20

V)

\

This is to certify that the
thesis entitled

Macroinvertebrate Community Response to Timber Harvest
and Spawning Salmon in Southeast Alaska Rainforest
Streams

presented by

Emily Yvonne Campbell

has been accepted towards fulfillment
of the requirements for the

MS. degree in Entomologv

 

 

 

Major Profes§6r’s Signature
5/3//0
/ V

Date

MSU is an Affirmative Action/Equal Opportunity Employer

 

 

l3 ARV
Michigan State

l U: .iversity

 

.-.ga-.-.-.-.--u-’--.-.-s—._.-

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5108 KIProj/Aoc8PrelelRC/DataDueJndd

 

Macroinvertebrate Community Response to Timber Harvest and
Spawning Salmon in Southeast Alaska Rainforest Streams

By

Emily Yvonne Campbell

A THESIS

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

MASTER OF SCIENCE
Entomology

2010

Abstract

Macroinvertebrate Community Response to Timber Harvest and Pacific Salmon in
Southeast Alaska Rainforest Streams

By

Emily Yvonne Campbell

This study examined the separate and interactive effects of timber harvest and
salmon spawning on benthic macroinvertebrate community composition and distribution
in Southeast Alaska streams. I predicted that l) spawning salmon disturb benthic
macroinvertebrate communities in riffles habitats and increase invertebrate drift, 2) the
magnitude of spawning salmon disturbance is greater in highly harvested watersheds
relative to less-impacted streams and, 3) that macroinvertebrates utilize refugia habitats
such as backwater pools, stream edges, and the hyporheic zone during the salmon run to
avoid riffle epilithic disturbances. Macroinvertebrates were collected quantitatively and
qualitatively in multiple habitats during salmon runs on Prince of Wales Island, Alaska in
2007 and 2008. Spawning salmon caused significant declines in riffle macroinvertebrate
density, biomass, and richness and the magnitude of this effect increased with increasing
timber harvest intensity. In less-impacted streams, macroinvertebrate density and
biomass increased. Macroinvertebrate density and richness significantly increased in
stream drift during spawning. Stream edges and the hyporheic zone appear to offer
refugia for macroinvertebrates during salmon spawning. This study demonstrated that
timber harvest activities intensify the effects of spawning salmon disturbance on
macroinvertebrate communities and that macroinvertebrates may utilize refugia in

response to salmon disturbance and in Southeast Alaska streams.

To my father, for always encouraging me to pursue my academic goals and

to my mother, who’s spirit will never fade.

iii

Acknowledgements

I am sincerely grateful to my adviser, Richard Merritt for his guidance, support, and
generous heart which gave me the invaluable experience to explore Alaskan streams,
enhance my knowledge and expertise in the field of entomology, and provided me with a
bright future, thanks Rich. Special thanks to my adviser Eric Benbow who has helped me
both in the field and laboratory, offering expertise in statistical methods and experimental
design, thanks Eric. I am especially grateful to my mentors Kenneth Cummins and
Peggy Wilzbach for believing in me and helping me get to graduate school at Michigan
State University to pursue my dreams. Thanks to Gabe Ording who is the best teacher of
pedagogy I have ever known, and not only taught me how to be a teacher, but a confident
leader. Thanks to my fellow ISB 201 lab teaching assistants for their support and
inspiration in teaching: Rachel Olson, Danielle Donovan, Sarah Willson, Andrew Pike,
Osvaldo Hernandez, Rodrigo Mercader, Megan Fritz, and Alex. I am grateful for the
entire Notre Dame crew: Janine Ritegg, Peter Levi, Gary Lamberti, Jennifer Tank,
Dominic Chaloner, Jim Junker, Susan Meyer, and especially to my boyfriend, Scott
Tiegs, for his love and professional support throughout this entire process. I am grateful
to Angeline Kosnik and Brittney Tanis for help in the lab sorting my endless invertebrate
samples. i would like to thank Winnie Winnikoff and Ann-Marie Larquier for their help
in the field. Special thanks to my fellow Merritt Lab graduate students for their
encouragement and support: Ryan Kimbraskus, Mollie McIntosh, Osvaldo Hernandez,
Todd White, Jaree Johnson, Kristie Zurawaski, Sarah Willson, and David Malakaskaus.
Thanks to the MSU. Department of Entomology for their financial, academic, and

personal support. I would like to acknowledge the Thorne Bay and Craig Ranger

iv

Districts (USDA Forest Service), Rick Edwards, Dave D’Amore, Aaron Prussian,
Katherine Prussian, Steve McCurdy and Erik Norberg for logistical support; John Hudson
for help with identification and being a great field work partner, Brigitte Kolouch for
calculating insect biomass; Megan Shoda for NMDS ordinations and analysis, and Wei
Wang for statistical consultation. This research was supported by the USDA Forest
Service Pacific Northwest Research Station and the USDA-CSREES National Research

Initiative Competitive Grants Program (Ecosystems Program 2006-35101-16566).

Table of Contents

List of Tables ....................................................................................... vii

List of Figures ....................................................................................... viii

Chapter 1. Timber Harvest Intensifies Spawning Salmon Disturbance of
Macroinvertebrates in Southeast Alaskan streams

Abstract ....................................................................................... 1
Introduction .................................................................................. 2
Methods ....................................................................................... 4
Results ....................................................................................... 10
Discussion .................................................................................. 23
Conclusions ................................................................................. 28

Chapter 2. Macroinvertebrates Utilize Refugia in Response to Spawning Salmon
Disturbance in Southeast Alaskan streams

Abstract .................................................................................... 30
Introduction ................................................................................ 31
Methods ...................................................................................... 34
Results ....................................................................................... 43
Discussion............................................ ...................................... 60
Conclusions ................................................................................. 67

Chapter 3. Macroinvertebrate Community Differences in Riffle and Backwater Pool
habitats in Response to Spawning Salmon in Southeast Alaskan streams

Methods .................................................................................... 69
Results and Discussion .................................................................. 72
Appendices .......................................................................................... 78
Literature Cited .................................................................................... 84

vi

List of Tables

Table 1. Characteristics of the seven study streams on Prince of Wales Island, Southeast
Alaska, USA. * = These values were taken from Tiegs et a1. 2008. Numbers in
parenthesis represent standard deviations ......................................................... 6

Table 2. Results of rmANCOVA analyzing the relationships between macroinvertebrate
density, richness, insect biomass; biomasses of Diptera, Ephemeroptera, Plecoptera,
collector-gatherers, scrapers, and predators with % timber harvest treated as a covariate
during as compared to before the salmon run. ................................................ 13

Table 3. Results of rmANCOVA analyzing the relationships between macroinvertebrate
density, richness, insect biomass; biomasses of Diptera, Ephemeroptera, Plecoptera,
collector—gatherers, scrapers, and predators with sediment size treated as a covariate
during as compared to before the salmon run. .................................................. 16

Table 4. Results of rmANCOVA analyzing the relationships between macroinvertebrate
density, richness, insect biomass; biomasses of Diptera, Ephemeroptera, Plecoptera,
collector-gatherers, scrapers, and predators with large woody debris (LWD) treated as a
covariate during as compared to before the salmon run. ...................................... 19

Table 5. Actual difference (‘during salmon’ — ‘before salmon’) of macroinvertebrate

density (no/m2), insect biomass (mg/m ), richness; and biomasses (mg/m2) of Diptera,
Ephemeroptera, Plecoptera, gathering-collectors, scrapers, and predators. ................ 22

Table 6. Characteristics of riffle, stream edge, and backwater pool habitats in Twelve
Mile Creek, Southeast Alaska, USA. Numbers in parenthesis represent standard
errors .................................................................................................. 36

Table 7. Results of rmANOVA analyzing macroinvertebrate density, richness, and insect
biomass during as compared to before the salmon run ......................................... 46

Table 8. Results of rmANOVA analyzing Ephemeroptera, Plecoptera, Trichoptera, and
Diptera biomasses (mg/m2) during as compared to before the salmon run .................. 50

Table 9. Results of rmANOVA analyzing Shredder, Collector-gatherer, Collector-filterer,

Scraper, and Predator biomasses (mg/m ) during as compared to before the salmon
run .................................................................................................... 54

Table 10. Results of rmANOVA analyzing Ameletus, Baetis, Cinygmula, Chironomus,

Sweltsa, and Suwallia densities (no/m2) during as compared to before the salmon
run ..................................................................................................... 58

vii

List of Figures

Figure 1. Map showing the locations of the seven study streams on Prince of Wales
Island, Southeast Alaska, USA ...................................................................... 5

Figure 2. Linear regressions of the difference (‘during salmon’ - ‘before salmon’) of
macroinvertebrate density (A), total insect biomass (B); and biomasses of Diptera (C),
Ephemeroptera (D), gathering-collectors (E), and scrapers (F) regressed against % timber
harvest ................................................................................................ 12

Figure 3. Linear regressions of the difference (‘during salmon’ - ‘before salmon’) of
macroinvertebrate density (A), total insect biomass (B); and biomasses of Diptera (C),
Ephemeroptera (D), gathering-collectors (E), and scrapers (F) regressed against sediment
size .................................................................................................... 15

Figure 4. Linear regressions of the difference (‘during salmon’ - ‘before salmon’) of
macroinvertebrate density (A), total insect biomass (B); and biomasses of Diptera (C),
Ephemeroptera (D), gathering-collectors (E), and scrapers (F) regressed against large
wood volume ......................................................................................... 18

Figure 5 Non-Metric Multi Dimensional Scaling ordination showing the separation of
macroinvertebrate community structure before and during the salmon run ................. 21

Fig 6. PVC tube, 1mm mesh net scooper and 250nm mesh net scooper used to collect

benthic macroinvertebrate samples ................................................................ 40
Fig 7. Macroinvertebrates sampling from the hyporheic zone with a bilge pump. . . . . . ....41
Fig 8. Drift nets used to collect drifting macroinvertebrates .................................. 42

Fig 9. Macroinvertebrate density (no/m2) (A), insect biomass (B), and macroinvertebrate
richness (C) differences before and during the salmon run in stream edges, riffles,
backwater pools, the hyporheic zone, and stream drift ....................................... 45

Fig 10. Ephemeroptera biomass (mg/m2) (A), Plecoptera biomass (B), Trichoptera
biomass (C), and Diptera biomass (D) differences before and during the salmon run in
stream edges, riffles, backwater pools, the hyporheic zone, and stream drift .............. 48

Fig 11. Shredder biomass (mg/m2) (A), Collector-gatherer biomass (B), Collector-filterer
biomass (C), Scraper biomass (D), and Predator biomass (E) differences before and

during the salmon run in stream edges, riffles, backwater pools, the hyporheic zone, and
stream drift ........................................................................................... 52

viii

Fig 12. Ameletus density (no./m2) (A), Baetis density (B), Cinygmula density (C),
Chironomus density (D), Sweltsa density (E), and Suwallia density (F) throughout the
salmon run in stream edges, riffles, backwater pools, the hyporheic zone, and stream

drift .................................................................................................... 56

Fig 13. Sampling backwater pool habitats (A) and riffle habitats (B) ....................... 69

Fig 14. Sampling technique used in riffle and backwater pool habitats in 2007. A 1m
long weighted PVC pipe was used to delineate a quantitative area of

0.315m2 ............................................................................................... 70

Figure 15. Non-Metric Multi Dimensional Scaling ordination showing the significant
separation of macroinvertebrate community structure in riffle and pool habitats .......... 72

Figure 16. Macroinvertebrate abundance in riffle and pool habitats in Maybeso
Creek .................................................................................................. 73

Figure 17. Macroinvertebrate abundance in riffle and pool habitats in Twelve Mile
Creek .................................................................................................. 74

Figure 18. Macroinvertebrate abundance in riffle and pool habitats in Indian
Creek .................................................................................................. 75

Figure 19. Macroinvertebrate abundance in riffle and pool habitats in Nossuk
Creek .................................................................................................. 76

Figure 20. Mean macroinvertebrate abundance in riffle and pool habitats before and
during the salmon run .............................................................................. 77

ix

Chapter 1. Timber Harvest Intensifies Salmon Disturbance of
Macroinvertebrate Communities in Southeast Alaskan Streams
Abstract

Natural disturbances and anthropogenic activities can interact to affect freshwater
ecosystems, but these two processes are typically studied separately. We addressed how
timber harvest can interact with salmon (Oncorhynchus spp.) spawning activities to
influence benthic macroinvertebrate communities in streams on Prince of Wales Island,
Alaska. We predicted that spawning salmon would cause greater disturbance to
macroinvertebrates in streams from harvested watersheds, relative to less-impacted
streams, because 1) finer sediments would be more readily dislodged by spawning
salmon, and 2) diminished in-stream large wood would limit macroinvertebrate refugia
from salmon activity. Benthic macroinvertebrates were sampled from 6 riffles within
each of 7 streams before and during the annual salmon run using a modified Hess
sampler. Diptera biomass was lower while Plecoptera biomass was higher during the
salmon run across all streams. Macroinvertebrate density, total biomass, and the biomass
of scrapers, predators, collector-gatherers and Ephemeroptera was higher during the
salmon run in less-impacted streams and was lower in more harvested watersheds.
Multivariate ordination demonstrated significant separation of macroinvertebrate
community structure before and during the run. Indicator species analysis identified
Epeorus longimanus (Ephemeroptera: Heptageniidae), Baetis (Ephemeroptera: Baetidae),
Seratella tibialis (Ephemeroptera: Ephemerellidae), Suwallia (Plecoptera:
Chloroperlidae), and the dipterans Chironomidae and Simuliidae as significant indicators

of before-salmon benthic communities; while the stoneflies Sweltsa (Plecoptera:

Chloroperlidae) and Zapada cinctipes (Plecoptera: Nemouridae) typified during-salmon
communities. Overall these results reveal that strong interactive effects can occur
between anthropogenic activities and natural disturbance and show that timber-harvest
activities can intensify the effects of spawning salmon disturbance on macroinvertebrates

in Southeast Alaska streams.

