braak 1994). As the gradient lengths were less than 3 standard deviation units (length of gradient = 0.597), the relationship between the species and the treatments was linear. Therefore, a linear method, redundancy analysis (RDA) was conducted on the data as opposed to the unimodal canonical correspondence analysis (CCA) (Palmer 1993, ter Braak 1994). In the R D A analysis, the data were log-transformed, scaled by inter-sample distances and centred by species. The D C A and R D A were done using C A N O C O (Windows version 4.02) and the figure is a correlation biplot.  RESULTS  Water Chemistry and Temperature The mean concentrations of nitrogen and phosphorus at the mesocosm site were 1.12 pg NO3N / L and 25.6 pg PO4-P/L respectively. The dissolved organic carbon did not differ significantly among treatments (P > 0.55) and ranged from 4.5 - 9.8 mg/L. The mean daily temperatures taken from the four locations differed from each other by < 0.2°C throughout the experiment, but  12  steadily decreased. The shade cloth used was found to reduce light levels by 85-90% in a previous experiment (Kim 1999).  Resource Treatments To determine the effectiveness of the detrital additions, a repeated-measures A N O V A was performed on each detrital category (Figure 2). Not only were the added detrital components of leaves and needles significantly higher (F^n =14.80, P < 0.0023), but the remaining C P O M (excluding leaves and needles) and the total organic matter were also greater in the high detrital treatments ( F = 5 . 2 1 , P < 0.05, F U2  U 2  = 22.2, P < 0.002 respectively). Although the F P O M was  higher in the high detrital treatments (Figure 2), differences were not statistically significant (P > 0.1).  The manipulation of light did not result in significant differences in chlorophyll a{P> 0.34, Figure 3a) or periphytic A F D M (P > 0.16, Figure 3b). Thus, the desired resource treatment of high and low levels of periphyton was not achieved in terms of these periphyton measures. However, the algal filaments on the clay tiles prior to the final sampling (Figure 3c) were found to be significantly shorter in the low light treatments (i i =61.58, P < 0.0001). 7  il2  Invertebrate Responses to Treatments Benthos Total numbers, functional feeding groups and individual taxa were all examined for treatment effects in terms of density and biomass. The net difference (post-treatment values minus pretreatment values) of total invertebrate density or total invertebrate biomass among the four treatments did not differ significantly (both P > 0.15). Invertebrate densities were higher in all  13  Jj  200  :  £  150  H  j  :  ¥  :  to  100 H  E o ^  50  H -  i  il I HD NS  i LD NS  i  i1  LD S  HD S  HD NS  Treatment  LD NS  LD S  HD S  Treatment  CD  E o  "E E o  i  E E  o  O  CL  o  HD NS  LD NS  LD S  HD S  HD NS  Treatment  LD NS  LD S  HD S  Treatment  300  Pre-treatment Post-treatment  c<rE  ^2 OT  w  250 H 200  CO  E o  HD NS: LD NS: HD S: LD S:  'n "co 2o  high detritus, no shade low detritus, no shade high detritus, shade low detritus, shade  150  H  100  HD NS  LD NS  LD S  HD S  Treatment Figure 2.  Pre- and post-treatment organic matter biomass as a function of resource treatments C P O M = coarse particulate organic matter (1.0mm - 4.0mm); F P O M = fine particulate organic matter (500 um - 1.0mm). Bars represent means + 1 S . E . (n = 4)  14  a)  so  0.30  b)  I 0.25  60  CM  J  3 Q LL <  5. CL  IT  I  HD NS  LD NS  LD S  0.20  3 40  o  g.  I_  20  -  0.15  >.  .c  L l 11 HD NS  LD NS  LD S  Q. P O  0.10  0.05  0.00  HD S  Treatment  HD S  Treatment  Pre-treatment Post-treatment  c)  E o c  HD NS: high detritus, no shade LD NS: low detritus, no shade HD S: high detritus, shade LD S: low detritus, shade  c 0  E  HD NS  LD NS  LD S  HD S  Treatment  Figure 3 .  a) Pre- and post-treatment periphytic ash-free dry mass ( A F D M g/m ) b) Pre- and post-treatment chlorophyll a (pg/cm ) c) Algal filament lengths found on the clay tiles at the end of the experiment. Bars represent means + 1 S.E. (n = 4). 2  2  15  troughs following the experiment relative to the pre-treatment samples, but the opposite trend was found for invertebrate biomass. When the invertebrates were assigned to functional feeding groups, again there were no significant differences in densities (all P > 0.14) or biomass (all P > 0.09).  Although there was no response from the invertebrates when grouped together, specific taxa within the functional groups did show significant responses (Figure 4a). A group of chironomids, Diamesinae (collector-gatherers), showed higher densities in the high detritus treatments ( F  = 24.37, P < 0.0003). As shown in Figure 4a, densities were actually lower in  1 : 1 2  the low detrital treatment troughs than densities at the start of the experiment. The litter addition also resulted in higher densities of the shredders Limnephilus sp. and Podmosta sp. (Nemouridae) (F  h 1 2  = 6.55, P < 0.03, F = hn  8.06, P < 0.015). Nymphs of the chironomid sub-  family, Orthocladiinae, were the only group to show a significant effect in terms of biomass where an interaction was found between the detrital and shade treatments (Fi  t 1 2  = 8.42, P <  0.015). Significantly lower orthoclad biomass was found in the low detrital and shade treatments relative to all other treatment combinations.  The results of the redundancy analysis are shown in a correlation biplot in Figure 5. The total variation explained (sum of all canonical eigenvalues) is 21.4%. Of this explained variation, the first axis accounts for 14.3% of the variation and the second axis accounts for 5.3%. The R D A analysis revealed that there were treatment effects due to the detrital addition when looking at the benthic assemblage as a whole. As shown in Figure 5 there is a boundary between the high and low detrital treatments. The shade treatments did not seem to elicit a response from the benthic assemblage. 16  140  400  300  200 4  100  A  HD NS  LD S  LD NS  HD S  HD NS  Treatment  LD S  LD NS  HD S  Treatment  700 600  E 500  i co  E  400  g  m CO c  300  JO  200 H  8 O  100  HD NS  LD S  LD NS  HD NS  Treatment  LD S  LD NS  HD S  Treatment  Pre-treatment Post-treatment  Figure 4.  Pre- and post-treatment in terms of densities (a, b, c) and biomass (d) of benthic macroinvertebrate taxa which showed a significant response to the resource treatments (HD N S : high detritus, no shade; LD N S : low detritus, no shade; L D S : low detritus, shade; HD S: high detritus, shade). Bars represent means + 1 S . E . (n = 4).  17  Lepidostoma -2.07 Podmosta -1.36 Diamesinae  -1.25  Limnephilus -1.12  Figure 5.  Correlation biplot based on a redundancy analysis of the species data from the treatment troughs. Eigenvalues of the first three axes are: 0.143, 0.053, and 0.018. The black line represents the border between the high and low detrital treatments. (HD S : high detritus, shade; HD N S : high detritus, no shade; LD S : low detritus, shade; LD N S : low detritus, no shade)  18  Drifting Invertebrates In the second week of the experiment, the immigration rates were higher than the emigration rates (Figure 6a), suggesting colonization of the channels. As a result, I did not analyze the net emigration data from week two because any trends detected could be the result of the higher numbers of individuals entering the troughs, and drifting directly through the channel. This resulted in 7 separate 2-way A N O V A analyses for the remaining three weeks, rather than repeated-measures A N O V A using time as the repeated-measure. There were no statistical differences found between treatments in overall net emigration (all P > 0.22). The numbers of invertebrates emigrating were low resulting in the analysis of only a few specific taxa, none of which showed a significant response (all P > 0.06).  Per capita immigration and emigration rates were examined in attempts to relate the number of invertebrates entering and leaving the troughs to the numbers found in the benthos. Figure 6b illustrates how low the per capita emigration rates were and that time did not affect this rate. As per capita immigration rates were similarly low, quantitative analysis of per capita drift was not appropriate. Macroinvertebrate functional groups and individual taxa were qualitatively examined to determine if there was one group or taxon that dominated the drift rates, but there were no apparent trends.  19  20  10 C\J  •=  0  CO i—  O)  'E E  -10  g '•*—'  CO CO  1  -20 H  LU -30 Weekl  Week2  Week3  Week4  Week 3  Week 4  Date  0.014  o  ^  o  0.012 H  0.010  0.008 H OJ  CO  0.006  c 0.004 CD -O  E  0.002  0.000 Weekl  Week 2  Date Figure 6.  a) Net emigration rates (emigration minus immigration) and b) per capita emigration (emigration in 24 h / total density of trough (per 1.5m ) for the four week time period of the experiment. Bars represent means + 1 S E . (n = 16). 2  20  DISCUSSION  Invertebrate Responses to Treatments Detritivores (Shredders and Collectors) The results of this food resource experiment indicated that the supplementation of leaf litter can lead to increased densities of some shredders. This suggests that the stonefly Podmosta sp., and the caddisfly Limnephilus sp., may have been food limited. In a similar mesocosm experiment, Richardson (1991) found that detrital supplementation increased densities, decreased emigration and increased mass at emergence for several detritivores. Therefore, changes in direct leaf litter inputs due to riparian harvesting have the potential to limit the detrital food resource of shredders and potentially collectors, which could affect secondary production.  The collector-gatherer Diamesinae, feeds on F P O M that is deposited on loose surface films or sediment (Merritt and Cummins 1996a). As relatively large pieces of organic matter are not believed to be a food resource for these invertebrates, the high densities found in the high detrital treatments suggests an indirect effect of the supplementation of leaves and needles. Although there were no statistical differences in benthic F P O M biomass found between treatments, the addition of leaves may have enhanced the retention of F P O M within the channels as was found by Richardson (1992). Any differences in F P O M may have gone undetected due to direct consumption by the greater densities of collectors in the high detrital treatments. Also, the large sieve size used when processing the benthic samples (500 urn) could have resulted in the loss of much of the F P O M , which would wash directly through the relatively large mesh.  21  In a similar mesocosm experiment, Richardson and Neill (1991) found that there was no difference in benthic F P O M related to added detrital treatments, but that the quantity of suspended F P O M exported was significantly greater. They also found that the densities of collectors increased with detrital additions and suggest that the activity of shredders feeding on the coarse detritus, producing FPOM, are largely responsible for the higher numbers of collectors. Although suspended F P O M was not measured in this experiment, this shreddercollector facilitation, a major assumption in stream ecosystem function (Cummins 1974, Anderson and Sedell 1979, Vannote et al. 1980, Heard and Richardson 1995) could have occurred since an increase in Diamesinae, a collector-gather, was detected. Dieterich et al. (1997) also found collector growth and development were enhanced in the presence of shredders.  Richardson and Neill (1991) give alternative explanations for the increase in collector densities when whole-leaf detritus is supplemented. The first is that although Diamesinae chironomids are believed to be collector-gatherers, they may have directly consumed the leaf litter itself. Secondly, dissolved organic carbon leached from the leaves can be another source of F P O M through physical-chemical flocculation (Merritt and Cummins 1996a), which could increase collector density or biomass. In this experiment, there were no differences found in DOC to support this and the stream channels themselves were probably not long enough to allow the formation of floes. Finally, the addition of detrital materials can alter the microhabitat, which can influence densities of invertebrates and may provide refuge from predators. In an experiment examining the colonization of alder and polyester leaf packs, it was found that F P O M concentration was equally important as changes in microhabitat to positive collector responses (Richardson 1992).  22  The Orthoclaniinae, also collector-gatherers, did not show a similar response as they had higher biomass under the no shade conditions, but with low amounts of detritus. This suggests that the high levels of detritus decrease the orthoclad's ability to obtain their food resource. One possible explanation for this is that under low detrital conditions, a release from competition with the macroinvertebrates that were feeding on the leaf litter may have occurred. The filamentous algae may have been a suitable food resource for the orthoclads and in the absence of competition, they were better able to obtain their food resource.  