«.‘mm a

 

I

#435

5:

:4:

ya "'
m

l'

5.; 5i f

n
w
I

‘1

‘e'
’nL

 

 

 

 

THfiIS

D
«AI:
(I!

733 W

This is to certify that the
dissertation entitled

INFLUENCE OF MARINE-DERIVED NUTRIENTS FROM
SPAWNING SALMON ON AQUATIC INSECT COMMUNITIES IN
SOUTH-EAST ALASKAN STREAMS

presented by

JoAnna Lynn Lessard

has been accepted towards fulfillment
of the requirements for the

degree m The Department of Entomology

 

 

 

%¢%

Major Wofesso 8 Signature

7/2 %%

 

Date

MSU is an Affimiative Action/Equal Opportunity Institution

LIBRARY l

MIChIQan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c:/CIRC/Dateoue.p65-p.15

 

INFLUENCE OF MARINE-DERIVED NUTRIENT S FROM SPAWNING SALMON
ON AQUATIC INSECT COMMUNITIES IN SOUTH-EAST ALASKAN STREAMS

By

JoAnna Lynn Lessard

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Entomology

2004

ABSTRACT

INFLUENCE OF MARINE-DERIVED NUTRIENTS FROM SPAWNING SALMON
ON AQUATIC INSECT COMMUNITIES IN SOUTH—EAST ALASKAN STREAMS

By
J oAnna Lynn Lessard

Alaska is the last region along the Pacific coast of the United States that still
enjoys large runs of spawning salmon. Salmon runs provide these oligotrophic systems
with a huge pulse of nutrients from the ocean. The retention of these nutrients in streams
potentially sustain this pulse of nutrients over many months. Annual nutrient delivery
from salmon to Alaskan streams range into the millions of tons of carbon, nitrogen,
phosphorous. These nutrients are termed marine-derived nutrients (MDN) and may be
tightly linked to secondary production in streams. The most important link between
MDN and production is in the production of juvenile salmon. In providing a positive
feedback mechanism to ensure viability of subsequent generations, intermediate steps
that connect MDN and juvenile salmon production include alterations of dissolved
nutrients, biofilm production and alterations to the macroinvertebrate community.
Changes due to MDN have far reaching implications in watersheds that receive salmon
and may shape the dynamics of stream communities seasonally.

This research focused on aquatic insect community responses to MDN in multiple
natural stream systems in southeast Alaska. The objectives were quantify insect
abundance, biomass, functional group proportions, richness, diversity, growth and
secondary production in relation to MDN inputs. These community attributes were
studied to establish if stream communities with exposure to MDN are organized

differently from non-anadromous streams in that region. This research will help to

elucidate the effect that MDN has on stream insect communities and will also help to
better understand the link between MDN and fish production.

Mayfly standing stock and secondary production was higher in control reaches.
Chironomid production was higher in stream sections that do receive large annual runs of
salmon. Richness, diversity and functional group composition was not impacted by
MDN over most of the year. These results suggest that the disturbance of salmon
spawning creates a dichotomy of response to MDN so that larger, univoltine taxa have
lower population levels in salmon spawning areas. Smaller, multivoltine taxa, however,
have higher populations levels, most likely due to their ability to recover and respond to
MDN. The links between MDN and juvenile salmon production may be more simplified

than many models predict.

Copyright by
JOANNA LYNN LESSARD
2004

This dissertation is dedicated to my parents, Kathy and Joe Lessard, who always
celebrate my victories and help me through my disappointments.

ACKNOWLEDGMENTS

I would like to thank all my friends and family for their love and support during
this research program. To my family, Mom, Dad, Jennifer, and Joe thanks for
understanding when I couldn’t always be around because of research trips and school and
thanks for your constant support and enthusiasm for my work. Thanks to Tom Passow
for giving me a great place to write, things to smile about when writing got tough and
understanding when my time was limited. Thanks to my labmates Mollie McIntosh,
Ryan Kimbirouskas, Ozvaldo Hernandez, Kelly Wessell, Christian Lesage, Eric Benbow
and Matt Wesener for their advice and support, and of course all the laughs. The
following people were instrumental in the field and laboratory and I thank them for all
their hard work and help: Todd White, Leia Watkins, Dusty Tazaaler, Holly Campbell,
Matt Wesener, Ryan Kimbirouskas, Christian Lesage, Kelly Wessell, Jessica Mistak, and
Eric Benbow. This research was a part of a large collaborative project and I would like to
thank the entire “MDN Crew” for their ideas, help and all the memories I have of them in
Alaska. Thanks to Dr. Dom Chaloner, Nicole Mitchell, Brittany Graham, and John
Hudson as well as Drs. Rich Merritt, Gary Lamberti, Peggy Ostrom, Rick Edwards, and
Mark Wipfli (The “MDN Crew”). Thank you to Dr. Marty Berg, Dr. Robert Waltz, Dr.
Peter Adler, Dr. Greg Courtney, and Dr. Ken Stewart for their advice and assistance with
taxonomic identifications.

I would like to especially thank my committee members for their dedication and
assistance in helping me be successful in my research program above and beyond the call
of duty. Thanks to Dr. Daniel Hayes for all your advice on the design and analysis

portion of this project. Thanks to Dr. Mike Kaufman for your advice, humor and of

vi

course the field help. Thanks to Dr. Ken Cummins for all of your time and effort on the
planning, executing and writing of this research. It has been an honor to work with you.
Thanks to Dr. George Bird for jumping in and helping out at a moments notice.

I need to say a very special thank you to my advisor Dr. Richard Merritt. Thank
you for this opportunity. All of us in your lab know how lucky we are to have such a
generous, fun and supportive mentor, and it has been amazing to be part of it. Doing

research in Alaska was like a dream and I thank you for making it possible.

vii

TABLE OF CONTENTS

LIST OF TABLES ............................................................................... viii

LIST OF FIGURES .................................................................................. x

CHAPTER 1

LITERATURE REVIEW AND RESEARCH INTRODUCTION ............................ 1
Introduction ................................................................................... 1
Study Sites .................................................................................... 7
Chapter Introductions ........................................................................ 9
Literature Cited ............................................................................. 12

CHAPTER 2

INFLUENCE OF MARINE-DERIVED NUTRIENT S FROM SPAWNING SAIMON
ON AQUATIC INSECT COMMUNITY PATTERNS IN STREAM SYSTEMS IN

SOUTH-EAST ALASKA ......................................................................... 14
Abstract ...................................................................................... 14
Introduction ................................................................................. 18
Study Sites .................................................................................. 20
Methods ..................................................................................... 23
Data Analysis ............................................................................... 24
Results ....................................................................................... 25
Discussion ................................................................................... 39
Literature Cited ............................................................................. 45

CHAPTER 3 ’

SECONDARY PRODUCTION OF MAYFLIES AND MIDGES IN RESPONSE TO

SPAWNING SALMON IN NATURAL ALASKAN STREAMS ........................... 48
Abstract ...................................................................................... 48
Introduction ................................................................................. 50
Methods ...................................................................................... 53
Results ....................................................................................... 55
Discussion ................................................................................... 62
Literature Cited ............................................................................. 68

CHAPTER 4

SPRING GROWTH OF CADDISFLIES (LIMNEPIIILIDAE: TRICHOPTERA) IN
RESPONSE TO MARINE-DERIVED NUTRIENTS AND FOOD TYPE IN A SOUTH-

EAST ALANKAN STREAM ..................................................................... 72
Abstract ...................................................................................... 72
Introduction ................................................................................. 74
Methods and materials ..................................................................... 74
Results ....................................................................................... 80
Discussion ................................................................................... 89
Conclusion .................................................................................. 92

viii

Literature Cited ............................................................................. 95

APPENDIX A ...................................................................................... 100

APPENDIX B ...................................................................................... 106

ix

LIST OF TABLES
CHAPTER 2
Table 1. Location, spawning run size and physical habitat data for the seven study

streams ................................................................................................ 22

Table 2. Results (p-values) of the repeated mixed modeling analysis by order. Period

and section are abbreviated in the 3-way interaction term .................................... 26

Table 3. Dominant families in each order by stream section (upstream and downstream)
and period (pre, during and post spawning). Percent of total density and biomass are

given for each family ............................................................................... 30

Table 4. Results (p-values) of the repeated mixed modeling analyses by family. Period

and section are abbreviated in the 3-way interaction term .................................... 31

CHAPTER 3
Table 1. List of taxa studies for secondary production. Taxa used were common in both

study streams ......................................................................................... 52

Table 2. Habitat data for the study streams by section. Mean temperatures are in

parentheses ........................................................................................... 54

Table 3. Standing stock biomass (dry mass), secondary production and maximum larval
body lengths for five mayfly genera and Chironomid midges from Fish Creek and Salmon
Creek. Values are given for each stream reach (downstream and upstream). Percent
difference is the difference between secondary production between the stream reaches
with positive values corresponding to higher production downstream and negative values

to higher production upstream ..................................................................... 56

Table 4. Dominant genera and the proportion of Chironomid subfamilies or tribes in each

stream and stream section (downstream or upstream of the waterfall barrier) .............. 59

xi

LIST OF FIGURES
CHAPTER 2
Figure 1. Map of southeast Alaska. Study streams were near J uneau-Douglas and on

Prince of Wales Island .............................................................................. 21

Figure 2. Natural log- transformed density means for each aquatic insect order by stream
section and run size for each period (A=pre, B=during, and C=post spawning).
Bars are: high run, downstream ( E), high run, upstream ( CI), low run, downstream ( I) and

low run, upstream (I). An * denotes significant differences by stream section within run size.
......................................................................................................... 27

Figure 3. Natural log- transformed biomass means for each aquatic insect order by stream
section and run size for each period (A=pre, B=during, and C=post spawning).
Bars are: high run, downstream ( a ), high run, upstream ( D ), low run, downstream ( I) and

low run, upstream (I). An * denotes significant differences by stream section within run size.
......................................................................................................... 28

Figure 4. Natural log- transformed density means for each aquatic insect family, dominant
in each order, by stream section and run size for each period (A=pre, B=during, and

=post spawning). Bars are: high run, downstream ( I), high run, upstream (CI ),
low run, downstream (I) and low run, upstream (I). An * denotes significant differences

by stream section within run size.

......................................................................................................... 32

xii

Figure 5. Natural log- transformed biomass means for each dominant aquatic insect family,
by stream section and run size for each period (A=pre, B=during, and C=post spawning).
Bars are: high run, downstream (I ), high run, upstream (CI ), low run, downstream ( I) and

low run, upstream). An * denotes significant differences by stream section within run size.

....................................................................................................... 33

Figure 6. Natural log— transformed density means for each functional feeding group, by
stream section and run size for each period (A=pre, B=during, and C=post spawning).
Bars are: high run, downstream ( I), high run, upstream (D ), low run, downstream ( I)

and low run, upstream (I). An * denotes significant differences by section within run size.

....................................................................................................... 34

Figure 7. Natural log- transformed biomass means for each functional feeding group, by
stream section and run size for each period (A=pre, B=during, and C=post spawning).
Bars are: high run, downstream ( J“ ), high run, upstream (CI ), low run, downstream ( I)

and low run, upstream (I). An * denotes significant differences by section within run size.
....................................................................................................... 37

Figure 8. Mean taxa richness (A) and Shannon-Weiner diversity (B) by stream section,
averaged across run size for each sample period (pre, during and post spawning).
Bars are: high run, downstream ( fl ), high run, upstream (D ), low run, downstream (I ) and

low run, upstream). An * denotes significant differences by stream section within run size.

....................................................................................................... 38

xiii

CHAPTER 3
Figure 1. Secondary production of five mayfly genera for upstream (blocked from

salmon) and downstream (open to salmon) sections of Fish Creek and Salmon Creek...57

Figure 2. Secondary production of Chironomidae for upstream (blocked from salmon)

and downstream (open to salmon) sections of Fish Creek and Salmon Creek .............. 58

Figure 3. Standing stock biomass of mayfly families across sample dates. Means (+/- SE)
for upstream (El ) and downstream (O ) sections for Baetidae (A), Ephemerellidae (B) and
Heptageniidae (C) are shown averaged across both study streams. The vertical dotted line

represents when the spawning run began in downstream sections.

......................................................................................................... 60
Figure 4. Standing stock biomass of chironomids across sample dates. Means (+/- SE)
for upstream (El ) and downstream (O ) sections are shown averaged across both study
streams. The vertical dotted line represents when the spawning run began in downstream sections.
......................................................................................... 61

CHAPTER 4

Figure 1. Schematic of growth experiment conducted in the Harris River. Identical set-
ups were used in upstream and downstream sections. Box cartoons show the
combinations of food type (conditioned alder leaves and rocks from the stream) and taxa

that were put in each box, and replicated three times .......................................... 78

Figure 2. Mean dry weight mg (+/- SE) of caddisflies from the in-stream samples (Free)

initially and at the end of the experiment, and from caddisflies confined to the boxes

xiv

divided by stream section (upst1=above the waterfall, downstr=below the waterfall) and
food type (leaves and rocks). MDN and growth box effects were determined by
comparing free and boxed individuals by section (significant differences indicated by
different letters) and food effects were determined by comparing food types (significantly
higher values indicated by *). All comparisons were done using the Mann-Whitney U

test with bonferroni adjustments made for multiple comparisons when appropriate. ..... 81

Figure 3. Mean % relative growth rate/ degree day (° C) (+/- SE) for caddisflies grown in
growth boxes averaged by food type and stream section. An * indicates significantly

greater %RGR from the Mann-Whitney U tests ................................................ 84

Figure 4. Mean % survival (+/- SE) of caddisflies in the growth boxes at the end of the
experiment by stream section (averaged across food type). Survival at the end of the
experiment is also shown by food type in each section (Leaves and Rocks). Significant
differences between stream sections (i.e., each pair of bars), determined using the Mann-

Whitney test, are indicated by an *. .............................................................. 85

Figure 5. Mean % relative growth rate/ degree day (° C) versus % Mortality in the
growth boxes for each caddisfly. The lines are the results of the regressions and the
symbols are from the data for each taxa. Adjusted R2 values for each model are given in

the legend ............................................................................................. 87

Figure 6. Wet length-dry weight relationships for the three caddisflies. Lines are the
results of the regression equations shown to the right of each graph and diamonds are the

individual data points from in-stream samples of each taxa ................................... 88

XV

Chapter 1
Literature Review and Research Introduction

Introduction

Nutrient transfers along streams and rivers from headwaters to the mouth are a
well-studied phenomenon. The utilization of terrestrially derived nutrients (e. g., nitrates,
phosphorous, etc.) from run-off, erosion, and riparian litter are considered to be the
drivers of stream productivity. Stream communities are hypothesized to be organized
spatially (Cummins 1974, Vannote et al. 1980) and temporally (Kaushik and Hynes 1971,
Cummins et al. 1989) to capitalize on the predictable influxes of these terrestrial
nutrients.

Coastal stream that serve as spawning ground for anadromous salmon have an
additional nutrient source that may cause organizational patterns that differ from streams
that rely solely on nutrients from the watershed. The life history of salmon is fairly
complicated and varies among the different species. The general scheme is for a salmon
to be semelparous (die after spawning) spending part of its early life growing in natal
streams before undergoing physiological changes needed to deal with salt water (i.e.,
smoltification). Smolts then migrate to the ocean where they grow to adult size by
feeding on marine-based nutrients (e. g., marine plankton and fish) until finally returning
to the stream where they were born to spawn and die. It is theorized that salmon evolved
anadromy around 25 million years ago to take advantage of newly cooled and productive
oceans where they could grow larger than if they remained in their comparatively un-
productive natal streams (Lichatowich 1999). By completing this circle, the salmon

create a nutrient pulse in the form of their excretion, eggs, sperm and carcasses that

travels in the opposite direction of normal river continuum theory (i.e. Vannote et al.
1980).

The role that these marine-derived nutrients (MDN) play in coastal systems has
been the object of study in recent years. It is of particular interest in the Pacific
Northwest of the United States, where salmon runs are extinct or threatened in many
streams along the coasts of Washington, Oregon and northern California (Lichatowich
1999, Gresh et a1. 2000). It has been hypothesized that the salmon provide an essential
nutrient source in these normally oligotrophic Northwestern systems and, by subsidizing
the nutrient base in their spawning grounds, increase the viability and production of their
own offspring fostering future generations of salmon (Kline et al. 1997, Lichatowich
1999). Without enough adults returning to spawn in these streams, salmon populations
may spiral into extinction and the entire stream community may lose a nutrient source
that it has relied on for thousands of years. Reversing this trend may be the key to saving
salmon in the Pacific Northwest (Bilby et al. 2000, Stockner et al. 2000).

Alaska is one of the few areas in the United States where salmon runs remain at or
near historic levels (Baker et al. 1996, Gresh et al. 2000). Southeast Alaska contains the
8.5 million hectare Tongass National Forest, most of which is pristine forest surrounding
5,200 anadromous salmon streams (Halupka et al. 1999). These streams collectively
support hundreds of millions of spawning salmon that annually transport millions of
kilograms of carbon, nitrogen, phosphorous, and other nutrients to freshwater streams
(Larkin and Slaney 1997, Halupka et al. 1999). This represents a considerable nutrient
load for one region when compared to streams in Washington, Oregon and California,

which collectively receive only 11.8-13.7 million kg of salmon annually (i.e., 360,000-

418,000 kg N and 43,000-49,000 kg P) (Gresh et al. 2000). Because the runs in Alaska
are still largely intact, this region provides the opportunity to study the structure and
function of stream systems with the salmon runs in a relatively pristine state. Research
on the role of MDN in stream ecosystems may elucidate conservation measures that
should be taken and also help direct restoration attempts in areas where salmon are
threatened.

Some impacts of salmon runs on stream communities have been investigated in
recent years, and these studies have revealed varying responses to MDN enrichment
(Cederholm et al. 1999). Some of this variation may be due to variability in the retention
of salmon carcasses in streams due to differences in flow rates and abundance of debris
jams and wood in stream channels (Cederholm and Peterson 1985, Cederholm et al.
1989). While there is variation in the level of response of stream communities to MDN
enrichment, several patterns of community responses are evident.

Increasing the production of lower trophic levels is an important enrichment
mechanism, because this provides the basis for higher production throughout the stream
community. Several studies have shown increased production of biofilm (i.e. mixed
assemblage of autotrophic and heterotrophic microbes set in a glycoprotein
polysaccharide medium attached to stream substrates) in the presence of MDN (Schuldt
and Hershey 1995, Cederholm et al. 1999, Peterson and Foote 2000). In Southeast
Alaska, Wipfli et al. (1998) found biofilm production to be 15 times higher in a MDN
enriched stream section compared to a non-anadromous upstream control section. Stable

isotope research on streams in Washington showed that biofilm in the presence of MDN

obtained up to 30% of nitrogen and 26.6% of carbon from salmon carcasses,
demonstrating the utilization of MDN by biofilm in these systems (Bilby et al. 1996).

The subsequent consumption of MDN enriched biofilm by invertebrate scrapers
and collectors is one way that MDN is transferred to the next trophic level. The
increased abundance of biofilm has been shown to coincide with an increase in
invertebrate abundance in enriched stream sections (W ipfli et al. 1998) and the
disproportionate incorporation of MDN into invertebrate grazers (Schuldt and Hershey
1995). Beyond the indirect enrichment of invertebrates from feeding on biofilm and
other insects that are enriched in MDN (Bilby et al. 1996), there is also the potential for
direct enrichment. Several researchers have found insects associated with salmon
carcasses themselves and also have found evidence for the direct consumption of carcass
tissue by aquatic insects (Piorkowski 1995, Kline et al. 1997, Bilby et al. 2000, Merritt
and Wallace 2001). While insects may feed on salmon flesh and benefit from these
nutrients, it is unclear if the salmon flesh is the attractant or the associated microbes and
fungi growing on the carcasses (Minakawa 1997). Most likely the salmon carcasses play
a dual role as substrate and concentrated food resource directly and indirectly, and as
such may be important ephemeral habitats.

Whatever the mechanism of enrichment (direct or indirect), invertebrate
communities have been shown to respond to salmon runs in some important ways.
Among these responses are short-term reductions in abundance and increased drift due to
the high level of benthic disturbance created during large spawning runs (Minakawa
1997, Peterson and Foote 2000). As discussed earlier, increased insect density following

salmon carcass decomposition has been documented (Minikawa 1997, Wipfli et al. 1998,

Kline et al. 1997), as well as indications that insect richness and diversity may increase
from salmon enrichment in central Alaskan streams (Piorkowski 1995). There is also
evidence that growth rates of certain taxa increase in the direct presence of salmon tissue
(Minakawa 1997, Chaloner and Wipfli 2002). Additionally, the source of nutrients for
invertebrate biomass has been traced using stable isotope analysis and these studies
reveal that MDN is an important contributor to the nutrition of many of the functional
feeding groups of insects in stream systems (Schuldt and Hershey 1995, Bilby et al.
1996).