Introduction

Human activities alter stream ecosystems worldwide and often the direct and
indirect consequences of these activities are not fiilly understood. In Southeast Alaska,
timber harvest in the form of clear-cut logging and associated road construction are major
anthropogenic impacts to streams. The most common direct impact to streams is erosion
and sediment deposition (Wood and Armitage 1997). Fine-sediment deposition can
reduce fish populations (Jones et al. 1999, Shaw and Richardson 2001 , Harvey 2009),
alter macroinvertebrates (Shaw and Richardson 2001, Zweig and Rabeni 2001), algal
communities (Schofield 2004), and other stream organisms including frogs (Dupuis and
Stevenson 1999) and salamanders (Harvey et al. 2009). Without appropriate riparian
management, logging can alter channel complexity by eliminating large wood
recruitment, thereby reducing debris jams, associated downstream pools and
macroinvertebrate habitats (Duncan and Brusven 1985). Over time, reductions in large
wood inputs can cause a shift in channel morphology toward wider, shallower channels
and finer sediment size due to fine particle erosion from unstable riparian banks (Barr and
Swanston 1970, Hawkins 1982). Finer sediments can affect habitat suitability for salmon

spawning and benthic communities.

Spawning salmon can have diverse effects on stream ecosystems. However,
spawners affect streams via two major recognized pathways: 1) as sources of nutrients
due to excretion, the release of gametes, and their decomposing carcasses, and 2) as
agents of disturbance through their spawning behavior and upstream migration (Moore et
al. 2004). As ‘ecosystem engineers’ (Jones et al. 1994, Moore and Schindler 2008)
salmon have been documented to induce the massive physical disturbance and
redistribution of benthic substrates during their upstream migration and redd construction
(Duncan and Brusven 1985, Minakawa and Gara 2003, Tiegs et al. 2008, 2009,
Monaghan and Milner 2009). In contrast to the disturbance effect, spawning salmon
have been documented to provide a major resource subsidy that can positively influence
stream food webs by the provision of nutrients and carbon (Gende et al. 2002, Minakawa
et al. 2002, Chaloner et al. 2004, Tiegs et al. 2008). The net outcome of this enrichment
and disturbance balance hinge on characteristics of the stream channel (Tiegs et al. 2008).
Tiegs et al. (2008) showed that spawning salmon enrich algal communities in low-harvest
watersheds with abundant wood and large sediments that likely retain salmon nutrients
efficiently, and that salmon disturb algal communities in high-harvest watersheds with
simplified channels and finer sediments that are more readily dislodged by spawners.

in Southeast Alaska, the legacy of logging, and healthy runs of spawning salmon,
offer a unique opportunity to study the interaction between timber harvest and salmon
disturbance. Others have demonstrated such an interaction for algal communities (Tiegs
et al. 2008), but no studies have evaluated this effect on other components of stream food
webs. I predicted that the greatest disturbance effect of macroinvertebrates due to

spawning salmon would be in streams with highly harvested watersheds. I hypothesized

that: 1) spawning activity of salmon would reduce macroinvertebrate density, biomass,
and taxonomic richness; 2) spawning activity would reduce disturbance-intolerant taxa
and favor tolerant taxa such as predators that may feed on salmon tissues; and 3)
increased timber harvest would intensify the effects of spawning salmon on
macroinvertebrate community structure. This research represents the first study to assess
macroinvertebrate responses to spawning salmon in relation to watershed harvest

intensity.

Materials and Methods
Study sites
This study was conducted between July and September 2007 in 7 streams on
Prince of Wales Island within the Tongass National Forest, Southeast Alaska, USA (Fig

1). Prince of Wales Island has a maritime climate with an annual precipitation of 0.25m

and a mean air temperature of 7 oC (US. Department of Agriculture, Forest Service

1997). Watersheds on Prince of Wales Island are composed of coniferous temperate
rainforest that has been primarily managed for timber harvest. Dominant tree species are
western hemlock (Tsuga heterophylla (Rafinesque)), Sitka spruce (Picea sitchensis
(Bongard)), and Western Red cedar (Chamaecyparis nootkatensis (D. Don)). Riparian
areas that were historically harvested of timber are dominated by red alder (Alnus rubra
(Bong)). Study streams were selected to provide a gradient of timber-harvest intensity,
measured as the percentage of the watershed harvested (ranging from 5.4% to 63.8%;

Table 1) but were otherwise similar in channel morphology, size, and slope. I used the

same quantifications of sediment size and large wood, and the same 300m study reaches

in each stream as delineated and described by Tiegs et al. (2008).

Figure 1. Map showing the locations of the seven study streams on Prince of Wales
Island, Southeast Alaska, USA.

 

Slide

. \.
O Nossuk

Maybeso 0

Indian O
Trocadero ‘ \Dog
. . Salmo
Twelve 4"
Mile _

l t. I

 

 

 

*Emfimqmzma “95.00 32292 2.8: 00.30:

*Emfimqmzma 32330 $3.303 Eooa< 9.83300 3mm: 00.7. 9032.”. 030.2

0283 3mm :33 $3 32330 0310300 A303 3.50.0.0 2.3130 30.33
280:: 00.0N 0.3 00000.00 30.0 0.00 0.30000 0003.03 0.00
Sam: 000.0 0.00 0030.3 30.8 0.3 0.01003 5.003.»: 0.00
0.380qu 3.00 HPNN 00930.0. 3.50 0.0~ 0.003000 00008.0: 0.00
000 0330: 00.00 3.3 00.0.0000 0.00 0.0~ 0.0050.on 5.33.00 0.00
33096 00.00 3.3 00.23.: 3.00 0.0N 0.02003 3.003.000 09w

30b

0.3m N000 0.0.0» 30000 3.00 0.- -- 5.00 3.03 0.~0
aim—5 25m 00.: 00.0 0~.0A0u.0v 3.0N 0.00 0.33.00 5,020.03 0.00

 

Table 1. Characteristics of the seven study streams on Prince of Wales Island, Southeast

Alaska, USA. * = These values were taken from Tiegs et al. 2008. Numbers in

parenthesis represent standard deviations.

Macroinvertebrate Sampling and Processing

Benthic macroinvertebrates were quantitatively sampled in the 7 streams once
before the salmon run (3-18 July) and once during the salmon run (15-20 September)
from 6 riffles in each of the study streams (except Trocadero Creek that was sampled
after the salmon run on 19 October). Within the 300 m delineated reach of each study
stream, the first 6 riffles within the reach were sampled. A haphazardly chosen area
within each riffle was selected and a single macroinvertebrate sample was collected from

that area. Macroinvertebrates were collected using a modified Hess sampler with a 500-

um mesh net and a total area of 0.1 m2. Samples were collected by agitating the benthos

by hand for 30 s to approximately 8-10 cm in depth. Samples were preserved in 70%
ethanol. Macroinvertebrates were counted and identified to the lowest reliable taxon:
insects were identified to genus or species (except the Chironomidae which were left at
family), and non-insects were identified to class or order. Insect taxa also were measured
for total length (nearest 0.5 mm) to estimate biomass based on published length-mass
relationships (Benke et al. 1999). Functional feeding groups were assigned to each taxon
using Merritt et al. (2008).
Salmon counts

Late-summer salmon runs were dominated by pink salmon (Oncorhynchus
gorbuscha) and chum salmon (Oncorhynchus keta). Salmon were quantified in 4 meter
wide belt transects perpendicular to stream flow every 10 m for the entire 300 m reach.
These counts were then up scaled to estimate the number of salmon present in the 300 m

stream reach on each date. Salmon were counted from the start of the spawning run

approximately weekly for the duration of the run until mostly carcasses remained in the
streams.
Statistical Analysis

I performed a repeated measures analysis of covariance (rmANCOVA) to
determine whether 1) salmon presence altered macroinvertebrate abundance and taxa
composition and 2) timber harvest intensity and habitat attributes (sediment size and large
wood volume) influenced the level of disturbance. Macroinvertebrate response variables
were density, total biomass, richness; and the biomasses of Diptera, Ephemeroptera,
Plecoptera, predators, scrapers, and collector-gatherers. Sampling from before and
during the salmon run was treated as the repeated factor, while, timber harvest, sediment
size, and large wood were treated as covariates in separate nnANCOVAs. Although
sediment size and large-wood volume are highly negatively correlated with percent
timber harvest, we observed non-redundant macroinvertebrate responses to these
predictor variables and thus present results of all three covariates. These analyses were
conducted separately to avoid multicolinearity statistical violations. Results were
considered significant when p<0.05. The 6 riffles in each stream were treated as random
effects, and a compound symmetric covariance structure was specified using SAS
(Release 9.1; SAS Institute, Cary, North Carolina, USA).

ANCOVA assumptions were evaluated from normal probability plots, Shapiro-
Wilk test statistics, and residual plots; natural logarithmic transformations were used to
correct violations. Shredders, collector-filterers, and Trichoptera were rare (<0.01%,
<0.02%, <0.01%, respectively, of total macroinvertebrates) and, when present, patchy in

distribution and were thus omitted from analyses because they did not meet ANCOVA

assumptions. All macroinvertebrate response variables were regressed against percent
timber harvest, sediment size, and large wood volume using SYSTAT (version 11;
SYSTAT software, Richmond, California, USA) to establish predictive power of changes
in macroinvertebrate variables due to the presence of salmon. In regression plots, the
differences (‘during salmon’ minus ‘before salmon’) in macroinvertebrate response
variables for each stream were used in analyses. Outlier data points in regressions were
identified and removed when the Studentized residual exceeded the SYSTAT default of
2.0 or greater.

A Non-Metric Multi-Dimensional Scaling (NMDS) ordination was used to
evaluate macroinvertebrate community structure differences in relation to salmon
disturbance and timber harvest (McCune 2002) using PC 0RD (version 5; MJM
software, Gleneden Beach, Oregon, USA). I ran a total of 250 iterations for both real
data and Monte Carlo analysis with a random seed start. A multiple response permutation
procedure (MRPP) using Serensen distances was performed to test for significant
differences in community structure in response to salmon disturbance and among
streams. When significant differences were found in macroinvertebrate community
structure, Indicator Species Analysis (ISA) was used to determine which
macroinvertebrate taxa were significant indicators of the respective communities. Taxa
were considered significant indicators when indicator values (% of perfect indication)
were >55% with p < 0.001. Higher indicator values demonstrate better predictive power
of that taxon for its assigned group as defined by the results of the MRPP analyses. All
aquatic insect taxa that represented > 3% of all samples were used in the ordination

procedures.

Results
Macroinvertebrate density, biomass, and richness
Macroinvertebrate density was greater during salmon spawning in streams

with low timber harvest, but was lower in streams with high timber harvest and the

magnitude of this effect Increased w1th increasrng timber-harvest intensrty (R =0.50,

p=0.07; Fig. 2A, Table 2 and 5). A significantly greater disturbance effect on

. . . . 2
macrornvertebrate densrty was observed 1n streams that had finer sediments (R =O.74,

p=0.02; Fig. 3A, Table 3) and lower volumes of large wood (R2=0.71, p=0.03; Fig. 4A,

Table 4).
Aquatic-insect biomass was significantly greater before than during the salmon
run (p=0.012; Table 2) and a significantly greater reduction in insect biomass during the

salmon run occurred in streams with high timber harvest as compared to those with low
timber harvest (R2=0.70, p=0.01; Fig. 2B, Table 2 and 5). In streams with finer
sediments, salmon had a significantly greater disturbance effect on insect biomass
(R2=0.78, p=0.02; Fig. 38, Table 3). Insect biomass also differed significantly before

and during the salmon run when large wood was treated as a covariate (p=0.014; Table
4), with biomass being significantly higher before the salmon run. Spawning salmon had

a highly significant disturbance effect on insect biomass in streams with low volumes of

large wood as compared to streams with higher volumes of large wood (R2=0.93,

p=0.002; Fig. 48, Table 4).

10

Macroinvertebrate taxa richness declined across all streams during the salmon
run. Significantly greater macroinvertebrate richness was observed before the salmon run
when timber harvest was treated as a covariate (p=0.041; Table 2 and 5). We found a
significant salmon-sediment size interaction (p=0.03 8, Table 3) suggesting a greater
disturbance effect on richness in streams that had finer sediments. A greater disturbance
effect of salmon on richness was observed in streams that had low volumes of large wood

as compared to streams with higher volumes of large wood (p=0.008, Table 4).

ll

Ephemeroptera (D), gathering-collectors (E), and scrapers (F) regressed against % timber

macroinvertebrate density (A), total insect biomass (B); and biomasses of Diptera (C),
harvest.

Figure 2. Linear regressions of the difference (‘during salmon’ - ‘before salmon’) of

 

 

 

   

 
       

    

 

 

 

 

 

 

 

 

 

 

 

 

  
   

  

 

 

 

 

 

 

 

 

 

”WI 0.0 )0 o 0 o ._ m.

I Oh i O I

o. enoosooioooo -r n . hoooomiooun o. Woowoiooso

n o.» . 0 MI NI m HO.Q.‘O

( m -o.oos -N . m -o.ooo noooo

0 p? . ouoo: ouoooo .. . . .

n . -o . .

m o.» 4 -w .

m .oo 2 L .

d_ -o.m 4 o o o -0 o -0 L o

n

I .oo - - i .o - i i - L i

.A.) 2.0 W 2.0 N0

m .. . 3.0. . 083.05vo 3.." .

m. 3 fioowosfioofi 9...: c.- .mouo ooo. . hooosxinfi

rm- 9...: 2 quoo . no 0.: r... . mouosoo

w . ouooa oo . . 8 . . ouoomo

an.' 0.0 -o.m A 9m . o

a .oo 4 . so . o.o . o

m LOL -90. .064

n_ . .no . . so . . .

I. -90 . . i . . . . -N0 . . 1 . 4 i . .00 . . . l . . l
o 8 ~o 8 .8 0o 0o .8 0o o o No uo o oo oo .3 9. o no ~o 8 .8 mo 8 «o oo

23004 00200" 05 .2309 0020.0. 05 .2309 0020.»; 0x0

12

collector-gatherers, scrapers, and predators with °/o timber harvest treated as a covariate

Table 2. Results of rmANCOVA analyzing the relationships between macroinvertebrate
during as compared to before the salmon run.

density, richness, insect biomass; biomasses of Diptera, Ephemeroptera, Plecoptera,

 

 

 

 

00002 a." .n 003.00 00.02 mm .0 0-<0.=m 3.000 2 .n 0-51:0
00:022.qu2

003.2 0.020 0.03000 0.03000

00.30: 0.0 0.00 0.000 00.30: 0.0 0.00 0.000 00.30: 0.0 0.00 0.000

230210200" 0.00 0.00 0.000 2302102000 0.00 0.00 0.000 2302102000 0.00 0.00 0.000

00.30: x 2302 00.30: x 2302 00.30: x 2302

102000 0. 00 0.00 0.00.. 102000 0. 00 00.00 0.000 0020.00 0. 00 00.00 0.000

3003200020

 

0.03000 0.03000 020.02 0.03000

00.30: 0.0 00.00 0.000 00.30: 0.0 0.00 0.000 00.30: 0.0 0.00 0.000
2302102000 000 0.00 0.000 2302102000 000 0.00 0.000 2302102000 0.00 0.00 0.30
00.30: x 2302 00.30: x 2302 00.30: x 2302

102000 0. 00 00.003000 10220 0. 00 0.00 0.000 102000 0. 00 000 0.000

0.20920

0.23000 0.03000 200002 0.03-0M0.