Both Diamesinae and Orthocladiinae are chironomid sub-families and it has been suggested that chironomids are an important component of stream food chains because of their high numbers and high turnover rates (Mundie et al. 1991). Although many species from these subfamilies may vary in their specific diets, these results suggest that they may be food limited some of the time. If the lack of direct input of leaf litter into a stream following riparian harvesting eventually decreases the detrital standing stock stored within the stream and potentially the F P O M produced from the detritus, the collector-gatherer productivity could be affected.  Other studies have found that macroinvertebrate responses that were detectable were probably due to increased survival of the larvae (Mundie et al. 1991, Richardson and Neill 1991, Perrin and Richardson 1997). In the absence of significant numbers of drifting animals in this experiment, both immigrating and emigrating, the responses of the macroinvertebrates are also probably due to increased survival.  Other Functional Groups There was not a large representaion of scrapers, relative to other functional groups in the mesocosm, as they numerically comprised only 9.5 % of the invertebrate assemblage. According 23  to the river continuum concept (Vannote et al. 1980) they should be a dominant group at the mesocosm site due to the extremely open canopy. Scraper densities may have been low because they are limited by the production of periphyton or by the type of algae within the periphyton matrix. It has been shown that there can be a close coupling between periphyton biomass and benthic invertebrate abundance controlled by a limiting resource (Perrin and Richardson 1997). In general, scrapers reject filamentous or blue-green algae (Gregory 1980) due to their thick cell walls (Mundie et al. 1991). From visual observations, filamentous algae was found in all treatments and may explain the lack of a dominant presence of the scraper functional feeding group.  There were no significant responses to the treatments from the collector-filterers or the predators, but the number of individuals in each of these groups was small. The filterers comprised only 3.5% and the predators comprised only 2.5% of the community assemblage. Thus, any changes within these groups would probably remain undetected due to the low densities.  Community Response As the results of the R D A analysis show, there is a separation of the invertebrate assemblage with respect to the detrital levels in the troughs. Although there were not many definitive responses from the macroinvertebrates when examining them separately, the food addition did affect the community assemblage. The taxa that seem to be driving this pattern are the shredders and gatherers. This is not surprising as these two groups make up nearly 85% of the macro invertebrate assemblage and that both of these groups seem to benefit from high detrital levels.  24  Effectiveness of Shade Treatments There was little evidence that the shade treatments had the desired effect of reducing levels of periphyton, as there were no differences between the shade treatments in terms of chlorophyll a and periphytic A F D M . Boston and Hill (1991) found that chlorophyll-specific photosynthetic rates of the periphyton community were not well related to the ambient light environment and tended to be lower for communities from high-light habitats. Periphyton communities which originate from high-light habitats and are subsequently shaded often do not demonstrate a detectable response (Bothwell 1988; Wellnitz et al. 1996a; P. Kiffney, National Marine Fisheries Service, personal communication). This suggests that some other factor was controlling periphyton levels. Nutrients, such as nitrogen and phosphorus, can also limit algal growth. If the one-time measurement of nutrients in this study is representative of the conditions throughout the experiment, nitrogen may have been a limiting factor to periphyton growth. Levels of phosphorus were high enough to saturate P-limited growth of periphyton, but nitrogen levels were below those needed to saturate N-limited growth (Bothwell 1989). Nitrogen limitation can affect the type of algae present in the system, as algae that can fix nitrogen (e.g. blue-green algae) are often associated with relatively high concentrations of phosphorus and nitrogen limitation (Perrin and Richardson 1997). Limitations to algal production can also be due to herbivory, but because of the low densities of scrapers in this experiment, it is unlikely that topdown control of the periphyton was occurring.  It has been reported that during nutrient enrichment of rivers, accrual of periphyton biomass increased logarithmically followed by a massive decline in biomass even with continuous enrichment (Perrin et al. 1987, Perrin and Richardson 1997). It is suggested that this occurs because of a nutrient diffusion gradient within the periphyton mat which increases as the mat itself increases in thickness. This results in the sloughing off of periphyton from within the mat 25  because of severe nutrient limitation (Perrin and Richardson 1997). The effects of light supplementation may have a similar result. In this experiment, under high light conditions, the periphyton mat may have grown, but the interior of the mat may have been severely light limited, as Hill and Harvey (1990) found. This would result in the sloughing off of the interior of the mat, which would decrease the periphytic biomass, potentially to levels similar to those found in low light conditions. In laboratory streams, Lamberti et al. (1987) found that algal senescence occurred after 32 days and that chlorophyll a increased until day 24 and subsequently decreased. In this experiment, the post-treatment algae was sampled on day 44 which may have been algal senescence occurred. This would result in lower chlorophyll a levels, which could be similar to those found in the shaded conditions.  The algal filament lengths were longer in the no shade treatments, suggesting that the filamentous algae present grew longer filaments under higher light conditions but this is not indicative of higher periphyton levels. As there were no differences in periphytic biomass between treatments, the productivity may have been the same but fewer, longer filaments may have grown under high light conditions while a greater number of shorter filaments grew in the shaded conditions. Another possibility could be that the algal composition of the periphyton matrix was different under the two light regimes. Wellnitz et al. (1996&) found that light level had a strong effect on the abundance of common algae. Filamentous algae were present in both the shaded and non-shaded treatments, but the dominance of this algal type may have varied under the two conditions. The filamentous algae may outcompete diatoms in the full light conditions, but in the shaded conditions diatoms may be the dominant algal type. Addressing the different algal types and quality of the periphyton resource is beyond the scope of this research.  26  It has been shown that ultraviolet radiation can inhibit algal growth and accrual rate (Boston and Hill 1991, Bothwell et al. 1994). If the shade cloth used in this study decreased not only P A R but also ultraviolet radiation, photoinhibition could have occurred in the unshaded troughs while light limitation occurred in the shaded troughs resulting in no differences in periphyton A F D M or chlorophyll a. A n alternate explanation could be due to the fine sediment found on the tiles, which did not differ between treatments and may have prevented the proliferation of certain algal types. As the fine sediments of lacustrine soils characterize the region where the experiment occurred, perhaps a smothering effect due to sedimentation overwhelms any potential benefits an increase in light may have on periphyton.  The various life-history patterns of aquatic insects enable species to emerge during favourable conditions and exploit their food resource during the season in which it is most abundant (Merritt and Cummins 1996a). The lack of an algal response and therefore a response from the grazer functional group could simply be due to the time of year as the intensity and duration of solar radiation would be the highest during the summer months (Bilby and Bisson 1992). In general, Ephemeropterans, many of which are scrapers as larvae, are better adapted to warmer waters in all life stages than plecopterans (Merritt and Cummins 1996a), which also could explain the lack of a response from the mayflies as their necessary growth period would be in the summer months.  The interior of British Columbia is characterized by autumn and winter snowfall, which traps leaves on the streambank, and therefore there is also a pulse of leaf litter into the streams in the spring during snowmelt (Mackay and Kalff 1972). This could explain why there was not a dramatic response from other shredders found in the system. As food limitation creates a potential for intense competition (Richardson 1991), some of the shredder macroinvertebrates 27  may have a life cycle timed to take advantage of this spring food resource instead of the leaf litter pulse in the fall, which could alleviate some competition for the detrital resource. Both Podmosta sp. and Limnephilus sp. are fall-winter taxa that have their major growth period during this time (Cummins et al. 1989).  MANAGEMENT IMPLICATIONS This experiment revealed that periphyton quantity was not limited by light in this system, but that light levels may influence periphyton quality (algal type). Nutrient manipulations in conjunction with shading may be necessary to increase algal production as studies have demonstrated interactive effects between light and nutrients on algal communities (Hill and Knight 1988, Rosemond 1993). Increasing light levels reaching a stream through forest harvest will not necessarily result in an increase in primary production.  Supplementation of leaf litter increased detritivore densities suggesting food limitation of some macroinvertebrates. Changes to detrital inputs following riparian forest harvesting have the potential to limit the detrital food resource of shredders and collectors, which could affect secondary production.  28  CHAPTER 2: RESPONSES OF MACROINVERTEBRATES TO CHANGES IN STREAM FOOD RESOURCES FOLLOWING C L E A R C U T LOGGING ON SMALL STREAMS IN T H E INTERIOR OF BRITISH COLUMBIA  INTRODUCTION  Clear-cut logging to a stream's edge can change many physical characteristics of a stream. One of the salient effects of riparian-harvesting is the reversal in the dominant energy source from allochthonous inputs to autochthonous production which can be stimulated with the increase in light (Hansmann and Phinney 1973, Murphy and Hall 1981, Murphy et al. 1981, Webster et al. 1983, Noel et al. 1986, Wallace and Gurtz 1986, Gregory et al. 1987, Bilby and Bisson 1992, Davies and Nelson 1994). Increased algal production may not occur if the stream is nutrient limited, but shifts in algal type may occur (Mundie et al. 1991).  It has been suggested that overall stream invertebrate productivity may increase following riparian-harvesting (Murphy 1984; Wallace and Gurtz 1986; Behmer and Hawkins 1986; Perrin et al. 1987; Bilby and Bisson 1992) due to an increase in algal biomass and because algae have a higher protein content and are more easily digested relative to allochthonous detritus (Triska et al. 1975). Some studies have documented a change from a univoltine, detritivore-dominated community to a multivoltine, grazer-dominated community (Webster et al. 1983; Wallace and Gurtz 1986; Gregory et al. 1987), but others found no difference in detritivore abundance (Hawkins et al. 1982) or even greater detritivore production in deforested streams (Stout et al. 1993). How riparian-harvesting affects the invertebrate community is still unresolved as some studies show that invertebrate densities are higher (Murphy et al. 1981; Silsbee and Larson 1983;  29  Noel et al. 1986) or lower (Davies and Nelson 1994; Vuori and Joensuu 1996) in clearcut streams relative to control streams.  It is assumed that the in-stream detrital standing stocks would decrease in a riparian harvested area due to the reduction of direct Utter input from the streamside trees (Ulrich et al. 1993). Bilby and Bisson (1992) found that allochthonous input at a coastal, old-growth stream reach was approximately five-fold greater than in a nearby eight-year old clear-cut site. They found relatively large quantities of Utter input and a greater diversity of Utter types throughout the year in the old-growth forested stream, whereas the Utter input in the harvested stream showed a distinct seasonal input that consisted of deciduous leaves from herbs and shrubs. Hetrick et al. (1998) found that aUochthonous input decreased following canopy removal, but stored organic matter in the substrate did not differ significantly between the relatively small (40-70m long reach) open- and closed-canopy sections.  