Macroinvertebrate communities often show temporal and spatial organization in
response to nutrient sources. Stream shredders emerge and grow at times which take
advantage of autumnal leaf fall and the associated microbial community in temperate
regions (Kaushik and Hynes 1971, Anderson and Cummins 1979, Cummins et al. 1989).
The vast amount of research that has been done on this phenomenon has concluded that
shredder communities have evolved with the predictable influx of leaves and their
subsequent conditioning, and that these factors drove the life history of shredders
(Anderson and Cummins 1979,Cumrnins and Klug 1979). Different populations of the
same insect species also have been shown to vary their growth rates in response to
different nutrient levels in streams in the same geographic area (Anderson and Cummins
1979). Filtering insects often congregate below lake outlets capitalizing on the high
quality seston in these areas (Wallace and Merritt 1980, Herlong and Mallin 1985,
Richardson and Mackay 1991). Because of the plasticity of insect communities to mold
themselves according to the productivity of the system, and the long historical

relationship that exists between marine enrichment and stream communities in Alaska, it

is possible that aquatic invertebrates in MDN enriched streams have systematically
different communities from un-enriched streams in the same region.

MDN enrichment may allow these cold, oligotrophic streams to sustain a greater
diversity and abundance of insect taxa compared to streams without salmon. The seasonal
presence of large numbers of spawning salmon and carcasses may alter the diversity and
abundance of taxa found in MDN enriched reaches on a seasonal basis as well. This
could happen as invertebrate communities re-structure themselves first around the
disturbance of spawning and then around the additional substrate and food resources
created by the carcasses themselves. Insects may also exhibit different growth rates in
enriched streams as the production of primary producers and microbes increase, which
would lead to greater condition, survival, fecundity and abundance of these insects. This
increase in diversity and abundance from MDN may provide the basis for increased fish
production in these streams. Fish communities, and in particular juvenile salmon, may
benefit initially from increased food during spawning in the form of salmon eggs and
increased invertebrate drift and then all year as invertebrate production remains elevated
due to the enrichment. The entire stream community may be structured seasonally
around the salmon spawning run, which will provide insight into another mechanism for
trophic linkages in stream ecology.

The objectives of this research were to:
1. Evaluate the effect of MDN on seasonal patters of aquatic insect abundance and
biomass in southeast Alaskan streams.
2. Evaluate the effect of MDN on seasonal patters of aquatic insect diversity and

richness in southeast Alaskan streams.

3. Evaluate the effect of MDN on annual secondary production of selected insect
taxa in southeast Alaskan streams.
4. Evaluate the effect of spring carry-over of MDN on growth rates of selected
insect taxa in southeast Alaskan streams.
I hypothesized, that natural stream systems that have a long history of natural MDN
enrichment via salmon spawners will:
1) Have aquatic insect communities with greater standing stock abundance and biomass
than streams that do not receive MDN.
2) Have aquatic insect communities with greater diversity and richness than
communities that are not subsidized by MDN.
3) Have aquatic insects that show greater annual secondary production than aquatic
insects living in non-anadromous streams
4) Have aquatic insects that show greater specific growth rates than aquatic insects
living in non-anadromous streams
Study sites
This research was conducted in southeast Alaska in the Tongass National Forest.
This area is described as having a maritime climate (average precipitation=1500-5000
mm, average July temperature=13°C) and dense coastal forest (Oswood et al. 1995).
The primary areas of study were streams in and around the Juneau area (Figure 1) and in
selected streams on Prince of Wales Island (Figure 1). The primary salmon runs in this
region are pink (Oncorhynchus gorbuscha) and chum salmon (0. keta) which spawn in
this region typically in early autumn (August-September). These streams also receive

runs from coho (0. kisutch), sockeye (0. nerka) and chinook (0. tshawytscha) salmon.

Juvenile coho, sockeye and chinook spend 1-3 years in their natal streams before
migrating to the ocean while pink and churn salmon migrate out to estuary waters soon
after emerging from the gravel as fry. Other common fish species found in these streams
are dolly varden (Salvelinus malma), sculpin (Cottus spp.), cutthroat trout (0. clarki), and
steelhead trout (0. mykiss).

This research was conducted in streams that we term ”legacy” streams, because they
are stream systems that have an anadromous section (connects to the ocean) and a non-
anadromous section that has been cut off from salmon migrations over eons of time
(since the little ice age) by a natural barrier (i.e. an impassable waterfall). Therefore
downstream sections have a “legacy” of annual MDN inputs. This disconnection of
stream sections allowed for the simultaneous comparison of the insect fauna that live in
areas with (treatment sections) and without (control sections) exposure to MDN, while
minimizing variation in other factors that could also impact insect communities (e. g.
riparian cover, water temperature etc.). Within this “category” of streams (i.e. legacy)
we have two general groups, the first being streams that receive large annual runs of
salmon and the second are streams that typically receive low numbers of spawners.
These studies will help us to understand if aquatic insect communities across multiple
natural stream systems respond to the cycle of natural MDN enrichment in a generalized
manner and also if there is some seasonality to their response. Specific experiments or
studies will address select community attributes (e. g. growth, secondary production) to
be studied in fewer systems but at a higher resolution than the studies across all streams.
The combination of all these studies fills gaps in the literature on the effects of MDN on

aquatic insect communities.

Chapter Introductions
Chapter 2: Influence of marine-derived nutrients from spawning salmon on stream
macroinvertebrate communities in southeast Alaska.

The current literature on insect community responses (e. g. standing stock
biomass) to MDN contains results from studies done either in mesocosms, from short-
terrn experiments and/or from natural stream observations with little or no replication.
Diversity, richness and functional group changes in association with MDN have not been
examined thoroughly in any study.

This chapter focuses on broad seasonal community patterns across all seven study
streams. Streams were sampled quantitatively and qualitatively in the spring (pre-run),
late summer (during-run), and mid-Autumn (post-run) in each stream section (both above
and below the barrier). Quantitative samples were used to examine trends in insect
abundance, biomass, diversity and functional groups around the salmon runs, using the
upstream sections as controls for each downstream treatment area. The qualitative
samples were used to examine community richness patterns.

Chapter 3: Influence of marine-derived nutrients from spawning salmon on mayfly and
rnidge secondary production in two southeast Alaskan streams.

One of the most important questions in research studies addressing MDN effects
on streams is the influence on secondary production, as this is the intermediate link
between the dissolved nutrients from salmon and fish production. Yet secondary
production has never actually been measured in MDN studies in streams, instead it had

been inferred from standing crop estimates.

In order to address the question of the influence of MDN on insect secondary
production, I selected two legacy streams to sample through time for the majority of the
growing season (May to October) in both stream sections (above and below the barrier).
The selection of taxa for this study was based on 1) taxa commonly cited in publications
as showing changes in standing crop from MDN, 2) taxa common in our study streams,
and 3) taxa with different life histories. For these reasons, secondary production
estimates were made for the five most common mayfly genera: Baetis spp. (Baetidae),
Epeorus spp., Cinygmula spp., Rhithrogena spp. (Heptageniidae) and Drunella spp.
(Ephemerellidae) and midges of the family Chironomidae. The richness of the rnidge
community in each stream section was estimated over the study period, but production
was calculated at the family level. This study determined what the influence of salmon is
on the secondary production of insects and how the results vary for univoltine and
multivoltine taxa.

Chapter 4: Growth of caddisflies (Limnephilidae: Trichoptera) in response to spring
carry-over of marine—derived nutrients and food type in a southeast Alaskan stream.

Previous studies have shown that insects grow faster in the direct presence of
MDN from salmon carcasses (Minakawa 1997,Chaloner and Wipfli in press). No
studies, however, have addressed the influence of MDN carry-over on insect
communities the following spring. In order for MDN to be of great importance to the
overall production of their natal streams the nutrients must be incorporated into the
community for a longer period than for the few weeks that carcasses are present.

I selected one legacy stream, Harris Creek, on Prince of Whales Island, to run a

growth experiment in May, after winter snow melt and at least six months after salmon

10

carcasses from the Autumnal pink and chum run would have disappeared. The
experiment was run on three limnephilid caddisfly genera including a scraper
(Dicosmoecus atripes) and two shredders (Onocosmoecus sp. and Psychoglypha spp) and
simultaneously compared the growth rates of these genera in anadromous and non-
anadromous stream sections. This experiment examined the effects of spring carry-over
of MDN into the stream system and its effect on insect growth of different functional
groups.

The following chapters, as introduced above, address one or more of the
objectives and hypotheses listed in this introductory chapter. Separately, they address
aspects of the influence of spawning salmon on aquatic insect community attributes that
are currently lacking in the MDN literature. Collectively they provide strong empirical
and experimental evidence for the dynamics of insect communities in southeast Alaskan
streams and question the role that MDN plays as a nutrient subsidy to insects in these

stream systems.

11

LITERATURE CITED

Anderson, N .H. and K.W. Cummins. 1979. The influence of diet on the life histories of
aquatic insects. J. Fish. Res. Bd. Can. 36:342-355.

Baker, T.T., A.C. Wertheimer, R.D. Burkett, R. Dunlap, D.M. Eggers, E.I. Fritts, A.J.
Gharrett, R.A. Holmes, and R.L. Wilmot. 1996. Status of Pacific salmon and
steelhead escapements in southeastern Alaska. Fisheries (Bethesda) 21 :6-18.

Bilby R.E., B.R. Fransen, and RA. Bisson. 1996. Incorporation of nitrogen and carbon
from spawning coho salmon into the trophic system of small streams: evidence
from stable isotopes. Can. J. Fish. Aquat. Sci. 53:164-173.

Bilby R.E., B.R. Fransen, J .K. Walter, C]. Cederholm, and W.J. Scarlett. 2000.
Preliminary evaluation of the use of nitrogen stable isotope ratios to establish
escapement levels for Pacific salmon. Fisheries 26: 6-13

Bonham, CD. 1989. Measurements for terrestrial vegetation. John Wiley and Sons,
New York, NY. 338 p.

Cederholm, C.J., and NP. Peterson. 1985. The retention of coho salmon (Oncorhynchus
kisutch) carcasses by organic debris in small streams. Can. J. Fish. Aquat. Sci.
42: 1222-1225.

Cederholm, C.J., D.B. Houston, D.L. Cole, and W.J. Scarlett. 1989. Fate of coho salmon
(Oncorhynchus kisutch) carcasses in spawning streams. Can. J. Fish. Aquat. Sci.
46: 1347-1355.

Cederholm, C.J., M.D. Kunze, T.M. Murota, and A. Sibatani. 1999. Essential
contributions of nutrients and energy for aquatic and terrestrial ecosystems.
Fisheries (Bethesda) 24:6-15.

Chaloner D.T., and MS. Wipfli. 2002. Influence of decomposing Pacific salmon
carcasseson macroinvertebrate growth and standing stock in southeastern Alaska
streams. Journal of the North American Benthological Society 21: 430-442.

Cummins, K.W. 1974. Structure and function of stream ecosystems. Bioscience
24:631-640.

Cummins, K.W., and M.J. Klug. 1979. Feeding ecology of stream invertebrates. Ann.
Rev. Ecol. Syst. 10:147-172.

Cumrrrins, K.W., M.A. Wilzbach, D.M. Gates, J .B. Perry and W.B. Taliferro. 1989.
Shredders and riparian vegetation. Bioscience 39:24-30.

Duncan, W.F.A., and M.A. Brusven. 1985. Benthic macroinvertebrates in logged and

12

unlogged low-order Southeast Alaskan streams. Freshwat. Invertebr. Biol. 4: 125-
132.

Gende S.M., R.T. Edwards, M.F. Willson, and MS. Wipfli. 2002. Pacific salmon in
aquatic and terrestrial ecosystems. BioScience 52: 917-928.

Gresh, T., J. Lichatowich, and P. Schoonmaker. 2000. An estimation of historic and
current levels of salmon production in the northwest Pacific ecosystem: evidence
of a nutrient deficit in the freshwater systems of the Pacific Northwest. Fisheries
(Bethesda) 25: 15-21.

Grubbs, S.A., and K.W. Cummins. 1994. Processing and macroinvertebrate colonization
of black cherry (Prunus serotina) leaves in two streams differing in summer biota,
thermal regime and riparian vegetation. Am. Midl. Nat. 132:284-293.

Halupka, K.C., M.D. Bryant, M.F. Wilson and EH. Everest. 1999. Biological
characteristics and population status of anadromous salmon in Southeast Alaska.
Gen. Tech. Rep. PNW-GTR-468. U.S. Dep. Agriculture, Forest Service, Pacific
Northwest Research Station, Portland, Oregon (In Press).

Hawkins, C.P., M.L., Murphy, N.H. Anderson, and M.A. Wilzbach. 1983. Density of
fish and salamanders in relation to riparian canopy and physical habitat in streams
of the Northwestern United States. Can. J. Fish. Aquat. Sci. 40: 1 173-1 185.

Herlong, DD, and M.A. Mallin. 1985. The benthos-plankton relationship upstream and
downstream of a blackwater impoundment. Journal of Freshwater Ecology 3:47-
59.

Hernandez, O. 2001. Benthic invertebrate community structure as affected by forest
succession after clear-cut logging on Prince of Wales Island, Southeast Alaska.
Master’s Thesis, Department of Entomology, Michigan State University, East
Lansing MI.

Kaushik, N.K., and H.B.N. Hynes. 1971. The fate of dead leaves that fall into streams.
Arch. Hydrobiol. 68:465-515.

Kline, T.C., J .J . Goering, and R.J. Piorkowski. 1997. The effect of salmon carcasses on
Alaskan freshwaters. Chapter 7 pgs 179-204, In Freshwaters of Alaska,
ecological syntheses. A.M. Milner and M.W. Oswood (Eds). Springer-Verlag,
New York.

Kondolf, GM. and S. Li. 1992. The pebble count technique for quantifying surface bed
material in instream flow studies. Rivers 3:80-87.

Larkin, G, and RA. Slaney. 1997. Implications of trends in marine-derived nutrient

l3

influx to south coastal British Columbia salmonid production. Fisheries 22: 16-
24.

Lichatowich, J .A. 1999. Salmon without rivers. Island Press, Washington DC. 317 p.

Merritt, R. W. and J. R. Wallace. 2001. The role of aquatic insects in forensic
investigations, pp. 177-221. In: Forensic entomology: The utility of arthropods
in legal investigations. CRC Press, Boca Raton, FL

Minakawa, N. 1997. The dynamics of aquatic insect communities associated with
salmon spawning. Ph.D. dissertation. University of Washington, Seattle.

Nakano, S., and M. Kaeiryama. 1995. Summer microhabitat use and diet of four
sympatric stream-dwelling salmonids in a Kamchatkan stream. Fisheries Science
61: 926-930.

Oswood, M.W., J .G. Irons, and A.M. Milner. 1995. River and stream ecosystems of
Alaska. Chapter 2 pgs 9-29, In Ecosystems of the world 22. CE. Cushing, KW.
Cummins and G.W. Minshall (Eds). Elsevier publishing, New York.

Peterson, DP, and OJ. Foote. 2000. Disturbance of small-stream habitat by spawning
sockeye salmon in Alaska. Transactions of the American Fisheries Society 129:
924-934.

Piorkowski, R.J. 1995. Ecological effects of spawning salmon on several southcentral
Alaskan streams. Ph.D. dissertation. University of Alaska, Fairbanks.

Rader, RB. 1997. A functional classification of the drift: traits that influence
invertebrate availability to salmonids. Can. J. Fish. Aquat. Sci. 54: 1211-1234.

Richardson, J .S., and R.J. Mackay. 1991. Lake outlets and the distribution of filter
feeders: an assessment of hypotheses. Oikos 62:370-380.

Schuldt, J .A., and A.E. Hershey. 1995. Effect of salmon carcass decomposition on Lake
Superior tributary streams. J. N. Am. Benthol. Soc. 14:2569-268.

Stockner, J.G., E. Rydin, and P. Hyenstrand. 2000. Cultural oligotrophication: causes
and consequences for fisheries resources. Fisheries (Bethesda) 25:7-14.

Vannote, R.L., Minshall, G.W., Cummins, K.W., Sedell, J .R., and Cushing, CE. 1980.
The river continuum concept. Can. J. Fish. Aquat. Sci. 37: 130-137.

Wallace, J .B., and R.W. Merritt. 1980. Filter-feeding ecology of aquatic insects. Ann.
Rev. Entomol. 25: 103-32.

Wallace, J.B., and AC. Benke. 1984. Quantification of wood habitat in subtropical
coastal plain streams. Can. J. Fish. Aquat. Sci. 41:1643-1652.

I4

Wilzbach, M.A., K.W. Cummins and J .D. Hall. 1986. Influence of habitat
manipulations on interactions between cutthroat trout and invertebrate drift.

Ecology 67 :898-91 1.

Wilzbach, M.A., K.W. Cummins and R.A. Knapp. 1988. Toward a functional
classification of stream invertebrate drift. Verh. Intemat. Verein. Limnol.

23: 1244-1254.

Wipfli, MS. 1997. Terrestrial invertebrates as salmonid prey and nitrogen sources in

streams: contrasting old-growth and young-growth riparian forests in
southeastern Alaska, USA. Can. J. Fish. Aquat. Sci. 54:1259-1269.

Wipfli, M.S., J. Hudson, and J. Caouette. 1998. Influence of salmon carcasses on stream

productivity: response of biofilm and benthic macroinvertebrates in southeastern
Alaska, USA. Can. J. Fish. Aquat. Sci. 55:1503-1511.

15

Chapter 2
Influence of marine-derived nutrients from spawning salmon on aquatic insect
community patterns in south-east Alaskan streams

Abstract

Studies investigating enrichment effects from spawning salmon, termed marine-
derived nutrients (MDN), have shown positive relationships between insect abundance
and biomass in artificial streams and in minimally replicated natural stream studies. To
better understand these relationships, we sampled seven streams seasonally in southeast
Alaska. Of the seven study streams four annually receive large runs of chum and pink
salmon, while three had low or no salmon runs. All the streams selected had a natural
waterfall barrier to salmon, which allowed for the simultaneous sampling of stream
sections with similar habitat characteristics but with the separation of the influence of
salmon and the barrier.

Nine'modified-Hess samples were taken before, during and after the fall salmon
run in each stream section of our seven study streams between 2001 and 2002.
Qualitative samples for taxa richness also were taken in each stream section. Samples
were analyzed for mean density, biomass (by taxon and functional feeding group), taxa
richness and Shannon-Weiner diversity for each stream section (upstream and
downstream) by period (pre, during and post) and run size (high and low).

High run streams had upstream sections with a greater abundance and biomass of
mayflies (dominated by Baetidae, Heptageniidae and Ephemerellidae) during the run, and
downstream sections had a greater abundance and biomass of Dipterans (dominated by

Chironomidae). Diversity and richness were similar across stream sections and run size

16

within each period, except for during the run when both were significantly lower in
downstream sections of high run streams. Functional feeding group patterns showed
higher abundance and biomass of collector-gatherers (primarily Chironomidae) and
shredders (primarily the nemourid Zapada) during the post spawning, carcass
decomposition period. Overall, this study suggests that a positive relationship between
MDN and stream insect abundance and biomass only exists for specific taxa with life

history attributes that allow them to take advantage of the MDN enrichment.

l7

Introduction

Alaska is the last region along the Pacific coast of the United States that still
receives large, annual runs of spawning salmon near historic levels (Baker et al. 1996,
Gresh et al. 2000). Upstream migrations of adult salmon provide these normally
oligotrophic systems with a huge pulse of nutrients from the ocean. These nutrients,
which enter the stream in the form of salmon eggs, sperm, waste and ultimately the adult
carcasses, are termed marine-derived nutrients (NflJN) and have been hypothesized to be
linked to the dynamics and structure of stream communities (algae to fish). Estimates of
annual nutrient delivery from spawning salmon to Alaskan streams range in the millions
of tons of carbon, nitrogen, phosphorous and other nutrients (Larkin and Slaney 1997,
Halupka et al. 1999).

Theories on the role of MDN in coastal stream communities predict that this
nutrient subsidy provides a positive feedback mechanism, linking adult anadromy and
semelparity with juvenile salmonid production (Kline et al. 1997, Lichatowich 1999).
Intermediate steps predicted to connect MDN and juvenile salmon include alterations of
dissolved nutrients, biofilm production and alterations to the macroinvertebrate
community (W ipfli et al. 1998, Cederholm et al. 1999, Chaloner et al. 2004). Stream
community responses to MDN have far reaching implications in these watersheds
affecting algal production to terrestrial vertebrate predators and have been suggested to
be a key factor for salmon recovery programs throughout the Pacific Northwest (Bilby et
al. 2000, Stockner et al. 2000, Gende et al 2002).