00.30: 0.0 0.00 0.000 00.30: 0.0 00.00 0.000 00.30: 0.0 0.00 0.000
230210200" 0.00 0.00 0.000 2302102000 0.00 0.00 0.000 2302102000 0.00 0.00 0.000
00.30: x 2302 00.30: x 2302 00.30: x 2302

102000 0. 00 0.00 0.000 1.02000 0. 00 0.00 0.000 102000 0.00 0.00 0.000

13

 

Macroinvertebrate community structure

The NMDS ordination and MRPP revealed a significant (T = 31.6; A = 0.08; p
<0.001) difference in macroinvertebrate community structure before and during the
salmon run (Fig. 5), but not among streams with different timber-harvest intensity. A

total of 69% of the variation in macroinvertebrate community structure was explained by

a three axes solution: lSt axis = 15.9%, 2"‘1 = 13% and 3rd = 39.5%. Mean stress was

19.0 for the ordination and 26.3 for the Monte Carlo solution. Six taxa were considered
significant indicators of macroinvertebrate communities before salmon spawning
disturbance: Epeorus longimanus (Eaton) (Ephemeroptera: Heptageniidae, indicator
value = 89%), Simuliidae (Diptera, 73%), Chironomidae (Diptera, 70%), Baetis
(Ephemeroptera: Baetidae, 67%), Suwallia (Plecoptera: Chloroperlidae, 60%), and
Seratella tibialis (McDunnough) (Ephemeroptera: Ephemerellidae, 57%) (Fig. 5).
However, Sweltsa (Plecoptera: Chloroperlidae, 83%) was the only significant indicator

taxon for communities sampled during the salmon run.

14

TT'P- -‘r‘v‘ ~

Figure 3. Linear regressions of the difference (‘during salmon’ - ‘before salmon’) of

macroinvertebrate density (A), total insect biomass (B); and biomasses of Diptera (C),

Ephemeroptera (D),

size.

gathering-collectors (E), and scrapers (F) regressed against sediment

(no./m2)

ln_differenoe
.6 .6 .6 .b .o .o .0 .°

.5
0|

(mg/m2)

.0
0

7‘
O

ln_differenoe
' ' 0
(II

-|
U‘

NON-5G

&

Q03

 

$003007. .. .3
mnuoio
0n0.000

   
  

 

 

 

 

 

$9803.00. .
annoy...
0u0000

   
  

 

 

00

s... 00 mo 00 0... a... .00 .8
009303000 .33.

 

 

$300030?
mouoio
ouoonm

   
  

 

 

.0

 

.0 .
0.0 .
0.0 .
.00 .
-..0 .
...0 .
.00 .
.00

 

$9923.30 .
000.000
ouooo

oU

    
  

-o.0 .
-..0 .

 

 

00

00

to 0.. a... 0..
000.303 0.00 .330

0000.00

-..0

 

 

$0039-00}
002.000
. ouo.o..

   
  

 

 

0.0

15

 

0.0 .
.0 .
..o .
0.0 .
0.0 .

 

 

$993003»;
mmuo0t
0u0.0. 0 .

    

 

 

 

00

4i

00

to al.... 0 0° 0° .8
02:30:00.5 .33.

 

 

 

000.303 0.~0
00.30: x 000.303
0.~0

0.03000

0030:

000.303 0.~0
00.30: x 000.303
0.~0

E

00.30:

000.303 0.~0
00.30: x 000.303
0.~0

0.00 0.00 0.000

0.50 0.00 0.000

0.0 00.00 0.000
0.00 0.00 0.000

0.00 0000 A0000
0.0 000.00 0.000
0.00 0.00 0.000

0.00 0.00 0.000

 

000.303 0.~0

00.30: x 000.303
0.0 0.00 P00 0.000

930303306
E

00.30:
000.303 0.~0
00.30: x 000.303

0.00 0. 50 00.00 A0000

208303F063000
00.30:

000.303 0.~0
00.30: x 000.303

0.8 0. 2.. 0.00 0.30

0.00 0.00 0.000

0.0 00.00 0.000
0.00 0.00 0.000

0.0 0.8 930
0.00 0.00 0.000

 

.302 0% .0 0.3.00 3.03 00. .0 0-51:0 00.000 0‘ 0.. 3.30:0
9:008? 0300‘

0030.2 0. 30m

0230: 0.0 0.00 0.00» 00.30: 0.0 0.00 0.000 00.30: 0.0 P00 0.000

000.303 0.~0 0. 00 0.00 0.000
0030: x 000.303

0.~0 0. 50 0.00 0.000

0.03000 0.03000

 

00.0
00.30: 0.0 0 0.000
000.3030.~0 0.00 0.00 0.000
00.30: x 000.303 00.0
0.~0 0. Am 0 A0000
30032 0.03pm..“
00.30: 0.0 0.00 0.000

000.3030.~0
0030: x 000.303
0.~0 0. 00 0.00 0000

0. 00 0.00 0.000

 

Table 3. Results of rmANCOVA analyzing the relationships between macroinvertebrate
density, richness, insect biomass; biomasses of Diptera, Ephemeroptera, Plecoptera,

collector-gatherers, scrapers, and predators with sediment size treated as a covariate

during as compared to before the salmon run.

16

Aquatic insect order-level responses
Diptera biomass was lower across all streams during the salmon run and a

significantly greater spawning salmon disturbance effect was observed in streams with

high timber harvest as compared to low timber harvest (R2=O.80, p=0.006; Fig. 2C, Table

2 and 5). Salmon had a significantly greater disturbance effect on Diptera biomass in

streams with finer sediments (R2=O.74, p=0.028; Fig. 3C, Table 3) and less large wood

(R2=0.71, p=0.035; Fig. 4C, Table 4). Diptera biomass was dominated by the family

Chironomidae (67%, of total Diptera) followed by the family Simuliidae (25%).
Ephemeroptera biomass increased during the salmon run in streams with low timber

harvest and decreased in streams with high timber harvest (Table 5). The magnitude of

the dlsturbance effect increased w1th1ncreasmgt1mber harvest intensrty (R =O.60,

p=0.041; Fig. 2D, Table 2). We observed a greater disturbance effect on Ephemeroptera

biomass in streams with finer sediments (R2=0.88, p=0.005; Fig. 3D, Table 3) and low

volumes of large wood (R2=O.72, p=0.031; Fig. 4D). Ephemeroptera biomass consisted

mostly of the family Heptageniidae (52%, of total Ephemeroptera), followed by the
family Baetidae (28%). Plecoptera biomass was greater across all streams during the
salmon run when timber harvest and large wood were treated as covariates (p=0.014;
Table 2 and p=0.0l 3; Table 4; respectively). Sediment size did not explain any of the
variation in the response of Plecoptera biomass to salmon (Table 3). Plecoptera biomass
was dominated by the family Chloroperlidae (90%, of total Plecoptera), followed by the

family Nemouridae (5%).

l7

 

 

 
 

capoomwxbaum
mmuoin
. $0.8m

 
 

 

 

 

(no/m2)

.0
o:

.0
o

25
on

I
3'"
o

 

In difference

.1.
at

  

393:.33
. m~uo$
$0.8»

 

 

 

macroinvertebrate density (A), total insect biomass (B); and biomasses of Diptera (C),
Ephemeroptera (D), gathering-collectors (E), and scrapers (F) regressed against large

Figure 4. Linear regressions of the difference (‘during salmon’ - ‘before salmon’) of
wood volume.

0 no 3 mo modooawoiodmodmonoo

roam Soon 3.53 A33

 

 

  

$9851.93.
mnuoflo

 
 

. buobwm

 

 

l4 1‘ .1 4 d 4‘

 

._.u
#0 #
Pm .
ob .
co 4
9° 4
...u A
.N.O .

 

Pm

 

$6.3»me gnu
mmuouwm
buobfi

 

 

o no #0 mo mo .8 so as ‘8 duo Noo

roam. Econ 3525 33

 

-u.

 

wuobouoxb 3mm
mwuoud
buohou

 

L

 

No
8 .
8 4
cm .
ob .
-00 .
-8 .

  

$338738
mNuoamm
buobd u

 

 

 

.._.u

o No 3 mo moaoodnoxoamoamowoo

roam Soon <oE3o 33

 

l8

 

 

 

 

 

 

 

 

l9

 

 

m a
m a
m m m
m m. a
.l. u
m m )
mm m
a,
L
m W (S\ 362 ax m 0.303. Gamma» am 0.. n-<2:m 3&3 &w m 0.515.
M m .m. Emnfioqfiarmqmq
.m m 4% E 9063 0638“,. 0.033%.
WWW. $.38 3 SE... 9on 3.32. km :3 98.» 3.38 5 ~23 83m
.mE m 250 33 E: 0.~0m 520 rpm 0&0 0.30 _.<<o p.00 0.5 0.00m
.m m, w m mm.30:x_.<<0 p.00 0.0» 0.00m $.33er0 p.90 H.~0 0.~m~ mm_303xr<<o Ram 0.3 0.30
m m m. n
m. D M m mozmamaocnmqm
.m cm .W. .w. E EloBImmm mnqmmmq 0.03am.
Mom m m 3.30: Em Hwhm 0.0: 3.30: Em 0:00 003 $56: rm who 0.0mm
Wm mm EB 33 93 $3 :20 {a B5 one. :20 {a pup o.~mm
mmmnm 3.303150 {—0 0.0.3 A003 mm_Bo:x_.<<U p.50 0.0m 0.0: $5632.20 ram who 0.000
b P e
; b
Wn mm +0.. E 2903qu 063mb 030qu 0.03%
m m flaw mango: Em 3.3 A000» 3.30: rm p.00 0000 3.30: rm 0.0» 0.80
MM WM. 520 Ram 0.00 0.0: EB 33 0.00 0.000 250 p.50 0.0H 0.0.0.0
mm m m 3.303.250 ram 800 0.000 mm.30=xE<o Ram mhm 0.05 mm_30_._x_.<<_u ram PS 003
S S C
01 n 3
mm m a
1m M... .m m.
a m a m
R .m mad
. r. e
4 v” m m
C uh C .
.l. S e m
.w m w m
T. d C C

Functional feeding group responses

The greatest disturbance effect on collector-gatherer biomass was observed in

streams with high timber harvest as compared to streams with low harvest (R2=0.61,

p=0.038; Fig. 2E, Table 2 and 5). A significantly greater disturbance effect on collector-

gatherer biomass was observed in streams with finer sediments as compared to streams

with larger sediments (R2=O.69, p=0.04; Fig. 3E, Table 3). Large wood volume was not

an important covariate in collector-gatherer biomass response to salmon (Table 4) and no
correlation was found between the disturbance effect on collector-gatherer biomass and
large wood (Fig. 4B).

A greater salmon disturbance effect on scraper biomass was observed in streams
with finer sediments (R2=O.84, p=0.01; Fig. 3F, Table 3) and low volumes of large wood
(R2=0.79, p=0.01; Fig. 4F, Table 4). Timber harvest and sediment size did not explain

any of the variation in predator biomass (Table 2). Large wood however was an
important covariate for predator biomass as indicated by the significant salmon-timber

harvest interaction (p=0.032, Table 4).

20

Figure 5 Non-Metric Multi Dimensional Scaling ordination showing the separation of
macroinvertebrate community structure before and during the salmon run.

Axis 3

 

 

 

 

21

0 Before Salmon
A After Salmon

2

, richness; and biomasses (mg/m ) of Diptera,

2
Ephemeroptera, Plecoptera, gathering-collectors, scrapers, and predators.

2

Table 5. Actual difference (‘during salmon’ — ‘before salmon’) of macroinvertebrate
density (no./m ), insect biomass (mg/m )

 

 

 

 

02003 00102000 003.2 0.030%. 0.1.5000 0.088 8303030800 0.08083 0030me 09.0003 300083
zommcr 0.00 000.00 0000.00 -0000 -0000 000.00 000.00 00.0 000.00 000.00
30.0: 0.00 -0000 000.0 0.00 00.00 000.00 000.0 -0000 000.0 000.00
0.08003 00.00 -0000 00.00 -0000 0.00 00.00 00.00 -0.00 00.00 .0000
00000.30: 00.00 000.00 0000.00 .000 -0000 000.00 000.00 00.00 0000.00 00.00
2.2080 00.00 -00000 000.00 -00.00 -00.0 000.0 000.0 00.00 -00000 00.00
0.50 00.00 -00000 000.00 -0000 -00000 -0000 00.00 -0000 0.00 000.00

._-¢<m.<mZ.=m 00.0 000.00 000.00 00.0 -0000 000.00 000.00 -0000 000.00 -00.00

22

 

Discussion
Macroinvertebrate responses to salmon

Pacific salmon are considered to be ecosystem engineers because they modulate
resource availability and reallocate biotic and abiotic materials (Jones et al. 1994). Upon
the arrival of spawning salmon, I documented changes in macroinvertebrate community
structure, richness, density, and biomass. Several studies in Alaska streams have
docmnented reductions in macroinvertebrate density and alterations to benthic
macroinvertebrate community organization during salmon spawning (Moore et al. 2004,
Lessard and Merritt 2006, Moore and Schindler 2008, Monaghan and Milner 2009). I
observed lower macroinvertebrate richness during the salmon run across all streams.
Studies in New Zealand and Michigan streams also showed that spawning salmon reduce
benthic macroinvertebrate richness (Hildebrand 1971, Field-Dodgson 1987), thereby
having a negative impact on stream assemblages. This response is likely due to the large
number of invertebrate taxa unable to survive the intense bioturbation impacts from
salmon during migration and spawning activities (see also Janetski et al. 2009).