Stream reaches lacking forest canopy may receive subsidies of leaf litter and wood transported from upstream forested reaches (Sweeney 1993). However, there is Uttle evidence of this and it has been suggested that non-forested reaches are lacking retentive structures sufficient to trap the debris (Sweeney 1993). Webster et al. (1990) reported a greater than 50% increase in export of particulate organic matter during storm events in riparian-harvested streams relative to undisturbed streams, and that this export greatly exceeded import, thus depleting organic matter standing stocks. This would have implications to the structure and productivity of the benthic community as it has been shown that detritivores are sometimes food Umited (Richardson 1991). Decreases in detrital standing stocks may result in greater food Umitation of detritivores and decreasing detritivore densities and production (Wallace et al. 1999).  30  Gurtz and Wallace (1984) found no significant change in benthic leaf detritus in cobble riffles in streams before and after clear-cutting. However, they did find lower standing stocks of detritus buried in pebble-riffle and sand substrates which may not have been available to the shredder populations. Although there were no differences in organic matter standing stocks, shredder invertebrates were rare at the clear-cut study site. This suggests some sort of limitation, i.e., food resources or competitive exclusion, as most of the insects found were scrapers and collectors (Gurtz and Wallace 1984). The effects of riparian forest harvest on detrital standing stocks are not clear and may be associated with substrate type.  Benthic macroinvertebrates are a major food source for some stream salmonids and other aquatic organisms. Benthos may also be useful indicators of stream productivity and environmental quality, and are a component of biodiversity (Richardson 1993). Therefore, it is extremely important to determine the impacts of shifts in the environment on macroinvertebrate communities (Richardson 1993).  Enhanced algal growth, which may occur following riparian-harvesting, is assumed to increase stream productivity by increasing the base of the food chain (Behmer and Hawkins 1986). Although it has been documented that salmonid production increased in streams flowing through harvested areas (Murphy and Hall 1981; Hawkins et al. 1982; Bisson and Sedell 1982; Bilby and Bisson 1992), other detrimental effects on the stream environment may offset this production (Bisson et al. 1987). Also, increased primary productivity may not result in increased productivity at higher trophic levels as hydrologic disturbance and other energy sinks can affect productivity (Richardson 1993). Prey availability can set an upper limit to fish production, and therefore, determining the productivity of benthic invertebrates and the factors that limit them will have great implications for fisheries management (Richardson 1993). 31  The present management guidelines for riparian areas in the interior of British Columbia are based on fish-forestry interaction research conducted in coastal areas. Although the riparian area management guidelines (BC M o F / M E L P 1995) for streams in the dry, continental area of the province are different from the coast, there is little scientific research upon which to base these guidelines. Streams in the dry, interior of B.C. may react differently from those on the coast as a result of differing climate, soil types, forest cover and a snowmelt dominated runoff regime (Heise 1997). This, coupled with the lack of riparian protection for fishless streams, results in a need for research on small, interior streams to provide information on the effects of riparian harvesting in these areas. The major objectives of this study were to determine how riparian forest harvest affects stream food resources and the relative composition and biomass of stream macroinvertebrates in small, interior streams of BC. I used a paired stream-reach approach, with one harvested and two upstream forested reaches in each of five streams.  MATERIALS AND METHODS  Study Sites and Study Design The study was conducted in the Sub-boreal Spruce biogeoclimatic zone (Meidinger and Pojar 1991) in the Cariboo Forest Region near the town of Horsefly, British Columbia (Latitude 52° 15'N, Longitude 121° 30'W). To determine the effects of riparian harvesting on small stream food resources and macroinvertebrates, five study streams were selected in the Horsefly Forest District.  32  The major criteria for stream selection were:  •  a 100m reach of stream which had clear-cut logging on at least one side, and  •  a 200m forested reach upstream of the clear-cut.  Locating the study streams proved to be difficult, as many small creeks in the interior tend to dry up during the summer and a fully intact upstream-forested area above the harvested area was not common.  The five, small study streams are classified as S5 to S6 streams according to the B C Forest Practices Code (BC M o F / M E L P 1995). These streams are defined as non-fish bearing streams with an average channel width of >3m for S5 streams and <3m for S6 streams and that do not occur within a community watershed (BC M o F / M E L P 1995). Neither S5 or S6 streams require a riparian reserve zone to be left along the streamside. Sucker Creek, Woodjam Creek and Moffat Creek #1 were harvested to the bank on only one side of the creek. Moffat Creek #2 was harvested on both sides of the creek and Deerhorn Creek was harvested on one side, but a steep embankment on both sides of the stream resulted in a small buffer zone where some riparian vegetation remained intact. The dominant vegetation at the study sites were Engelmann x white spruce (Picea glauca x engelmannii), trembling aspen (Populus tremuloides), black cottonwood (Populus trichocarpd) and mountain alder (Alnus tenuifolid). Thimbleberry (Rubus parviflorus), black twinberry (Lonicera involucratd) and cowparsnip (Heracleum sphondylium) dominated the understorey. Agriculture and timber harvesting are the major land use practices in the area, with much of the logging conducted in the winter months. The characteristics of the study streams are found in Table 1.  33  T3  I  8  CO  'co  8  O  d o d o  O  S3  o  •d U  > S3  8  .d  c  d o  ,0  £>  +^  -l->  o  s o  £2 O 8 > 53  o  8  > S3  •d  rO  .d  O 8 > S3  •d  8 > CN  CO  CN  U  I  IT) 00  la oo  -8 6  VO  00  VO  VO  VO  00  00  00  o  o  o o  CN  o  43 o  H  CO cj  o o o  O CN  as  CN CN  00 CN  CN i-i  U  u M o  00  a  u  '5' o o  u  I 34  Three sites were located on each of four of the five streams; two upstream forested sites and one downstream clear-cut site. Only 2 sites were located on Moffat Creek #1: one upstream forested site and one downstream clear-cut site, as the stream area above these sites did not contain riffle habitat. Each site, or sampling unit, consisted of a 30m stream reach with 6- 5m stream sections. One riffle sample was taken from 3 of the 6- 5m reaches, totalling 9 riffle samples per stream. A l l stream measurements and sampling took place within a one-week period in each of the summer and fall (after leaf fall). Sites on the same stream were sampled the same day.  Field Sampling Methods In both July and October 1997, current velocity and water depths were measured at each of the five streams using a current meter and converted into discharge. At each of the three sites within each of the five streams, aspect was measured using a clinometer and temperature (°C) was taken using a thermometer. Water samples were collected at each site for nutrient analysis (nitrogen and phosphorus). Mean substratum particle size was estimated in each 5m stream section visually as percentage composition within a 0.09m quadrat of fine (<2mm), gravel/pebble (22  65mm) and large (>65mm) particle sizes. Habitat type was visually assessed at each 5m stream section (i.e. 5m intervals) and expressed as the percentage of pool and riffle within each stream reach and then averaged for combined stream reach values. Canopy cover was visually estimated every 5m and expressed as the percentage of large woody debris (LWD) and vegetation covering the stream and then averaged for combined site values. Bankfull and wetted width of the streams were measured at each sampling site. Solar irradiance levels were measured at each site on September 15, 1997 using a PAR-quantum light sensor (LI-COR, Nebraska) and values were converted to quantum units (pE'm" s" ). 2,  35  1  Analytical Methods  Periphyton In both July and October 1997, periphyton samples were taken by scraping an area of 12.57cm  2  from each of three rocks, which were then combined to produce one sample. The end of a toothbrush was attached to a drill bit and placed within a tubular template secured to a rock from the stream. A small amount of water was added to the template and the toothbrush scraped the rock with the operation of the drill resulting in a slurry. Each periphyton sample was split into two equal parts and filtered onto pre-ashed 47mm glass fibre filters (Whatman GF/F), one for periphytic biomass analysis and the other for chlorophyll a analysis. The filters were allowed to air dry in darkness and then placed in the freezer. A total of 3 periphyton samples were taken at each site resulting in 9 samples per stream. The periphytic biomass filters were oven dried (60°C for at least 24 h), weighed, ashed (500 °C for at least 1 h) and reweighed to determine ash free dry mass by weight loss on ignition. The chlorophyll a samples were soaked in 90% acetone for 24 h and processed according to the extraction procedure outlined in Strickland and Parsons (1972) using a Turner fluorometer (model 10-005 R).  Macroinvertebrates and Detritus On both sampling periods, macroinvertebrate collections were taken at each site resulting in 9 riffle samples per stream, per season. The benthos was sampled using a 900cm Surber sampler 2  with 250 u.m mesh. The substratum within the Surber frame was manually disturbed to a depth of 10 cm. Samples were preserved in the field in a buffered formaldehyde solution.  The riffle samples were washed through 3 nested sieves (4mm, 1mm and 500urn). A l l animals were removed from each sieve size and identified, counted and digitized for animal biomass 36  (Hopcroft, R. 1995, Zoobenthos Biomass Digitizing Program, Department of Zoology, University of Guelph, Ontario). A l l animals were classified into functional feeding groups (scrapers, shredders, filterers, gatherers or predators) according to Merritt and Cummins (1996) classifications. The organic matter was retained from the benthic invertebrate samples and used to determine organic matter standing crop. The detritus from the 4mm sieve was sorted into wood, leaf, needle, fruit & flower, and moss categories. The detritus from the 1mm sieve was classified as coarse particulate organic matter (CPOM l-4mm) and the detritus from the 500pm sieve was classified as fine particulate organic matter (FPOM). Organic matter was oven dried (60°C for at least 24 h), weighed, ashed (500 °C for at least 1 h) and reweighed to determine ash free dry mass by weight loss on ignition.  Nutrients Nitrogen and phosphorus concentrations were measured to the nearest 0.35 pg NO3-N/L and 2.4 pg PO4-P/L respectively according to the methods of Wood et al (1967) for nitrogen and Hager et al. (1968) for phosphorus. The analysis was done on previously frozen water samples taken at each site within each creek using a Technicon AutoAnalyzer II.  Statistical Analysis Three-way analysis of variance (ANOVA) was used to examine any differences between streams, between the sites within the stream (harvested and forested), and between seasons on the bio tic measurements taken. Any significant differences were further tested using Tukey's test. Periphyton, detritus, and the density and biomass of macroinvertebrate functional feeding groups and the ten most numerically dominant taxa were examined. The data from the two forested sites were used as replicates to compare to the harvested site. Significant differences 37  between the harvested and forested sites should only be revealed if those differences are greater than the differences between the two forested sites alone. A l l analyses were done using the General Linear Models procedure in SAS version 6.12 (SAS 1996). Type III sums of squares were used in all analyses. The assumption of normality was tested on the residuals of all of the data sets using the conservative Shapiro-Wilk statistic and where necessary, data were logio (x +1) transformed before analysis (Zar 1984). Figures, resulting from any statistically significant analyses, display the raw data. In all tests the critical level of significance was a=0.05.  