Aquatic macroinvertebrates are important organisms for studying the impact of

MDN on the overall stream community because they integrate bottom-up pathways of

18

enrichment to juvenile salmonids, and are good indicators of stream ecosystem structure
and function. Previous studies have shown aquatic insect communities to respond to
salmon runs in several important ways. Among these responses are short-term reductions
in abundance and increased drift due to the disturbance created during large spawning
runs (Minakawa 1997, Peterson and Foote 2000). Increased insect density following
salmon carcass decomposition has been documented (Minikawa 1997, Wipfli et al. 1998,
Kline et al. 1997), as well as indications that insect richness and diversity may increase
from salmon enrichment in central Alaskan streams (Piorkowski 1995). These previous
studies provide important insights into the potential impact of MDN from salmon on the
invertebrate communities, but the majority of the evidence linking MDN and insect
enrichment comes from artificial stream studies or short term, poorly replicated natural
stream studies. N 0 study has addressed the relationship of MDN and aquatic
macroinvertebrate community dynamics across seasons or stream systems. The objective
of this study was to evaluate the impact of MDN from spawning salmon on aquatic insect
communities seasonally and across several natural stream systems. The community
attributes I focused on were insect abundance and biomass, community richness and
diversity, and functional feeding groups. Seasonal samples were directed at times of the
year that would allow me to address the long-terrn effects of MDN into the spring (pre
fall spawning run), the impact of the spawning disturbance (during the fall run) and the
responses to the major nutrient subsidy (post fall salmon run and during carcass

decomposition).

l9

Study Sites

Southeast Alaska contains the 8.5 million hectare Tongass National Forest, most
of which is pristine forest surrounding 5200 anadromous salmon streams (Halupka et al.
1999). Because the runs in Alaska are still intact, these streams provide the opportunity
to study the structure and function of these systems around the salmon runs from a fairly
pristine state. The primary salmon (Oncorhynchus) runs in terms of biomass entering the
stream in this region are pink (0. gorbuscha) and chum salmon (0. keta), but these
streams also receive runs from coho (0. kisutch), sockeye (0. nerka) and chinook ( 0.
tshawytscha) salmon. Other common fish species are dolly varden (Salvelinus malma),
sculpin (Cottus spp.), cutthroat trout (0. clarki), and steelhead trout (0. mykiss).
Southeast Alaska is described as having a maritime climate (average precipitation=1500-
5000 mm, average July temperature=13°C) and dense coastal forest (Oswood et al. 1995).
Study streams were located in and around the J uneau-Douglas area (Figure 1) and in
selected streams on Prince of Wales Island (Figure 1). Study streams around Juneau-
Douglas were: Fish Creek (58°19'N, 134°35'W), Sheep Creek (58°16’N, 134°18’W),
Salmon Creek (58°19'N, 134°27'W), Peterson Creek (58°28’N, 134°44’W), and Bessie
Creek (58°35’N, 134°54’W). Study streams on Prince of Wales Island, near Hollis were:
Harris River (55°27’N, 132°42’W) and Sata/Gulch Creeks, which are both tributaries to
the Trocadero River (55°21’N, 132°52’W). Most study streams were systems that have
an anadromous section (connects to the ocean) with healthy salmon runs and a non-
anadromous section that had been cut off from salmon migrations for thousands of years
(since the little ice age) by a natural waterfall barrier, which is enough time for adaptive

changes due to MDN to have occurred. This disconnection of stream sections allowed

20

for the simultaneous comparison of the insect fauna that occur in areas with and without
exposure to MDN, while minimizing variation in other factors that could also impact
insect communities (e.g., substrate, riparian cover etc.). To test this, we also selected
stream systems with barriers but with lower reaches that didn’t receive large salmon runs.

Table 1 provides a description of the seven study streams.

 

Figure 1. Map of southeast Alaska. Study streams were near Juneau-Douglas and on

Prince of Wales Island.

21

Table 1. Location, spawning run size and physical habitat data for the seven study

 

 

streams.
Width Depth Water Riparian
Stream Location 56000“ Run Size (m) (cm) Temp (°C) Substrate Canopy
Bessie Juneau Downstream Low 6.7 24.9 10.6 Cobble Conifer
Upstream 4.7 39.5 10.6 Cobble Conifer
Fish Douglas Is. Downstream High 20.52 27.27 8.0 Cobble/Boulder Conifer
Upstream 1 1.93 35.38 8.0 Cobble/Boulder Conifer
Harris POW Downstream High 29.38 49.53 8.0 Cobble/Boulder Conifer/Alder
Upstream 13.56 33.80 9.0 Cobble Conifer
Peterson Juneau Downstream High 13.68 9.38 10.9 Cobble Conifer/Alder
Upstream l 1.90 41.25 10 Cobble Conifer
Salmon Juneau Downstream High 1 1.38 26.58 7.3 Cobble Conifer/Alder
Upstream 9.72 30.89 6.30 Cobble Conifer
Sata/Gulch POW Downstream Low 8.70 30.42 N/A Cobble Alder
Upstream 5.58 31.6 N/A Cobble Conifer/Alder
Sheep Juneau Downstream none 12.5 45.0 8.0 Boulder/Cobble Alder
Upstream 8.9 46.8 6.5 Cobble Alder

 

22

Methods

Streams were sampled between late May and early June before the large autumnal
run of pink and chum salmon (“pre”), again in late August once the autumnal run had
begun and spawning was occurring (“during”), and once more in late September-early
October before snow fall but after decomposition of the salmon carcasses was well
underway (“post”). I took three modified-Hess samples (0.04 m2, mesh size 250 um)
from each of three riffle reaches from each stream section (upstream and downstream of
the barrier) pre, during and post spawning for a total of 54 samples from each study
stream. Samples were washed into labeled ziptop bags with 90% ethanol and transported
to the laboratory. Hess samples were used for abundance, biomass, functional group and
diversity estimates. For taxa richness, a field crew of three to four people took a
combination of modified-Hess or D-net samples from all macro-habitats (riffle-cobble,
run-gravel, pool-fines, submerged wood and vegetation, and detritus) in one reach (~100-
300m long) in each stream section (upstream and downstream of the barrier). Richness
samples were picked in the field from white pans and processed as above. Sampling was
conducted for approximately two hours and was terminated when no new taxa were
collected for at least 30 minutes.

Selected physical properties were measured from each stream section
once, during the pre-run sampling period. Parameters measured included channel width,
depth, water temperature, substrate, and riparian vegetation. All habitat parameters were
measured at five transects in each stream section. Transects were selected randomly
along a 300m reach. The pebble count method was used to analyze mineral substrate size

(Kondolf and Li 1992). Riparian vegetation was assessed using the line intercept method

23

from the stream bank out 30 m (Bonham I989, Grubbs and Cummins 1994). Run size
for each stream (high or low/no run) were determined from published data on these
streams (Chaloner et al. 2004), communications with the Alaska Department of Fish and
Game, and direct observations. The habitat data for the seven study streams is
summarized in Table 1. Upstream were the control sites and refer to above the barrier
while downstream are the treatment sites and refer to below the barrier and open to the
ocean. There were four high run streams that consistently received large runs (i.e.
thousands to hundreds of thousands) of salmon and three low or no run streams that
received virtually no spawning run. All streams had similar temperatures and substrate,
but there was variation in stream size (i.e. width) and riparian canopy.

In the lab, samples were picked and sorted under 10x magnification.
Identification was done down to the lowest taxonomic unit possible, which was generic
level for most orders. Species designations were confirmed by taxonomic experts. Insect
abundance and total lengths (nearest 0.5 mm) were recorded for all taxa in each
modified-Hess sample. Published length-weight regressions were used to calculate
biomass (Benke et al 1999). Diversity was compared using the Shannon-Wiener
diversity index. Functional feeding group designations were made using Merritt and
Cummins (1996). O
Data Analysis

The data were analyzed using repeated mixed model analyses with stream as a
random effect, section within stream as the repeated factor (because each stream section
was sampled three times pre, during and post spawning) and compound symmetric

covariance structure. Density and biomass data were normalized with natural-log

24

transformations that are referred to as ln_density and ln_biomass on all figure axes. For
simplicity, however, the results and discussion will refer to these data as simply density
and biomass. Trichopteran density and biomass, however, was too rare to be normalized
with any transformation and so trichops were omitted from order and family level
analyses, but were included in all the community comparisons (i.e. diversity, richness,
and functional feeding groups). Separate repeated mixed models were run on the density
and biomass of each insect order, the dominant families in each order, functional feeding
groups, and also taxa richness and diversity. Each model tested the effects of section
(upstream vs. downstream), period (pre, during and post), run size (high run vs. low/no
run) and interactions of these three factors. Due to the high occurrence of significant
(alpha < 0.05) three-way interactions in these models, significant differences between
stream sections (our primary treatment factor for the effects of MDN) were analyzed
separately with regular (i.e., non-repeated) mixed models within period and run size.
Results

The results of the mixed modeling analyses by order are given in table 2 and
figures 2 and 3. Density and biomass varied significantly by period and run size for
Ephemeropterans, while significant three way interactions for Dipterans only existed for

the density data, and for Plecopterans for the biomass data (Table 2).

25

Table 2. Results (p-values) of the repeated mixed modeling analysis by order. Period

and section are abbreviated in the 3-way interaction term.

 

 

Order Transformation Section Period Run Period*Section Per*Sec*Run
Epemeroptera ln_density 0.086 <0.0001 0.514 0.150 0.002
ln_biomass 0.51 l <0.0001 0.448 0.307 0.009
Diptera ln_density 0.085 <0.0001 0.512 0.715 0.046
ln_biomass 0.112 <0.0001 0.442 0.730 0.488
Plecoptera ln_density 0.653 0.050 0.392 0.769 0.087
ln_biomass 0.734 0.033 0.880 0.752 0.017

 

 

26

=post spawning).

=pre, B=during, and C

27

 

 

 

i.‘

1_________ .2i4_._ _~_______

$5: £8512.—

Bars are: high run, downstream ( I), high run, upstream ( [3), low nm, downstream ( I) and

Figure 2. Natural log- transformed density means for each aquatic insect order by stream
low run, upstream (I). An * denotes significant differences by stream section within run size.

section and run size for each period (A

       

    
           

   

     

     

    

 

 

A B _. . C
“\WQ\\

765432107654321076543210

€395 anaemia

EPH

Order
transformed biomass means for each aquatic insect order by stream

DIP

Figure 3. Natural log-

section and run size for each period (A=pre, B=during, and C=post spawning).

Bars are: high run, downstream ( ), high run, upstream ( D), low run, downstream ( I) and

low mn, upstream (I). An * denotes significant differences by stream section within run size.

28

Dipteran density was on average higher in the high run streams compared to the low run
streams, and was significantly higher in downstream (treatment) sections of the high run
streams across all three time periods (Fig. 2). Dipteran density in the low run streams
was significantly higher in downstream sections for only the post spawning time period
(Fig. 2). Dipteran biomass also was significantly higher in downstream sections of high
run streams across all three time periods, with upstream sections of the high run streams
being similar to both sections of the low run streams (Fig. 3). On average, biomass of
ephemeropterans was also greater in the high run streams compared to the low run
streams (Fig. 3). Ephemeropteran density and biomass were significantly greater in
upstream sections of high run streams during the salmon run only (Figs. 2 and 3).
Plecopteran density and biomass were not different by section for streams in either run
size category during any of the three sampling periods (Figs. 2 and 3). Plecopteran
biomass was greater in both upstream and downstream sections in the high run streams
during the salmon run.

To examine these data at a finer resolution, family level analyses were run on the
dominant families in each order (Table 3). Families were considered dominant if they
comprised 20% or more of the density or biomass in any stream section for any period.
The dominant mayfly families were Heptageniidae (primarily Epeorus, Cinygmula and
Rhithrogena), Baetidae (primarily Baetis) and Ephemerellidae (primarily Drunella). The
dominant Dipteran family was Chironomidae (primarily Orthocladiinae, Tanytarsini and
Tanypodinae). The dominant plecopteran family was Chloroperlidae (primarily Sweltza

and Suwallia). A synoptic list of all taxa identified in each stream by section and period

29

are given in Appendix A. The results of the repeated mixed analyses by family are

shown in table 4 and figures 4 and 5.

Table 3. Dominant families in each order by stream section (upstream and downstream)
and period (pre, during and post spawning). Percent of total density and biomass are

given for each family.

 

 

Upstream Downstream
Period Order Family % Density % Biomass % Density % Biomass
Pre DIP Chironomidae 91.1 77.9 90.7 73.9
EPH Heptageniidae 38.5 51.6 61.6 57.1
EPH Baetidae 52.4 25.2 35.6 38.6
EPH Ephemerellidae 1.9 22.1 1.4 3.9
PLE Chloroperlidae 66. 1 76.4 92.3 95 .3
TRI Rhyacophilidae 28. 1 3 1 .6 26.7 68.4
TRI Limnephilidae 23.4 13.5 26.7 1.2
TRI Glossosomatidae 25.0 39.1 67 5,1
During DIP Chironomidae 86.9 53.1 97.8 87.9
EPH Heptageniidae 39.9 51.8 30.0 41.0
EPH Baetidae 36.7 24.3 44.3 38.0
EPH Ephemerellidae 4.6 21.2 5.8 6.6
PLE Chloroperlidae 69.5 81.7 90.8 96.3
TRI Rhyacophilidae 19.2 43.9 12.1 22.0
TRI Limnephilidae 26.7 25.3 57.8 57.0
TRI Glossosomatidae 41.7 10.8 28.4 16.3
Post DIP Chironomidae 81.0 49.8 93.2 73.0
EPH Heptageniidae 44.6 83.8 65.1 79.1
EPH Baetidae 43.1 13.4 29.1 17.9
EPH Ephemerellidae 6.9 1.5 2.3 0.9
PLE Chloroperlidae 55 .8 62.4 70.3 94.9
TRI Limnephilidae 39.6 71.8 20.9 37.1
TRI Lepidostomatidae 5 ,7 45 46.3 40.9
TRI Rhyacophilidae 39.6 71.8 9.0 9.8
TRI Glossosomatidae 30.2 15 .3 20.9 10.7

 

30

Table 4. Results (p-values) of the repeated mixed modeling analyses by family. Period

and section are abbreviated in the 3-way interaction term.

 

 

Family Transformation Section Period Run Period*Section Per*Sec*Run
Heptageniidae ln_density 0.126 <0.0001 0.193 0.006 0.002
ln_biomass 0.486 <0.0001 0.588 0. 108 0.050
Baetidae ln_density 0.132 <0.0001 0.890 0.064 0.001
ln_biomass 0.546 <0.0001 0.353 0.295 0.002
Ephemerellidae ln_density 0.076 0.002 0.223 0.1 12 0.047
ln_biomass 0.607 0.131 0.408 0.1 1 1 0.347
Chironomidae ln_density 0. 102 <0.0001 0.640 0.344 0. 13 l
ln_biomass 0.093 <0.0001 0.335 0. 192 0.705
Chloroperlidae ln_density 0.968 0.132 0.541 0.687 0.018
ln_biomass 0.984 0.010 0.843 0.605 0.034

 

 

31

_\. ,3... y, ., .

      

§

       

 

       

  

\

      

   

 

 

 

_ 3.. ____.________a_<_
87654321087654321087654.3210

Asa-ht bacon—I73

 

Figure 4. Natural log- transformed density means for each aquatic insect family, dominant

in each order, by stream section and run size for each period (A=pre, B=during, and

C=post spawning). Bars are: high run, downstream ( E), high run, upstream ( D ),

). An * denotes significant differences

low run, downstream (I ) and low run, upstream (

by stream section within run size.

32

  

B . x\\\

\
..\ .N

\
.\

.\..\.\ xxx . x 8... \.\\. .V\.\.\ \M
“I

     

 

“I

 
   

m, \. ... 3...». \x

    

  

 

“I

     

 

VRQ

\\
\

\9

 

“I “I

  

 
 

 

 

sense 83:52..

33

Bars are: high run, downstream (I ), high run, upstream (El ), low run, downstream ( I) and

Figure 5. Natural log- transformed biomass means for each dominant aquatic insect family,
low run, upstream). An * denotes significant differences by stream section within run size.

by stream section and run size for each period (A=pre, B=during, and C=post spawning).

Due to the high proportion of chironomids (50 to 97%) versus all other dipterans, the
patterns for density and biomass by family were very similar to the order level analyses
for Diptera, with greater density and biomass occurring in downstream sections of the
high run streams in all three sampling periods (Figs. 4 and 5). Likewise, Chloroperlidae
results were also very similar to what was shown for plecopterans, with no significant
differences by stream section related to the salmon run (Fi gs. 4 and 5). Both baetid and
heptageniid mayflies showed significant differences by stream section, with both density
and biomass being greater in upstream (control) sections of the high run streams during
the salmon run. Baetids and ephemerellids both had significantly lower densities in
downstream sections of the high run streams post spawning, but this was also true in the
low run streams.

When biomass and density data at both order and family levels are considered
across sections and run size, the trend is for insect density and biomass to be higher
before and during the run and at a minimum during the post spawning period (Fi gs. 2-5).
These differences were greatest for the biomass data. This seasonal variation was
consistent for both low and high run streams and both upstream and downstream sections,
indicating factor(s) other than MDN enrichment influenced these communities post
spawning.

I investigated the potential roles of stream size and riparian canopy (the only
habitat parameters measured that showed variation across streams) on insect community
abundance and biomass. Neither factor (width or riparian canopy), however, resulted in

models with significant predictive power for insect density or biomass for any of the taxa

34

in these streams. Since these factors did not interfere with the interpretation of the results
in terms of salmon or MDN impacts, they will not be discussed further.

Mixed repeated analyses examining functional feeding groups by period, stream
section and run size showed patterns similar to the family level analyses. Overall
biomass declined in the autumn during the post-spawning time period. High run stream
differences in density were seen for: 1) collector-gatherers, which were higher in
downstream sections pre-spawning; 2) predators, which were higher in downstream
sections during spawning; 3)and shredders, which were higher in downstream sections
post-spawning (Fig. 6). High run stream differences in biomass were seen for collector-
gathers, which were significantly lower in downstream sections during spawning and
higher in downstream sections post-spawning (Fig. 7). Mixed analyses on taxa richness
and Shannon-Weiner diversity resulted in significantly lower richness and diversity in
downstream sections of high run streams during the spawning runs, with upstream
sections of high run streams being more similar to both sections of the low run streams
(Fig 8). There were no differences in diversity or biomass pre- or post-run in either high

or low run streams.

35

xi .xx.\\\\\ .

SR .3\\\\\§ \,

        
   

    

“I

 

 

1“ .|W

0076543210876543210876543210

cit E254:

SC

PR
Functional Group
=pre, B=during, and C=post spawning).
Bars are: high run, downstream ( ), high run, upstream (El ), low run, downstream ( I)
36

Figure 6. Natural log- transformed density means for each functional feeding group, by
and low run, upstream (I). An * denotes significant differences by section within run size.

stream section and run size for each period (A

          

B L»... .. n v .....\.\ \\.\\
\\\\\.\\\\\\\\\\ \ \\ \k

“I

I

\\\\\\\\.\\«..\\
V\\\\\u \\\\\.

 

"I

        

I
.3.

 

    

was...

CF

 
 

 

 

 

0076543210876543210876543210

£35 3893:...—

SH

SC

Functional Group
37

PR

CG
Bars are: high run, downstream ( *3" ), high run, upstream (D ), low run, downstream ( I)

Figure 7. Natural log- transformed biomass means for each functional feeding group, by
stream section and run size for each period (A=pre, B=during, and C=post spawning).
and low run, upstream ( fi“ ). An "‘ denotes significant differences by section within run size.

Number of Taxa

 

 

 

Pre During Post

Figure 8. Mean taxa richness (A) and Shannon-Weiner diversity (B) by stream section,
averaged across run size for each sample period (pre, during and post spawning).
Bars are: high run, downstream (I ), high run, upstream (D ), low run, downstream (I ) and

low run, upstream (I). An * denotes significant differences by stream section within run size.

38

Discussion

This study was the first to address aquatic insect community patterns in relation to
MDN from spawning salmon in multiple natural stream systems across multiple seasons.
The sampling periods in this study spanned not only most of the growing season, but also
included the spawning run and well as salmon carcass decomposition. The inclusion of
low/no run streams allowed for the simultaneous comparison of barrier effects in similar
stream systems without the confounding influence of the salmon run.

The order level analyses illustrate an immediate dichotomy in response to salmon.
Dipteran density and biomass, which was driven by the dominant family Chironomidae,
was greater in downstream sections of the high run streams in each sampling period.