I found that taxonomic groups responded differently to salmon spawning. Diptera
biomass was lower across all streams during the salmon run and the magnitude of this
effect was greatest in streams with high timber harvest. Chironomidae were the dominant
family in this order, comprising 70% of the biomass of all dipteran larvae collected.
Other studies in Alaska streams have observed declines in Chironomidae density and
biomass during salmon spawning (Peterson and Foote 2000, Moore et al. 2004, Lessard
et al. 2009). Ephemeroptera biomass consisted mostly (52%) of heptageniid mayflies,

which declined during the salmon run in streams with high timber harvest and increased

23

in streams with low harvest. Most genera in the family Heptageniidae are scrapers and
several taxa such as E. longimanus, Rhithrogena (Ephemeroptera: Heptageniidae), and
Cinygmula (Ephemeroptera: Heptageniidae) dominated before salmon, but then were less
abundant during the salmon run. This response is most likely due to their feeding
behavior, as these scrapers forage on biofilm from the tops of rocks in the active channel
where redd construction is prevalent and thus are highly vulnerable to bioturbation
impacts. In addition, biofilm is generally reduced in the main channel of these salmon
streams during spawning (Janetski et al. 2009), a result also found in these same 7
streams in 2006 (Tiegs et. al 2008). Other mayfly taxa highly affected by salmon
spawning were Baetis, S. tibialis, Drunella grandis (Eaton) (Ephemeroptera:
Ephemerellidae) and Drunella doddsi (N eedham) (Ephemeroptera: Ephemerellidae),
consistent with others from Southeast Alaska that found significant declines in these taxa
during the salmon run (Lessard and Merritt 2006, Lessard et al. 2009). Plecoptera
density and biomass increased during the salmon run across all streams, mainly
predaceous Sweltsa and Suwallia, which may consume the abundant salmon tissues and
salmon eggs (Ellis 1970).

Multivariate ordination demonstrated significant separation of communities
before and during the salmon run, with several taxa characterizing macroinvertebrate
communities before the arrival of salmon and Sweltsa as the only significant indicator of
communities associated with spawning activity. This result explains the significant
decline in richness of riffle macroinvertebrates with the arrival of salmon, and suggests
that sensitive taxa are unable to survive the massive bioturbation impacts due to salmon.

Disturbance intolerant taxa may also actively move to refugia habitats such as backwater

24

pools or stream edges during bioturbation impacts. It is possible; however, that this
separation may be due to evolved phenologies that are timed around the annual fall
salmon run or macroinvertebrates. For example, some predatory taxa such as Dicranota
(Diptera: Tipulidae) and Sweltsa were more abundant during salmon spawning and may
be adapted to feeding on salmon tissues and eggs (Ellis 1970). Other shredder species,
such as Zapada cinctipes (Banks) (Plecoptera: Nemouridae), could time their life
histories around leaf litter input that occurs concomitantly with the salmon run. Lessard
and Merritt (2006) reported that Z. cinctipes proliferated during the salmon run and
speculated that they may benefit from both salmon nutrients and leaf litter inputs in late
summer and fall. Some dipteran families, such as Chironomidae and Simuliidae, and
ephemeropteran families, such as Heptageniidae and Baetidae, may time their life
histories to avoid spawning salmon by emerging before the arrival of salmon (Moore and
Schindler 2008, Lessard et al. 2009). The emergence of certain sensitive taxa before the
salmon run may be evolutionarily favored by the historical legacy of thousands of years
of spawning salmon disturbance. However, mechanistic experiments and longer-term
data are needed to resolve these potential mechanisms.
Macroinvertebrate response to salmon and timber harvest interactions

The greatest reduction in macroinvertebrate density and insect biomass during the
salmon run was observed in streams with high timber harvest as compared to less-
impacted streams. In low harvest streams, macroinvertebrate density and biomass
increased during the salmon run, consistent with responses in benthic algae in these same
streams (Tiegs et al. 2008). in high-harvest streams, algal biomass and chlorophyll a

declined during the salmon run but increased in more pristine streams. My results for

25

macroinvertebrate communities also suggest that the ecological role of salmon can be
modified by stream structure in response to human disturbance.

Tiegs et al. (2008) documented that sediment size in our study streams was
negatively related to percent timber harvest in the watershed. Other studies have shown
similar patterns in other watersheds (Barr and Swanston 1970, Hawkins 1982, Murphy
and Milner 1996, Dupuis and Stevenson 1999, Jones et al. 1999). Such shifts to finer
sediments can have several deleterious effects on macroinvertebrate habitats, such as
filling in interstitial hyporheic habitats, increasing abrasion and scour, and altering food
web dynamics by influencing primary production (Allan 2004). The greatest decline in
macroinvertebrate density and biomass occurred in streams with finer sediments, perhaps
because smaller sediments are more readily dislodged by spawning salmon. For
example, meta-analysis showed that the effect of salmon on macroinvertebrates was
positive in streams with large sediments (>32 mm), but negative in streams with small
sediments (<32 m; Janetski et al. 2009). In my study, all streams had mean sediment
sizes greater than 32 mm and yet we still observed a similar trend where salmon had a
negative effect on macroinvertebrates in streams with small sediments and a positive
effect in streams with larger sediments.

Reductions in regionally-important collector-gatherers (Cushing et al. 1995) in
managed streams during salmon spawning could affect the secondary production of these
streams and thus alter food web dynamics. The 3 dominant collector-gatherers in my
streams were the mayflies Baetis and S. tibialis, and the flatworm Planaria (Seriata:
Planariidae). Salmon redd construction in harvested streams may increase the mortality

of collector—gatherers due to very fine sediments that may smash them by substrate

26

mobilized by salmon, or force them into stream drift. Collector-gatherers typically
scavenge for food resources in benthic interstices and it is also likely that the finer
sediments in high harvest streams may fill in these areas during salmon spawning and
interfere with feeding. Scraping macroinvertebrates also were significantly affected by
the interaction of salmon and timber harvest, with the greatest decline in scraper biomass
observed in the high-harvest streams with small sediment sizes. The 3 dominant scraper
macroinvertebrates in our streams were the mayflies Cinygmula and E. longimanus, and
the caddisfly Glossosoma (Trichoptera: Glossosomatidae). Scrapers were perhaps the
most vulnerable functional feeding group affected by spawning salmon as they typically
scrape biofilm directly from the tops of rocks in the main channel where disturbance
effects are likely to be greatest.

The volume of large wood was negatively related to percent watershed harvest,
which has also been documented in other studies (Gregory et al. 1991, Montgomery and
MacDonald 2002). Reductions in large wood can directly affect stream organisms by: 1)
diminishing substrate for cover, attachment, and feeding (Ehrman and Lamberti 1992); 2)
reducing shade and increasing stream temperatures (Bourque and Pomeroy 2001); and 3)
altering flow dynamics, which effects both habitat distribution and heterogeneity (Allan
2004). The change in insect biomass over time was positively related to the volume of
large wood with the greatest declines observed in streams with the lowest volume of large
wood. This finding could be due to reduced habitat heterogeneity, which may limit the
refugia available to insects and other macroinvertebrates during spawning periods. The

abundance and diversity of macroinvertebrates, which are a major food resource for

27

juvenile salmonids, often increase with habitat complexity (Crowder and Cooper 1982,

Robson and Barmuta 1998, Taniguchi and Tokeshi 2004).

Conclusions

In the relatively pristine streams, biomasses of scrapers, predators, collector-
gatherers, and Ephemeroptera increased during salmon spawning whereas they declined
in more harvested streams. The increased retentive capacity of salmon nutrients in
complex pristine streams may allow for salmon nutrients to be more fully incorporated by
the stream food web. Pristine streams may also provide macroinvertebrates with greater
habitat diversity and more refugia, such as backwater pools, to buffer disturbance effects,
as also observed in other non-salmon studies (Death and Winterbourn 1995, Gjerlov et al.
2003). Backwater pools are also more prevalent in pristine streams and may be important
refuge habitats for fish such as overwintering salmonids (Heifetz et al. 1986). Greater
channel complexity can also positively influence benthic algal production during annual
salmon runs in Southeast Alaska (Tiegs et al. 2008).

My study is the first to explicitly examine combined salmon and timber harvest
impacts on stream macroinvertebrate communities. As such, this study represents an
important step toward understanding these interactive effects on macroinvertebrate
communities and potentially salmon populations. Overall, the greatest salmon-induced
reductions in macroinvertebrate density and insect biomass were observed in streams
with a high degree of timber harvest, small sediments, and low volumes of large wood.
Timber harvest operates through multiple mechanisms to reduce channel complexity and

thereby modify the effect that spawning salmon have on benthic communities. I

28

demonstrate that changes at the watershed level due to timber harvest can amplify local
disturbances from spawning salmon and elicit declines in macroinvertebrate density and

biomass, potentially altering the productivity of stream food webs.

29

Chapter 2. Macroinvertebrates Utilize Refugia in Response to Spawning
Salmon Disturbance in Southeast Alaskan streams
Abstract
Spawning salmon create patches of disturbance through redd digging and
upstream migration which can alter the abundance and community structure of
macroinvertebrate communities. We investigated how the presence of salmon spawners
alter the distribution, abundance, and community composition of macroinvertebrates
among habitats with different degrees of spawning disturbance activity including riffles,
backwater pools, the hyporheic zone, and edge habitats, as well as stream drift in Twelve
Mile Creek on Prince of Wales Island, Alaska, USA. We predicted that spawning salmon
would 1) reduce the abundance and richness of macroinvertebrates in riffles; 2) increase
macroinvertebrate abundance and richness in backwater pool, stream edge, and hyporheic
zone refuge habitats during the salmon run; and 3) increase the magnitude of stream drift.
We quantitatively sampled benthic macroinvertebrates from the four benthic habitats and
collected 30 min drift samples six times before and four times during the salmon run.
Spawning salmon significantly reduced the density (p<0.001), biomass (p<0.001), and
richness (p=0.009) of macroinvertebrates occurring in riffle habitats compared to before
the run. Within backwater pools, most taxa declined during spawning, especially the
Limnephilidae (Trichoptera), but Chironomus (Diptera: Chironomidae) and Sweltsa
(Plecoptera: Chloroperlidae) densities increased during spawning. Stream edges
appeared to offer refugia for certain taxa such as Limnephilidae, Ostracoda, Simuliidae
(Diptera), and Sweltsa stoneflies. The hyporheic zone appeared to offer refugia for

certain invertebrate taxa including Chironomus, planaria flatworms, Apatanidae

3O

(Trichoptera) and Limnephilidae, and overall insect density increased in the hyporheos
during salmon spawning. Macroinvertebrate density (p=0.05) and richness (p=0.001)
increased in stream drifi during salmon spawning. This research elucidated some of the
mechanisms of benthic macroinvertebrate persistence despite massive annual main-
channel benthic disturbance by spawning salmon. Biomonitoring and other benthic
studies should include sampling from several habitat types to provide more
comprehensive information on how macroinvertebrates respond to salmon disturbance.
Habitat heterogeneity offers refugia for macroinvertebrates and may be a crucial

determinant of macroinvertebrate survivability during salmon spawning.

Introduction

Spawning salmon have been documented to induce massive physical disturbance
reported to redistribute of benthic substrates during upstream migration and redd
construction (Duncan and Brusven 1985, Minakawa and Gara 2003, Tiegs et al. 2008,
2009, Monaghan and Milner 2009). Spawning disturbance can play a critical role in
influencing nutrient transfer and nutrient availability to benthic communities (Moore and
Schindler 2008, Monaghan and Milner 2009). Benthic disturbances from spawning
salmon can alter the distribution, abundance, and community composition of benthic
organisms and cause significant reductions of macroinvertebrates in riffle habitats where
salmon activities are often greatest (Peterson and F oote 2000, Chaloner et al. 2004,
Moore et al. 2004, Lessard et al. 2009). Not all benthic habitats are equally impacted by
spawning salmon. The most common areas for spawning activities are main-channel

riffle and run habitats, and areas that receive less impact by spawners are stream edges

31

which are too shallow for redd construction, and backwater pools or other slack water
habitats that do not offer sufficient oxygen important to salmon egg survival (Quinn
2005). Macroinvertebrates may utilize these less impacted habitats as refugia to avoid
bioturbation impacts during periods of intense salmon spawning (Minakawa and Gara
2003)

Habitat heterogeneity is important for stream biota as it offers diverse habitat
types that sustain diverse and unique taxa (Kerans and Karr 1992, Gjerlov et al 2003).
For example, riffle habitats are often dominated by scrapers, such as heptageniid
mayflies, that feed on abundant biofilm within riffles and collector-filterers, such as
simuliid dipterans, that collect fine particulate organic matter from stream drift.
Backwater pools typically sustain shredders, such as limnephilid case-building
caddisflies, and collector gatherers, such as some genera of the family Chironomidae
(Merritt et al. 2008). Backwater pools represent important reach scale heterogeneity in
stream flow and substrate composition, which may offer efficacy as refugia for
macroinvertebrates and other stream organisms (Lancaster I993). During salmon
spawning, macroinvertebrates in highly-disturbed riffle habitats may migrate, or be
displaced, into low-disturbance habitats such as backwater pools, stream edges, or the
hyporheic zone. These habitats could be important for the completion of growth and
development of certain species that would otherwise be continually displaced
downstream resulting in increased mortality.

Habitat refugia can be defined as distinct habitats that sustain communities that do
not normally become disturbed, or display resilience to disturbances (Sedell et al. 1990,

Winterbottom et al. 1997). Biota that are transient or different than the typical

32

communities may inhabit refugia as they move in from disturbed areas (Sedell et al.
1990). Stream edges may offer refuge to macroinvertebrates during flood events
(Negishi et al. 2002) and possibly during salmon spawning as edges are areas of the
stream channel that receive less salmon spawning activity compared to main-channel
riffles where disturbance is often greatest. The hyporheic zone is recognized as an
important refugium for aquatic macroinvertebrates and other river organisms during
hydrological disturbances, such as floods or droughts (Williams and Hynes 1974, Oliver
et al. 1997, Rosario 2000), and may also offer macroinvertebrates refuge in response to
spawning salmon disturbance.