To determine if there was a response in community structure to riparian harvesting, multivariate analysis of variance ( M A N O V A ) was used. A l l of the categories of periphyton and detritus were examined. The density and biomass of the functional feeding groups and that of the ten most common taxa were also tested. The Wilks' Lambda statistic was used in all analyses to determine significance.  To determine if the harvested and forested sites differed with respect to species composition and environmental variables combined, a direct multivariate approach was used. A detrended correspondence analysis (DCA) was conducted to determine if the response was linear or unimodal (ter Braak 1994). As the gradient lengths were less than 3 standard deviations (length of gradient = 1.62), the relationship between the species and environmental variables was linear. Therefore, a linear method, redundancy analysis (RDA) was conducted on the data as opposed to the unimodal canonical correspondence analysis (CCA) (ter Braak 1994).  As season was shown to have a strong influence on the results from the univariate analyses, two separate redundancy analyses were performed on each of the two seasons to display differences in species composition and environmental variables between sites. In both analyses, the data 38  were log-transformed, scaled by inter-species correlation and centred by species. Monte Carlo permutation tests were used to determine significant relations with environmental variables. The D C A and RDAs were done using C A N O C O (Windows version 4.02) and the figures are correlation biplots.  To determine if the macroinvertebrate community assemblage was more different between the forested and harvested sites relative to the two forested sites the sample scores of the R D A analyses were examined. The weighted Euclidean distance between the sample scores of the two forested sites within a stream was calculated. Also, the distance between the mean of the sample scores of the two forested sites and the sample score of the harvested site within that same stream was calculated (Appendix I). Moffat Creek #1 was not included in this analysis as there was only one forested site and therefore, there were not two distances to compare. An A N O V A was then performed on these distances and the ratios of the harvested to forested site distance and the two-forested sites distance was plotted to further elucidate the magnitude of differences between sites.  RESULTS  Physical Data The mean values for the physical measurements taken in both July and October are shown in Table 2. The stream temperatures in the summer were only slightly, but consistently higher in the forested areas (mean = 12.5 °C) relative to the harvested sites (mean =11.6 °C), but in the fall there did not appear to be differences (harvested mean =1.4 °C; forested mean =1.7 °C). In all of the streams, except Moffat Creek #1, nitrogen ( N O 3 - N) concentrations were higher in the 39  52  8  - I CN  VO  CN VO  Q CN  ll  o ON  u  CN CN  CN  VO  5 VO  1*  5  CN  Q  o\ oo  fa  8 fa  3  e o  •S  VO  CN  Q CN  CM  fa  101Q u fa u CO  CN  o  ON  u  CN  DO  I  *3  o  3  hi  CO  3 CU  '  3 00  u O  i CU  H  O Z  00  i  a  s  03  "C  40  §  ^ S  •  - J CN  3 00  •  >/">  .8 <^ •Q  s  oo  ^—'  I« , •  lO  +1 VO jg .A.I S 00  forested areas in both seasons and nitrogen concentrations within a stream were higher in the summer than in the fall. Phosphorus ( P 0 - P) concentrations were much more variable between 4  stream reaches and between seasons. The creeks were consistently wider in the harvested stream reaches (mean = 4.34m) compared with the forested upstream reaches (mean = 2.63m). The total riparian cover, including both L W D and vegetation, was greater in forested sites relative to the riparian harvested sites in both seasons. The distribution of riffles and pools seem to follow opposite trends as the percentage of stream area consisting of riffles is greatest in the harvested sites (the furthest downstream) and the percentage of stream area consisting of pools is greatest in the forested sites (furthest upstream). The percentage of fine (<2mm), gravel/pebble (265mm) and large (>65mm) particle sizes did not seem to show a trend and was quite variable within a stream, between streams and between sample periods. Figure 1 shows the light measurements taken in September 1997. There were obvious differences in three of the five streams, as the light levels reaching the streams are consistently higher in the harvested sites. The photosynthetically active radiation (PAR) reaching Deerhorn Creek does not seem to differ between the forested and harvested sites and Moffat Creek #1 actually has higher P A R in the forested site due to local gaps. Although the light measurements were all taken on the same day, both Deerhorn Creek and the harvested site at Moffat Creek #1 were taken under hazy conditions. Figure 2 shows the discharge of the five streams in the summer, showing considerable variation among streams and greater similarity in autumn.  Periphyton and Detritus Analysis of variance (ANOVA) showed that there were no statistical differences found in periphytic ash-free dry mass (AFDM) between streams, between positions within a stream and between seasons (P > 0.118). However, higher concentrations of chlorophyll a were found in the summer (Fi, = 21.28, P < 0.0003) (Figure 3a). Both chlorophyll a and periphytic A F D M were 1 5  41  Sucker Figure 1.  Woodjam  Moffat #2  Deerhorn  Moffat #1  Photosynthetically active radiation (PAR) at the five streams in the harvested and forested stream sections.  0.18 0.16  0 ^  0.14  Summer discharge Fall discharge  0.12 H 0.10  CD £>  0.08  g 5  0.06 -|  0.04 0.02 0.00  Sucker  Figure 2.  Deerhorn  Woodjam  Moffat 1  Moffat 2  Stream discharge for the five creeks based on measurements from a single day in July and October 1997. 42  Fall  Summer  Figure. 3a.  Seasonal chlorophyll a concentration (pg/cm ) for all sites combined. Bars represent means + 1 S.E. (n = 45).  1.0  3. to  =  .c  #  Chlorophyll a  0>  Periphytic A F D M  CM  E  0.5 H  Q LL  CL  2 o  <  O  h1  0.0  Figure 3b.  i  1  Sucker  Deerhorn  r—  Woodjam  1  1  Moffat #1  Moffat #2  Chlorophyll a (pg/cm ) and periphytic AFDM (g/m ) in the five streams for all seasons and sites. Symbols represent means + 1 S.E. (n = 18).  43  examined together using multivariate analysis of variance ( M A N O V A ) and this also showed a significant season (F , =18.00, P < 0.0001) and stream effect (F =2.968, P < 0.016) (Figure 2  14  g|28  3). There was no difference found in the periphyton community between the harvested and forested sections of the streams. When examining the periphyton results on a stream by stream basis, it becomes obvious that the streams did not respond in a similar manner to riparian harvesting (Figure 4). In the summer season, Woodjam Creek was the only stream with higher chlorophyll a concentrations and periphytic A F D M in the harvested stream sections. In the fall season, both Sucker and Woodjam Creeks demonstrated greater chlorophyll a levels in the harvested sites, but only Sucker Creek displayed higher A F D M in the harvested site.  Some of the detrital categories varied by season. Wood biomass was found to be greater, but only slightly, in the summer sampling season (Fx, = 4.65, P < 0.05) and leaf biomass (F  1(1 6  1 6  = 8.45, P < 0.01), whereas both needle (Fj,  1 6  =14.43, P < 0.002) were greater in the fall season  (Figure 5). Moffat Creek #2 had significantly greater moss biomass than both Sucker and Deerhorn Creeks (F  4 |u  = 6.24, P < 0.003) and forested stream reaches supported higher moss  biomass than harvested stream reaches (Fi, i6= 5.19, P < 0.037) (Figure 6). There were no statistical differences found in C P O M (lmm-4mm) or F P O M (500pm-1mm) (P > 0.06).  A significant site*stream interaction was found for wood (F , 4  1 6  = 4.16, P < 0.02), needle (F  4|1 6  =  12.78, P < 0.0001), fruit and flower ( F , = 4.05, P < 0.02) and total detrital biomass ( F , = 4  16  4  16  8.51, P < 0.0007) (Figures 7 & 8). Although detrital biomass in all of the above categories seems higher in the fall, similar trends in each stream were found in both seasons. Moffat Creek #2 clearly had a greater accumulation of all detrital categories in the forested site relative to the harvested site. Moffat Creek #1 showed a similar trend, but the needle biomass was only slightly  44  CD  5_  3  CD  E T<o *  CM  O  O  C0  *  CO  45  Summer  Fall p  Figure 5.  Mean detrital biomass (g/m ) in the summer and fall seasons. +1 S.E.  a)  E 3  E g  <n o  Forested  Harvested  Sucker  Figure 6.  Deerhorn  Woodjam  Moffat #1  Moffat #2  Moss biomass (g/m ) a) found in the harvested and forested sites and b) found in the five streams. +1 S.E. 47  CD o C 3 CO D 5 CO Q  +  T- CM  CO *  *  s  >ffat  £  £  >ffat  £  s  p  CM  E  00  CD  •*  CM  ( UJ/6) ssBOjqq 8 | 8 8 5  P  O  O  00 ^  N  48  CD  ^mom  Tl" J 9 M 0 | d  CM  ^  1 | m d  O  §  c-  E  CM  ! Ii s s  Hill  + <> f  T o CO CD  CM  </) CO  E o !a  "co CD  TJ « O  •a c CO C/)  co CO  E o O O  O  *  O  to  E > CO i2? £  00 T J  ° 9 §""»  ~  CD  "co . g CD C/3 CD  . C O  o. <2 CD g co CD  ( w/6) ssBUiojg poo/v\  ( UJ/B) 2  49  ssBUioiq |B}uiep IBJOI  greater in the forested site in both seasons. Although needle biomass was greater in the forested site of Woodjam Creek, all other detrital categories showed only small or no differences between the two sites. Deerhorn Creek showed little difference between forested and harvested sites with respect to all detrital categories. Although slightly lower needle and fruit and flower biomass were found in the forested sites of Sucker Creek in the summer, wood and total detrital biomass were lower in the harvested sites. In the fall, few differences between the two sites could be detected. When examining all of the detrital categories at once using M A N O V A , a significant stream*site interaction was again revealed (F =2.54, P < 0.004). 4 >16  Macroinvertebrates There were no statistically significant differences found in the total macroinvertebrate density or biomass between streams or between positions within a stream (P > 0.094). Macroinvertebrate densities ranged between 600 and 50,000 individuals/m and biomass ranged between 260 and 2  91,000 g/m  2  Higher total densities (F  h 1 6  = 16.0, P < 0.001) and biomass (Fi, i = 4.75, P < 6  0.045) were found in the fall season.  Functional Feeding Groups The scrapers, shredders, and gatherers all had much higher densities in the fall season (all P < 0.031) (Figure 9a). Although the predator feeding group had higher total densities in the fall, there was a significant season*site interaction (Fi, = 5.33, P < 0.035) (Figure 10). Predator J 6  densities in the forested sites greatly increased from the summer to the fall, but a decrease in predator density was observed in the harvested sites. There were no significant differences found in the densities of the filterer functional feeding group (P > 0.09). There were no significant effects found in the M A N O V A analyses (all P > 0.19)  50  a) 6000  4000  'co  c Q CD  Summer 500  b)  l Filterer 400  E  3  3 Shredder  A  Predator  300 H  % CO  E  o in  200 H  100  Summer  c)  E  i  Fall  - [-  30000 20000 H 10000 / 3000 / 2500 2000 H  CO  E  o m  1500 1000 500 H 0 Deerhorn  Figure 9.  ID I Sucker  i  W o o d j a m Moffat #1  Moffat #2  Functional feeding group a) seasonal densities b) seasonal biomass, c) biomass in each stream. +1 S.E. 51  900 D  800 700  -•— Harvested sites • o - Forested sites  600 CO  a5  500  .g  400 300 200 H 100  fall  summer  Figure 10. Predator density changes between seasons and sites. Lines shown for visual effect only.  52  With respect to macroinvertebrate biomass, the shredders, filterers and predators had much higher biomass in the fall season (all P < 0.003) (Figure 9b). Scraper biomass was greater in Deerhorn Creek than in all the other creeks, but was only statistically different from both Moffat Creek 1&2 (F , i = 6.17, P < 0.034) (Figure 9c). Both of the Moffat Creeks had higher gatherer 4  6  biomass than the other creeks, but only Moffat Creek #2 was significantly greater than Deerhorn and Woodjam Creeks (F , u = 3.42, P < 0.033) (Figure 9c). The M A N O V A analyses also 4  revealed a significant seasonal effect when all of the functional group's biomass were examined together ( F , = 3.87, P < 0.03) 5  12  Individual Taxa Individual taxa from many of the functional feeding groups showed a significant seasonal response as higher densities and biomass were found in the fall (Figure 11). Specifically, the scraper Cinygma (Fi, i6 = 10.14, P < 0.006), the shredders Capnia (Fi, i6 = 9.37, P < 0.008) and Zapada (Fi,  1 6  = 13.14, P < 0.002), the predator Sweltsa (Fi, i = 7.44, P < 0.015), and the 6  gatherers, Microspectra (F , a  ] 6  = 5.10, P < 0.038) and Oligochaetes (Fi,  J 6  = 14.32, P < 0.002) had  much higher densities in the fall season. With respect to biomass, Cinygma (Fi, 0.002), Capnia (Fi, u = 23.42, P < 0.0002) and Sweltsa (F , a  a 6  1 6  = 13.55, P <  = 11.00, P < 0.004) were found to  have significantly higher biomass in the fall, as well as the shredder/collector, Paraleptophlebia (F ] = 5.62, P < 0.031). A M A N O V A was conducted on the densities and biomass of the 10 lt  6  most dominant taxa, which included the scrapers Baetis and Orthocladiinae, and the filterer Simulium, as well as the taxa mentioned above. A significant seasonal effect was detected from the assemblage with respect to density (Fin, 7 = 4.56, P < 0.03) and biomass ( F , 7 = 4.56, P < 10  0.03).  