This appeared to be unique to the high run streams as chironornid biomass in the
upstream (control) sections of the high run streams was very similar to both sections of
the low run streams. Ephemeropterans, however, showed different trends with greater
biomass and density in upstream sections of the high run streams during the run. These
differences, however, were more likely due to a decrease in ephemeropterans in the
downstream sections during the spawning run rather than an increase in upstream
sections, because comparisons of pre-run and during-run data show that upstream
sections were similar between these two time periods. The reduction in mayflies was due
to the decrease of heptageniids and baetids in downstream sections. Salmon spawning
activities as a disturbance to stream benthos is well documented (Hildebrand 1971,
Peterson and Foote 2000, Minikawa 1997, Chaloner et al. 2004). Minikawa (1997) found
decreases in Chironomid midges, heptageniids and baetids during coho salmon redd

excavation in a Washington stream. Chaloner et al. (2004), whose study included three

39

of the streams sampled in this study (Fish Creek, Salmon Creek and Peterson Creek),
found the same patterns during salmon spawning with higher biomass of chironomids in
downstream sections, higher biomass of heptageniids in upstream sections, but, contrary
to this study, they detected no difference in baetid biomass between sections.

The premise that MDN from salmon acts as a nutrient subsidy to stream
communities, must assume that the preceding disturbance from spawning either left
enough benthos intact to respond to the nutrient subsidy or the benthos can recover in
time to take advantage of this subsidy. The light and temperatures in this region are at a
maximum in the spring and summer. The primary salmon runs in these streams (pink and
chum salmon), however, occur in the late summer and early fall, when flows are
increasing due to autumnal spates, and temperatures and light are in sharp decline.
Salmon MDN seem to be delivered at the time when the stream community, particularly
aquatic macroinvertebrates, is least able to capitalize on them. Seasonal factors
.combined with the disturbance caused by the large salmon run sizes in this region, are
likely important mortality factors for benthic organisms, as evidenced by the decline in
many taxa during spawning in this and other studies (Peterson and Foote 2000, Minikawa
1997, Maier 2001, Chaloner et al. 2004). Minikawa (1997) noted a recovery of the
Chironomid densities approximately 45 days post spawning, while in these study streams
chironomids did not exhibit a reduction during spawning. This indicated that
chironomids in these streams are able to avoid the mortality factors associated with the
benthic disturbance of spawning and therefore may have been able to take advantage of
the MDN influx during decomposition. Chironomid standing stock densities and biomass

were elevated, even during pre-spawning sampling periods, in downstream treatment

40

sections above what was measured for both upstream sections of high run streams and
both sections of low run streams. The nutrient enrichment by salmon may allow for
greater fecundity and/or winter survival of midges in these streams, which would explain
these differences remaining even 5-6 months after the carcasses have disappeared from
the streams (i.e., the pre-run period). Another factor that may promote rnidge response to
MDN is that, compared to mayflies, they seem to be less tied to bottom-up, algal
mediated pathways in order to utilize MDN, because they have been shown to feed on the
carcasses directly (Minakawa 1997, Chaloner and Wipfli 2002, Chaloner et al. 2002).
By feeding on carcasses, rrridges would be less reliant on sunlight and temperature
compared to epilitic biofilm feeders. Its reasonable for larger-bodied, univoltine, taxa
like the heptageniid mayflies, to be most negatively affected by spawning. Unlike the
small multivoltine chironomids, large univoltine fauna would be less able to escape the
disturbance behaviorally and to re-colonize disturbed areas with new cohorts, before the
salmon carcasses are decomposed and the MDN are lost to the ocean. Even if these
mayfly taxa have larvae in the stream post-spawning, if their food is mediated by primary
production they may be light and/or temperature limited in their growth during this part
of the year, no matter how nutrient rich the waters are (Rosemond et al. 2000).

The biomass data for these streams supports the theory of seasonal production
limitations, because biomass in all streams (high and low run) and sections (upstream and
downstream) were lowest during the post-spawning period for all the dominant families
and functional groups (except shredders). It was interesting that standing stock of
mayflies in the post-spawning period were similar across stream sections. It doesn’t

appear that this similarity was due to downstream sections “catching up” to upstream

41

population density and biomass, but rather during this time of the year mayfly
populations in all stream sections were in decline. The most important factor, therefore,
for predicting the influence of MDN on aquatic insects in stream systems may be the
timing of the enrichment with the insect’s life cycle. If in-stream insect production is
reduced, due to emergence or diapausing stages predominating during salmon
decomposition, the importance of MDN to these fauna would be severely limited. Only
Chironomid midges (which comprised most of the post-spawning collector-gatherers) and
shredders (primarily Zapada stoneflies) maintained elevated densities and/or biomass
during the post-spawning, salmon decomposition period. These may be the only aquatic
insects that can be said to “benefit” from the influx of MDN and to then transfer these
nutrients into higher trophic levels of the stream food webs. If this is the case, it would
reduce the extent that MDN influences the stream communities as a whole, but it may not
reduce the importance of MDN to production of fish, especially juvenile salmonids.
Many studies have shown the high dependence of juvenile salmonids on Chironomid
larvae (Frolenko 1973, Loftus and Lenon 1977, Kaeriyama et al. 1978, Dauble et al.
1980, Armitage et al. 1995)

The fact that these streams overall, across run size and section, had very similar
taxa richness and diversity values, implies how similar in general the insect communities
are across study streams. The only factor that can be attributed to the decline in richness
and diversity, during the salmon run was the spawning disturbance. However, while this
may be considered a negative effect, it appears to be short-lived, as both richness and
diversity were again similar post spawning. Piorkowski (1995) found marginal to

significant increases in macroinvertebrate richness and diversity related to salmon

42

enrichment in south-central Alaskan streams. In this study, however, the enrichment of
MDN appeared to have no long-term effects in terms of stimulating richness or diversity
of the aquatic insect communities.

In conclusion, it appears that the influence of salmon-mediated marine-derived
nutrients on stream insect communities is more complicated and more simplistic than
theories have often predicted. The complexity lies in the variation of the timing of
salmon runs across salmon species and the coupled disturbance that must precede any
natural enrichment. The primary runs in Southeast Alaska occur at the time of the year
when stream productivity may be limited by factors other than nutrients and, therefore,
would be less able to respond to MDN in all the complicated mechanisms that previous
mesocosm studies have predicted. This may be very different for other regions or other
spawning runs (i.e., summer spawners), and in streams where physical actors (e.g.,
spates) are not as pronounced. Due to the natural decline of most univoltine aquatic
insect fauna, and the disturbance-mediated benthic community structure created in stream
reaches containing salmonids, the influence of MDN on insect communities and the
transfer of these nutrients into higher foods webs may be from only one or two families.
Chironomidae (and possibly also the shredding nemourid Zapada) appear to be the only
insects in these stream systems that respond to MDN in a detectable way, with their
standing stock density and biomass. Therefore, if juvenile salmonids must rely on
invertebrate—mediated access to MDN from spawners, it may be solely from these taxa.
This research points to MDN as having no long term effect on overall stream richness,
diversity or standing stock of any insect taxa or functional group, besides midges and

shredders. Standing stock only provides a snap shot of these patterns, and so annual

43

production studies should be conducted to better understand the dynamics of insect
communities in these streams. This research provides strong evidence that conceptual
models predicting the role of salmon and MDN in stream ecosystems must include
disturbance as well as enrichment in order to better predict the mechanisms that will lead

to better stream management and salmonid recovery programs.

LITERATURE CITED

Armitage, P.D., P.S. Cranston, and L.C.V. Pinder (Eds). 1995. The Chironomidae:
biology and ecology of non-biting midges. Chapman & Hall, New York, NY.

Baker, T.T., A.C. Wertheimer, R.D. Burkett, R. Dunlap, D.M. Eggers, E.I. Fritts, A.J.
Gharrett, R.A. Holmes, and R.L. Wilmot. 1996. Status of Pacific salmon and
steelhead escapements in southeastern Alaska. Fisheries (Bethesda) 21:6-18.

Bilby R.E., B.R. Fransen, J .K. Walter, C]. Cederholm, and W.J. Scarlett. 2000.
Preliminary evaluation of the use of nitrogen stable isotope ratios to establish
escapement levels for Pacific salmon. Fisheries 26: 6-13.

Bonham, CD. 1989. Measurements for terrestrial vegetation. John Wiley and Sons,
New York, NY. 338 p.

Cederholm, C.J., M.D. Kunze, T.M. Murota, and A. Sibatani. 1999. Essential
contributions of nutrients and energy for aquatic and terrestrial ecosystems.
Fisheries (Bethesda) 2426-15.

Chaloner, D.T., and MS. Wipfli. 2002. Influence of decomposing Pacific salmon
carcasses on macroinvertebrate growth and standing stock in southeastern Alaska
streams. J. North. Amer. Benthol. Soc. 21: 430-442.

Chaloner, D.T., M.S. Wipfli, and J .P. Caouette. 2002. Mass loss and macroinvertebrate
colonization of Pacific salmon carcasses in south-eastern Alaskan streams.
Freshwater Biology 47: 263-273. ,

Chaloner D.T., G.A. Lamberti, R.W. Merritt, N.L. Mitchell, P.H. Ostrom, and MS.
Wipfli. 2004. Variation in responses to spawning Pacific salmon among three
south-eastem Alaska streams. Freshwater Biology 49:587-599.

Dauble, D.D., R.H. Gray, and T.L.Page. 1980. Importance of insects and zooplankton in
the diet of 0-age chinook salmon (Oncorhynchus tshawytscha ) in the central
Columbia River. Northwest Science 54: 253-258.

Frolenko, LA. 1973. Feeding of chum and pink salmon juveniles migrating downstream
in the main spawning rivers of the northern coast of the sea of Okhotsk. Fisheries
Research Board of Canada Translated Serial no. 2416: 22 pp.

Gende S.M., R.T. Edwards, M.F. Willson, and MS. Wipfli. 2002. Pacific salmon in
aquatic and terrestrial ecosystems. BioScience 52: 917-928.

Gresh, T., J. Lichatowich, and P. Schoonmaker. 2000. An estimation of historic and

45

current levels of salmon production in the northwest Pacific ecosystem: evidence
of a nutrient deficit in the freshwater systems of the Pacific Northwest. Fisheries
(Bethesda) 25: 15-21.

Grubbs, S.A., and K.W. Cummins. 1994. Processing and macroinvertebrate colonization
of black cherry (Prunus serotina) leaves in two streams differing in summer biota,
thermal regime and riparian vegetation. Am. Midl. Nat. 132:284-293.

Halupka, K.C., M.D. Bryant, M.F. Wilson and RH. Everest. 1999. Biological
characteristics and population status of anadromous salmon in Southeast Alaska.
Gen. Tech. Rep. PNW-GTR-468. U.S. Dep. Agriculture, Forest Service, Pacific
Northwest Research Station, Portland, Oregon.

Hildebrand, 8.6. 1971. The effect of coho spawning on the benthic invertebrates of the
Platte river, Benzie County, Michigan. Transactions of the American Fisheries
Society 100: 61-68.

Kaeriyama, M., S. Sato, and A. Kobayashi. 1978. Studies on a growth and feeding habit
of the chum salmon fry during seaward migration in the Tokachi River system. 1.
Influence of thaw on a growth and feeding habit of the fry. Scientific Report of
Hokkaido Salmon Hatchery no. 32: 27-41.

Kline, T.C., J .J . Goering, and R.J. Piorkowski. 1997. The effect of salmon carcasses on
Alaskan freshwaters. Chapter 7 pgs 179-204, In Freshwaters of Alaska,
ecological syntheses. A.M. Milner and M.W. Oswood (Eds.). Springer-Verlag,
New York.

Kondolf, G.M. and S. Li. 1992. The pebble count technique for quantifying surface bed
material in instream flow studies. Rivers 3:80-87.

Larkin G., and RA. Slaney. 1997. Implications of trends in marine-derived nutrient flux
to south coastal British Columbia salmonid production. Fisheries 22: 16-24.

Lichatowich, J .A. 1999. Salmon without rivers. Island Press, Washington DC. 317 p.

Loftus, W.F. and H.L. Lenon. 1977. Food habits of the salmon smolts, Oncorhynchus
tshawytscha and O.keta, from the Salcha River, Alaska. Transanctions of the
American Fisheries Society 106: 235-240.

Lopea, ES, 1. P_ardo, N. Felpeto. 2001. Seasonal differences in green leaf breakdown
and nutrient content of deciduous and evergreen tree species and grass in a
granitic headwater stream. Hydrobiologia 464: 51-61.

Maier, KJ. 2001. The influence of floods on benthic insect populations in a Swiss

mountain stream and their strategies of damage prevention. Archiv fuer
Hydrobiologie 150: 227-247.

46

Merritt, R.W. and K.W. Cummins (Eds). 1996. An Introduction to the Aquatic Insects of
North America. Kendall/Hunt, Dubuque, IA.

Minakawa, N. 1997. The dynamics of aquatic insect communities associated with
salmon spawning. Ph.D. dissertation. University of Washington, Seattle.

Oswood, M.W., J .G. Irons, and A.M. Milner. 1995. River and stream ecosystems of
Alaska. Chapter 2 pgs 9-29, In Ecosystems of the world 22. CE. Cushing, K.W.
Cummins and G.W. Minshall (Eds.). Elsevier publishing, New York.

Peterson, DP, and C]. Foote. 2000. Disturbance of small-stream habitat by spawning
sockeye salmon in Alaska. Transactions of the American Fisheries Society 129:
924-934.

Piorkowski, R.J. 1995. Ecological effects of spawning salmon on several southcentral
Alaskan streams. Ph.D. dissertation. University of Alaska, Fairbanks.

Rosemond, A.D., P.J. Mulholland S.H Brawley. 2000. Seasonally shifting limitation of
stream periphyton: response of algal populations and assemblage biomass and
productivity to variation in light, nutrients, and herbivores. Canadian Journal of
Fisheries and Aquatic Sciences 57: 66-75.

 

Stockner, J.G., E. Rydin, and P. Hyenstrand. 2000. Cultural oligotrophication: causes
and consequences for fisheries resources. Fisheries (Bethesda) 25:7-14.

Triska, F.J., A.P. J ackman J .H. Duff, R.J. Avanzino. 1994. Ammonium sorption to
channel and riparian sediments: A transient storage pool for dissolved inorganic
nitrogen. Biogeochemistry 26: 67-83.

 

Wigingtonfl Jr, M.R. Church T.C. Strickland K.N. Eshleman J. Van Sickle. 1998.
Autumn chemistry of Oregon Coast Range streams. Journal of the American
Water Resources Association 34: 1035—1050.

 

 

 

Wipfli, MS. 1997. Terrestrial invertebrates as salmonid prey and nitrogen sources in
streams: contrasting old-growth and young-growth riparian forests in
southeastern Alaska, USA. Can. J. Fish. Aquat. Sci. 54:1259-1269.

Wipfli, M.S., J. Hudson, and J. Caouette. 1998. Influence of salmon carcasses on stream
productivity: response of biofilm and benthic macroinvertebrates in southeastern
Alaska, USA. Can. J. Fish. Aquat. Sci. 55:1503-1511.

47

Chapter 3
Secondary production of mayflies and midges in response to spawning salmon in

natural Alaskan streams

Abstract

Theories on the relationship between marine-derived nutrients (MDN) from
spawning salmon link MDN with juvenile salmonid production via bottom-up pathways.
Many studies have used short-term standing stock biomass of aquatic macroinvertebrates
to infer relationships between MDN and secondary production in streams that receive
spawners. No study, however, has actually measured secondary production in relation to
MDN. To assess the relationship between MDN and aquatic insect production, we
measured secondary production of the five dominant mayfly genera (B_aeti_s spp.,
Drunella spp., Cinygmula spp., M spp., and Rhithrogena spp.) and Chironomid
midges throughout the primary growing season in two southeastern Alaskan streams.
Both streams had upstream control reaches blocked from spawning salmonids by a
waterfall barrier and downstream treatment reaches that received large spawning runs of
pink and chum salmon. Four of the mayfly genera (Drunella spp., Cinygmula spp.,
Em spp., and Rhithrogena spp.) had significantly greater production in upstream
control sections. Secondary production of Ems spp. was not significantly different
between sections. Chironomid production was significantly greater in downstream
treatment sections. Biomass of each taxon was maximized, however, in the spring and
summer, before the primary time of MDN input. These patterns point to spawning

disturbance and fish predation as the primary drivers of mayfly and rnidge production in

48

these streams. If this is a common pattern in this region, then in-stream secondary
production mediated links between MDN and juvenile salmonid production most likely

result from chiromonid midges.

49

Introduction

Nutrient transfers in lotic systems occur in a variety of ways. A well-
known example is the utilization and transfer of terrestrially derived nutrients (e.g., leaf
litter) from headwaters to downstream areas, which drives stream productivity and the
spatial/temporal organization of stream communities (Kaushik & Hynes 1971, Cummins
1974, Vannote et al. 1980, Cummins et al. 1989). Nutrients also are transferred from
marine systems into freshwater via fish migrations (Polis et al. 1997). Coastal streams
that are spawning grounds for salmon receive these nutrients, termed marine-derived
nutrients (MDN), in the form of salmon eggs, sperm, metabolic waste and adult
carcasses. The role that MDN plays in stream systems has been the object of study in
recent years. Most of these studies have dealt with either tracing MDN through surface-
stream and riparian food webs, or comparisons of stream communities with and without
salmon (Bilby et al. 1996, Kline et al. 1997, Wipfli et al. 1998, Cederholm et al. 1999,
Chaloner et al. 2002a). These relationships are of particular interest in the Pacific
Northwest of the United States, where salmon runs are extinct or threatened in many
streams along the coasts of Washington, Oregon and northern California.

It has been suggested that salmon provide an essential nutrient source to the
typically oligotrophic, anadromous streams of the Pacific Northwest region and, by
subsidizing the nutrient base in their spawning grounds, increase stream productivity and
the viability of their own offspring (Kline et a1. 1997, Lichatowich 1999). There is
evidence of the incorporation of MDN into stream communities (Schuldt and Hershey
1995, Bilby et al. 1996) and the short-term stimulation of primary production and

increases in certain fauna (i.e. Chironomid midges) (Kline et al. 1997,Wipfli et al. 1998,

50

Wipfli et al. 1999, Chaloner et al. 2002a), however, the influence of MDN on the
productivity of these streams remains unclear. In fact, while many studies discuss the
implications of their results in terms of production, the influence of MDN on secondary
production (i.e. accrual of biomass over time) has yet to be measured (Gende et al. 2002).

Alaska is one of the few areas in the United States where salmon runs remain at or
near historic levels (Baker et al. 1996, Gresh et al. 2000). Southeast Alaska contains the
8.5 million hectare Tongass National Forest, with 5200 anadromous salmon streams that
collectively support millions of spawning salmon (e. g. annual transport of over 100
million kg carbon, 10 million kg nitrogen, 2 million kg phosphorous and other nutrients
to freshwater streams) (Halupka et al. 1999, Gresh et al. 2000). The objective of this
study was to measure secondary production of selected aquatic insects to evaluate the
influence of MDN on their annual production in these systems. By conducting research
in Alaskan streams we were able to take advantage of the relatively pristine state of the
MDN transfer cycle in streams in this region. This study also takes advantage of the fact
that southeast Alaska contains many streams with reaches open to the marine
environment that provide spawning habitat for annual migrations of salmon, but also
have natural waterfalls that block salmon from reaches further upstream (i.e. natural
control), and has done so for thousands of years.

In order for these nutrients to be of real importance to overall stream productivity,
MDN must extend a significant distance upstream and be retained long enough for the
bottom-up response of the fauna. We hypothesized that if MDN does provide an
important nutrient subsidy to these streams, then aquatic insects living below the

waterfall barriers (i.e., with MDN in the system) will exhibit higher annual production

51

rates than aquatic insects living above the barriers (i.e., without MDN). To understand
how MDN influences secondary production of different types of insects, we selected
insects that are common and abundant in southeast Alaskan streams and have varied life

histories (Table 1).

Table 1. List of taxa studied for secondary production. Taxa used were common in

both study streams.

 

 

Order Family Genus Dominant Species Voltinism
Ephemeroptera Baetidae Baetis bicaudatus Bi-voltine
Ephemerellidae Drunella doddsi Univoltine
Heptageniidae Epgorus Univoltine
Cinygmula Univoltine
Rhithrogena Univoltine

Diptera Chironomidae1 Multivoltine

 

1 Chironomids were grouped at the family level for secondary production analyses.

See table 4 for more detailed taxonomic information on nridges.

52

Methods
Study Area

Fish Creek (58°19'N, 134°35'W) and Salmon Creek (58°19'N, 134°27'W) are both
anadromous streams in the J uneau-Douglas area in Southeast Alaska. Both streams are
characterized by the cool, clear, oligotrophic appearance, typical of streams in the Pacific
Northwest. Fish Creek (watershed area: 36 kmz) is on Douglas Island and receives
annual runs of salmon (Oncorhynchus) including: chum (Q. m), chinook (Q.
tshawflscha), coho (Q. EM, and pink (9. gorbuscha). The largest runs are the pink
and chum spawning migration, which normally take place between July and September.
Salmon Creek (watershed area: 26 kmz) is located near downtown Juneau and receives
pink, chum and coho salmon, with pink and chum also being the largest runs. Both
study streams have natural waterfall barriers that block salmon migration from
“upstream” reaches. Previous habitat sampling showed that upper and lower reaches
were similar for both streams (e. g. substrate, canopy cover) (Chaloner et a]. 2004,
Lessard, unpublished data) (Table 2).