Spawning-related fluctuations in benthic topography and community structure are
a form of substrate disruption which likely increases invertebrate drift (Waters 1972).
Much research has been done concerning the three different types of macroinvertebrate
drift: 1) constant (casual) drift; 2) behavioral (predictable) drift; and 3) catastrophic
(sudden) drift that is most often a response to physical or chemical factors (Waters 1972,
Hynes 1975, Chutter 1975, Waters and Hokenstrom 1980). Spawning salmon can cause
catastrophic drift of macroinvertebrates due to bioturbation impacts (Peterson and Foote
2000), or salmon can cause behavioral drift if benthic macroinvertebrates have evolved
life history traits to drift in avoidance of spawning disturbance due to thousands of years
of exposure to annual salmon runs. Most research regarding macroinvertebrate drift and
fish populations has focused on predation pressures as the causal mechanism for
increased drift (Ringler 1983, Skinner 1985, Bowles et al. 1988). Few studies have tested
the effect of spawning salmon disturbance as a means of dislodgement potentially

increasing the magnitude of macroinvertebrate drift (see Peterson and Foote 2000,

33

Minakawa and Gara 2003). Macroinvertebrates that are dislodged into stream drift may
be displaced into slow moving waters such as stream edges or backwater pools, or they
may actively move into edges, pools, or the hyporheic zone to avoid bioturbation impacts
after resettlement in runs or riffles.

This research quantified macroinvertebrate community composition throughout a
salmon run evaluating temporal shifts in community composition among four in-stream
habitats. We investigated how the presence of salmon spawners altered the differential
distribution, abundance, and community composition of macroinvertebrates among
riffles, backwater pools, the hyporheic zone, and edge habitats, as well as stream drift.
We hypothesized that spawning salmon would: 1) reduce the abundance and richness of
macroinvertebrates in riffles; 2) increase macroinvertebrate abundance and richness in
backwater pool, stream edge, and hyporheic zone refuge habitats during the salmon run;

and 3) increase the magnitude of macroinvertebrates in stream drift.

Materials and Methods
Study Sites
This study was conducted between June and September 2008 within a 300 meter
reach of Twelve Mile Creek on Prince of Wales Island within the Tongass National

Forest, Southeast Alaska, USA. Prince of Wales Island has a maritime climate with a

mean annual precipitation of 25cm and a mean air temperature of 7 0C (US. Department

of Agriculture, Forest Service 1997). Watersheds on Prince of Wales Island are
composed of coniferous temperate rainforest that has been primarily managed for timber

harvest and Twelve Mile Creek has had 68% of its watershed harvested for timber.

34

Dominant riparian tree species of the stream include: red alder (Alnus rubra (Bong)),
western hemlock (Tsuga heterophylla (Rafinesque)), and Sitka spruce (Picea sitchensis

(Bongard)). Biotic and abiotic characteristics of Twelve Mile Creek are listed in Table 6.

35

 

 

 

 

Emma 2mm: 02082.2: 3mm: 03.0635.
><mBom 333888. woeamamfim .x. 0263 m 3633 m 3633
13:3 .53 34¢ Ad .x. 00 moo a... SS: 096.. macaw 3.30: 95.3 9.30:
Emu» 8.3 mi. 6
25.6 Bug who 8.8V 6.9: 3N3 8.8 8.8V mob» bev 8 8.ro 93 : .va 89A 3 g 9 3
. 3am imam mbu
mama 3.9 Ba... 8?: 6.3V : .me BbQ Na .8 a; 9 mmh 3.9: no; 3 .3 3.»; a .8
mmbw In .3 may

_uoo_ wwhw 8N3 Pam 8; 3 6.3 3N3 Nam 8.8 Nwhm 33$ 3 ~on 49mm 2 N3 8N0 2 .03

 

Table 6 Characterlstlcs of nffle, stream edge, and backwater pool habitats 1n Twelve
Mile Creek, Southeast Alaska, USA. Numbers in parenthesis represent standard errors.

Macroinvertebrate Sampling and Processing

Macroinvertebrate samples were collected from 5 replicate riffles, backwater
pools, stream edges, and hyporheic wells every 10 days from 27 June until 20 September
2008, for a total of 10 sampling dates, 6 pre—salmon and 4 during salmon sampling dates.
There was a substantial flood that prevented sampling on 23 August which resulted in 10
instead of 11 sampling dates. For riffles, backwater pools, and stream edges, benthic

macroinvertebrates were collected quantitatively using a large PVC tube with a diameter

of 36cm and an area of 0.4m2. The tube was sealed within the benthic sediments so that

there was no water movement from outside the tube. Samples were collected by swirling
and mixing the benthos within the tube at a depth of 8cm for 305. A 1mm mesh net was
then used to scoop up large organic matter and macroinvertebrates for 30s, followed by
mixing for 303, and then scooping again with a 250nm mesh net for another 303 to collect
smaller pieces of organic matter and macroinvertebrates (Fig 6). The collected materials
were rinsed through a 250nm sieve, placed in plastic bags and preserved in 70% ethanol
for later processing. We defined the edge of a stream as the low-velocity area along the
wetted stream bank that was 1/10 the width of the entire stream channel. Backwater
pools were defined as low-velocity areas lateral to the stream channel that were
connected to the main channel at low flow, caused by large wood debris or boulders, and
were of suitable depth (no deeper than 70 cm) to allow for macroinvertebrate sampling.
Hyporheic macroinvertebrates were collected using 60 cm long hyporheic wells
installed 30 cm into the stream bottom at randomly selected areas within each riffle. All
wells were capped to avoid surface and water column invertebrates from getting into the

wells. The bottom 15cm of the wells had 40 small holes (5mm diameter) drilled into the

37

sides to allow invertebrates to be pulled from a larger volume of water surrounding the
bottom of the wells. A bilge pump was used to pump 2 liters of hyporheic water and
invertebrates per sample (Fig 7). Hyporheic water was then rinsed through a 250nm
sieve, placed in plastic bags and preserved in 70% ethanol for later processing.

Macroinvertebrate drift was collected by placing three 250nm drift nets evenly
spaced across the channel and perpendicular to stream flow at noon once every 10 days
from June thru September corresponding to benthic habitat sampling dates (Fig 8). We
elevated nets 3cm above the sediment to avoid collecting benthic invertebrates that were
not part of the drift. Drift nets were left out for 30 minutes and discharge was measured
in front of each net. After 30 minutes, the nets were rinsed; the collected materials put
through a 250nm sieve, placed in plastic bags, and preserved in 70% ethanol for later
processing. Macroinvertebrates were counted and identified to the lowest reliable taxon:
insects were identified to genus or species (except the Chironomidae which were left at
family), and non-insects were identified to class or order. Insect taxa also were measured
for total length (nearest 0.5 mm) to estimate biomass based on published length-mass
relationships (Benke et al. 1999).
Habitat characteristics

Late-summer salmon runs were dominated by pink salmon (Oncorhynchus
gorbuscha) and chum salmon (Oncorhynchus keta). Salmon were quantified in 4 meter
wide belt transects perpendicular to stream flow every 10 m for the entire 300 m reach.
These counts were then scaled up to estimate the number of salmon present in the 300 m

stream reach on each date (Tiegs et al. 2009). Salmon were counted from the start of the

38

spawning run and approximately weekly for the duration of the run until mostly carcasses
remained in the streams.

In each habitat we measured pH, % dissolved oxygen, and conductivity using a
Hydrolab MS-5 Mini—Sonde, habitat area, % canopy cover using a spherical densiometer,
sediment size using a Wentworth scale gravelometer, water velocity using a digital flow
meter, and water depth using a meter stick (Table 1). These measures were taken at three
different times throughout the study. To establish in-habitat autochthonous production
we measured benthic biomass (as chlorophyll a) four times over the course of the study.
Three rocks were randomly selected from each habitat, placed in plastic bags, and
brought back to the lab for processing. In the lab, rocks were scraped entirely with a
bristled brush and then filtered onto pre-ashed glass fiber filters with a SASSNXGTE—
4870 Filtering Tower and chlorophyll a measured with a Trilogy Turner Design

Fluorometer.

39

Fig 6. PVC tube, 1mm mesh net scooper and 250nm mesh net scooper used to collect
benthic macroinvertebrate samples.

1 mm
mesh net

Invertebrate
Sampling
Tube

 

40

Fig 7. Macroinvertebrates sampling from the hyporheic zone with a bilge pump.

 

41

Fig 8. Drift nets used to collect drifting macroinvertebrates.

 

Statistical Analysis

A repeated measures analysis of variance (rmAN OVA) was performed to
determine whether salmon presence altered macroinvertebrate abundance and taxa
composition. The two factors were salmon and habitat, salmon (before versus during)
was treated as the repeated factor, and habitat number (replicates of each habitat) were
treated as random effects, and compound symmetric covariance structure was specified
using SAS (Version 11; SAS Institute, Cary, North Carolina, USA). There were n=6

42

before salmon sampling dates, and n=4 during salmon sampling dates, and a total of 230
samples were collected. Macroinvertebrate response variables were density, total
biomass, richness; and the-biomasses of Ephemeroptera, Plecoptera, Trichoptera, Diptera,
shredders, collector-gatherers, collector-filterers, scrapers, predators and biomass of the 7
most dominant taxa: Chironomus, Sweltsa, Ameletus, Baetis, Cinygmula, and Suwallia.
Results were considered significant when p<0.05. ANOVA assumptions were evaluated
from normal probability plots, Shapiro-Wilk test statistics, and residual plots; natural
logarithmic transformations or transformations to the power of 2 or 3 were used to correct
violations. All macroinvertebrate response variables were plotted as bar charts before
and during the salmon run using Sigma Plot (version 11; Sigma Plot software, San Jose,
California, USA) to establish changes in macroinvertebrate variables in relation to

salmon presence.

Results

Macroinvertebrate density, biomass, and richness

Macroinvertebrate density significantly declined in riffles (p<0.001) during the
salmon run, did not change in pool and edge habitats, and significantly increased in the
hyporheic zone (p=0.019) and stream drift (p=0.052) during the salmon run (Fig 9A,
Table 7). Insect biomass significantly decreased in riffles (p<0.001) and backwater pools
(p=0.002) during the salmon run (Fig 93, Table 7). There were non-significant increases
in insect biomass within edge habitats and in drift and no changes in hyporheic habitats.
Macroinvertebrate richness significantly decreased in riffle habitats (p=0.009), did not a

significantly change in pool and edge habitats, and significantly increased in stream drift

43

(p=0.001) during salmon spawning (Fig 9C, Table 7). There was an increase in
macroinvertebrate richness in the hyporheic zone during salmon spawning, though this

was not statistically significant (Table 7).

44

Fig 9. Macroinvertebrate density (no/m2) (A), insect biomass (B), and macroinvertebrate
richness (C) differences before and during the salmon run in stream edges, riffles,
backwater pools, the hyporheic zone, and stream drift.

— Before Salmon

 

‘19 8 ‘ A 12:] During Salmon
6 —I—

5 'k —I—

se- *
a ' +
o

8

1:

9

.EZt

E

8

5 l

5'0 —- ~—

 

 

 

 

 

 

 

 

 

N

t_lnsect Biomass (mg/m2)

O —I N (a) A 01 O)
a”
l Fla .

I +

 

 

 

 

 

N
OI

Macroinvertebrate Richness (no. taxa)
o o- 8 a B

I Fla

I Fla

i 4+ 4

 

 

 

 

 

 

 

Riffles Pools Edges Hyporheic Drift
45

Table 7. Results of rmANOVA analyzing macroinvertebrate density, richness, and insect
biomass during as compared to before the salmon run.

 

 

 

 

Effect df F-value P-value
Minvertebrgte Densm

Edge 1, 218 0.02 0.877

Riffle 1, 218 29.11 <0.001
Drift 1, 218 3.82 0.052

pad 1, 218 0.09 0.766

Hyporheic 1, 218 5.55 0.019

Insect Biomass

Edge 1, 218 0.79 0.376

Riffle 1, 218 43 <0.001
Drift 1, 218 1.14 0.287

Pool 1, 218 9.99 0.002

Hyporheic 1, 218 0.51 0.475

Macroinvertebrate Richness

Edge 1, 218 0.59 0.442

Riffle 1, 218 6.92 0.009

Drift 1, 218 15.06 0.001

pool 1, 218 0.16 0.693

Hyporheic 1, 218 3.42 0.066

 

Insect Order-Level Response to Spawning Salmon

Ephemeroptera biomass significantly decreased in riffle habitats (p<0.001) during
the salmon run but did not significantly change in any of the other habitats or stream drift
during spawning (Fig 10A, Table 8). The most dominant mayfly families were:
Heptageniidae (34%, of total Ephemeroptera larvae collected), Ameletidae (30%), and
Baetidae (29%). Plecoptera biomass did not significantly change in any habitats during

salmon spawning, though there was an appreciable yet non-significant increase in

46

plecopteran drift during salmon spawning (Fig IOB, Table 8). The most dominant
plecopteran families were: Chloroperlidae (96%; of total Plecoptera larvae collected) and
Leuctridae (1%). Trichoptera biomass did not change in riffle or pool habitats,
significantly increased in stream edge habitats (p=0.028), did not change in hyporheic
habitats, and significantly decreased in stream drift (p=0.023) during the salmon run (Fig
10C, Table 8). The most dominant caddisfly larvae were: Limnephilidae (61%; of total
Trichoptera larvae collected) and Apataniidae (15%). Diptera biomass significantly
decreased in riffles (p<0.001), did not change in pools, significantly decreased in edges
(p=0.042) and did not change in the hyporheos or stream drift during salmon spawning
(Fig 10D, Table 8). The most dominant fly families were: Chironomidae (82%; of total

Diptera larvae collected) and Simuliidae (11%).

47

Fig 10. Ephemeroptera biomass (no/m2) (A), Plecoptera biomass (B), Trichoptera
biomass (C), and Diptera Biomass (D) differences before and during the salmon run in
stream edges, n'ffles, backwater pools, the hyporheic zone, and stream drift.

3.5

A — Before Salmon
3.0 1 [:1 During Salmon
2.5 4 F}

2.0 -

1.0 -

0.5 ~

t_Ephemeroptera Biomass (mg/m2)

 

 

 

0.0 . —

ptera Biomass (mg/m2)

ln_Pleco

 

 

 

 

 

Riffles Pools Edges Hyporheic Drift

48

Fig 10. Continued

 

 

 

 

 

6 -

61‘ C — Before Salmon
E 5 _ :21 During Salmon
U)

.5,

8 4 -
(0
E

a 3-
8
a 2 -
O

.C
-‘:’
l-| 1
E

0 _
2.5 -
D
a?

g 2.0 -

O)

E

3 1.5T *

(0

E

.9

m 1.0 -

‘2

2

0.