53  6000 5500  /  V  3500/ 3000 E  2500  •5 c CD  2000  Q  l ^ B Cinygma FMm Capnia w@m Zapada i i Sweltsa — Microspectra Oligochaete  1500 1000 500 H 0  Summer  Fall  500  b)  400 175^ 150  Cinygma Capnia Paraleptophlebia Sweltsa  125  co % E O S  100 75 H 50 25 0  Summer Figure 11.  Fall  Seasonal differences of individual taxa with respect to a) densities and b) biomass. +1 S.E.  54  Responses of specific taxa were also stream dependent (Figure 12). The scraper Cinygma was found in much lower numbers in Moffat Creek #2 than in Sucker and Woodjam Creeks (F , = 3.42; P < 0.033). Significantly higher numbers (F , = 7.71; P < 0.001) and 4 x6  4 X6  biomass (F \ = 4.26; P < 0.016) of the shredder/collector Paraleptophlebia were found in 4)  6  Deerhorn Creek relative to the other creeks although Woodjam Creek was not significantly different in terms of biomass (Figure 12). (The mayfly, Paraleptophlebia, has been classified as both a shredder (Mattingly 1987) and a collector (Dieterich and Anderson 1995) as both C P O M and F P O M are utilized for growth). The oligochaetes were found in much higher densities in both Moffat Creeks #1 and #2 relative to Woodjam and Deerhorn Creeks and statistically higher densities were found in Sucker Creek relative to Deerhorn Creek as well (F  4i] 6  = 10.51; P <  0.0002). Moffat Creek #2 had significantly higher oligochaete biomass than Deerhorn, Sucker and Woodjam creeks (F  4 ) 16  - 3.64; P < 0.027). Although there was quite a bit of variability with  respect to Capnia biomass, this shredder seems to have higher biomass in Deerhorn and Sucker Creeks (F u = 3.64; P < 0.03), but there were no differences found in the Tukey's test. The 4>  predator, Sweltsa, was found to have significantly higher biomass in Woodjam Creek than the Moffats and Deerhorn Creeks, and significantly lower biomass was found in Moffat Creek #2 relative to Deerhorn, Sucker and Woodjam Creeks (F , i = 1 1 . 1 5 ; P < 0.0002). 4  6  Macroinvertebrate Communities The results of the redundancy analyses on the two seasons are shown in Figure 13. In the summer season, the total variation explained (sum of all canonical eigenvalues) is 47.5%. Of this explained variation, the first axis accounts for 21.4% of the variation and the second axis accounts for 15.4%. Both Woodjam and Moffat Creek #1 have a higher percentage of pool habitats, whereas the other three creeks have a higher percentage of riffle habitats. Temperatures and phosphorus concentrations were greater in Deerhorn Creek relative to all 55  Deerhorn  4000 A  2000 800 £T  E  S  Sucker  Woodjam  Moffat #2  I  Capnia Paraleptophlebia Sweltsa Oligochaeta  y  Moffat #1  I  /  700 600 500 400 300  I  200 100 0  Deerhorn  Figure 12.  i Sucker  Woodjam  Moffat #1  Moffat #2  Individual taxa differences between streams with respect to a) densities and b) biomass. + 1 S.E.  56  V  Summer 3 -i  •  Deerhorn Sucker Woodjam Moffat 1 Moffat 2  o T V  •  7  _FD  FU  FDVi  2 A  .% Pool  Phosphorus Temperature  FD  'FU  Fall FD.  FIL  |FU  1  IFD  i  Nitrogen HV  -2  • o T V  •  •1 Leaf biomass  1  Temperature Deerhorn Sucker Woodjam Moffat 1 Moffat 2  FQ_  -1  ^FU  FD  O H  o  c5u  -2 Figure 13.  Correlation biplot based on redundancy analyses of the species data from the five study streams in the summer and the fall. Significant quantitative environment variables are indicated by the arrows. H = harvested; FD = forested downstream; FU = forested upstream. Eigenvalues of the first three axes are: a) summer analysis: 0.214, 0.154 and 0.107 and b) fall analysis: 0.40, 0.196 and 0.025.  57  others. Differences between harvested and forested sites do not seem as obvious. In Woodjam, Moffat Creek #1 and Moffat Creek #2, the forested sites have more habitat area consisting of pools than the harvested sites. Deerhorn and Sucker Creeks do not seem to show any trends with respect to sites.  In the fall season, the total variation explained is 62.1%. Of this explained variation, the first axis accounts for 40% of the variation and the second axis accounts for 19.6%. Both Moffat Creek #1 and #2 supported higher leaf biomass than the other creeks. Deerhorn Creek had higher nitrogen levels and temperatures were warmest in Woodjam Creek. There do appear to be site differences in the fall season. The harvested sites of all of the creeks are lower on the y-axis than the forested sites for the same creek. This suggests that the harvested sites are more influenced by temperature and nitrogen concentrations.  Figure 14 shows the ratio of the Euclidean distance between the harvested and forested sites and the distance between the two forested sites in the two seasons. In the summer season, the harvested site is more different from the forested sites than the forested sites are from each other in both Woodjam and Moffat Creek #2. The opposite was found for Deerhorn and Sucker Creeks. In the fall season, the harvested sites in all streams were consistently more different from the forested sites relative to any differences found between the forested sites. The A N O V A analysis which was performed on the distance data from the R D A analyses for both seasons revealed a significant stream*distance interaction (F  3 ]1 5  = 5.6; P < 0.036) and a significant  season*distance interaction (F^ = 10.00; P < 0.02). This suggests that although there were 1 5  significant differences between the harvested and the two forested sites, the direction of this difference was dependent on the stream and season.  58  Summer  Figure 14.  Deerhorn  Sucker  Woodjam  Moffat 2  Deerhorn  Sucker  Woodjam  Moffat 2  The ratio of the distance between the harvested and forested sites (H-F) and the distance between the two forested sites(F-F) from the seasonal RDA analyses. A ratio value above 1 indicates that the harvested site is more different from the forested sites than the forested sites are from each other. 59  DISCUSSION  The Physical System One of the major differences between coastal and interior stream systems in B C is the discharge regime. Although the discharge measurements in this study were just a snapshot, they still show more variation among streams in summer flow and a low, stable fall flow, characteristic of a snowmelt-dominated system. This differs from coastal systems where low, stable flows are found in the summer and the variable flows are found in the winter, characteristic of a raindominated system (Rothacher 1970).  Greater amounts of streambank erosion may be occurring in the harvested stream reaches, as they were consistently wider than the upstream-forested stream reaches. The soils in this area are highly erodible and lacustrine in origin, which may explain the greater stream width found in the harvested areas (Larkin et al. 1998). In a recent study done in the Stuart-Takla and WillowBowron watersheds, also in the central interior of BC, Fuchs (1999) found that creeks that had been harvested 20 years previously had greater stream widths and bankfull heights (i.e. depth). In that same study, it was found that the bankfull widths and heights in recently cut streams (less than 5 years old) were intermediate relative to uncut and 20-year old cut streams. This suggests that effects of erosion may not be detected within the first five years after riparian harvesting, but that even following riparian regeneration, streambank erosion may still occur.  Changes to the ratio of pool and riffle areas has been noted following riparian harvesting (Hartmann et al. 1996). In coastal streams, the number of pools eventually decreases without a sustained supply of obstructions such as large organic debris (Montgomery et al. 1995). In this study, the percentage of riffles was greatest in the harvested stream sections and that of pools 60  was greatest in the forested stream sections. There tended to be less L W D over the stream channel in the harvested sections, which suggests that it was taken away when the harvesting occurred. This depletes the future supply of L W D to the stream channel, which may affect the distribution of pool and riffle habitat. If L W D is stable, it allows the persistence of riffle-pool systems, as pools are created immediately downstream of the L W D and riffles immediately upstream (Bisson et al. 1987).  Periphyton Many studies have demonstrated that open stream sections support higher algal standing crops and primary production than more shaded stream sections (Lyford and Gregory 1975, Noel et al. 1986, Hawkins et al. 1982, Murphy and Hall 1981, Murphy 1984, Behmer and Hawkins 1986, Hetrick et al. 1998, Kim 1999). In B C interior streams, it has also been suggested that an increase in solar radiation reaching the stream and a potential release of nutrients following riparian harvesting would stimulate primary production (Richardson 1994). In a previous study, also conducted in the B C interior, it was found that recently cut streams (within 5 years) had the highest levels of benthic algae relative to streams with 20-year-old riparian forest or control streams (Fuchs 1999). In my study, there were no statistically significant responses found between the forested and harvested sites within the streams, and any responses seem stream dependent. In the fall season, Woodjam and Sucker Creeks did display the predicted trend that higher chlorophyll a concentration and/or periphytic A F D M would be found in the harvested stream sections. Both Sucker and Woodjam Creeks showed the largest difference in P A R reaching the stream between the harvested and forested streams sections, which may explain the differences in periphyton. In an experiment manipulating both light and nutrient levels conducted in two coastal streams, it was found that light levels were of greater importance than nutrient levels in terms of periphyton limitation (Kim 1999). Woodjam Creek also displayed 61  higher periphyton levels in the summer season, but Sucker Creek did not. Heavy rains fell two days prior to the start of the summer sampling session, which may have increased the discharge of the creeks enough to erode periphyton. Sucker Creek, which had the lowest periphyton values, was the first stream to be sampled and may not have had sufficient time to allow for the recolonization of the periphyton. Flood disturbance is an important factor influencing periphyton and can explain a substantial amount of annual variation in mean monthly biomass of algae in streams (Biggs 1995).  Two of the study streams (Deerhorn and Moffat Creek #1) did not have higher solar irradiance in the harvested site relative to the forested sites. The harvested section of Deerhorn Creek had fairly steep banks on both sides and a small riparian buffer. This partial shading of the harvested stream reach resulted in only a small difference in P A R between the forested and harvested reaches. In Moffat Creek #1 an open marsh area with only a few sparse trees characterizes the upstream, forested site. This resulted in higher P A R reaching the stream in this site than in the harvested site, which was harvested only to one side of the creek. Although Moffat Creek #2 had higher P A R reaching the stream in the harvested site, the quantity of light was less than half that of Sucker or Woodjam Creeks. This may not have been a large enough difference between the harvested and forested site to elicit a detectable response from the periphyton in Moffat Creek #2.  Nutrients can also limit periphyton production and it has been suggested that phosphorus is commonly the limiting nutrient (Schindler 1977). Bothwell (1988) found that in southwestern BC, specific growth rates of unicellular periphytic diatoms were saturated at phosphorus concentrations of 0.3-0.6 pg P/L. Phosphorus concentrations in this study ranged from undetectable -1.73 pg P/L suggesting that phosphorus limitation may have been occurring some 62  of the time. Potential light limitation of periphyton production should be studied when information about nutrient limitation is known.  Bothwell (1988) also found that algal growth rates were not affected by the seasonal changes in solar insolation, but were highly correlated with temperature. This may explain the greater concentrations of chlorophyll a found in the summer season when temperatures were higher. In the fall sampling period, some factor seemed to be limiting periphyton production (i.e. temperature, nutrients, herbivory), but a release from this limitation occurred in the summer. The greater number of grazers found in the fall season may have decreased the standing stock of periphyton resulting in a high turnover of algal production. Herbivory can be a primary factor that structures periphyton communities relative to that of competition, predation and abiotic factors (Rosemond et al. 1994, Feminella and Hawkins 1995), but if periphyton is limited by abiotic factors, grazers usually have little influence (Wellnitz et al. 1996&).  