Samples of benthic invertebrates were taken using a modified Hess-sampler (0.04
m2, mesh size 250 pm). On each sample date, nine samples were collected from riffle
areas in each stream section (upper and lower) from each study stream. Benthos samples
were collected in each stream approximately every two weeks from May to September,
2002 (18 May, 14 June, 29 June, 12 July, 28 July, 10 August, and 27 September) for a
total of 252 samples. Samples could not be collected all year due to high flows and
logistical constraints in the late fall and winter. The samples collected, however, should

have captured most of the production for the year as light and temperature were at a

53

Table 2. Habitat data for study streams by section. Mean temperatures are in parentheses.

 

 

Stream Section Canopy Substrate Water Mean Mean
Temp. (°C) Width (m) Depth (cm)

Fish Creek Downstream Conifer cobble/boulder 5- 12 (7) 20.5 27.3

Upstream Conifer cobble/boulder 5-12 (7) l 1.9 35.4

Salmon Creek Downstream Conifer/Alder cobble/boulder 4-10 (7.3 11.4 26.6

Upstream Conifer cobble/boulder 4- 10 (7.3) 9.7 30.9

 

maximum, and our sample period extended through the major fall runs of pink and churn
salmon and well into the period of carcass decomposition.

Samples were washed into labeled zip-top bags, fixed with 90% ethanol in the
field and transported back to the laboratory for processing. In the laboratory, samples
were picked under magnification and sorted. Insects were identified and measured for
total length. Biomass (i.e., dry mass) was calculated using length-weight regressions
from Benke et al. (1999). Preliminary data analysis revealed overlapping cohorts,
therefore secondary production was calculated using the size-frequency method (Benke
1996). Cohort production intervals were estimated from either size frequency histograms
of individual taxa or were taken from the literature. Standing stock biomass means and
standards errors for each taxon, over the study period, were calculated using SYSTAT

statistical software.

54

Results

Secondary production patterns between upstream and downstream sections were
similar for both Fish Creek and Salmon Creek (Table 3, Fig.1). _B_2§_t_i§ production was
similar between upstream and downstream sections, where as the other mayfl y genera
(Drunella, Cinvgmula, Epeorus and Rhithrogena) had consistently higher production in
the upstream sections of both study streams (Table 3, Fig.1). Chironomid production
showed the opposite trend with production rates over 800% and 600% higher in
downstream sections of Fish and Salmon Creeks, respectively (Table 3, Fig. 2). The
differences in production of each taxon by section represent not only a difference in
number of individuals, but also differences in individual body size. Although maximum
larval length of m was similar between sections, all other mayfly taxa were larger in
upstream sections, where as chironomids were larger in downstream sections (Table 3).

Chironomids were only analyzed for production at the family level, but the
proportion of subfamilies and the number of dominant taxa were documented for each
stream and stream section (Table 4). Overall, the Orthocladiinae comprised between
95% and 99% of the total midges in these streams, with the Tanytarsini making up the
remainder of the Chironomidae. The richness of dominant chironomid genera tended to
be higher in upstream sections, and overall it was higher in Salmon Creek.

Because the production patterns were similar across both study streams and
genera within a family, we averaged the standing stock biomass data for each sample date
across streams and genera. Figures 3 and 4 show the average standing stock biomass
over time by stream section for the three mayfly families (Baetidae, Ephemerellidae and

Heptageniidae) and one dipteran family (Chironomidae) studied.

55

 

0 m6 mg. : n42 We AEEV 5mg..— EEQA x32 8.3me3

 

 

 

 

 

 

 

 

 

 

«_ ms 2 a n: 3 255 593 3:3 52 885238
«.:. an 3. «.2 n.« «.m 3: ME 5333
9: to «.o «a 3. no 3: mi 585238
33 a.«m- «.3- 4.3- «.5- 2- 855:5 s
«SE 4.3. 32 24$ an: «.34 Aaaaeuev 88:85 88:8:
332 3% E: 2% 3m 25. Aaaaeué 58:95 585238
«.9. «.8 4.2 22 9% «a: $35 3285 Sosa:
32 «.3 Ea Sn 2 OS @595 £82m Engage
anEQSHC dam «EEMNEU dam «Eoglm dam «cowegsm dam £255 dam 30am x020 flea—am
c N. m : n.«_ 0 A85 593 3:3 .32 88:8:
h «a 3 ca ca 3 3:5 593 323 .32 538558
«.3 2: 2. mi 3 «.o 3: ME Essa:
«.3. «.n «.e 2: an .3 3: ma 685238
33 «.2- 3c- «.2- 33- 2: 855:5 8
2E 23 _.m_« «.«m«_ 38 m.m«m Eob«§«§ 88:85 8853:
343 «.2 9% an: «.2 «a? Aaobauaué 88:85 835238
«.2 3.« n.«« 32 o«_ 4.«n €35 £52m 5353:
32 4.2 _.: a: m _.«m $55 3285 anaemia
omEEocohzu dam Esfiwxfiu dam macaw dam «cowEEEM dam 2.2.39 d% 325 «.995 .3..—
33,—.

 

.8359: 5:26on 5:an 9 $3.? gamma: USN 8359526 nouosuoa «BE: 8 36:88on 82? “Eamon
55 8:88 833 503qu cowoscoa 5:88 5 8:88.56 2: mm connect? :8ch .3883? Ea 8855503
some 833 :08 «8 53w 8a 828/ £020 nogmm Ba 3080 in Sch 8ch 2:88qu ES 805%

3.32: 2:.“ H8 mfimafl .309 332 SEEKS: was «88355 $288 .238 .35 32:05 Moo; wEuaSm .m 033.

56

Production (mg/mz/y)

1400.0
1200.0
1000.0
800.0
600.0
400.0
200.0

0.0
1400.0

1200.0

1000.0

800.0

600.0

400.0

200.0

0.0

 

 

 

 

   

 

 

 

 

 

 

 

<z- Q‘ . .
39¢? 0;? 65% f
O
«3% (0

Fish Creek

_J | .J | m Downstream
' ' E] Upstream

Salmon Creek

 

99?.

of

Figure 1. Secondary production of five mayfly genera for upstream (blocked from salmon)

and downstream (open to salmon) sections of Fish Creek and Salmon Creek.

57

Chironomidae
12000

10000 *

8000 r

6000‘

4000—

Production (mg/ m2/y')

 

 

 

 

 

Fish Creek Salmon Creek

Downstream
D Upstream

Figure 2. Secondary production of Chironomidae for upstream (blocked from salmon)

and downstream (open to salmon) sections of Fish Creek and Salmon Creek.

58

Table 4. Dominant genera and the proportion of chironomid subfamilies or tribes

in each stream and stream section (downstream or upstream of the waterfall barrier)

 

Stream Section

Subfamily/Tribe Proportion

Dominant Genera

 

Fish Creek Downstream

Downstream
Upstream

Upstream

Orthocladiinae 96.32

3.68
99.29

Tanytarsini
Orthocladiinae

Tanytarsini 0.71

Cognoneura

Cricotopus
Eukiefferiella

Orthocladius

Micropsectra

Corynoneura
W12

Orthocladius

Para haenocladius
Tvetenia

 

 

 

Salmon Creek Downstream

Downstream

Upstream

Upstream

Orthocladiinae 96.55

Tanytarsini 3.45

Orthocladiinae 90.63

Tanytarsini 9.38

 

Cormoneura
Eukiefferiella

mm
Thienemanniella
Tvetenia

MW
Tanflarsus

Corygoneura
Eukiefferiella

Q. (Euorthoclafl)
Orthocladius
Paraphaenocladius
Paratrissocladius
Psectrocladius
Thienemanniella
Tvetenia

Micropsectra
Tan rsus

 

59

300*
250‘

150‘

 

 

 

Biomass (mg/m2)

 

p—A
Ur
o
fur
q.—-————-

   

..I
0..
'0

  

 

 

S\\% 61‘“ 509 107' 10% %no 90."

Figure 3. Standing stock biomass of mayfly families across sample dates. Means (+/- SE)
for upstream (Cl ) and downstream (O ) sections for Baetidae (A), Ephemerellidae (B) and
Heptageniidae (C) are shown averaged across both study streams. The vertical dotted line

represents when the spawning run began in downstream sections.

60

600 ' Chironomidae

   
  

Biomass (mg/m2)

  

 

I
-r‘*'r-§"T’“'I' ‘-

5\\% 60‘ 6'19 1N)“ qflfz’ %\\° 9‘11

 

Sample Date

Figure 4. Standing stock biomass of chironomids across sample dates. Means (+/- SE)
for upstream (Cl ) and downstream (O ) sections are shown averaged across both study

streams. The vertical dotted line represents when the spawning run began in downstream sections.

61

Mayfly biomass in both stream sections was similar early in the year and then diverged
with biomass increasing at a greater rate in upstream versus downstream sections for both
Ephemerellidae and Heptageniidae (Fig. 3). Baetidae showed a similar trend, but had
greater variation and was not notably different on most sample dates. The Chironomidae
showed the opposite trend, and had greater differences in biomass between sections in
May with biomass increasing at a faster rate in downstream sections. All mayfly families
showed a decline in standing stock biomass just before and during the salmon run in
downstream sections. Upstream sections also showed a general decline in mayfly
biomass after the run, but not as severe as in downstream sections (Fig. 3). Chironomid
biomass peaked just before the run in downstream sections and then declined steeply
during the run (Fig. 4). In late September, after the run and once carcass decomposition
was well underway, biomass between stream sections was similar for each taxon.
Discussion

Theories about the linkages between nutrients transported by spawning salmon
and juvenile salmonid production include many intermediate steps as these nutrients flow
through stream communities to salmon fry via bottom-up processes (Kline et al. 1997,
Lichatowich 1999, Gende et al. 2000). These theories have been bolstered by studies
using stable isotopes to trace MDN into stream biofilm, aquatic insects, and resident fish
(Bilby et al. 1996, Chaloner et al. 2002a). Studies looking at standing stock comparisons
of selected taxa in both natural and artificial streams have found differences under
various circumstances that also have implied a nutrient subsidy effect of MDN (Kline et
al. 1997, Wipfli et al. 1998, Wipfli et al. 1999). However, the incorporation of MDN into

stream communities and even higher standing stock biomass in salmon treatment areas

62

does not automatically imply a stimulation of production for stream communities, as
many studies have concluded. The nature of salmon spawning runs demands that prior to
any real nutrient enrichment there is a large disturbance of the stream benthos (Minakawa
1997, Peterson and Foote 2000). It follows then, that this disturbance should increase
with the number of spawners so that disturbance effects increase with enrichment
potential. In order for enrichment influences of MDN in natural stream systems to be
understood, it is necessary to consider the influence of the physical act of spawning,
something mesocosm studies are unable to do. Similarly, pre/post standing stock studies
of insect abundance and biomass do not give a realistic picture of what is occurring
during the primary growing season. This study is the first to document secondary
production differences in relation to MDN enrichment.

This study reveals the complex relationship between salmon and energy flow in
streams. For larger sized, univoltine taxa such as ephemerellid and heptageniid mayflies,
the influence of spawning disturbance appears to overshadow any potential enriching
effect of the salmon. Upstream sections represent areas of refuge that allow for greater
production than in downstream sections. Average biomass was initially higher in the
spring in upstream sections for these mayflies, which may be due to higher numbers of
adults laying eggs in this area. There are certain species of mayflies (e. g. Rhithrogena
spp.) that have an over-wintering cohort, and these likely are severely reduced or even
eliminated in downstream sections during the run. Both Drunella spp. and Rhithrogena
spp. had large individuals sampled from upstream sections in May, where as only small
individuals were sampled from downstream sections on the same date (Lessard,

unpublished data). Because insect biomass naturally declined in both sections in the fall,

63

even in upstream sections, there may have been selective processes over time that have
driven insect life histories to have emergence periods occurring prior to the big salmon
runs in this region. This type of influence on aquatic communities would be in direct
opposition to MDN enrichment theories.

Chironomid patterns in these streams follow the predictions of MDN enrichment
theories. Production was consistently much higher in downstream “treatment” sections
of these two streams. This production primarily took place over the spring and summer,
however, and similar to the ephemeropteran taxa, rnidge biomass declined during the run
and remained low well into the carcass decomposition period. In fact, rnidge biomass
was at a minimum during this period. For the midges to take advantage of salmon
nutrients they would need to remain in the stream in adequate numbers to capitalize on
this carcass decomposition period. There were several potential factors that could have
influenced the midge patterns observed here. Although the focus of this research was not
to determine which mechanism(s) were influencing these communities, it is still pertinent
to discuss what could be driving these patterns. Midges may be able to avoid the
disturbance of spawning by virtue of their small size allowing them to take advantage of
hyporheic or interstitial spaces, thereby maintaining higher population levels in these
areas, ultimately resulting in more individuals emerging and reproducing. However, this
does not explain why rnidge populations are lower in upstream sections, where no
spawning occurs. It may be that the lack of mayflies in downstream reaches reduces
invertebrate competitive and predatory (Drunella are considered predators at larger sizes)
interactions so that midges are able to grow and produce at much higher rates in

downstream sections through the spring and summer. This seems contrary to other

64

studies that have shown salmon spawning to increase drift and fish predation on insects,
primarily midges (Peterson and Foote 2000). However, studies of fish predation effects
on chironomid midges have found that the indirect effects of predatory fish on
invertebrate rnidge competitors and predators, benefited chironomid populations more
strongly than the negative direct effects of fish predation on the midges (Power et al.
1992, Batzer et al. 2000, Rosenfeld 2000). The indirect effects of spawning and
predation on midges may be very important mechanisms in these systems as well. This is
further supported by the fact that the fish community richness in downstream sections
differ from upstream sections with sculpin, resident and ocean run Dolly Varden as well
as juvenile salmonids occurring downstream and only Dolly Varden occurring upstream
in these systems (J. Hudson, US Forest Service, Juneau, Alaska, unpublished data).
Chironomids also have been identified in some streams as primary colonizers of salmon
carcasses during decomposition (Chaloner et al. 2002b), so it may be that the rnidge
community is temporarily diverted to the carcasses from other areas in the stream
resulting in reduced densities on mineral substrates. This explanation could not be
confirmed because carcasses were not sampled in this study. If midges are able to
capitalize on MDN during carcass decomposition before winter, a net effect would be
greater production of over-wintering cohorts the following year. Another closely related
but distinct potential mechanism that could be driving the production patterns in these
streams, would be based on the functional feeding groups that dominate each stream
section. The upstream sections, that do not receive spawning runs, are assumed to be less
disturbed over the year and so would support more consistent periphyton communities. It

seems reasonable that scrapers (like heptageniids and ephemerellids) would do better in

65

more stable habitats (i.e., upstream sections). The downstream sections, however, not
only are less stable (due to spawning activities) but also contain decomposing salmon
particles during the fall-winter period. This combination of unconsolidated sediments
and high quality FPOM, should benefit collector-gatherers, such as midges of the tribe
Orthocladiinae. Facultative collector-gatherers, such as baetids, would be expected to do
fine under either scenario, and therefore would be expected to show little difference in
their production between stream sections.

This research points to spawning disturbance, and possibly fish predation, as
major driving forces organizing the benthic community so that areas upstream of salmon
barriers support much greater production of univoltine, grazing, mayfly genera. It also
suggests that this production of mayflies imposes constraints on the rnidge community in
upstream sections that is released by the cycle of spawning and the presence of fish
(resident species, juvenile salmonids and oceanic transient species) in sections open to
the ocean. It is not clear if salmon impact midges by indirectly limiting invertebrate
competitors and predators and/or stimulate midge production with the nutrients provided
by the spawners. It is also not clear if the lower production of mayflies in spawning areas
is due to increased predation by fish and/or the spawning disturbance. It seems likely that
all these mechanisms may be acting simultaneously resulting in the production patterns
seen in these streams. The link between MDN and juvenile salmonid production in these
systems, therefore, may rely solely on chironomid production. Both pink and chum
salmon have been shown to feed mainly on midges during their seaward migration
(Frolenko 1973, Loftus and Lenon 1977, Kaeriyama et al. 1978). Chinook and coho

juveniles are also known to feed predominantly on midges (Loftus and Lenon 1977,

66

Dauble et al. 1980). The linkages between anadromy and juvenile salmon production
may have less to do with enrichment than with spawning and fish community
characteristics for streams in southeast Alaska. These factors need be considered and

incorporated into MDN theory.

67

LITERATURE CITED

Baker, T.T., A.C. Wertheimer, R.D. Burkett, R. Dunlap, D.M. Eggers, E.I. Fritts, A.J.
Gharrett, R.A. Holmes, and R.L. Wilmot. 1996. Status of Pacific salmon and
steelhead escapements in southeastern Alaska. Fisheries 21: 6-18.

Batzer, D.R., C.R. Pusateri, and R.Vetter. 2000. Impacts of fish predation on marsh
invertebrates: Direct and indirect effects. Wetlands 20: 307-312.

Benke A.C., A.D.Huryn, L.A. Smock, and J .B. Wallace. 1999. Length-mass relationships
for freshwater macroinvertebrates in North America with particular reference to the
southeastern United States. Journal of the North American Benthological Society 18:
308-343.

Bilby R.E., B.R. Fransen, and RA. Bisson. 1996. Incorporation of nitrogen and carbon
from spawning coho salmon into the trophic system of small streams: evidence
from stable isotopes. Canadian Journal of Fisheries and Aquatic Sciences 53:164-
173.

Cederholm, C.J., M.D. Kunze, T.M. Murota, and A. Sibatani. 1999. Essential
contributions of nutrients and energy for aquatic and terrestrial ecosystems.
Fisheries 24:6-15.

Chaloner D.T., K.M. Martin, M.S. Wipfli, P.H. Ostrom, and GA. Lamberti 2002a. Marine
carbon and nitrogen isotopes in southeastern Alaska stream food webs: evidence from
artificial and natural streams. Canadian Journal of Fisheries and Aquatic Sciences 59:
1257-1265.

Chaloner, D.T., M.S. Wipfli, and J .P. Caouette. 2002b. Mass loss and macroinvertebrate
colonisation of Pacific salmon carcasses in south-eastern Alaskan streams.
Freshwater Biology 47: 263-273.

Chaloner,D.T., G.A. Lamberti, R.W. Merritt, N.L.Mitchell, P.H. Ostrom, and MS.
Wipfli. 2004. Variation in responses to spawning Pacific salmon in three
southeastern Alaska streams. Freshwater Biology 49: 587-599.

Cummins, K.W. 1974. Structure and function of stream ecosystems. Bioscience
24:631-640.

Cummins, K.W., M.A. Wilzbach, D.M. Gates, J .B. Perry and W.B. Taliferro. 1989.
Shredders and riparian vegetation. Bioscience 39: 24-30.

Dauble, D.D., R.H. Gray, and T.L.Page. 1980. Importance of insects and zooplankton in

the diet of 0-age chinook salmon (Oncorhynchus tshawytscha ) in the central
Columbia River. Northwest Science 54: 253-258.

68

Frolenko, LA. 1973. Feeding of chum and pink salmon juveniles migrating downstream
in the main spawning rivers of the northern coast of the sea of Okhotsk. Fisheries
Research Board of Canada Translated Serial no. 2416: 22 pp.

Gende S.M., R.T. Edwards, M.F. Willson, and MS. Wipfli. 2002. Pacific salmon in
aquatic and terrestrial ecosystems. BioScience 52: 917-928.

Gresh, T., J. Lichatowich, and P. Schoonmaker. 2000. An estimation of historic and
current levels of salmon production in the northwest Pacific ecosystem: evidence
of a nutrient deficit in the freshwater systems of the Pacific Northwest. Fisheries
25: 15-21.

Halupka, K.C., M.D. Bryant, M.F. Wilson and EH. Everest. 1999. Biological
characteristics and population status of anadromous salmon in Southeast Alaska.
Gen. Tech. Rep. PNW-GTR-468. U.S. Dep. Agriculture, Forest Service, Pacific
Northwest Research Station, Portland, Oregon.

Kaeriyama, M., S. Sato, and A. Kobayashi. 1978. Studies on a growth and feeding habit
of the chum salmon fry during seaward migration in the Tokachi River system. 1.
Influence of thaw on a growth and feeding habit of the fry. Scientific Report of
Hokkaido Salmon Hatchery no. 32: 27-41.

Kaushik, N .K., and H.B.N. Hynes. 1971. The fate of dead leaves that fall into streams.
Arch. Hydrobiol. 68:465-515.

Kline, T.C., J .J . Goering, and R.J. Piorkowski. 1997. The effect of salmon carcasses on
Alaskan freshwaters. Chapter 7 pgs 179-204, In Freshwaters of Alaska,
ecological syntheses. A.M. Milner and M.W. Oswood (Eds.). Springer-Verlag,
New York.