5| 0.5 -

0.0 U

 

 

 

 

 

 

 

Riffles Pools Edges Hyporheic Drift

49

Table 8. Results of rmANOVA analyzing Ephemeroptera, Plecoptera, Trichoptera, and
Diptera biomasses (mg/m2) during as compared to before the salmon run.

 

 

 

Effect df F -value P-value
W1:

Edge 1' 213 0.09 0.758
Riffle 1' 213 18.15 <0.001
Drift 1' 213 0 0.959
Pool 1' 213 0.01 0.918
Hyporheic 1' 218 0 0.946
W

Edge 1: 213 0.25 0.617
Riffle 1: 213 0.41 0.521
Drift 1: 213 3.65 0.057
Pool 1' 213 0.01 0.943
Hyporheic 1' 213 1.01 0.315
W

Edge 1. 213 4.96 0.028
Riffle 1. 213 0.02 0.875
Drift 1. 213 5.33 0.023
Pool 1: 213 1.39 0.241
Hyporheic 1: 213 0.69 0.408
D' E‘

Edge 1' 218 4.18 0.042
Riffle 1' 213 11.73 <0.001
Drift 1. 218 0.66 0.419
Pool 1' 213 0.01 0.925
Hyporheic 1: 213 0.2 0.653

 

50

F unctional-F eeding Group Response to Spawning Salmon

Shredder biomass increased significantly in riffles (p=0.023), decreased
significantly in pools (p<0.001), and did not change in edges, the hyporheos, or stream
drift during the salmon run (Fig 11A, Table 9). The most dominant shredders were
Limnephilidae (88%; of total shredder larvae collected) and Tipulidae (9%). Collector-
gatherer biomass decreased significantly in riffles (p<0.001), did not change in pools and
edges, increased significantly in the hyporheic zone (p=0.034), and did not change in
stream drift during salmon spawning (Fig 118, Table 9). The dominant collector-
gatherers were Chironomus (32%; of total collector-gatherer larvae collected) and
Planaria (12%). Collector-filterer biomass significantly decreased in riffles (p<0.001)
and backwater pools (p<0.001), and did not significantly change in edges, the hyporheos,
or stream drift during the salmon run (Fig 1 1C, Table 9). However, there was an
appreciable, yet non-significant, increase in collector-filterer biomass in edge and
hyporheic habitats during the salmon run (Fig 11C). The dominant collector-filterers in
our samples were Simuliidae (73%; of total collector-filterer larvae collected) and
Ostracoda (21%). Scraper biomass significantly decreased in riffle habitats (p<0.001)
and did not show considerable changes in the other habitats or stream drift during salmon
spawning (Fig l1D, Table 9). The dominant scrapers were Heptageniidae (84%; total
scraper biomass collected) and Ameletidae (11%). Predator biomass significantly
decreased in riffle habitats (p=0.002), did not significantly change in pools or edges, and
Significantly increased in stream drift (p<0.001) during the salmon run (Fig 11E, Table
9)- The dominant predators in the stream were Chloroperlidae (94%; of total predator

larvae collected) and Dytiscidae (3%).

51

 

Fig 11. Shredder biomass (mg/m2) (A), Collector-gatherer biomass (B), Collector-filterer
biomass (C), Scraper biomass (D), and Predator biomass (E) differences before and
during the salmon run in stream edges, riffles, backwater pools, the hyporheic zone, and
stream drift.

5 .
01" A — Before Salmon
E I: During Salmon
TE” 4

ln Shredder Biomass
N

 

 

: Collector-Gatherer Biomass (mg/m2)

 

 

 

(mg/m2)

 

 

In Collector-Filterer Biomass

Riffles Pools Edges Hyporheic Drift

52

Fig l 1. Continued

0)

_ Before Salmon
D :1 During Salmon

per Biomass (mg/m2)
N 0) A U!

t_Scra

 

 

 

L

 

(mg/m2)
A
+
_'_+_'

t Predator Biomass

 

 

 

 

 

 

0 . _. _
Riffles Pools Edges Hyporheic Drift

 

53

Table 9. Results of rmANOVA analyzing Shredder, Collector-gatherer, Collector-filterer,
Scraper, and Predator biomasses (mg/m2) during as compared to before the salmon run.

 

 

 

Effect df F—value P-value
Was:

Edge 1' 213 1.5 0.222
Riffle 1- 213 5.27 0.023
Drift 1. 213 0 0.954
Pool 1: 213 16.28 <0.001
Hyporheic 1. 218 1.41 0.238
W

Edge 1: 213 0.16 0.687
Riflie 1, 213 36.13 <0.001
Drift 1:213 2.41 0.122
Pool 1. 213 0 0.949
Hyporheic 1' 218 4.57 0.034
Wanna

Edge 1. 213 6.14 0.014
Riffle 1. 213 21.89 <0.001
Drift 1. 218 0.38 0.539
Pool 1. 213 21.11 <0.001
Hyporheic 1- 213 0.6 0.441
Wanna:

Edge 1' 213 0.78 0.377
Rifile 1. 213 50.71 <0.001
Drift 1. 213 0.39 0.535
Pool 1: 213 1.94 0.166
Hyporheic 1. 213 0.26 0.614
We:

Edge 1. 213 0.35 0.556
Rifile 1. 213 9.55 0.002
Drift 1: 213 11.7 <0.001
Pool 1' 218 2.6 0.108
Hyporheic 1' 218 0.34 0.563

 

54

Specific Taxa Responses to Spawning Salmon

The most dominant insect taxa (from most abundant to least abundant) were:
Chironomus (Meigan) (Diptera: Chironomidae; 19.9%; of total taxa), Sweltsa (Yurock)
(Plecoptera, Chloroperlidae; 7.6%), Ameletus (McDunnough) (Ephemeroptera:
Heptageniidae; 5.6%), Baetis (Umpqua) (Ephemeroptera: Baetidae; 5.3%), Cinygmula
(Dodds) (Ephemeroptera: Heptageniidae; 5.1%), and Suwallia (Ricker) (Plecoptera:
Chloroperlidae; 3.2%). Chironomus biomass significantly declined in riffles (p<0.001)
and pool (p=0.014), and significantly increased in the hyporheic zone (p=0.002) and drift
(p=0.024) (Fig 12A, Table 10). Sweltsa were not present prior to salmon arrival and then
significantly increased in all habitats (p<0.001) as well as stream drift (p<0.001) during
the salmon run, and were most prominent in edge and riffle habitats (Fig 12B, Table 10).
Ameletus biomass did not significantly change in any habitats or drift after the arrival of
salmon (Fig 12C, Table 10). Baetis biomass significantly declined in riffles (p<0.001),
did not change in pools, significantly declined in edges (p<0.001), and did not
significantly change in the hyporheos or stream drift during salmon spawning (Fig 12D,
Table 10). Cinygmula biomass significantly declined in riffles (p<0.001), did not change
in pools, significantly declined in stream edges (p=0.048) and remained fairly low and
uniform in the hyporheos and stream drift during spawning (Fig 12E, Table 10).
Suwallia were present prior to salmon arrival and then significantly declined in riffles
(p<0.001), pools (p<0.001), edges (p<0.001), the hyporheic zone (p=0.004), and did not

Significantly change in stream drift during spawning (Fig 12F, Table 10).

55

2
Fig 12. Chironomus biomass (mg/m ) (A), Sweltsa biomass (B), Ameletus biomass (C),
Baetis biomass (D), Cinygmula biomass (E), and Suwallia biomass (F) throughout the
salmon run in stream edges, riffles, backwater pools, the hyporheic zone, and stream
drift.

N
at

A — Before Salmon
* [:1 During Salmon

f}

(mg/m2)
1°
C)

t Chironomus Biomass

 

 

 

9.419199
octooto
fl‘
I
I

t_Swe/tsa Biomass (mg/m2)
O
'01
I

 

 

 

 

 

o

0
L1
r—l
F
H

 

(mg/m2)
7" .-‘ 1° 1° 9°
0 0| 0 0| 0

.0
01

t_AmeIetus Biomass

 

 

 

_o
o
I

Riffles Pools Edges Hyporheic Drift

56

Fig 12. Continued

 

 

 

 

n n
O O
m m m
a .l
S a.“ D
m m. .m
cm .H e
m m m
y
g H
m.
d
E
B
O
O
P
s
e
m
R
D
1865005060500505060
2 1. 1. 0. 0. 3 2 2 1 1. 0. 0. 3 2 2 1 1 0 0
AuEBEV 0.0.0805 mzommlu ANEBEV 80:55 EzEmAEOJ ANEBEV 39:05 «.363sz

57

Table 10. Results of rmANOVA analyzzing Ameletus, Baetis, C inygmula, Chironomus,
Sweltsa, and Suwallia biomass (mg/m ) during as compared to before the salmon run.

 

 

 

Effect df F-value P-value
2! . E"
Edge 1. 218 1.8 0.181
Riffle 1. 218 34.7 <0.001
Drift 1. 218 5.17 0.024
Pool 1. 218 6.13 0.014
Hyporheic 1, 218 10.25 0.002
W311
Edge 1. 218 200.14 <0.001
Riffle 1. 218 170.56 <0.001
Drift 1. 218 55.78 <0.001
Pool 1. 218 115.57 <0.001
Hyporheic 1. 218 43.42 <0.001
Ameletusflinmass
Edge 1. 218 2.46 0.118
Riffle 1. 218 1.96 0.163
Drift 1. 218 0.05 0.831
Pool 1. 218 1.28 0.259
Hyporheic 1. 218 0.7 0.403
E . 8'
Edge 1. 218 24.21 <0.001
Riffle 1. 218 30.29 ~ <0.001
Drift 1. 218 0.02 0.88
Pool 1. 218 0.11 0.738
Hyporheic 1: 213 2.26 0.134
:’ l 3'
Edge 1. 218 3.95 0.048
Riffle 1. 218 35.61 <0.001
Drift 1. 218 3.65 0.058
Pool 1. 218 1.16 0.283
Hyporheic 1: 213 0.11 0.74
W11
Edge 1. 218 20.76 <0.001
Riffle 1. 218 45.71 <0.001
Drift 1. 218 1.81 0.18
Pool 1' 218 15.01 <0.001
Hyporheic 1, 218 8.13 0.005

 

58

Discussion
Macroinvertebrate density, biomass, and richness

We found that macroinvertebrates in riffle habitats were the most highly disturbed
communities among those evaluated. Macroinvertebrate density, biomass, and richness
significantly declined in riffle habitats during the salmon run. Other studies in Alaska
streams have documented reductions in riffle macroinvertebrate abundance and richness
during salmon spawning (Moore et al. 2004, Lessard and Merritt 2006, Moore and
Schindler 2008, Monaghan and Milner 2009). Studies in New Zealand and Michigan
streams have also shown that spawning salmon reduce riffle macroinvertebrate richness
(Hildebrand 1971, Field-Dodgson 1987). This macroinvertebrate response is likely due
to the large number of invertebrate taxa unable to survive the intense bioturbation
impacts from salmon during migration and spawning activities in main-channel habitats
(see also Janetski et a1. 2009).

Several studies have shown that pool refugia habitats retain more
macroinvertebrates compared to other habitats during flood disturbance events (Lancaster
and Hildrew 1993, Winterbottom et al. 1997, Lancaster 2000). However, in response to
salmon disturbance, we found that total insect biomass significantly decreased in
backwater pools, and only one taxon, the midge Chironomus significantly increased in
pools during the salmon run. Perhaps the high historical timber harvest activity of the
Twelve Mile Creek watershed (whereby 68% of the total watershed was harvested)
diminished the quality of backwater pools. Without appropriate riparian management,
logging can alter channel complexity by eliminating large wood retention, reducing

overall stream habitat heterogeneity and consequently altering pools and other habitats

59

(Duncan and Brusven. 1985). Over time, reductions in large wood inputs can cause a
shift in channel morphology toward wider, shallower channels and finer sediment size
due to fine particle erosion from unstable riparian banks (Barr and Swanston 1970,
Hawkins 1982). Perhaps the reduced habitat heterogeneity of Twelve Mile Creek
provides backwater pools that are not suitable refugia for macroinvertebrates that may
have otherwise colonized pools during salmon disturbance activities. Alternatively,
changes in the overall stream channel may also reduce the resettlement efficacy of
dislodged and passively drifting macroinvertebrates into lower quality pools if the flow
conditions are inappropriate. Macroinvertebrate density and taxonomic richness
significantly increased in the hyporheic zone during salmon spawning, suggesting the
hyporheos is a refuge for certain macroinvertebrates during salmon spawning. Other
studies have recognized the hyporheic zone as an important refugium for aquatic
macroinvertebrates and other river organisms during hydrological disturbances such as
floods and droughts (Williams and Hynes 1974, Oliver et al. 1997, Rosario 2000).
Macroinvertebrate density and richness of drift increased significantly during
salmon spawning. A study in Alaska found that total invertebrate drift was 2-4 times
greater during sockeye salmon (Oncorhynchus nerka) spawning as compared to pre and
post drift (Peterson and F oote 2000). Another study in coastal Alaska found that the
massive streambed disturbance due to salmon significantly increased macroinvertebrate
drift density (Monaghan and Milner 2009). Spawning salmon likely caused catastrophic
drift of macroinvertebrates. Alternatively behavioral drift may have been for a
component of increased drift if these benthic macroinvertebrates have evolved life history

traits to avoid spawning disturbance. The literature on drift propensity shows that

60

behavioral drift is a crepuscular activity (Waters 1972, Hynes 1975, Chutter 1975, Waters
and Hokenstrom 1980). We sampled drift at noon to avoid behavioral drift, and thus,
drift in this study should represent primarily those individuals that have been dislodged
due to spawning activities.
Insect Order-Level Response to Spawning Salmon

Ephemeroptera and Diptera biomass significantly declined in riffles during
salmon spawning. Others studies from Southeast Alaska found significant declines in
riffle mayfly taxa during salmon spawning such as Baetis, Seratella (Ephemeroptera:
Ephemerellidae), Drunella grandis (Eaton) (Ephemeroptera: Ephemerellidae),
Cinygmula, and Drunella doddsi (N eedharn) (Ephemeroptera: Ephemerellidae) (Lessard
and Merritt 2006, Lessard et al. 2009). The most dominant dipteran family in Twelve
Mile Creek was the Chironomidae, comprising 82% of total Diptera. Other studies in
Alaskan streams have observed declines in Chironomidae density and biomass in riffle
habitats during salmon spawning (Peterson and F oote 2000, Moore et al. 2004, Lessard et
al. 2009). Plecoptera and Trichoptera biomass was not significantly affected by
spawning salmon in riffles, perhaps because they have been reported to feed on abundant
salmon tissues associated with post-spawning activities (Ellis 1970, Walter et al. 2006).
Plecoptera biomass increased during the salmon run in riffle habitats, and the most
dominant plecopteran was the predaceous Sweltsa, which may consume the abundant
salmon tissues and salmon eggs (Ellis 1970). Trichoptera biomass significantly increased
in riffles, and another study found that Ecclisomyia conspersa (Trichoptera:
Limnephilidae) fed on salmon tissues and increased in abundance in riffle habitats during

spawning due to the increased food availability (Walter et al. 2006).