Detritus The higher biomass of leaves and needles found in the streams in the fall season was expected simply due to the phenological changes in the streamside vegetation. Also, because decomposition rates are a temperature dependent process (McArthur et al. 1988), decomposition of detrital material should occur at the highest rate in the summer months (Richardson 1992) resulting in a low detrital standing stock. In small, shaded streams, mosses are the only significant macrophytes (Anderson and Sedell 1979) and in this study, the greatest moss biomass was found in the forested stream sections. Filtered sunlight and plenty of moisture found in some of the forested stream reaches may be more suitable for moss than direct sunlight which can cause desiccation. Also, larger, less mobile substrates are more suitable moss habitat (Vuori  63  and Joensuu 1996) and more of this substrate was found in Moffat Creeks #1 and #2 where the greatest amount of moss was found.  Differences in detrital biomass between the harvested and forested stream sections were not obvious and seem stream dependent. Small headwater streams are particularly retentive (Bilby and Likens 1980) and it has been suggested that it can take years to detect differences in detritus because of storage in the system (B. Wallace, Univ. of Georgia, personal communication). Therefore, the harvested stream sections may still contain detritus that has been stored over the years or transported from the forested sections. Hetrick et al (1998) found that although allochthonous inputs were significantly higher in closed canopy stream sections relative to open ones, the amount of stored organic material available as food for benthic invertebrates did not differ between the relatively small (40-70m long stream reach) open- and closed-canopy sections. This suggests that further research should include determining if sufficient amounts of detritus are transported to harvested reaches, how long stored organic material is available to consumers following riparian harvesting and how long it takes for the riparian vegetation to return to a state similar to pre-harvest conditions. The role of detrital transport and export to downstream reaches lacking a riparian canopy should also be investigated.  In the dry, continental area of the province, the majority of the leaf litter would enter the stream in the fall during relatively stable flows, allowing for a high detrital standing crop near the point of entry. Throughout the fall and winter, when temperatures are low, decomposition rates of the leaf litter should be slow leaving a substantial food resource for macroinvertebrates feeding on the detritus until the spring snowmelt. In the Horsefly area, the snow cover may have reduced inputs in the fall, which could result in a pulse of leaf litter with the spring snowmelt. How the  64  detrital dynamics in the interior of the province differ from those on the coast has yet to be determined.  Macroinvertebrates Many studies have shown that higher densities and/or biomass of benthic invertebrates occur following riparian harvesting (Newbold 1980; Murphy et al. 1981; Hawkins et al. 1982;Gregory et al. 1987; Anderson 1992; Hetrick et al. 1998) and this increase is believed to be due to an increase in periphyton production. This was not shown in this study, as there were no differences found between the forested and harvested sites in both periphyton and macroinvertebrates. This is in agreement with a study done on continental streams in Ontario where Giroux (1994) did not find any differences in macroinvertebrate biomass or density between streams. Conversely, Fuchs (1999) found greater macroinvertebrate biomass and higher chlorophyll a biomass in recently logged streams relative to 20-year old riparian-harvested and unlogged streams in the interior of BC. Many stream characteristics were similar in both the study conducted by Fuchs (1999) and this study, except for the stream size. The larger streams used by Fuchs (1999) may have allowed more solar radiation to reach the stream and a greater surface area for periphyton production. Higher primary production is often associated with higher macroinvertebrate biomass as was found by Fuchs (1999). In this study, the macroinvertebrate densities and biomass and periphyton parameters were not found to be different between the continental streams. Thus, although an increase in macroinvertebrate abundance and/or biomass is a common finding following riparian harvesting in coastal systems, a common trend in continental streams does not seem apparent.  Seasonal differences were found in macroinvertebrate density and biomass, as both were greater in the fall. This is coincident with higher detrital biomass, but lower chlorophyll a 65  concentrations. Thus, the greater densities and biomass of macroinvertebrates in the fall season, which feed on detrital material, may be due to the increase in quantity of their food resource. The responses of macroinvertebrates could also be due to life-history characteristics if their major growth period occurred in the fall when food resources were greatest. Richardson (1991) found that some nemourid stoneflies in southwestern B C had a life cycle timing where the major growth period coincided with the autumn leaf fall. Although it was not apparent that there was an increase in the algal resource, macroinvertebrates which feed on periphyton may have decreased the periphyton standing stock due to grazing pressure from the higher densities found in the fall season (i.e. top-down control).  Functional Feeding Groups Population limitation is believed to be mediated by food supply in some macroinvertebrate communities, but the conditions under which food limitation occurs are not well known (Richardson 1993). To examine food limitation, information must be known concerning which resources certain taxa use. Most macroinvertebrates are opportunistic feeders (Anderson and Sedell 1979) or trophic generalists, where they are capable of utilizing more than one resource for growth (Mihuc and Minshall 1995). Thus, it is difficult to categorize macroinvertebrates into specific categories. Functional feeding groups are based on the association between the basic nutritional resource used and the limited set of feeding adaptations (Merritt and Cummins 1996b). Although there are still difficulties classifying specific taxa, this method allows for the determination of the degree to which macroinvertebrates are dependent on a particular food resource (Merritt and Cummins 1996£>).  Shredder productivity seems to have tracked the productivity of their food supply. Higher densities and biomass were found in the fall that coincides with the greater biomass of leaves and 66  needles. Although the relationship between biomass and productivity is not clear, it does seem to be positively related (Richardson 1993). A seasonal release from food limitation could be occurring as it has been shown that shredders may be food limited at certain times of the year in coastal systems (Richardson 1991). Although the variation in the availability of detrital food may be different in systems with a discharge regime dominated by snowmelt, i.e. greater detrital retention in the fall and winter seasons, the food supply is still lowest in the summer and subject to high export during the spring freshet. Thus, seasonal food limitation of detritivores would still be probable, as there is a large difference in the availability of detritus to consumers between the seasons.  Deerhorn Creek seemed to support higher densities of the shredder/collector, Paraleptophlebia, and higher biomass of both Paraleptophlebia and the shredder, Capnia. The R D A analyses shows that temperatures and phosphorus concentrations were higher in Deerhorn Creek in the summer months, which may result in greater macroinvertebrate growth. In the fall, high nitrogen content was found in this stream which may be due to the release of nutrients from a high quantity or quality of decomposing detrital material. Types of leaf litter with a high nitrogen content are considered high quality (Triska et al 1975) and thus can affect macroinvertebrate growth (Anderson and Sedell 1979). Although detrital quality is beyond the scope of this study, it could be a possible explanation for differences seen between streams.  Collectors showed a positive response to the changing of the seasons from the summer to the fall, which may be due to an increase in their food resource. Fine particulate organic matter found in streams is generated from animal consumption (e.g. shredders), physical breakdown of larger particles, the sloughing off of periphyton, flocculation of dissolved substances, microbial processes and terrestrial inputs (Wotton 1984). As greater leaf and needle biomass was found in 67  the fall, these terrestrial inputs may have increased the quantity of F P O M in the system. Differences in benthic F P O M were not detected, but the surber sampler used for F P O M collection included a relatively large mesh size (250u.m) which can underestimate benthic organic matter standing crops up to 65% (Minshall et al. 1982) and suspended F P O M was not sampled. One would expect that the increase in leaf and needle biomass found in the streams would increase the retention of benthic F P O M , but consumption by macroinvertebrates and the sampling technique may explain this.  The subgroups of collectors, the gatherers and filterers, responded differently to the seasonal increase in detritus. The gatherers as a group showed greater densities in the fall season, again coinciding with the higher detritus. The gatherer, Microspectra, showed higher densities, but not biomass suggesting that this may be an intraspecific density-dependent response as greater individual biomass was not obtained due to exploitative competition for their food resource or due to a low quality resource. Oligochaetes, also gatherers, showed a stream dependent response in that higher biomass was found in the creeks with the higher number of depositional habitats and the greatest amount of accumulated detritus. There were no specific filterer taxa that showed a response to season, site or stream, but as a group, greater biomass was found in the fall. Larger filterer taxa could have replaced smaller ones before fall sampling began or larger instars of the same taxa may have been sampled. They could also be tracking their food resource if greater amounts of F P O M were produced in the fall and remained in suspension. These results suggest that food limitation of collectors may occur at certain times of the year. t  Although higher chlorophyll a concentrations were found in the summer, the greater density of scrapers were found in the fall. This result was largely due to a heptageniid mayfly, Cinygma. If Cinygma emerged in the spring or early summer, only the eggs or tiny larvae would have been 68  present in the stream during the summer sampling period, resulting in low densities. A n alternate explanation for the high fall densities could be that the competition with other consumers for the periphyton resource may have been released in the fall season if there were a shift to whole leaf detritus by some taxa. This would allow for greater numbers of scrapers even if there were not a greater periphyton food supply. Cinygma could also be an opportunist and may be capable of not only scraping periphyton from the stream bottom, but also consuming other organic matter when it is in great supply. Mihuc and Minshall (1995) found evidence that many lotic invertebrates are trophic generalists and that food switching may occur when resource patterns change.  The river continuum concept predicts that in small, heavily shaded headwater streams the dominant functional groups would consist of shredders and collectors (Vannote et al. 1980). It was also expected that scraper abundance would be low in the forested stream sites. Thus, it was surprising to see that although there were no differences found between the forested and harvested sites, there was a substantial number of scrapers found in the streams in the fall. Dieterich et al. (1997) found the scraper, Ameletus andersoni, to have high densities and biomass in a light limited, temporary stream and suggest that it is unlikely that this mayfly is dependent on algae as it probably ingests bacterial aufwuchs as well. A similar explanation could apply to the large numbers of the mayfly, Cinygma, in this study.  