Lichatowich, J .A. 1999. Salmon without rivers. Island Press, Washington DC. 317 p.

Loftus, W.F. and H.L. Lenon. 1977. Food habits of the salmon smolts, Oncorhynchus
tshawytscha and O.keta, from the Salcha River, Alaska. Transanctions of the
American Fisheries Society 106: 235-240.

Minakawa, N. 1997. The dynamics of aquatic insect communities associated with
salmon spawning. Ph.D. dissertation. University of Washington, Seattle.

Peterson, DP. and OJ. Foote. 2000. Disturbance of Small-Stream Habitat by Spawning
Sockeye Salmon in Alaska. Transanctions of the American Fisheries Society 129:
924-934.

Polis G.A., W.B. Anderson, and RD. Holt. 1997. Toward an integration of landscape and

food web ecology: the dynamics of spatially subsidized food webs. Annual
Review of Ecology and Systematics 28: 289-316.

69

Power, M.E., J .C. Marks, and MS Parker. 1992. Variation in the vulnerability of prey to
different predators: Community-level consequences. Ecology 73: 2218-2223.

Rosenfeld, J .S. 2000. Contrasting effects of fish predation in a fishless and fish-bearing
stream. Archiv fuer Hydrobiologie 147: 129-142.

Schuldt, J .A., and A.E. Hershey. 1995. Effect of salmon carcass decomposition on Lake
Superior tributary streams. Journal of the North American Benthological Society
14:2569-268.

Vannote, R.L., G.W. Minshall, K.W.Cummins, J .R. Sedell, and CE. Cushing. 1980.
The river continuum concept. Canadian Journal of Fisheries and Aquatic
Sciences 37:130-137.

Wipfli, M.S., J. Hudson, and J. Caouette. 1998. Influence of salmon carcasses on stream

productivity: response of biofilm and benthic macroinvertebrates in southeastern
Alaska, USA. Canadian Journal of Fisheries and Aquatic Sciences 55:1503-
151 1.

Wipfli M.S., J .P.Hudson, D.T. Chaloner, and J .P. Caouette. 1999. The influence of

salmon spawner densities on stream productivity in Southeast Alaska. Canadian
Journal of Fisheries and Aquatic Sciences 56: 1600-1611.

70

Chapter 4
J .L. Lessard, R.W. Merritt, and K.W. Cummins. 2002. Spring growth of caddisflies

(Limnephilidae: Trichoptera) in response to marine-derived nutrients and food type in a
South-east Alaskan stream. Ann. Limnol.-Int. J. Lim. 39: 3-14.

71

Chapter 4
Spring growth of caddisflies (Limnephilidae: Trichoptera) in response to marine-

derived nutrients and food type in a South-east Alaskan stream

Abstract

The short-term stimulation of production, due to marine-derived nutrients (WN)
from spawning salmon, for certain trophic levels in stream communities (e. g. algae and
insect biomass) is well documented. The effect of these nutrients on the stream
ecosystem as a whole, however, especially later in the year remains unclear.
Trichopterans have been shown to feed on salmon and other fish carcasses and there is
evidence for greater growth rates in the presence of salmon tissue. To address the
question of long-term MDN subsidy on Trichopterans, we investigated the growth of
three limnephilid caddisflies in the spring in the Harris River on Prince of Wales Island,
Southeast Alaska. The Harris River has a natural waterfall barrier to salmon and receives
large runs of pink and churn salmon each fall. We selected two shredding caddisflies
(Onocosmoecus unicolor) and (Psychoglypha sp.) and one facultative scraper,
(Dicosmoecus atripes) for our study. We had two objectives: 1) compare the spring
growth of larval caddisflies in a stream section that receives a large autumn run of salmon
with their growth in a stream section that is blocked from receiving salmon (due to an
impassable waterfall), and 2) compare the growth of shredders with that of a facultative
scraper when provided either leaves or biofilm on rocks as food.

Insects were placed in growth boxes 18 May 2001 with either conditioned alder

leaves or rocks from the stream in the boxes. The boxes along with temperature loggers

72

were placed in both the salmon (below the waterfall) and non-salmon (above the
waterfall) reaches. On 26 June 2001, 40 days later the boxes were removed. In-stream
samples were taken of each caddisfly initially and at the end of the experiment to
establish in-stream growth versus growth in the boxes. All larvae were coaxed from their
cases from the rear with a blunt probe, measured for total wet length, dried and weighed.
Only Q. m and Psychoglypha were growing during our experiment and both
showed very high relative growth rates in the Harris River. We found a strong food
effect with both Psychoglvphg and Q. Mg being significantly larger in the leaf boxes
and Q. m being significantly larger in the rock boxes. Both Q. a_tg'_pe_s and
Psvchoglyplm had significantly greater relative growth rates between food types (on
biofilm on rocks and leaves respectively). These results support that Q. m are most
likely facultative scrapers at least in their first year of growth. None of these caddisflies
showed differences in their final mean weights or relative growth rates between stream
sections, indicating no effect of MDN on their spring growth in the Harris River. Further
research on caddisfly growth in the fall and winter will help clarify if MDN has an
influence on the life history of these species closer to the salmon run. This study
questions the long-terrn influence of MDN on stream communities, particularly those
parts that do most of their production in the spring, months after salmon carcasses are no

longer visible.

73

Introduction

The utilization of terrestrially derived nutrients (e. g., leaf litter) is considered to
be the driver of productivity and the spatial/temporal organization of many stream
communities (Kaushik & Hynes 1971, Cumrrrins 1974, Vannote et al. 1980, Cummins et
al. 1989). Coastal streams, that are spawning ground for salmon, have an additional
nutrient source termed marine-derived nutrients (MDN) in the form of salmon eggs,
sperm, metabolic waste and adult carcasses. The role that these nutrients play in coastal
systems has been the object of study in recent years. It has been suggested that the
salmon provide an essential nutrient source to the typically oligotrophic anadromous
streams of the Pacific Northwest of the United States and, by subsidizing the nutrient
base in their spawning grounds, increase the viability of their own offspring (Kline et al.
1997, Lichatowich 1999). Most studies have dealt with tracing MDN through various
surface-stream and riparian trophic levels or quantitative comparisons of stream
communities with and without salmon (e.g. Bilby et al. 1996, Kline et al. 1997, Ben-
David et al. 1998, Wipfli et al. 1998, Cederholm et a1. 1999, Chaloner et al. 2002). There
is ample evidence of the incorporation of MDN into stream communities (Schuldt &
Hershey 1995, Bilby et al. 1996) and the short-term stimulation of primary production
and increases in certain fauna (i.e. chironomid midges) (Kline et al. 1997,Wipfli et al.
1998, Wipfli et al. 1999, Chaloner et al. 2002). However, the long-terrn influence (i.e.
through the winter and into the spring the following year) of MDN on the productivity of
these streams is still unclear.

The importance of MDN to a stream community will depend on how far upstream

the enrichment extends and the length of time these nutrients are retained in the system.

74

This is particularly true for insects not growing in the streams immediately following the
major salmon run(s). Insect responses to enrichment take several forms including
changes in population size, voltinism, fecundity and growth rates (e. g Anderson &
Cummins 1979, Sweeney 1984, Arsuffi & Suberkropp 1986, Peterson et al. 1993).
Several studies as well as observations by the authors, have indicated that
Trichopterans are attracted to and feed on salmon and other fish carcasses (Brusven &
Scoggan 1969, Chaloner et al. 2002). Minakawa (1997) found greater growth in
Trichopterans directly feeding on MDN (i.e. salmon carcasses). The indirect influence of
NEDN on insect growth, however, is poorly understood. The objectives of this study were
to: 1) address the question of long-term MDN subsidy (via indirect pathways) by
comparing the spring growth of caddisflies above and below a natural barrier to salmon
(i.e. impassable waterfall) in one stream in Southeast Alaska; and 2) compare the growth
of selected caddisflies who are known to have different feeding modes (i.e., shredding
and scraping) on two food sources, leaves and biofilm (defined as a matrix of algae,
bacteria, fungi, protozoans and other organic matter). We hypothesized that if MDN does
in fact provide an important long-terrn nutrient subsidy to these streams, then organisms
grown below the waterfall barrier (i.e. with MDN in the system) would exhibit higher
growth rates than organisms grown above the barrier (i.e. without MDN). The
mechanism for increased growth was assumed to be greater food quality in the
downstream section. We also hypothesized that when provided with either leaf litter or
biofilm on rocks as food, shredders would grow better on leaves and scrapers would grow
better on biofilm. The organisms selected have at least a one-year life cycle and in some

circumstances take two years before metamorphosis, therefore, they have the ability to

75

integrate the influence of MDN over the entire year. The timing of this study, however,
was to determine if MDN influences caddisfly growth in the spring, eight months after
the autumn salmon run.
Methods and materials
Study site

Harris River is an anadromous salmon stream on Prince of Wales Island, which is
approximately 45 miles west of Ketchikan in Southeast Alaska. It receives annual runs
of coho (Oncorhynchus kisutch) pink (Q. gorbuscha) and chum (Q. m) salmon. The
largest runs are the pink and chum spawning migration, which normally take place
between late July and early September. Harris River received an estimated 920,000
chum and pink salmon the autumn of 2000. previous to the experiment (Alaska
Department of Fish &Game, pers. comm.) Harris River has a natural waterfall that
blocks salmon migrations from “upstream” reaches, and has done so since the last ice
age. Preliminary habitat sampling showed that reaches above and below the waterfall
(termed upper and lower Harris) were of similar mean depth (~40 cm), had similar
bottom substrate (primarily small-large cobble and boulders) and riparian vegetation
(mixtures of conifer and alder trees, shrubs, ferns and herbaceous plants), but lower
Harris was wider and had less wood debris than upper reaches (Lessard, unpublished
data).
Study Organisms

The limnephilid caddisflies we selected were Onocosmoecus unicolor (Banks),
Dicosmoecus m (Hagen) and Psychoglvpha sp. (Banks), all of which are fairly

common fauna in anadromous streams of southeast Alaska. All three of these genera

76

have also been associated with fish carcasses (Brusven & Scoggan 1969, Chaloner et a].
2002). These species, however, vary in their primary modes of feeding. Both Q.
M and Psvchoglvphg sp. are considered obligate shredders (Wiggins & Mackay
1978, Merritt & Cumrrrins 1996), while Q. m larvae are facultative scrapers and
shredders as early instars and shredders-predators during later instars (Wiggins &
Richardson 1982, Gotceitas & Clifford 1983). While all species were found in both
stream sections of the Harris River, preliminary sampling revealed that downstream of

the barrier there was a greater abundance of Q. unicolor and Q. atripes while upstream

 

there were more Psychoglypha sp.
Growth study

The experiment was initiated on 18 May 2001 during which time caddisflies were
sampled from both upper and lower Harris River. Due to the lack of a large enough
population of all three taxa in both reaches, all Q. Mg and Q. gt_rip_e§ used in the
experiment were from the lower reach and all Psvchoglvpha sp. used were from the upper
reach. There also were some apparent instar differences among the Q. m larvae: two
general groups of 2'“d-3rd instars, and 3:44:11 instars. Because of this, each of these size
class groups was grown separately and the number of individuals varied between 5 and

10 per growth box (Figure 1). Q. unicolor and Psychoglypha had only one size class.

 

Insects were captured from cobble, side areas and pools using D-frame nets.
After sampling each reach, caddisflies were sorted and placed into growth boxes
(modified 25x18x5 cm Tupperware ® containers with the top, sides and bottoms cut out

and replaced with 2mm plastic mesh).

77

Harris River

  
 

 

 

 

 

 

 

 

 

 

 

Growth Box Re licates
Upstream
5 Dicosmoecus & Boxes
5 Psychoglypha
Per box
5 Dicosmoecus & Waterfall
5 Psychoglypha Barrier
Per box
Downstream
10 Dicosmoecus & axes
10 Onocosmoecus
Per box | Area
10 Dicosmoecus & I 0f
10 Onocosmoecus I MDIY
Per box Ennchment

 

 

 

 

 

 

 

Mouth

Figure 1. Schematic of growth experiment conducted in the Harris River. Identical
set-ups were used in upstream and downstream sections. Box cartoons show the
combinations of food type (conditioned alder leaves and rocks from the stream) and

taxa that were put in each box, and replicated three times.

78

Food for the growth boxes consisted of either 12 conditioned (incubated in the stream for
2 weeks) red alder leaves (m r_ul_)ra Bong) or biofilm on 4-5 small rocks from the
stream, both were inspected and had any insects removed before being put into the
grth boxes. Caddisflies from each section (Q. m and Q. unjgo_lor from
downstream, Psychoglypha from upstream) were grown in both sections (above and
below the waterfall) and on both food types (conditioned leaves or biofilm on rocks).
Each box contained Q. m and either Q. unicolor or Psychoglvnha, so that one of the
obligate shredders and the facultative scraper were in each box. Five 3“'-4th instar Q.
m and 5 Psvchoglvpha were placed in each of 6 boxes (3 filled with leaves and 3

filled with rocks), while 10 Q. unicolor and 10 2nd —3rd instar Q. atrip_es were placed in

 

another 6 boxes (3 filled with leaves and 3 filled with rocks). There were a total of 12
boxes placed in each stream section (Figure 1) which were cable tied to rebar (metal rods
~1.25 cm diameter) imbedded in the stream bottom in a slow flow area near the stream
bank. Onset® temperature loggers were also placed in each stream section and set to
take hourly readings. The remaining organisms from the stream samples of each
caddisfly (30-50 per taxa) were used to obtain an initial population mean wet length and
dry weight for each. These samples were brought back to the lab, coaxed from their
cases from the rear with a blunt probe, measured to the nearest 0.5 mm, oven dried (30°C
for four days) and weighed using a Cahn-Electrobalance (to the nearest 0.01 mg).
Throughout the experiment boxes were monitored to ensure no tampering or
sedimentation had occurred. The rocks were changed twice and the leaves once in the
boxes and conifer needles also were added. This was done to provide case making

material and ensure access to food would not lead to a growth box effect not

79

representative of in-stream growth. On 26 June, after 40 days, the experiment was
terminated. All insects were removed from the boxes, counted and kept separate for
processing. At this time another large, in-stream sample (20-30) of each caddisfly was
taken from the Harris River to obtain a final in-stream population mean wet length and
dry weight for each. All insects (from growth boxes and upstream and downstream in-
stream samples) were brought back to the lab alive and processed the same way the initial
samples were (i.e., coaxed from cases, measured for length, dried and weighed). The
temperature loggers were also removed and the data downloaded.

Instar determinations were made by measuring the width of the head, dorsally
across the eyes, using an ocular micrometer at 70x to the nearest 0.05 mm (Wiggins &
Richardson 1982, Gotceitas & Clifford 1983, Wiggins & Richardson 1986). Descriptive
statistical calculations, regressions and Mann-Whitney U tests were all computed using
Systat statistical software. Growth box effects were examined by comparing average
final weights in the boxes in each section versus average final weights of free-living
larvae in the stream. MDN effects on these caddisflies were determined by comparisons
between stream sections (i.e., above versus below). Food preferences were determined
by comparing growth on leaves versus biofilm on rocks. Statistical significance was
determined using a=0.05 for independent comparisons and bonferroni adjustments to
alpha were made for any multiple comparisons.

Results

Both Q. M and Psychoglypha showed significant growth during the

experiment and had similar final mean weights in the boxes as in the stream (i.e. no

growth box effect) (Figure 2).

80

D. atripes

’63
gr 0. unicolor
fl
.=
.9
d)
3
C
I“
d)
E
Psychoglypha

 

 

0 _

Initial Final Upstr Downstr Leaf Rock
Free Free Boxes Boxes Boxes Boxes

Figure 2.

81

Figure 2. Mean dry weight mg (+/- SE) of caddisflies from the in-stream samples (Free)
initially and at the end of the experiment, and from caddisflies confined to the boxes
divided by stream section (upstr=above the waterfall, downstr=below the waterfall) and
food type (leaves and rocks). MDN and growth box effects were determined by
comparing free and boxed individuals by section (significant differences indicated by
different letters) and food effects were determined by comparing food types (significantly
higher values indicated by *). All comparisons were done using the Mann-Whitney U

test with bonferroni adjustments made for multiple comparisons when appropriate.

82

Q. M were not growing at all in the stream during this time and were significantly
smaller in the boxes than in the stream (xzdf=1=13.349, p<0.0001). None of the caddisfly
taxa showed any differences in growth between stream sections (i.e. no MDN effect). All
three taxa, however, showed varied growth by food type with both Q. grim and
Psvchoglypha growing significantly more on leaves (xzdf=1=10.829, p=0.001;
xzdf=|=10.763, p=0.001, respectively) and Q. a_trip§ growing significantly more on
biofilm on rocks (xzdle=l4.578, p<0.0001) (Figure 2).

The upstream reach accumulated 282 degree-days (°C) throughout the 40-day
experiment, while downstream accumulated 215 degree-days (°C). Normalizing for
temperature, that is comparing relative growth rates per degree-day [calculated as:
100*(mean initial weight-mean final weight) /(mean initial weight * #of degree-days)]

revealed similar results. Q. unicolor and Psvchoglvpha had significantly greater growth

 

rates on leaves (xzdf=1=10.98, p=0.001; xzdf=1=10.345, p=0.001, respectively), and Q.
m had a significantly greater growth rate on biofilm (xzdf=1=l4.295, p<0.0001)
(Figure 3). Again, there was no MDN effect found when above and below mean growth
rates were compared for each caddisfly taxa.

Nearly all the boxes had some loss (assumed to be mortality) of caddisflies
(Figure 4). A greater percentage of Q. m compared to the other two taxa were lost
during the experiment in both stream sections, but significantly more Q. m survived
in boxes from the downstream reach (xzdf=]=4.669, p=0.031). Psvchoglvpha was the only
taxon that showed a food effect on mortality, with a significantly greater percentage

surviving on leaves (xzdf=1=8.539, p=0.003).

83

 

‘ Psychoglypha

  

% RGR/ degree day

1.6 _ 0. unwolor

1.2 '
0.8 '

 

 

-0.2 ‘ Up- Down-

Leaves Rocks stream stream
Food Section

Figure 3. Mean % relative growth rate/ degree day (° C) (+/- SE) for caddisflies grown in
growth boxes averaged by food type and stream section. An * indicates significantly

greater %RGR from the Mann-Whitney U tests.

84

1:1 Upstream Boxes

Downstream Boxes

es

D.

 

 

 

 

0. unicolor

 

 

 

 

 

 

Psychoglypha

 

 

 

 

 

 

 

100 -

80*

60*

40'-

20*

 

100 -

_ _ _ _
0 O 0 0
00 6 4 2

33.5 a

 

100 -

40a

20*

 

Leaves Rocks

Final

Figure 4. Mean % survival (+/- SE) of caddisflies in the growth boxes at the end of the

experiment by stream section (averaged across food type). Survival at the end of the

experiment is also shown by food type in each section (Leaves and Rocks). Significant

differences between stream sections (i.e., each pair of bars), determined using the Mann-

Whitney test, are indicated by an *.

85

Comparisons of mean % RGR/degree-day versus % mortality in the growth boxes,
revealed significant negative models for Q. m and Psvchoglyplla, but the high degree
of variation made these relationships quite weak, especially for Q. m (Adjusted R2:
0.19 and 0.45, p=0.022 and 0.010 respectively) (Figure 5). There was no relationship
between mean

% RGR/degree-day and % mortality for Q. unicolor (Adjusted R2: 0.00, p=0.895)

 

(Figure 5).

The large number of animals handled during this experiment enabled us to
develop strong wet length-dry weight relationships for Q. am and Psychoglypha
(Adjusted R2: 0.95 and 0.91 respectively) (Figure 6). However, the small range in sizes
for Q. 211219.12 (due to their lack of growth) made it difficult to establish a good length-
weight equation for this species (Adjusted R2=0.49). Preliminary analysis revealed that
the length-weight data were not linear so a natural log transformation was performed to

produce a linear fit. The regression equations for these data were:

 

Q. atripgs (n=221) Ln(w)=0.421L -1.865 [1]
Psvchoalvpha (n=62) Ln(w)=0.331L + 0.04 [2]
Q. unicolor (n=126) Ln(w)=0.269L + 1.716 [3]

Where Ln(w), is the natural log of the dry weight (mg) and L is the total wet length

(mm).

86

.N
u.

N

 

Mean %RGR/degree-day

O D. atripes (R2=0.188)
A Psychaglypha (Rzfl.454)
. l 0. unicolor (Rz=0.000)

 

 

0.0

 

 

0.5 0.6 0.7 0.8 0.9 1.0
% Mortality

0.1 0.2 0.3 0.4

Figure 5. Mean % relative growth rate/ degree day (° C) versus % Mortality in the

growth boxes for each caddisfly. The lines are the results of the regressions and the

symbols are from the data for each taxa. Adjusted R2 values for each model are given in

the legend.