61

The efficacy of stream edge habitats as refugia appears to be taxa specific.
Diptera biomass significantly decreased in edge habitats during spawning. Dipterans
appear to be the least tolerant of spawning disturbance as they significantly decreased in
both riffle and edge habitats and did not increase in any potential refuge habitats during
the salmon run. Trichoptera biomass significantly increased in edge habitats during
salmon spawning. Caddisfly response is likely species-specific, with some species that
migrated to edge habitats to avoid salmon disturbance while others were feeding on the
abundant salmon carcasses along stream edges as the salmon run progressed (Walter et
al. 2006). Trichoptera biomass significantly increased in the hyporheic zone during
salmon spawning. Early instar caddisflies may migrate vertically into the more protected
hyporheic zone during spawning to feed on organic particles in less disturbed areas of the
substrata. Trichoptera biomass significantly decreased in stream drift during the salmon
run. This is likely because most caddisflies appeared to migrate to stream edges during
salmon spawning. Plecoptera biomass increased appreciably in stream drift during
salmon spawning. Plecopterans are perhaps highly susceptible to bioturbation due to
spawning salmon and may be very easily dislodged from the benthic substrates.

F unctional-F eeding Group Response to Spawning Salmon

Scrapers, collector-filterers, collector-gatherers, and predators all significantly
declined in riffle habitats during salmon spawning. This response is most likely due to
their feeding behavior, as scrapers forage on biofilm from the tops of rocks in the active
channel where redd construction is prevalent and thus are highly vulnerable to
bioturbation impacts (Lamberti 1996, Quinn 2005). However, chlorophyll a actually

increased in riffles during the salmon run, and yet scraper biomass still significantly

62

declined, a response likely due to abrasion and dislodgement. Another study in Southeast
Alaska found declines in riffle macroinvertebrates despite increases in chlorophyll
abundance (Chaloner et al. 2004). Collector-filterers are likely easily disturbed by
spawning salmon due to their sensitive filter feeding mouthparts that may be abraded by
fine sediments dislodged by spawners, or physical displacement by salmon movement.
Collector-gatherers may also be easily disturbed by the dislodged sediments and
increased turbidity that occurs during spawning that might fill in benthic interstices where
collector-gatherers feed. Predators may be low in abundance in riffles due to little prey
availability; or they may be feeding on salmon carcasses in low-disturbance habitats such
as main—channel pools or stream edges. Shredders significantly increased in riffles
during salmon spawning, a response likely due to the concomitant occurrence of leaf
litter fall during the late summer salmon run.

Shredders and collector-filterers significantly declined in backwater pools during
spawning. Shredders may be moving from pools into riffle habitats to feed on
allochthonous leaf litter inputs within riffles. Shredders and collector-filterers may also
be susceptible to anoxic conditions that can occur in pools as the salmon run progresses
and salmon carcasses build up and decompose in backwater pools. Collector-filterers
increased, though not significantly, in edge habitats during salmon spawning. This is
likely due to decreased disturbance from spawners in edge habitats as collector-filterers
appeared to decline in riffle habitats and increase in edge habitats during spawning.

The hyporheic zone is an important ecotone between surface and ground water
which supports several resident fauna, typically meiofauna less than 1mm, such as water

mites, early instars of aquatic insects, rotifers, segmented worms, flatworms, and

63

crustaceans (Boulton et al. 1998). Shredder, collector-gatherer and collector-filterer
biomass significantly increased in the hyporheic zone during spawning. These functional
feeding group taxa may migrate vertically into the hyporheic zone during spawning to
feed on organic particles in less disturbed areas of the substrata. Also, predator and
collector-gatherer biomass increased appreciably, though not significantly, in stream drift
during the salmon run. Predators and collector-gatherers are perhaps highly susceptible
to bioturbation due to spawning salmon and may be very easily dislodged from the
benthic substrates. Collector-filterers decreased in drift during the salmon run and
instead may be migrating or be displaced into stream edge and hyporheic habitats as
refuge from spawner disturbance.
Specific Taxa Responses to Spawning Salmon

Midges belonging to the genus Chironomus significantly decreased in riffles and
backwater pools, and significantly increased in the hyporheos and drift during the salmon
run. Perhaps these collector-gatherers were disturbed in riffles and pools due to lack of
organic matter in benthic interstices that spawners dislodged during spawning activities.
Chironomus biomass may have also declined in riffles and pools due to the fact that
Twelve Mile Creek is a heavily harvested watershed with relatively small sediments that
may be readily dislodged by spawners. In contrast to our findings, another study in
Southeast Alaska found increases in Chironomidae biomass during salmon spawning and
postulated that this may be due to well developed dispersal ability along with fast
reproduction and development (Armitage et a1. 1995, Chaloner et al. 2004).

The three most dominant mayfly taxa in Twelve Mile Creek were Ameletus,

Baetis, and Cinygmula; Baetis and'Cinygmula significantly decreased in riffles during the

64

salmon run. Others studies from Southeast Alaska found significant declines in riffle
mayfly taxa during salmon spawning (Lessard and Merritt 2006, Lessard et al. 2009).
Another study in the Pacific Northwest found significant declines of Baetis and
Cinygmula mayflies during salmon redd excavation in riffle habitats (Minakawa and
Gara 2003). Baetis and Cinygmula mayflies significantly decreased in edge habitats
during the salmon run. Edge habitats are not a refuge for these two genera and Baetis and
C inygmula may be especially vulnerable to spawning activities due to their modes of
feeding. Baetis is a collector-gatherer that feeds on fine particulate organic matter within
benthic sediments which may become dislodged during spawning activities, thus
reducing Baetis food availability. Cinygmula is a scraper that scrapes benthic biofilm
from the tops of rocks in the main-channel. A study looking at benthic biofilm within
Twelve Mile Creek found spawning salmon reduced algal abundance, thus also reducing
the major potential food resource for Cinygmula (Tiegs et al. 2008).

Sweltsa density increased while Suwallia density decreased in riffles, the
hyporheos, backwater pools, and stream edges during the salmon run. These results may
be due to evolutionary competitive exclusion of these very similar predacious
Chloroperlidae stoneflies. Prior to the salmon run, Suwallia was the dominant
plecopteran in Twelve Mile Creek, and then their abundance dropped to nearly zero in all
habitats during the salmon run. Sweltsa stoneflies were not present at all in any habitats
prior to salmon arrival and then became overwhelmingly dominant in all habitats during
the salmon run. It is quite possible that such similar genera have adapted life histories
that limit direct competitive interactions and minimal physical overlap with Sweltsa being

the most abundant predator to feed on salmon tissues during the spawner run (Ellis 1970);

65

the ghost of competition past (Connell 1980). While Suwallia likely feeds on
invertebrates prior to salmon arrival and then emerges before the salmon run to avoid
competition with Sweltsa stoneflies. Sweltsa stoneflies significantly increased in stream
drift during salmon spawning, a result likely due to their high abundance in all habitats

during the salmon run and they may be easily dislodged from benthic substrates.

Conclusions

Disturbance is a central organizing factor in stream ecology (Resh et al. 1988) and
is fundamental to the concept of patch-dynamics whereby ecosystems are viewed as both
temporally and spatially variable mosaics, in which patchiness is established by the
heterogeneous impacts of disturbance (White and Pickett 1985). Spawning salmon create
patches of disturbance through redd digging and upstream migration which shifts the
colonization and community structure of benthic macroinvertebrates (Milner et al. 2008).
We demonstrated that benthic disturbances from spawning salmon altered the
distribution, abundance, and community composition of benthic organisms and caused
significant reductions of macroinvertebrate communities in riffle habitats where salmon
activities are often greatest, a result found in several other studies (Peterson and Foote
2000, Chaloner et al. 2004, Moore et al. 2004, Lessard et al. 2009). However, few
studies have investigated the efficacy of macroinvertebrate refugia habitats in response to
spawning salmon disturbance.

Studies regarding macroinvertebrate utilization of refugia in response to
disturbance have largely focused on flow (Palmer et al. 1991, Lancaster et a1. 1993,

Negishi et al. 2002, Olsen and Townsend 2005) and the use of refugia in response to

66

spawning salmon is poorly understood. This research elucidated some of the
mechanisms of macroinvertebrate persistence throughout the annual benthic disturbance
due to spawning salmon. This research also demonstrated that the observed
macroinvertebrate response to disturbance depends on where samples are taken within a
stream channel. Most studies that show reductions in macroinvertebrate abundance in
response to spawning activity were sampled from riffle habitats (Duncan and Brusven
1985, Minakawa and Gara 2003, Tiegs et al. 2009, Monaghan and Milner 2009).
However, if refugia habitats were sampled instead, one might observe increases in
macroinvertebrate abundance and richness during spawning disturbance. Biomonitoring
and other benthic studies should include sampling from several habitat types in order to
get a broader view of how macroinvertebrates are responding to disturbance impacts.
This study supports the theory that stream habitat heterogeneity is a vital feature that
mediates spawning salmon disturbance impacts (Monaghan and Milner 2009). The
function of certain in-stream habitats such as backwater pools, stream edges, and the
hyporheic zone are effective refugia for certain macroinvertebrate taxa. Refugia habitats
may be a crucial determinant of macroinvertebrate survivability during the massive
annual benthic disturbance from spawning salmon in Southeast Alaskan rainforest

streams.

67

Chapter 3. Macroinvertebrate Community Differences in Riffle and
Backwater Pool habitats in Response to Spawning Salmon in
Southeast Alaskan streams
Methods

This study was conducted between July and September 2007 in four streams:

Nossuk Creek, Indian Creek, Maybeso Creek, and Twelve Mile Creek on Prince of Wales

Island, Southeast Alaska, USA (Fig 1). Benthic macroinvertebrates were collected semi-

quantitatively within a defined 0.315m2 area with a D-frame kick net from riffle and pool

habitats (Fig 13). This was accomplished by placing a 1m long weighted PVC pipe on
the benthos and collecting a sample within the 1m length and the 0.315m width of the D-
Net (Fig 14). Each riffle sample was collected for 30 seconds by agitating the benthos at

a depth of 8-100m within a randomly chosen area. The pool samples were collected by

sweeping the D-Net within a 0.315m2 randomly chosen area repeatedly for 30 seconds.

The samples were rinsed through a 500 m sieve and then preserved in 70% ethanol for
later processing. All insects were identified to genus or species (except the family

Chironomidae, which were left at family) and non-insect taxa were identified to order or

family. Samples were collected twice before the salmon run (15th June and 12nd July)

and twice during the salmon run (17th August and 20th September) from 6 riffles and

pools within a 300m reach of each of 4 streams. Two of the streams were from low-
harvest watersheds: Indian Creek and Nossuk Creek, and two of the streams were from
high-harvest watersheds: Maybeso Creek and Twelve Mile Creek. Several abiotic factors
were collected in each habitat within each stream including: water temperature, %

dissolved oxygen, conductivity, pH, % canopy cover, water velocity, and water depth.

68

Fig 13. Sampling backwater pool habitats (A) and riffle habitats (B).

 

69

Fig 14. Sampling technique used in riftle and backwater pool habitats in 2007. A 1m
long weighted PVC pipe was used to delineate a quantitative area of 0.315m2.

 

70

Results and Discussion

A non-metric multi-dimensional scaling ordination demonstrated a significant
separation of the macroinvertebrate communities in riffle and backwater pool habitats
(Fig 15). There was a significant decline in macroinvertebrate abundance in riffle
habitats upon salmon arrival within Maybeso Creek (Fig 16), Twelve Mile Creek (Fig
17), and Indian Creek (Fig 18). There was a non-significant decline of macroinvertebrate
abundance within riffle habitats during the salmon run in Nossuk Creek (Fig 19). There
was no significant change of macroinvertebrate abundance within backwater pool
habitats upon the arrival of salmon in any stream. These data suggest that backwater
pools may offer refuge for macroinvertebrates during spawning in that pools are stable
and protected habitats that sustain macroinvertebrate communities that are not disturbed
by spawning salmon. Overall, when the data are pooled, macroinvertebrate abundance
declined in riffles and increased in backwater pools (Fig 20). These data also supports
the prediction that backwater pools offer refuge to macroinvertebrates during salmon
spawning disturbance. Habitat heterogeneity is important as it offers diverse habitat types
that sustain diverse and unique taxa (Kerans and Karr 1992, Gjerlev et a1 2003). Riffle
habitats are often dominated by scrapers, such as heptageniid mayflies, that feed on
abundant biofilm and collector-filterers, such as simuliid dipterans, that collect fine
particulate organic matter from stream drift. Backwater pools typically sustain shredders,
such as limnephilid case-building caddisflies, and collector gatherers, such the
Chironomidae (Merritt et al. 2008). Backwater pools represent important reach scale
heterogeneity in stream flow and substrate composition, which may offer refugia for

macroinvertebrates and other stream organisms (Lancaster 1993).

71

Figure 15. Non-Metric Multi Dimensional Scaling ordination showing the significant
separation of macroinvertebrate community structure in riffle and pool habitats.