Higher densities and biomass of predators, mainly due to the stonefly Sweltsa, were found in the fall. In general, greater densities and biomass of macroinvertebrates were also found in the fall, suggesting that more prey items or larger, more nutritious prey items were available. Hawkins and Sedell (1980) showed positive relationships between the densities of predators and their prey. Although there were no correlations found between the predators and prey in this study, 69  the much larger predator densities found in the forested sites relative to the harvested sites in the fall could result from having a better food resource in the forested sites. There are many alternate explanations for the high predator densities found in the forested sites in the fall, such as better habitat or environmental conditions, but determining which factors are responsible for the higher densities are beyond the scope of this study.  Management Implications Although differences between the harvested and forested sites in this study did not appear to be large, the RDAs did suggest that there were differences. In both Moffat Creek #2 and Woodjam Creek, the harvested sites were different from the forested sites in both seasons, but for Deerhorn and Sucker creeks, the harvested sites were only different in the fall. This shows that riparian forest harvesting does have an effect on the stream biota and physical environment, but those effects may only be apparent in certain seasons. There was no gradient found between the streams in relation to the degree of riparian modification (i.e. from a small buffer to a cut across the stream) and the magnitude of the differences between the forested and harvested streams seems unrelated to the degree of forestry impact. The ordination diagrams also suggest that any effects were stream dependent as harvested and forested sites within the same stream were clustered closer together than those of different streams.  Many studies have observed that functional recovery of instream processes can occur years before taxonomic recovery following a disturbance (Wallace et al 1986; Lugthart and Wallace 1992; Whiles and Wallace 1992; Whiles et al 1993; Hutchens et al. 1998). Lugthart and Wallace (1992) reported up to a 5-year time interval between disturbance and taxonomic recovery. As this study was conducted in riparian harvested creeks in which the harvesting occurred within the  70  previous 5 years, taxonomic recovery may have already been reached even though the riparian vegetation had yet to recover.  Although this research has revealed information about small B C interior streams, highlighting differences between seasons, the results did not show consistent effects of riparian harvesting. It is difficult to separate the response of the macroinvertebrates to the actual disturbance from the recovery from the physical alterations to the stream that resulted from the disturbance (Hutchens et al 1998). It has been suggested that canopy removal may be a potential management tool to increase the abundance of benthic invertebrates as food for juvenile salmonids in some locations (Hetrick et al 1998). In this study, an increase in periphyton and macroinvertebrate density and biomass, which often follows canopy removal in coastal systems, was not seen. This suggests that factors other than light may limit the primary productivity of the system or that changes in the physical environment following riparian harvesting overwhelm any increases in productivity. In a study conducted in New Zealand, it was shown that not only flood frequency, but sediment instability, greatly increases disturbance intensity for periphyton and thus, degree of bed movement should be addressed when attempting to quantify the effects of disturbance (Biggs et al. 1999). As the interior of B C is characterized by easily erodible, lacustrine soils, it is probable that some sediment would be mobilized even at relatively low discharges. This could abrade against periphyton and increase the sloughing of algae (Peterson 1996). Thus, even if riparian harvesting had the potential to increase primary production due to the decreased canopy cover, an increase in streambank erosion may overwhelm any positive effects.  71  G E N E R A L CONCLUSION Headwater streams (small first- to third-order streams) make up about 85% of the total length of running waters (Leopold et al. 1964). These small streams would be the most greatly affected by riparian harvesting as shading and organic matter inputs are at a maximum because of the high length of bank to stream bottom ratio (Anderson and Sedell 1979). This study reveals that not all small streams respond the same way and generalizations concerning riparian harvesting may not be adequate when determining management guidelines. Effects on the stream biota and physical environment are complex and as this study illustrates, effects are stream dependent and may only be detectable in certain seasons.  An increase in primary production commonly found following riparian harvesting in coastal areas did not occur in some of my study streams or my experiment. This suggests that other factors may be limiting periphyton production in my study area. The experiment showed that light can affect qualitative characteristics of periphyton, but that other limiting factors (e.g. nitrogen) can prevent quantitative changes. Thus, although there were no detectable changes in periphyton quantity in the field study, an increase in light following riparian harvesting may have altered the algal composition. The lack of an algal response to the experimental light manipulations make it difficult to determine the relative importance of algae and detritus as food resources and how differing levels of algae may affect the macroinvertebrate community. Further research should include the examination of how light and other potentially limiting factors (nutrients, herbivores) interact and affect the periphyton community.  72  Both the field sampling and the experiment suggest that detritus is important to secondary production as both shredder and collector macroinvertebrates were shown to be food limited some of the time. In the experiment, the supplementation of leaf litter greatly increased the densities of some detritivores suggesting food limitation. A similar result was found in the field study in that greater detritivore densities were found when the quantity of detrital resources was greatest, which was in the fall season. Thus, detritivore productivity seems to track the productivity of their food supply and any changes to detrital standing stocks in small streams due to forest harvesting may alter secondary production.  The field study gave correlative information relating forest harvesting, stream food resources and macroinvertebrates. By conducting the experiment I was able to determine how differing levels of food resources directly affect the macroinvertebrate community and relate this back to the results from the field study. M y results demonstrate that detritivorous invertebrates are food limited some of the time in the continental area studied. Any changes to detrital standing stocks due to forest harvesting were stream dependent, which could affect macroinvertebrate secondary production. The role of detrital export to downstream reaches and the duration of detrital storage should be investigated to determine how riparian harvesting affects detrital dynamics.  73  LITERATURE CITED  Anderson, N . H . and J.R. Sedell. 1979. Detritus processing by macroinvertebrates in stream ecosystems. Ann. Rev. Entomol. 24: 351-377. Anderson, N.H. 1992. Influence of disturbance on insect communities in Pacific Northwest streams. Hydro bio logia 248: 79-92. B.C. Ministry of Forests (MoF) and Ministry of Environment, Lands and Parks (MELP), 1995. Riparian Management Area Guidebook. Victoria, British Columbia. Behmer, D.J. and C P . Hawkins. 1986. Effects of overhead canopy on macroinvertebrate production in a Utah stream. Freshwater Biol. 16: 287-300. Biggs, B.J.F. 1995. The contribution of flood disturbance, catchment geology and land use to the habitat template of periphyton in stream ecosystems. Freshwater Biol. 33: 419-438. Biggs, B.J.F., R.A. Smith and M.J. Duncan. 1999. Velocity and sediment disturbance of periphyton in headwater streams: biomass and metabolism. J. N . Am. Benthol. Soc. 18: 222241. 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. Bilby, R.E. and P.A. Bisson. 1992. Allochthonous versus autochthonous organic matter contributions to the trophic support of fish populations in clearcut and old-growth forested streams. Can. J. Fish. Aquat. Sci. 49: 540-551. Bisson, P.A. and J.R. Sedell. 1982. Salmonid populations in streams in clearcut vs. old-growth forests of western Washington, p. 121-130. In W.R. Meehan, T.R. Merrelljr., and T.A. Hanley [eds.]. Fish and wildlife relationships in old-growth forests. Amer. Inst. Fish. Res. Biol., Juneau, Alaska. 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. Pages 143-190 in E.O. Salo and T.W. Cundy (Eds.) Streamside Management: Forestry and Fishery Interactions. Contribution 57, Institute of Forest Resources, University of Washington, Seattle, WA. Boston, H.L. and W.R. Hill. 1991. Photosynthesis-light relations of stream periphyton communities. Limnol. Oceanogr. 36: 644-656. Bothwell, M . L . 1988. Growth rate responses of lotic periphytic diatoms to experimental phosphorus enrichment: the influence of temperature and light. Can. J. Fish Aquat. Sci. 45: 261270. Bothwell, M . L . 1989. Phosphorus-limited growth dynamics of lotic periphytic diatom 74  communities: areal biomass and cellular growth rate responses. Canadian Journal of Fisheries and Aquatic Sciences 46: 1293-1301. Bothwell, M . L . , D . M . J. Sherbot and C M . Pollock. 1994. Ecosystem response to solar ultraviolet-B radiation: Influence of trophic-level interactions. Science 265: 97-100. Cummins, K.W. 1974. Structure and function of stream ecosystems. Bioscience 24: 631-641. Cummins, K.W., M . A . Wilzbach, D . M . Gates, J.B. Perry and W.B. Taliaferro. 1989. Shredders and riparian vegetation: Leaf litter that falls into streams influences communities of stream invertebrates. Bioscience 39: 24-30. Davies, P.E. and M . Nelson. 1994. Relationships between riparian buffer widths and the effects of logging on stream habitat, invertebrate community composition and fish abundance. Aust. J. Mar. Freshwater Res. 45:1289-1305. Dieterich, M and N.H. Anderson. 1995. Life cycles and food habits of mayflies and stoneflies from temporary streams in western Oregon. Freshwater Biol. 43: 47-60. Dieterich, M . , N.H. Anderson and T . M . Anderson. 1997. Shredder-collector interactions in temporary streams of western Oregon. Freshwater Biol. 38: 387-393. Feminella, J.W. and C P . Hawkins. 1995. Interactions between stream herbivores and periphyton: a quantitative analysis of past experiments. J. N . Am. Benthol. Soc. 14: 465-509. Fuchs, S. 1999. Responses of macroinvertebrate community composition to changes in stream abiotic factors after streamside clear-cut logging in the central interior of British Columbia. M.Sc. Thesis, University of British Columbia, Vancouver, BC. 125 p. Fuller, R.L., J.L. Roelofs and T.J. Fry. 1986. The importance of algae to stream invertebrates. Journal of the North American Benthological Society 5: 290-296. Gessner, M.O. and M . Dobson. 1993. Colonisation of fresh and dried leaf litter by lotic macroinvertebrates. Arch. Hydrobiol. 127: 141-149. Giroux, P. 1994. Correlations between catchment disturbance, stream chemistry, invertebrate and riparian communities in northeastern temperate forests near Temagami, Ontario: a test of the biomass accumulation model. MSc Thesis, Trent University, O N . Gregory, S.V. 1980. Effects of light, nutrients, and grazing on periphyton communities in streams. Ph.D. Dissertation. Oregon State University, Corvallis, Oregon, USA. Gregory, S.V., G.A. Lamberti, D . C Erman, K . V . Koski, M . L . Murphy and J.R. Sedell. 1987. Influence of Forest Practices on Aquatic Production. Chapter 7 In E.O. Salo and T. Cundy [eds.]. Proceedings of an interdisciplinary symposium on streamside management: Forestry and Fisheries Interactions. University of Washington Press Seattle, W A .  75  Gregory, S.V., F.J. Swanson, W.A. McKee and K.W. Cummins. 1991. An ecosystem perspective of riparian zones. Bioscience 41: 540-551. Gurtz, M.E. and J.B. Wallace. 1984. Substrate-mediated response of stream invertebrates to disturbance. Ecology 65: 1556-1569. Hager, S.W., L.I. Gordon, and P.K. Park. 1968. A practical manual for the use of Technicon Autoanalyzer in seawater nutrient analysis. A final report to B.C.G., Contract 14-17-0001-1759, October 1968, Reference 68-33. Victoria B.C. Hansmann, E.W. and H.K. Phinney. 1973. Effects of logging on periphyton in coastal streams of Oregon. Ecology 54: 194-199. Hartmann, G.F. and J.C. Scrivener. 1990. Impacts of forestry practices on a coastal stream ecosystem, Carnation Creek, BC. Can. Bulletin Fish. Aquat. Sci. 223. Hartmann, G.F., J.C. Scrivener, and M.J. Miles. 1996. Impacts of logging in Carnation Creek, a high-energy coastal stream in British Columbia, and their implication for restoring fish habitat. Can. J. Fish. Aquat. Sci. 53:237-251. Hawkins, C P . and J.R. Sedell. 1980. Longitudinal and seasonal changes in functional organization of macroinvertebrate communities in four Oregon streams. Ecology 61: 387-398. Hawkins, C P . , M . L . Murphy, and N.H. Anderson. 