87

 

 

 

 

 

12 ‘ D .
. atnpes
10 ‘
, Ln(w)=0.421L -1.865
8 Adjusted R2: 0.95
6 4 =221
4 'i
2 J
0
A 10 . 0. unicolor
on
g 8 Ln(w)=0.269L + 1.716
3 Adjusted R2=o.49
E” 6 N=126
'3’ 4
5‘
7'5 2
A
O T f f a
l
10
Psychoglypha
8 Ln(w)=0.33lL + 0.04
Adjusted R2: 0.91
6 N=62
4
2
0 u

 

 

0 5101520253035
Wet Length (mm)
Figure 6. Wet length-dry weight relationships for the three caddisflies. Lines are the

results of the regression equations shown to the right of each graph and diamonds are the

individual data points from in-stream samples of each taxa.

88

Discussion

There was no detectable influence of MDN on the spring growth of these three
limnephilid caddisflies in Harris River. Q. M hatch in the fall and grow quickly to
fifth instars and then remain at that stage over the winter and spring, pupating and
emerging during the summer (Wiggins & Richardson 1986, Irons 1988). Head capsule
width measurements revealed that the Harris River’s population of Q. unic_olor larvae
were fifth instars throughout our experiment. There were not enough Q. M found
above the waterfall to make in-stream above and below comparisons of weights.
However, the presence of such a large population of this species in areas that receive
salmon might be related to MDN. The life cycle of this species is timed to obtain the
greatest advantage from the salmon with most larval growth occurring in autumn during
salmon carcass decomposition (Irons 1988). Chaloner et al. (2002) found Onocosmoecus
to be one of the most common invertebrate taxa associated with salmon carcasses after
spawning runs in Southeast Alaskan streams. However, this also is the primary time for
deciduous leaf fall so this shredder’s life history may be keyed to the salmon, leaf litter
inputs or both.

Q. at_ripe_s also were found in greater numbers downstream. Q. a_trip_§§ was the
only taxon that showed mortality variation between stream sections, with significantly
more surviving in the downstream boxes. It is possible that niche partitioning in this

stream relegates Q. atripes and Q. unicolor to downstream reaches and Psychoglypha to

 

upstream areas, with further separation temporally of Q. atripes and Q. unicolor larval
growth periods. However, the growth boxes did not support this hypothesis because

more Q. atripgs were lost in boxes with Q. unicolor, their natural physical companion, but

 

89

these boxes also had twice as many and younger Q. m individuals in the boxes.
Lamberti et al. (1995) found evidence for density dependent growth of the scraping Q.
gm in laboratory streams. In our experiment there was a weak negative relationship
between %mortality and %growth for Q. m, so it appears that boxes that lost more
individuals were somehow also systematically poorer environments for growth. Due to
the complexity of this experiment, however, and the number of potentially interacting
factors, we were not able to determine inter- or intra-specific competitive interactions in
the stream or in our growth boxes.

Q. atfigs; main growth period is during the spring and early summer (Wiggins &
Richardson 1982, Gotceitas & Clifford 1983), and while they grew significantly during
our experiment we were not able to detect an influence of MDN on the spring growth of
this species either. Because Q. m when supplied conditioned leaves grew less than
when supplied biofilm in the boxes, this species is most likely a scraper, at least as earlier
instars. Other feeding and growth studies on late and middle instar Q. m larvae
concluded that this species is a shredder and sometimes a predator (Wiggins &
Richardson 1982, Mihuc & Mihuc 1995). Our data support those of Gotceitas & Clifford
(1983), in Alberta, Canada who found in the first year of their life cycle Q. m utilize
algae as their main food source

During the second sampling period we found several very large Q. m larvae
that had switched their case making from organic to mineral. Head width analysis
revealed that during our original sampling we had primarily 2nd and 3rd instars and during
the second sampling we had 3”, 4m, and 5‘h instars (Wiggins & Richardson 1982). Q.

atripes larvae have been shown to have two-year life cycles in cold streams (Gotceitas &

90

Clifford 1983). Q. m in the Harris River also appeared to have a two year life-cycle
which accounted for 5th instars being found during the second sampling. Second year Q.
m larvae over-winter as diapausing fifth instars attached in large aggregations under
large boulders, and then become active again in the spring (Gotceitas & Clifford 1983).
Dicosmoecus larvae were shown to be important processors of fish carcasses in a study of
squawfish eradication in Idaho (Brusven & Scoggan 1969). Since Q. m larvae
remain in this stream throughout the spawning run and carcass decomposition, before the
winter diapause, the importance of MDN to this species may occur during that time (i.e.
between its first and second year of development) rather than at the beginning of the life-
cycle.

Psvchoglypha_§p_. are shredding caddisflies limited to the western montane
regions of North America (Denning 1970, Wiggins & Mackay 1978, Merritt & Cummins
1996). Not much is known of the life history of this genus except that they are cold
adapted and adults have been found between late fall and early spring in southeast Alaska
and from April to August in the Yukon (Ellis 1978, Wiggins & Parker 1997). Growth
studies of Psychoglypha sp. found that these caddisflies grew differently on varying food
quality and exhibited high relative growth rates (0.51% body weight/degree day) on the
highest quality food (Arsuffi & Suberkropp 1986). Psychoglvmha in our study grew on
average at a similar rate on the biofilm and at over twice that rate on the stream-
conditioned alder leaves, indicating the generally high growth achieved by this taxon in
the Harris River. These high growth rates were similar between stream sections and
slightly higher in upstream sections, which again points to the lack of importance of

MDN to the spring growth of this caddisfly. Psychoglypha also were associated with

91

squawfish decomposition in Idaho (Brusven & Scoggan 1969), indicating the potential
for a direct relationship between MDN and this caddisfly. The dominance of
Psychoglypha_sp_. above the barrier, however, and their high growth rates upstream and
probable emergence prior to salmon decomposition indicates less of a reliance on these
nutrients for this caddisfly.

Conclusions

Salmon have an obvious effect on anadromous stream communities, from the
huge disturbance caused by their numbers and spawning activity to the pulse of nutrients
released following this activity and ultimately to the stream fish community by the
addition of their offspring (Kline et al. 1997, Minakawa 1997, Wipfli et al. 1998,
Peterson & Foote 2000). Initial post-spawning increases, however, are mainly due to the
incorporation of salmon metabolic waste, sperm, and eggs into stream communities, all
of which are temporary forms of NflDN. The question is how is the community impacted
the rest of the year? Is this enrichment retained in the system long enough to impact
fauna that are not active in the stream during the initial autumn MDN pulse?

Once spawning has ended and salmon begin dying off, the retention of these
blocks of MDN (i.e. carcasses) are now a function of hydrologic retention and stream
structure (e. g. wood debris, surface topography), as are all sources of CPOM (coarse
particulate organic matter)(Speaker et al. 1984, Cederholm & Peterson 1985). Retention
would have to involve sequestration of nutrients in the surface and subsurface community
and the slow release and cycling of these nutrients throughout the year. The easiest slow
release mechanism to visualize would be from the burial and subsequent aerobic and/or

anaerobic decomposition of whole carcasses and pieces, an area yet to be researched. A

92

more difficult and complicated mechanism would be from the stimulation of production
of autumnal flora and fauna that would cycle these nutrients via the surface and
hyporheic food web at a magnitude and rate that allows for the benefit of other organisms
months later. To our knowledge no research has been able to trace nutrients through a
stream system for that long.

By comparing growth of these caddisflies on both biofilm on rocks and fungi on
leaves we were able to look indirectly at variations in food quality between stream
sections as well as establishing which food type each taxa preferred. This study revealed
strong food preferences for all three caddisflies, which followed what was hypothesized.
This study also suggests that MDN enrichment in Harris River may not have extended
into subsequent generations of caddisflies and therefore growth of larvae in the spring
was not different between areas with and without salmon. Further research in this area
should continue, as well as studies on these insects in the fall and winter to help clarify if
MDN has a role in their growth and life history more proximal to the large salmon input.
The MDN story seems to be much more complicated than the typical model that links
salmon enrichment, stream productivity and ultimately a positive feedback for juvenile
salmon survival via direct and indirect pathways. The complications come from the
coincidental disturbance that precedes MDN enrichment, varied life histories of aquatic
insects, and the lack of knowledge of the fate of MDN in streams over longer periods of
time. The links to fish production may take place over the few months directly following
the large salmon run and therefore the influence on the stream community as a whole

may be more limited than many hypotheses predict.

93

Acknowledgements

This manuscript was improved greatly from the reviews and advice of R. Wotton, J.
Richardson, J. Wallace and G. Lamberti. C. Lesage and R. Kimbirouskas were
invaluable with setting up this experiment. Thanks to M. Wipfli for support and the
USPS Pacific Northwest Research Station for the use of their housing and research
facilities in Hollis, Prince of Wales Island, AK. This research was supported by the
following: USDA National Needs Water Science Fellowship Program, USDA-CSEES
National Research Initiative Competitive Grants Program (Ecosystem Science Program)

and Sigma Xi Grants in Aide of Research.

94

LITERATURE CITED

Anderson N .H. & Cumrrrins K.W. 1979.—The influence of diet on the life histories of
aquatic insects. J. Fish. Res. Bd. Ca, 36: 342-355.

Arsuffi T.L. & Suberkropp K. 1986. -Growth of two stream caddisflies (Trichoptera) on
leaves colonized by different fungal species. J. N. Am. Benthol. Soc.. 5: 297-
305.

Baker T.T., Wertheimer A.C., Burkett R.D., Dunlap R., Eggers D.M., Fritts E.I.,
Gharrett A.J., Holmes R.A., & Wilmot R.L. 1996. -Status of Pacific salmon and
steelhead escapements in southeastern Alaska. Fisheries (Bethesda), 21: 6-18.

Ben-David M., Hanley T.A., & Schell D.M. 1998.-Fertilization of terrestrial vegetation
by spawning Pacific salmon: the role of flooding and predator activity. Oikos,
83: 47-55.

Bilby R.E., Fransen B.R., & Bisson RA. 1996. —Incorporation of nitrogen and carbon
from spawning coho salmon into the trophic system of small streams: evidence
from stable isotopes. Can. J. Fish. Aqugt. Sci., 53: 164-173.

Brusven M.A. & Scoggan AC. 1969. ~Sarcophagous habits of Trichoptera larvae on
dead fish. Ent. News, 80: 103-105.

Cederholm C.J. & Peterson NP. 1985. -The retention of coho salmon (Oncorhynchus
kisutch) carcasses by organic debris in small streams. Can. J. Fish. Aquat. Sci.,
42: 1222-1225.

Cederholm C.J., Kunze M.D., Murota T.M., & Sibatani A. 1999. -Essential
contributions of nutrients and energy for aquatic and terrestrial ecosystems.
Fisheries, 24: 6-15.

Chaloner D.T., Wipfli M.S. & Caouette J .P. 2002. -Mass loss and macroinvertebrate
colonization of Pacific salmon carcasses in south-eastem Alaskan streams.
Freshwat. Biol., 47: 263-273.

Cummins K.W. 1974. -Structure and function of stream ecosystems. Bioscience,
24: 631-640.

Cummins K.W., Wilzbach M.A., Gates D.M., Perry J .B. & Taliferro W.B. 1989.
-Shredders and riparian vegetation. Bioscience, 39: 24-30.

Denning D.G. 1970. -The genus Psychoglypha (Trichoptera: Limnephilidae). Can.
1733,,102: 15-30.

Ellis R.J. 1978. -Seasonal abundance and distribution of adult caddisflies of Sashin

95

creek, Baranof Island, Southeastern Alaska. The Pan-Pac. Ent., 54: 199-206.

Fisher Wold A.K. & Hershey A.E. 1999.—Effects of salmon carcass decomposition on
biofilm growth and wood decomposition. Can. J. Fish. Aquat. Sci., 56: 767-773.

Gotceitas V. & Clifford HF. 1983. The life history of Dicosmoecus atripes (Hagen)
(Limnephilidae: Trichoptera) in a rocky mountain stream of Alberta, Canada.
Can. J. Zool., 61: 586-596.

Gresh T., Lichatowich J ., & Schoonmaker P. 2000. -An estimation of historic and
current levels of salmon production in the northwest Pacific ecosystem: evidence
of a nutrient deficit in the freshwater systems of the Pacific Northwest. Fisheries
(Bethesda), 25: 15-21.

Grubbs S.A., & Cummins K.W. 1994. —Processing and macroinvertebrate colonization
of black cherry (Prunu_§ serotina) leaves in two streams differing in summer biota,
thermal regime and riparian vegetation. Am. Midl. Na_t., 132: 284-293.

Halupka K.C., M.D. Bryant, M.F. Wilson and EH. Everest. 1999. —Biological
characteristics and population status of anadromous salmon in Southeast Alaska.
Gen. Tech. Rep. PNW-GTR-468. U.S. Dep. Agriculture, Forest Service, Pacific
Northwest Research Station, Portland, Oregon.

Irons J .G., 111. 1988. Life history patterns and trophic ecology of Trichoptera in two
Alaskan (U .S.A.) subarctic streams. Can. J. Zool., 66: 1258-1265.

Kaushik, N.K. & Hynes H.B.N. 1971. -The fate of dead leaves that fall into streams.
Arch. Hydrobiol., 68: 465-515.

Kline T.C., Goering J .J ., & Piorkowski R.J. 1997. —The effect of salmon carcasses on
Alaskan freshwaters. In: Milner A.M. & Oswood M.W. (Eds.) Freshwaters of
Alaska, ecological smtheses, Springer-Verlag, New York., : 179-204,

Lamberti G.A., Gregory S.V., Ashkenas L.R., Li J .L., Steinman A.D., & McIntire CD.
1995. -Influence of grazer type and abundance on plant-herbivore interactions in
streams. Hydrobiologi a, 306: 179-188.

Lichatowich, J .A. 1999. -§§lmon without rivers, Island Press, Washington DC. 317 p.

Merritt R.W. & Cummins K.W. 1996. -An introduction to the mtic insects of North
America, Kendall/Hunt Publishing Co. 862 p.

Minakawa N. 1997. —The dynamics of aquatic insect communities associated with

salmon spawning. Unpublished Ph.D. dissertation. University of Washington,
Seattle.

96

Mihuc T.B. & Mihuc J .R. 1995. -Trophic ecology of five shredders in a rocky mountain
stream. J. Fresh. Ecol., 10: 209-216.

Peterson B.J., Deegan L., Helfrich J ., Hobbie J .E., Hullar M., Moller B., Ford T.E.,
Hershey A., Hiltner A., Kipphut 6., Lock M.A., Fiebi g D.M., McKinley V.,
Miller M.C., Vestal J .R., Ventullo R., & Volk G. 1993. -Biological responses of
a tundra river to fertilization. Ecology, 74: 653-672.

Peterson D.P. & Foote C.J. 2000. -Disturbance of small—stream habitat by spawning
sockeye salmon in Alaska. Trans. Am. Fish. Soc., 129: 924-934.

Schuldt J .A. & Hershey A.E. 1995. -Effect of salmon carcass decomposition on Lake
Superior tributary streams. J. N. Am. Benthol. Soc.. 14: 2569-268.

Speaker R., Moore K. & Gregory S. 1984. -Analysis of the process of retention of
organic matter in stream ecosystems. Verh. Intemat. Verein. L_imnol., 22: 1835-
1841.

Sweeney B.W. 1984. -Factors influencing life-history patterns of aquatic insects. In:
Resh V.H. & Rosenberg D.M. (Eds), The ecologv of aquatic insects. Praeger
Publishers, New York, N Y., : 56-100.

Vannote R.L., Minshall G.W., Cummins K.W., Sedell J .R. & Cushing CE. 1980.
—The river continuum concept. gm. J. Fish. Aquat. Sci., 37: 130-137.

Wiggins G.B. & Mackay R.J. 1978. -Some relationships between systematics and
trophic ecology in nearctic aquatic insects, with special reference to Trichoptera.
Ecology, 59: 1211-1220.

Wiggins G. B. & Parker CR. 1997. —Caddisflies (T richoptera) of the Yukon, with
analysis of the Beringian and Holarctic species of North America. In: Danks
H.V. & Downes J .A. (eds.), Insects of the Yum. Biological survey of Canada
(Terrestrial Arthropods), Ottowa, : 787-866.

Wiggins G.B. & Richardson J .S. 1982. -Revision and synopsis of the caddisfly genus
Dicosmoecus (T richoptera: Limnephilidae; Dicosmoecinae). Aguat. Insects, 4:
181-217.

Wiggins G.B. & Richardson J .S. 1986. —Revision of the Onocosmoecus unicolor group
(Trichoptera: Limnephilidae, Dicosmoecinae). Psyche, 93: 187-216.

Wipfli M.S., Hudson J. & Caouette J. 1998. —Influence of salmon carcasses on stream
productivity: response of biofilm and benthic macroinvertebrates in southeastern
Alaska, USA. Qarn. J. Fish. Aquat. Sci., 55: 1503-1511.

Wipfli M.S., Hudson J .P., Chaloner D.T & Caouette J .P. 1999. Influence of salmon

97

spawner densities on stream productivity in Southeast Alaska. Can. J. Fish.
Aquat. Sci., 56: 1600-1611.

98

APPENDICES

99

APPENDIX A

100

 

X X X X X X + + + 0 0800200500 00088820
X X X X X X + + + + + + n 0800200500 000808830
X X X X X X + + + + + + + N 08:00—00:80 0000:8950
X X X X X X + + + + + + + D 080020055 000880000
X X X X X + + + + + 8880020 000805000
X 08300:: 05:800zm 0:0tonoasu
X X X X + + + «8:305 00080309800
X + «0.0—imam 00080woao~800
X + Em €080? :050w< 0005098805
X X + 08.020800 :050w< 00200088005
X + xt05< 0006502 8085
x x + + 808383:
x + 3088: 808.02
X X + + + + 090:2 0:055:80»:
X + :90w005 0:0_088m
X + 88—05% 00085
X X X + 0:902 00085
X X X + 38> D 0:08:85
X X + + 0003005 0:0_8_§Q
x + 83085 3280.5
X X + 38>80>E 000800.?—
X + 3000005 000800.?—
X + 0.503.. 0:0_000.AD
X + 03:..— 000823080
X + 8:282 0:080:85 80800.00
D D D D D Q am 0% 0m m I m m 000.2% 3:00 3:80.: 000:0
8an :0000m :0000m 8:0:m >08m
“mom mEEQ 05

 

.823 00:00.0 :0 08:00 50: 008083000 0003 9005.2.— 0:::0£005:O .wdv

0080:: o. :80: 00: 88:. d0osmunm 0:: .:0_:O\Sumu0\m .:08_0mn0m 808000de 0000:": .nmfinm .0_mm0mnm ”0:: 800030500 80080 z08m

.A800bmasuD .00.:83 mo 8:080:300HQV :00000 0:: O8: 0:: wES0 .003 02:0: .3 800:: :000 E 800:0 0000.8 02.8 05 :8 a: 88:. .< 80:09?

101

 

X X + 08:30: 0:08.09 0005000808”:
X X X X X X + + + + + + + 60000 0:0:EQ 002:0:08080
X X X X X + + + + + 280008.00 0:003 0005000808”,—
X X X X X + + + + + + + 0000000 0:300:00 00050008085
X X X + 800% 0:300:00 00080008085
X X + 000000000000: 0:200:00 002:0:0808m
X X X X X X + + + + + + + 08000008 0000m 000000m
X + x0_0> 080.08 0020.05
X X X X X X + + + + + + + 080.08< 00200—0800 00080008085
x + 030:. 32.25
X X + + 08085 002802.
X X X X X + + + + + + 080:0x0m 00288.:
X X + + 008000000003 00288:.
X + 000000.05 00288:.
X X X X X X + + + + + + + 008005 00288:.
X X X X X + + + 0:00.: 00288:.
X + 8:903 002:9:m
X X X X X + + + + + 82885 002.88%
X X X X X + + + + + 8088005 0022885
X X X + 88008800 :00: 82885 002885
X X X X + 008800 8300000880 :00203 002882m
X + 08880023 0022:8m
X X X X X X + + + + + + :000w005 002208.":
X + 80008.0 002208..“—
x x x x x + + + + + 800.30 002280
X X X X + 00.80082 0808000809 0028080000809
X X X X X X + + + + + + 86:03:00. 002880.80
X X X X X X + + + + + 008000808 002880080
X + 00880—00500 002880080
X X X X X X + + 008820055 002880.80
D D D O D Q 5 gm 0m m E m m 880%.. 8:00 380% 00000
:0000m :0000m :0000w 800.:m >005
000m 8.85 05

 

8800 ._ 5:28

102

 

X X X + + 883me caugoaoz
X X X X X + + + «50:82 03:50:82
X X + Engaging 03:50:52
X X X + + «EEO—hum 32583
X + Smswsa «can—mun 26528:
X X X X X X + + + + + + 2.8369 3252.3
X + «ton—SD 32.590820
X X X X X X + + + + + + 3.33".” exam—“39020
X + «3988 5:555 82139020
X X X X X X + + + + + + 3:555 32138020
X X X + 38>? «toga—3m 32139020
X X + gunman 33.2.9020
X X X + Saxon «coach—SM 32.63520
X X X X + + + + «coach—SM 32.898020
X + atone—an: 32.58520
X X X X + + + + + 2:95 32:—many 82885
X X X X X X + + + + + «Bo—:QQQQEEA 8252:8on—
X X X X X X + + + + + + «couch—HEM anEoanom
X X X + $328: omEEomSaoI
X + mover—o; anEoquom
X X + 358%:2 :0: 823%.:qu
+ £53306 :0: anEowSmom
X + «Bow—ago: anEowSaom
X X X X X X + + + + + + 3025 ouEEomSgom
X X X X X X + + + + + + «38350 owEEoanom
X X X X + + + + «3.36 anEowSaom
X X X X X + + + + 2.32% owns—8082mm
X + 23883.5 omflzoBEosam
X X X X + + 30259“ 2.0.5.5 oaggofionam
X X X X X + + + Eucflw £3.55 oavzfihofionam
D D D O D Q :m 0% am A I n— 86on 8:00 Dish .020
.8303 cocoom nouoom .525 33m
30m war—=0 2m

 

9.88 ._ 5282

103

 

X X X X X X + + + + + 085wosoxmm 002—20085
X X + + + 8.02:: 80008000000 0022000824
X + 8:08:04 002—20883
X + 20:05 002—20884
X X X X X X + + + + + 03802.05 002202.84
X 0:30 800080000805 002—20085
X X X X + + 8000800002334 002—2088:
X X + 00800 800080005 002—20885
X + + + + 00000080005 002—2088:
x + 88:80 88285:
X X X X X + + + 0800002004 0020080602004
X X X X X + + + + 05930824 0025308222
X 2808 0.2030082 002503008022
X X X X + + + 05.30082 002003008022
X X X X X X + + + + 080000020 0020080000005
X X X X + + + + 0800822 00280093005
X X X + 0:02.302 0028030<
X X X + 080:8? 002E080< 0:080:25.
X X X X + 0880200,“. 00238808002.
X X 8800 0:00—22 00200102
X X X X X X + + + + 80:80 928032 00200:?—
X X + + 0200004 00200102
X 0060808503 00200194
X X X X + 00500: 0100000000: 002:0m
X X X X X X + + + + + 20:0:0w08 00000N 00280802
X X X X + + + 00000080 00000N 00280802
x + 5.85 808852
X + 008.5 00280802
X + 298$ 00280802
X X + 8008094 00280802
0 0 0 0 0 0 :0 0:0 «0 m m 802.0 380 :80: .020
:0000m :0000m :0500m 830m .305
.002 880 05

 

3.808 .— 50:09?

104

 

 

X X + 00000020fl=0 0020:0D
X X + 0003:0002 0020:0D
X X X X X X + + + + + + 0.2000022 0022000005—
X X X + + + 00000800060 00220009800300
x + 00.0583 88.3822:
x x + + 832860 8083822:
D D D D D Q .5 gm 0m 0 m 0 00:000m 00:00 3800 0080
:0:00m :280m :280m 800:5 025m
0000 mESQ 0:0

 

€088 ._ 00:83

105

APPENDIX B

106

Appendix 8
Record of Deposition of Voucher Specimens‘
The specimens listed on the following sheet(s) have been deposited in the named museum(s) as
samples of those species or other taxa, which were used in this research. Voucher recognition
labels bearing the Voucher No. have been attached or included in fluid-preserved specimens.
Voucher No.: 2004-04

Title of thesis or dissertation (or other research projects):

Influence of marine-derived nutrients from spawning salmon on aquatic insect communities in
south-east Alaskan streams

Museum(s) where deposited and abbreviations for table on following sheets:

Entomology Museum, Michigan State University (MSU)

Other Museums: none

Investigator’s Name(s) (typed)
JoAnna L. Lessard

 

 

Date 8/25/04

*Reference: Yoshimoto, C. M. 1978. Voucher Specimens for Entomology in North America.
Bull. Entomol. Soc. Amer. 24: 141-42.

Deposit as follows:
Original: Include as Appendix 1 in ribbon copy of thesis or dissertation.

Copies: Include as Appendix 1 in copies of thesis or dissertation.
Museum(s) files.
Research project files.

This form is available from and the Voucher No. is assigned by the Curator, Michigan State
University Entomology Museum.

107

Appendix B

Voucher Specimen Data

$030 000...“. fit

_of Pages

Page

0.00

.050

 

 

0.00

 

.2 .60 0.0m

 

0.0.9.5 0.0.0 0095.2 9.. c. .088
.0. 000050000 00.0.. 0>000 0.... 0020000

 

 

3-08m .02 .0...o:o>

20000.. 00046..

.08....

00.002 0..o.0o..00>0.

3.0000000 .. 0.0000 0005000 00:.

 

 

 

 

 

 

 

 

 

x v.< 50002. 0.00.0 5003 .00 00.00005 ”000.0.0Ew ”0. 5n.
x v.4. 500021.020 00.“. .00 00005.0 ”0050.005 K. 0.0
x 02 :20. 8.0.5 .o 8:00 0.8.0 00.0... .00 0.0020 000.2050 0. 0.0
x .2 5000:... 200.0 00.". 0.00026. 050.000.0500 ”000.30.080.30 2... 0.0
x X... ..0.0_ 00.03 .0 005.0 0.00.0 00.0... .0_0.0.>00.. ”03.800055 ”3 0.0
x v.< :20. 00.0.5 .0 005.0 0.00.0 0.00... 005000.60... ”0050.000500 ”0.. 0.0
x 0.0. ..0.0. 00.02. .0 005.0 0.00.0 0.00: 0.. 005.00.09.00 .000.E000..00 N. 0.0
x v.< ..0.0. 00.0.... .0 005.0 0.00.0 0.00... 0.. 005.00.09.00 ”000.500920 .3 0.0
x x< ..0.0. 00.0.5 .0 005.... 0.00.0 0.00... 0.. 005.00.09.00 H000.E000..00 no. 0.0
x v.4. 50002. 0.00.0 0.0000 0.. 005.00.09.00 ”0058000....0 .0 0.0
x 0.... 50003.. 0.00.0 0.0000 0.. 005.00.09.00 ”000.805....0 .0 0.0
x v.< 50002. 0.00.0 0.0000 .0 005.00.09.00 000.0.000..00 K 0.0
x 0.0. 5000:... 500.0 000.20.... .50.000..00 000.0.000..00 ”0 0.0
x 0.0. ..0.0_ 00.02. .0 005.0 0.00.0 00300.00 000.2000: 00.00.83 ”0000000000 .0 0.0
x v.< ..0.0_ 00.02. .0 005.0 0.00.0 0.00... 000..000.._00 ”v 0.0
x v.< 50002. 200.0 0.008 .00 0.000090 000.080.80.00 ”0 0.0
x v.< 5000:... 0.00.0 00.". 00.00.0800 000.09. ”000000000005 ”N 0.0
x 02. 5005.10.85 00.“. .00 x02... 600.0000... n. 0.0

M. m m 09.08% 9.0

w m. w W? 000: .0 00.02.00 00950000 .0. 0.00 .000. 00x0. .050 .0 00.0000

A N L E

”.0 .00052 _

 

108

Appendlx B

Voucher Specimen Data

Pages

_of

Page

0.00

.9050

 

0.00022 30.90900

 

Gumm-

 

02025 0.0.0 00002.2 2. 0 .0800
.0. 00950000 00.0.. 0>000 00. 0020000

 

 

3-30m .02 .000:o>

0.00000 00000..

.08....

0.0.002 0..0.00..00>0.

5.0000000 .. 0.0000 .000...000 00:.

 

 

 

 

 

 

 

x v.< ..0.0. 00.0.5.0 005.0 0.00.0 0.00... 00.020 0:900:00 ”005.000.0000 .0 00m
x v.< 0.0002. 0.00.0 00.0 .0000. 0:200:00 ”000.__0.0E000w ”0 00m
x v.< 50002. 0.00.0 00.0.00 0.00:000.0.00 0:900:00 ”005.000.0000 0.. 00m
x v.< 5000:... 0.00.0 00.... 00.00200... 00000 ”0000000 "0 00m.
x v.< 5000210020 00.0 .00 090.084. ”000..0.0E< N 000
x v.< 50000.. 0.00.0 00.0 x0_0> 03.0.9.2 ”00000.00... u. 000
x v.< 5000:... 0.00.0 0.0000 00 0.30... ”000.30.... 50 0.0
x v.< 50002. 0.00.0 000.00 .00 0.095.. ”000.30.... ”mm 0.0
x .2 50002. 0.00.0 00.0 .00 00.0.0on ”000.30; ”00 0.0
x 0.0. 50003.. 0.00.0 090.00 .00 00000090000... ”000.30: #0 0.0
x V... :20. 00.0.5 .0 005.0 0.00.0 0.00... .00 0.0.0000 ”000.30.... ”00 0.0
x .2 :22 00.02. .o 80.0 0.000 530.000 .00 0.2.00.0 000.30... 00 0.0
x v.< 50002. 0.00.0 00.0 .00 0000.0< ”000.30... ”00 0.0
x v.< ..0.0_ 00.02. .0 005.0 0.00.0 0.00... .00 0:00.50 ”000.0050 ”mm 0.0
x v.< 50003. 0.00.0 00.0 .000 05.00.0008 .000 0.5.30.0 ”000.....005 ”mm 0.0
x v.< 50002. 0.00.0 0.0000 .00 0.0.3.5020 ”000...:E.w ”.m 0.0
x 0.... 5000210090 00.0 06.00.00 00500000500 0000.0: ”000...:0..0 now 0.0
x v.< 5000210020 0.0000 .00 0.000E00.0>> ”000.0.0Em ”m. 0.0

0+ 8 I J.

m m m s 8.08% 0:0

m Numv m m 000: .0 00.02.00 000050000 .0. 0.00 .0000 00x0 . . 00.0 .0 00.0000

 

”.0 .00052

 

109

Appendix B

Voucher Specimen Data

of Pages

Page

Gama bougso

 

$5822 >mo_oE2:m

 

2mm

 

€525 28m c8222 2: s ESE

 

.2 22.89% 8%.: o>onm 85 8288:
voéoom .oz Econo>

 

 

 

3.888: z 285. 65:68 82

 

X

x< :22 88>) :0 855 £85 2an
x x< 58:2. £85 :mE
x< ..m_m_ mm_m>> 6 8::n. £85 9an
x< 58:2. £85 288m

¥< 58:2. £85 868m

x< :22 8_m>> 6 855 £85 280
x< 58:2. £85 :mE

x< 58:2. £85 988m

x< 58:2. £85 0695

v.< 58:5... £85 983

v_< 58:3... £85 :mE

x< ..m_m_ 8.25 .o moctn. £85 9.5:
x< 58:2. £85 :9:

v.< .87.. 9225 B 855 £85 wEwI
x< 58:51 £85 :mE

¥< 58:2. £85 :2“.

x< 58:2. £85 :mE

x< 58:2. £85 :mE

 

938.. «:52.
68.8 3952 $089.85.

«3: mo ”82: mo ”N 8E

 

~20an 3:95 “82:95 “P 85

5% 3:62:55

5% mcmmoigé
5% 88:9.

5% £89281
5% 9:83

5% m_:Em>:_o
.Qam mEQEG

5% 36:28 u
5% £68,8ch “
&m 96:55 N
90:53 9.855 H
mficma 38:20 ”

”825:235

5% «582598.528 “8259:5984

”82:80.98:
”82:89:81
”82:88:81
“82:88qu
88:83am:
88:83:81
82:28:23
wmu___2me:am
8288523
82:98.:23
82:98Emcqw

”mm Em
“a Em
”om Em
”2 Em
a: Em
5 Em
”S Em
”m? Em
i Em
”2 Em
”NF Em
i Em
”9 Em

fiuuoc 28:20 “82:88Emcam “m :am
8:28:28 96:55 68:.28Emcnm ”a cum
5% £672.80 “89:98.83 K :am

 

 

I.

8:88: :5.
8m: 5 880:8 mcmEEmuw :8 98 6:3

Eggs

ll

 

, Adults

 

:93 8:6 5 86me

 

 

 

 

E
DLaNanXXXXXXXXXXXXXXX

 

m I‘yIIIPI ID

.—
O
m

.2

 

110

Appendix B

Voucher Specimen Data

of Pages

Page

Sun.

8850

$88.2 58.2.8.5

 

Ema

 

3.82:: 98 595.5. as c. 888
8. m8E.8:m 88.. o>onm 85 8288:

 

 

3-§N .oZ 8:0:0>

 

 

v.< 58:2. £85 :2:

v.< 58:5... £85 :888:

v.< 58:2. £85 :2...

v.< 58:2. £85 :2:

v.< 58:5... £85 :oEEm

v.< :22 8.25 8 oo::: £85 8.50825
v.< :22 8.25 .0 8::: £85 50.50825
v.< :22 8.85 .o 8::: £85 2.8...

v.< 58:2. £85 88.2w

v.4. :22 825 8 8::: £85 2an

v.< :22 8.25 .0 8::: £85 :o.:088w
v.< 58:2. £85 :2“.

v.< 58:2. £85 :2:

v.< 58:2. £85 :2:

v.< 58:2. £85 :2:

v.< 58:2. £85 :2:

v.< 58:2. £85 :2:

v.< 58:2. £85 :2:

    

 

288.. 8:52.

€82. .882 859.82.
3:888: = 28% 88:68 83
m 2 o 2< 88.8.8: now 8.:
8:88 28:988.. 88:8: ”9 8.:
288:0on mumamN 88:05:82 8.. 8.:
898:8 828M 88.50682 N: 8.:
.28 2.85 88:59:82 How 8.:
dam 825 88:50:82 Um: 8.:
:8 2868: 88:8sz “3 8.:
:8 888.80 88.50582 “or 8.:
:8 «5982 88:50:82 ”NF 8.:
d8 S:oEo:.::E< 88:50:82 ”I 8.:
:8 2>Eo_8: 88:88.. ”or 8.:
d8 2.88.88: 88:88.. “a 8.:
285:: 2x880 88:88.. ”m 8.:
d8 :2.mb 88.88820 K 8.:
.98 2.835 88.88825 8 8.:
28.222835 88....89520 ”m 8.:
882: 28:.E5_: 88.:89020 ”v 8.:
2:8: 28:953. 88:888.:0 “m 8.:

 

 

 

 

 

[Larvae XXXXXXXXXXXXXXXXXX

 

 

_ Adults 9

u...
0

5 Nymphs

.o
E
3

85.88: :8
88 8 88.2.8 89:8on 8. 28 .88..

 

:92 8:8 8 8.8%

 

 

4 I Eggs

 

 

lll

Appendix B

Voucher Specimen Data

Page

Gama hung—30

 

250022 30.9.25.

2.0.9.5 290 $0.50.: 9.. c. .088
.2 0:06.080 :22. 0>0:0 0:. :0>.000....
3.38 .02 .0:050>

 

2WD.

 

 

20000... 0:50..
60:3. 6.0802 2.209.003.
5.000000: .. 0.00:0 .0:0...::0 0B...

 

of Pages

 

 

 

 

 

 

 

x v.< 500:2. £005 :2: .::0 0:303:50: .00:...::0:E... .3 :2. ...
x v.< 500:2. £005 :2: 5.02:: 05000800030 602.285.... .9 :2. ..
x v.< 500:2. £005 0.003 .::0 05802.05 002.235.... .3 :0: ..
x v.< 500:2. £005 :2: 0.500 0000060000280: 602.285.... .2 :2. .r
x x< 500:2. £005 983 00:...0 05000:.0005 .00:...::0:E... .9 :2. ...
x v.< :90. 00.05 .0 02...: £005 2:0: .::0 2:02:55 .00:...::0:E... .m :2. ...
x v.< :90. 00.0.5 .0 02...: £005 9:00 .::0 0E200:.:0.. .00:..0E200:.:0.. .0 :2...
x v.< 500:2. £005 :2: .::0 0:050:20: .00:.:0>0:0.:>... K :0: k
x v_< 500:2. £005 :2: .::0 0:050:20? .00:.:0>0:0.:>... .0 :0: h
x v.4. 500:2. £005 :2: 2::05 0:050:20... .00:.:0>0:0.:>.._ .m :2. ..
x :2 500:2. £005 233 .::0 0E0000005 .00:..0E000000.0 .v :2...
x v.< 500:2. £005 233 .::0 05000.22 002.200.3005 .0 :2;
x v.< 500:2. £005 :0:._0m .::0 0:05.002). .00:..:0.0:< .w :2...
x v_< :90. 00.0.5 .0 002.: £005 2:0... .::0 2:209: .00:_.:0.0:< .p :2. ..
x v.< 500:2. £005 :2: ::0 0E0:0.:00P .00:.a>.2:0.:00.r .vw 00.:
x v.< 500:2. £005 0.003 .::0 0.00.2: 60:62.0: .3 00.:
x v.< 500:2. £005 :2: 20:90 050.0905. .00:.:0..0: .mm 00.:
anII x v.< 500:2. £005 825 .::0 0:0:00. .00:.:0..0: ...N 002.:
S
M W m s :2.00:0: ::0
m m. m w .80: .o 8.8.8 0:05.080 .o. 3% .80.. :93. .05.. .o 8.080
A N L E
.:0 .0252

 

 

112

Ban .2530

 

23
.5333: 3295.5 J:

b.2225 29m :85... 2.. c. 2258
.2 2.958% 322. o>onm 2.: 8282...
3.38 .oz 820...;

 

 

233.. 2.56..

63.5 22252 2562396.
9.338: 2 232m .mcoauum can.

 

Appendix B

Voucher Specimen Data

Pages

_of

Page

 

 

 

 

 

 

 

 

 

x x< :22 m2m>> .0 ooctn. .235 2:22 $2235.35 .2 .00
x 2< 58:2. .285 .22". dam «8805 68.5.22 ”NF .00
x v.< 585.1285 288 dam .222 ”32.23.22: u: .00
x v.< :22 8.35 .0 mecca. .235 223.92% dam 26.955 622:...w ”or .5
x x< :22 «033 S 855 .285 2an dam maemz “$2.2m ”m .5
x v.< :22 8.25 .o eoctn. .235 2.5: 5.3 mans: ”mmu.o2.>o no .5
x v.< :22 m2m>> .o mourn. .285 2an dam $562.0 ”omv.o2.>o K .5
x x v.< :22 m2m>> .o 8.25 .235 2an dam «35>: ”massage ”m .5
x v.< 53:32. .285 23mm dam 2:922... 68.0225 ”m .00
x 2< :22 «225 .o mecca. .285 223558 dam 25mg ”$202.5 ”m .5
x x< :22 «22$ .o 8.25 .285 223553 dam max... 6228.39.30 ”w .00
x 2<:22 «22$ .o 825 .2mo5 530 due mo~...aE< ”mugging up .00
x 2< 53:2. .285 825 .3» 83922825 “222083 How so:
x v.< :22 m2m>> .o 8.25 .235 223033 dam x2>caomz ”$295: .9 22F
x 2< 53:3... .285 230m dam 2.2883,”. ”$2288.25 ”op .22....
x v.< :22 m2m>> .o 85.1 .235 2.250 dam 38:52.0; 68.082.50.26... .5 :2....
x v.< 535.. .285 23mm dam 2u_mE._o>> 63.5.2822.“ .9 :2;
III x 22.2.... 8.2... .o 855L286 6.8 an... 88.28.00 58.52822... ”m. 5.:
S
s m m 8588 can
u“ S .
m m. m m 88 .o 83.8 «85.8% .2 58 3%.. :98. .26 .o 8.88
A N L E
go .3832

 

 

113

Appendix B

Voucher Specimen Data

of Pages

Page

Qumo 205.30

.Ezomas. 3202.95

 

2mm.

 

3.29.2: 99m 8222.2 82. 2. 282.8

 

.o. 28.2.88 82. o>onm 82. 8282".
3.38 .oz .m2o:o>

2882 856..

683. @982 2.0.8282.
3.888.. 2 282m 820.228 8:.

 

114

 

 

 

 

 

 

 

 

x v.< .3822. .285 28.22“. 8.2.32.2:5 ”m :5
x v.< :22 8_m>> .o 8....n. .285 22.553 dam 26282828.. ”82.2.2225 ”N ..00
x x< :22 8.m>> .o 8...... .285 2.8: dam 2.8% 88......20E22m 2. =5
R. m. m s 8288.. 08
.w m. m w 82. .0 882.8 9.22.88 .2 28 .85 :98. 828 .o 8.83
A N L E
”.o .8532

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIII

1|ngmwrunmgwnWW