1982. Effects of canopy, substrate composition, and gradient on the structure of macroinvertebrate communities in Cascade Range streams of Oregon. Ecology 63: 1840-1856. Heard, S.B. and J.S. Richardson. 1995. Shredder-collector facilitation in stream detrital food webs: is there enough evidence? Oikos 72: 359-366. Heise, B. 1997. The use of artificial substrates to measure effects of timber harvesting and road construction on stream invertebrates in the Sicamous Creek Watershed: Preliminary report. In Sicamous Creek silvicultural systems project: Workshop proceedings Apr. 24-25, 1996, Kamloops, B.C. C. Hollstedt and A.Vyse [eds.] pp. 272 - 283. Hetrick, N.J., M . A . Brusven, T.C. Bjoran, R . M . Keith and W.R. Meehan. 1998a. Effects of canopy removal on invertebrates and diet of juvenile Coho salmon in a small stream in Southeast Alaska. Trans. Am. Fish. Soc. 127: 876-888. Hill, W.R. and B.C. Harvey. 1990. Periphyton responses to higher trophic levels and light in a shaded stream. Can. J. Fish. Aquat. Sci. 47: 2307-2314. Hill, W.R. and A.W. Knight. 1988. Nutrient and light limitation of algae in two northern California streams. J. Phycol. 24: 125-132. Hill, W.R., M . G . Ryon and E . M . Schilling. 1995. Light limitation in a stream ecosystem: responses by primary producers and consumers. Ecology 76: 1297-1309.  76  Hutchens, J.JJr., K.Chung and J.B. Wallace. 1998. Temporal variability of stream macroinvertebrate abundance and biomass following pesticide disturbance. J. N . Am. Benthol. Soc. 17: 518-534. Johnson, S.L. and A.P. Covich. 1997. Scales of observation of riparian forests and distributions of suspended detritus in a prairie river. Freshwater Biol. 37: 163-175. Kim, M . A. 1999. The influence of light and nutrients on interactions between a tadpole grazer and periphyton in two coastal streams. M.Sc. Thesis, University of British Columbia, Vancouver, B C . 74 p. Lamberti, G.A., L.R. Ashkenas, S.V. Gregory and A.D. Steinman. 1987. Effects of three herbivores on periphyton communities in laboratory streams. J. N. Am. Benthol. Soc. 6: 92-104. Lamberti, G.A., S.V. Gregory, L . R. Ashkenas, J.L. L i , A.D. Steinman and C D . Mclntire. 1995. Influence of grazer type and abundance on plant-herbivore interactions in streams. Hydrobiol. 306: 179-188. Larkin, G.A., P.A. Slaney, P. Warburton and A.S. Wilson. 1998. Suspended sediment and fish habitat sedimentation in central interior watersheds of British Columbia. Province of British Columbia, Ministry of Environment, Lands and Parks, and Ministry of Forests. Watershed Restoration Management Report No. 7:31p. Lugthart, G.J. and J.B. Wallace. 1992. Effects of disturbance on benthic functional structure and production in mountain streams. J. N . Am. Benthol. Soc. 11:138-164. Lyford, J.H. Jr., and S.V. Gregory. 1975. The dynamics and structure of periphyton communities in three Cascade Mountain streams. Internationale Vereinigung fur Theoretische und Angewandte Limnologie: Verhandlungen 19: 1610-1616. McArthur, J.V. J.R. Barnes, B.J. Hanson and L.G. Leff. 1988. Seasonal dynamics of leaf litter breakdown in a Utah alpine stream. J. N . Am. Benthol. Soc. 7: 44-50. Mackay, R.J. and J. Kalff. 1972. Ecology of two related species of caddis fly larvae in theorganic substrates of a woodland stream. Ecology 54: 499-511. Mattingly, R.L. 1987. Handling of coarse and fine particulate organic matter by the aquatic insects iParaleptophlebia gregalis and iP. temporalis (Ephemeroptera: Leptophlebiidae). Freshwater Biology 18: 255-265. Meidinger, D. and J. Pojar. [Eds.] 1991. Ecosystems of British Columbia. B.C. Ministry of Forests. Merritt, R.W. and K.W. Cummins. 1996a. An introduction to the aquatic insects of North America. Third Edition. Kendall/Hunt Publishing Co., Dubuque, Iowa. 862pp. Merritt R.W. and K.W. Cummins. 1996&. Trophic relations of macroinvertebrates. p. 453-474 in F.R. Hauer and G.A. Lamberti (eds.). Methods in stream ecology. AcademicPress, San Diego. 77  Mihuc, T.B. and G.W. Minshall. 1995. Trophic generalists vs. trophic specialists: implications for food web dynamics in post-fire streams. Ecology 76: 2361-2372. Minshall, G.W., J.T. Brock, and T.W. LaPoint. 1982. Characterization and dynamics of benthic organic matter and invertebrate functional feeding group relationships in the Upper Salmon River, Idaho (U.S.A.). International Revue Gesamten Hydrobiologie 67: 793-820. Montgomery, D., J. Buffington, R. Smith, K . Schmidt and G. Pess. 1995. Pool spacing in forest channels. Water Resources Research 31:1097-1105. Mundie, J.H., Simpson, K.S. and Perrin, C.J. 1991. Responses of stream periphyton and benthic insects to increase in dissolved inorganic phosphorus in a mesocosm. Can. J. Fish. Aquat. Sci. 48: 2061-2072. Murphy, M . L . and J.D. Hall. 1981. Varied effects of clearcut logging on predators and Their habitat in small streams of the Cascade mountains, Oregon. Can. J. Fish. Aquat. Sci. 38: 137-145. Murphy, M . L . , C P . Hawkins, and N.H. Anderson. 1981. Effects of canopy modification and accumulated sediment on stream communities. Trans. Am. Fish. Soc. 110: 469-478. Murphy, M . L . 1984. Primary production and grazing in freshwater and intertidal reaches of a coastal stream, southeast Alaska. Limnol. Oceanogr. 29: 805-815. Newbold, J.D., Erman, D.D. and Roby K.B. 1980. Effects of logging on macroinvertebrates in streams with and without buffer strips. Can. J. Fish. Aquat. Sci. 37: 1076-1085. Noel, D.S., C.W. Martin and C A . Federer. 1986. Effects of forest clearcutting in New England on stream macroinvertebrates and periphyton. Environmental Management 10: 661-670. Palmer, M.W. 1993. Putting things in even better order: the advantages of canonical correspondence analysis (CCA). Ecol. 74: 2215-2230. Perneger, T.V. 1998. What's wrong with Bonferroni adjustments. British Medical Journal 316: 1236-1238. Perrin, C.J., M . L . Bothwell, and P.A. Slaney. 1987. Experimental enrichment of a coastal stream in British Columbia: effects of organic and inorganic additions on autotrophic periphyton production. Can. J. Fish. Aquat. Sci. 44: 1247-1256. Perrin, C J . and J.S. Richardson. 1997. N and P limitation of benthos abundance in the Nechako River, British Columbia. Can. J. Fish. Aquat. Sci. 54: 2574-2583. Peterson, B.J. and 16 co-authors. 1993. Biological responses of a tundra river to fertilization. Ecology 74: 653-672.  78  Prescott, C.E., J.P. Corbin and D. Parkinson. 1989. Input, accumulation, and residence times of carbon, nitrogen, and phosphorus in four Rocky Mountain coniferous forests. Canadian Journal of Forest Research 19: 489-498. Raven, J.A. 1992. How benthic macroalgae cope with flowing freshwater: Resource acquisition and retention. Journal of Phycology 28: 133-146. Richardson, J.S. 1991. Seasonal food limitation of detritivores in a montane stream: an experimental test. Ecology 72: 873-887. Richardson, J.S. and W.E. Neill. 1991. Indirect effects of detritus manipulations in a montane stream. Canadian Journal of Fisheries and Aquatic Sciences 48: 776-783. Richardson, J.S. 1992. Coarse particulate detritus dynamics in small, montane streams of southwestern British Columbia. Can. J. Fish. Aquat. Sci. 49: 337-346. Richardson, J.S. 1993. Limits to productivity in streams: evidence from studies on macroinvertebrates, p.9-15 In R.J. Gibson and R.E. Cutting [ed.] Production of juvenile Atlantic salmon, Salmo salar, in natural waters. Can. Spec. Publ. Fish. Aquat. Sci. 118. Richardson, J.S. 1994. Forest-stream interactions: potential effects on productivity and biodiversity of benthic communities in interior British Columbia, pp.9-14 in Proceedings of the Takla Fisheries/Forestry Workshop: a two year review. April 1, 1993, Prince George, B.C. Can. Tech. Rep. Fish. Aquat. Sci. 2007: 104p. Rosemond, A . D . , P.J. Mulhulland and J.W. Elwood. 1993. Top-down and bottom-up control of stream periphyton: effects of nutrients and herbivores. Ecology 74: 12641280. Rosemond, A . D . 1993. Interactions among irradiance, nutrients, and herbivores constrain a stream algal community. Oecologia 94: 585-594. Rothacher, J. 1970. Regimes of streamflow and their modification by logging in Proceedings of a Symposium Forest land Uses and Stream Environment. Krygier, J.T. and Hall. J.D. Eds. October 19-21,1970. SAS Institute. 1996. SAS® Procedures guide. Release 6.1.2. Cary, North Carolina, USA. 705 pp. Schindler, D.W. 1977. The evolution of phosphorus limitation in lakes. Science 195:260-262. Silsbee, D.G. and G.L. Larson. 1983. A comparison of streams in logged and unlogged areas of Great Smoky Mountains National Park. Hydrobiologia 102: 99-111. Sinclair, A . R . E . and R.P. Pech. 1996. Density dependence, stochasticity, compensation and predator regulation. Oikos 75:164-173.  79  Stockner, J.G. and K.R.S. Shortreed. 1976. Autotrophic production in Carnation Dreek, a coastal rainforest stream on Vancouver Island, British Columbia. Journal of the Fisheries Research Board of Canada 33: 1553-1563. Stout, B . M . H I , E.F. Benfield and J.R. Webster. 1993. Effects of a forest disturbance on shredder production in southern Appalachian headwater streams. Freshwater Biol. 29: 59-69. Strickland, J.D.H. and T.R. Parsons. 1972. Pigment Analysis: Spectrophotometric determination of chlorophylls and total carfotenoids. Section IV.3.I In A practical handbook of seawater analysis. Buletin 167 (2 Ed.) Fisheries Research Board of Canada. Ottawa. nd  Sweeney, B.W. 1993. Effects of Streamside Vegetation on Macroinvertebrate Communities of White Clay Creek in Eastern North America. Proc. Acad. Nat. Sci. Philadelphia. 144: 291-340. ter Braak, C.J.F. 1994. Canonical community ordination. Part I: Basic theory and linear methods. Ecoscience 1: 127-140. Triska, F.J., J.R. Sedell and B. Buckley. 1975. The processing of conifer and hardwood leaves in two coniferous forest streams: II. Biogeochemical and nutrient changes. Verh. Internat. Verein. Limnol. 19: 1628-1639. Ulrich, K.E., T . M . Burton and M.P. Oemke. 1993. Effects of whole-tree harvest on Epilithic algal communities in headwater streams. J. Freshwater Ecol. 8: 83-92. Vannote, R.L., G.W. Minshall, K.W. Cummins, J.R. Sedell and C E . Cushing. 1980. The river continuum concept. Can. J. Fish. Aquat. Sci. 37: 130-137. Vuori, K. and I. Joensuu. 1996. Impact of forest drainage on the macroinvertebrates of a Small boreal headwater stream: Do buffer zones protect lotic biodiversity? Biol. Conserv. 77:87-95. Wallace, J.B. and M . E . Gurtz. 1986. Response of Baetis mayflies to catchment logging. Amer. Midi. Nat. 115: 25-41. Wallace, J.B., D.S. Vogel and T.F. Cuffney. 1986. Recovery of a headwater streamfrom an insecticide-induced disturbance. J. N . Am. Benthol. Soc. 5:115-126. Wallace, J. B., S. L . Eggert, J. L . Meyer, J. R. Webster. 1999. Effects of resource limitation on a detrital-based ecosystem. Ecological Monographs 69: 409-442. Webster, J.R., M.E. Gurtz, J.J. Hains, J.L. Meyer, W.T. Swank, J.B. Wallace and J.B. Waide. 1983. Stability of stream ecosystems, p. 355-395. In Barnes, J.R. and G.W. Minshall [ed.]. Stream Ecology: Applications and testing of general ecological theory. Plenum Press, N . Y . . Webster, J.R., S.W. Golladay, E.F. Benfield, D.J. D'Angelo and G.T. Peters. 1990. Effects of forest disturbance on particulate organic matter budgets of small streams. J. N . Am. Benthol. Soc. 9: 120-140.  80  Wellnitz, T. A., R.B. Rader and J.V. Ward. 1996a. Importance of light and nutrients in structuring an algal community in a Rocky Mountain stream. J. Freshwater Ecol. 11:399-413. Wellnitz, T.A., R.B. Rader, and J.B. Ward. 1996£>. Light and a grazing mayfly shape periphyton in a Rocky Mountain stream. J. N . Am. Benthol. Soc. 15: 496-507. Whiles, M.R. and J.B. Wallace. 1992. First year benthic recovery of a headwater stream following a 3-year insecticide-induced disturbance. Freshwater Biol. 28: 81-91. Whiles, M.R., J.B. Wallace and Keun Chung. 1993. The influence of Lepidostoma (Trichoptera: Lepidostomatidae) on recovery of leaf-litter processing in disturbed headwater streams. Am. Midi. Nat. 130: 356-363. Wood, E.D., F.A. J. Armstrong, and F.A. Richards. 1967. Determination of nitrate in seawater by cadmium-copper reduction to nitrite. J. Mar. Biol. Assn.: 47: 23-31. Wotton, R.S. 1984. The importance of identifying the origin of microfine particles in aquatic systems. Oikos 43:217-221. Zar, J.H. 1984. Biostatistical analysis. Second edition. Prentice Hall, Englewood Cliffs, New Jersey.  81  APPENDICES  82  APPENDIX I.  An example calculation for determining the weighted Euclidean distance between sample scores from a Redundancy Analysis. % Forested O Harvested ®  Y 1  I  X  2  Mean Forested  2  T Eigenvalue X-axis = Ex  Eigenvalue Y-axis = E  y  Distance between the 2 forested sites: Z\ = Ex (Xi ) + E (Yi ) 2  2  y  Distance between the mean forested site and the harvested site: Z2 = E ( X ) + E (Y ) 2  2  x  83  2  2  y  2  

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.831.1-0089578/manifest

Comment

Related Items