 

 

 

 

A
A
A A A
A
A A AA
‘AA A Z: A
A A A AAAAAA A
A 49A A
AA A A A
A A
AA A AA A
AA A A
AAAA
AAA A AA
AA A AA
N
.SL’A 0
E A A A O
A 00 0
A .
5.: "g’
0
A 0. ’0. .
0%.. ’.
00 0'
0 '0 0.
.0
0 . .0 '0 0.
0. g .
. O
0
9 0
or
' 0
Axis1

72

Habitat

o Riffle
A Pool

Fig 16. Macroinvertebrate abundance in riffle and pool habitats in Maybeso Creek.

Macroinvertebrate Abundance in Riffle and Pool Habitats
during the 2007 Salmon Run in Maybeso Creek, AK

 

   

 

 

 

 

51200 0.6
g - Riffle
\. - Pool
2,1000 1 — Fish Density ~ 0.5 A
U) N
.9 E
9 \
3 800 - ~ 0.4 5
t 10:
2 as
'9 600 . ~ 0.3 t“?
8 8
2 o
3 400 - ~ 0.2 i
.8 LL
E .3
:3 _J
z 200 - ~ 0.1
I:
8 l
2

o 4 . . . ., . ' . . .. . , . . . . 00

Jun Jul Aug Sep Oct Nov

Date

73

Fig 17. Macroinvertebrate abundance in riffle and pool habitats in Twelve Mile Creek.

Macroinvertebrate Abundance in Riffle and Pool Habitats
during the 2007 Salmon Run in Twelvemile Creek, AK

 

  

  

 

 

 

A2500 0.7
3 - Riffle
g - Pool . 0.6
£2000 - — Fish Density NE
*‘ - 0.5 \
8 a
t 1500 . :
9 - 0.4 v
..=. .é‘
(I)

§ 03 5
2 1000 . ' g
2 _0, t.
g ' .?>.’
3 500 « .1
z i - 0.1
C
(0 .
g ' A i

0 - l ' . .. . 3 . 0.0

Jun Jul Aug Sep Oct Nov

Date

74

Fig 18. Macroinvertebrate abundance in rifile and pool habitats in Indian Creek.

)

Mean Number of Macroinvertebrates (+I- s.e

2000

Macroinvertebrate Abundance in Riffle and Pool Habitats

during the 2007 Salmon Run in Indian Creek, AK

 

.A

0"

8
1

1000 .

500 4

 

0

Jun

 

Jul

- Riffle
- Pool

—— Fish Density

 

 

.—

p—

 

 

Aug Sep
Date

75

 

Nov

0.4

0.2

0.1

0.0

Live Fish Density (# fish I m2)

Fig 19. Macroinvertebrate abundance in riffle and pool habitats in Nossuk Creek.

)

Mean Number of Macroinvertebrates (+/- s.e

Macroinvertebrate Abundance in Riffle and Pool Habitats
during the 2007 Salmon Run in Nossuck Creek, AK

 

 

 

 

 

 

76

.1000
- Riffle
- Pool "
800 . — Fish Density
600 s
400 J "
200 - I I
1
Jun Jul Aug Sep Oct Nov
Date

0.14

0.12

- 0.10

~ 0.08

0.06

0.04

0.02

0.00

Live Fish Density (# fish / m2)

Fig 20. Mean macroinvertebrate abundance in riffle and pool habitats before and during
the salmon run.

70

 

- BEFORE

- DURING
60 4

50-1

40‘

30-

20-

10-

 

 

Mean macroinvertebrebrate individuals (+l- s.e.)

 

 

 

RIFFLE POOL
Habitat Type

77

Appendix 1
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.: 2010-01

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

Macroinvertebrate Community Response to Timber Harvest and Spawning Salmon in

Southeast Alaska Rainforest Streams

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

Entomology Museum, Michigan State University (MSU)

Other Museums:

lnvestigator's Name(s) (typed)

 

Emflv Yvonne Campbell

 

Date May 1. 2010

*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.

78

i 28 5.
Q3“ $8.1m.
m3). Em

E293: 23m 59:2: as 5 e88“.
3. «85.08» BE. uses as e939".

 

 

9-524 28

 

 

.338 85> 2.2.“.

 

5.9.8 .02 5:030)

68.8 Easez 303939;.
Seances: e 38% .8868 one

 

Appendix 1.1

Voucher Specimen Data
Page 1 of 5 Pages

v-IONNV'V'v-FNv-N

ooomRmB .m .03. v_< .onO 052 9.03...

8835 .m sage»... x4. .385 2:2 29¢

Bowen. a .08 x4. £85 use...
583% .m .08 x< £86 8352
8835 .m 68 x< 286 2:... 29:
8833 N Sew x< £86 2;. 22¢
s8~§a .v .8“. x< {20 2:2 29:
583% .m 68 x< .xoeooeeessz
NSNBS .0 .8e x< {so cues
385% .F .08 x< .0320 cheeses.
58823 .F .8“. v2 820 0.0382
NSNBE .F .08 x< {so 3352
Bogs .e .08 v2 £86 sees
88:3 .m seem x< {20 2:2 29¢
8853 .F .oos v2 {so 8832
8835 .m .08 x< £86 0.0382
583% .Em see x4. .0.er 88.52
889:. .m .8n_ vi .0620 0:2 9.2:

3852.8 520
cheese: i=5“. $58.8 520.
338.5 ”macaw .owv_ct>0 ”35mm
33055 3:53

£562 ”3:00 .omBEE “2.6m“.
3.52 ”8:00 .omBEE ”2:5“.
8%on ”880 $28.3 SEE
8:820 ”380 323:5 mafia”.
meQEEI ”350 63.0sz ”2E5“.
wuoaoaoo $3.095 60033.0 ”mam-o
$25398 ”2E5“. .eoaoeoo 5.20
«Reggae. .320 .8855 330
@63896 huge

grew 90289.0 ago
E925: .BeoEofiEez “£225
853.: ”e86

emo< M8655 3:594. ago
Saw 52582 $90

 

 

 

 

 

 

 

Pupae
Nymphs
Larvae
Eggs

deposited

Museum

3588 see
use: b 38:8 ecoéeeem .2 See .33

 

coxE .050 3 020on

 

 

 

3 Adults

.... Other
E
3
z

O

 

79

Appendix 1.1

Voucher Specimen Data

5 Pages

of

Page 2

San. .2830

2:585. 30.02.25
3.595: 85 5852.2 e... c. .588
b. 83.8% 8.2. 908 2.. 59.88
6.98 .e2 .88.;

 

 

 

 

 

2.5.24 28
.5858 85; 2.5m.
683. @882 382.85.

2.388.. 2 950% .8368 $3

 

 

 

 

 

 

 

 

 

 

 

 

 

m 882.2 .o 58 £< £85 2.2 9.8: a 85 .23: .885 82.2.: 2.58
P mooNaNR .N .08 v.< .850 2.2 9.85 2855 .880 82.2.: 2.58
N 8832. .N 88 £< £85 2.2 9.85 eases: .885 82.2.: ”2.58
N 8832,. .P .8“. £< £85 2.2 28¢ 288.5 .880 82.3.: 2.58
m. 5835 .N 2...”. x< £85 2.2 9.8: 82.555 ”2.58
N 3852. .5 .8n. £< .85 88.52 82.88 .880 828588 2.58
N 88.83 .F 88 £< £85 2.2 9.2.5 85850 880 6820.8 ”2.58
N 2823 .. .8n. £< £85 8.2.. s.eseeed .880 82588258 2.58
m 8835 .m 2...... v2 £85 88.62 8.22.8 2.58
P 8853 .m 88 2... .85 2.2 9.23 82.22.50 “2.58
F 8852.. .N 8.. £< £85 0896.2 £58.58 .885 82588258 2.58
m 58.85 .N .85 v.< £850 0896.2 59.82.“. 82.50525 “2.25...
m 88.82. .:.N 2...”. V2 £85 e828: 88.50.9880 82589.5 2.58
N BoNaNa .EN 2...”. v2 .855 8882 528.50.55.82“. 68.58550 “2.58
N 3853 .N .8n. £< £85 88222 82228 .880 82.22.55 “2.58
e 882% .m .8e x< .850 6882 82525.5 2.58 .5220 820
F BONRNB .m 58 £< £85 .5882 822.585 2.58 5280.8 585
e 8882. .m 2...... £< £85 .e...2 9285 822.585 2.58 528.8 820
m m s 8.588 2:...
m m m m m m m m w .8: .e 88:8 8558 .2 s8 .83 Sea. .85 .e 8.8m
m m 1% Nu. M cm W m E
up 59:52

80

Appendix 1.1

Voucher Specimen Data

Page 3 of 5 Pages

2.59.5 2.25 58222 8. 2 2858

38

.8830

2:322 >mo.oEoEw

 

2-524

850

 

 

.8 952.5QO no.2. 935 05 32.85”.

 

5.05m .02 .oco:o>

=mnaEmo 92.0; 2:25

682:

55552 8.28.525.

3.5389. 2 220% 55.228 3:.

 

.835 2. .85 2... .285 2852
.888 .5 222 v2 .285 .0882

2.2.5585 2.58 585028.. .85
828:0 2.58 23850: .220

 

 

 

 

 

 

 

 

F 58.8... .F .8... v.2. .285 2.2 9.2.: 82.885 2.58
F 88.3... .m 55 x... .285 2.2 9.2...» 8229.. 2.58
9 88.8.8 .m 88 .2 .285 2.2 9.2.5 8252588.. 2.58
v 888... .F 28.82.. v2. .825 2.2 9.23 222285258 .880 82.8.2853 2.58
8 8°35 .m 88 2... .285 2.2 9.8.» 565...: 82.8822. 2.58
F 88.3 .m 8.8 v2 .285 2.2 9.25 8855...... 2.80 82.888: 2.58
v 8855 .v 222 v.< .2850 2852 858...”. 880 82508.8: 2.58
2 88.85 .EN 22.. v2 .285 88.52 2353.5 880 82.8852. 2.58
F moowauk .m 2&3 x< .2280 2.5. 9.02.5 56:25: 652.8250ch ”2.68.
5 883:» .8 22m. V... .550 88.52 2.258 2.80 82.20583 2.58
m 58.85 .EN 2...”. v2 .285 88.52 2.855 ”8.085 2.855 880 82.2388 2.58
N 58.25 .N 2.5. v2 .820 20882 58.52. 52850. 2.855 .880 82.50588 2.58
N 58.55 .F 2...”. V2. .285 522.. .88.. ”8.085 2.855 880 82:28.88 2.58
m NSNBNB .EN 822 v2. .285 8852 2.8... .880 .828m 2.58
V 58.95 .m .3... x< .2080 203532 922.25.. ”9:00 .3253sz ”2:88.
F 38.93 .N 0.25.82: x< .2020 9:2 9.03... N 82... 5.23.» “3:20 .3259... “2.25“.
m m s 8288.. 25
m m m m. m m m m w .88 .o 852.8 885.88 .2 .28 .83 595. .85 .o 8.085
m m .m m M cu. m. m E

 

 

«o .3532

 

 

81

Appendix 1.1

Voucher Specimen Data

Page 4 of 5 Pages

£225 9.2m 5222.2 05 2 588

28

.6650

.8395. 620E025

 

0:65: 2mm.

 

 

.2 80268» 32. 93a 9: 3:082

 

rod—hm .oz :ocoso>

:3228 25; 2:5
682: :mEEmz £52592
$3800: 2 902m .mcoEocm 3.3

 

 

 

 

 

 

 

 

 

N BoQomB ._. .00.... x< .230 003222 38258.5 “3:00 .mwu:Eam:E_._ ”2:23“.
N 88:3 .2 2.22 x< .286 :22. 2228.86 ”258 32.892: 2:53
N 2835 .v .82 x< .2020 0896.2 $320.: ”328 $358253 2:59.
n 882:» a 2.22 2< .286 2:2 9.2,: 2=m§< H380 .82_cm§< ”2:58
P 883% .< :5 x< .285 2:2 9.9.: 2288mm :25
m: 8035 .. 25¢ x< .230 2:2 9.22:. £2.22 222025 32:55:25.2 “5222.:
m 383:: .v 222 x< .285 0826.2 ”228.22 “8:8 .8285: “2.5m“.
m 8833 .2 £3 2,: .285 089222 «8qu “2.80 .oaoEoEoz “2:53
n Bowen .o 25m x< .285 20382 £03358 “2.50 3288.. ”2:53
V BomBEm .m 0:5: x< .2020 003222 32:89.25 ”2E3“.
m 88:86 .m 55 x< .2020 2:2 9.23... 3:26 “macaw 63.2.8225 “gem”.
m 3833 .EN 252 x< .2020 88.3.2 2:38 “2.80 322.8825 2:59.
2 @8283 .m 8cm x< .286 2:2 9.25: @328 ”380 .8238 2:58
F 883m .m .82 v2 .285 2:2 9.2;: $288 2:25“. 228022 :85
F 8033 .33 252 x< .286 089%: $8ng 286
m 885m .m 25m vi .2080 2:2 9.25» moo €08.22 3552.580
2 885. .o :8 x< .285 2:2 22¢ «8:258 :85 68330 ago
: 3083B .86 058 v_< .2020 0826—2 «.295 ”$30
m m W 8:828 Ba
m m. m m m m m w 82:0 828:8 836822 9% .83 SE 5520 868m
M m M nu. M, m E

 

 

«0 28:52

 

 

82

Appendix 1.1

Voucher Specimen Data

Pages

Page 5 of 5

3.29.5 82w 28.222 2.. 2 2828
a. 2.8288 8.2. 2,82 9.. 82.822

320

23250

.8335. 30.02.25

 

2-222.. 28

 

 

 

5.05m .02 .5502;

.6858 829.: 2.22m.
.82.... .2282 2:28:29...
3888.. : 28% .2288 8:.

 

 

 

 

 

 

 

 

 

m 882.5 .m 22:82: x< .2220 2:2 9.2;» 822:2: 2......22 222882.: 28.0
m Sonata .m .002 x< .2020 203.82 mquomxfi 82.20 .32.an825: ”22.52
F 88.3: .m 22.82: x< 28.0 88222 829822.22 .880 828882.22 ”2:82
. a 283:: .m .82 2< .285 22822.2 822.8..ng .880 82:28:52 ”2:82
.2. 28282 .m .82 x< .285 88222 882E885 .880 82.2285: ”2.82
F 28.2.23 22.2. 2.22 V? .285 8822.2 28.8223 ”2.80 82522828.. ”2:82
.2 2.833 .m .82 v2 .285 2.8282 «8.828.. .380 82.28.8223 ”2:82
.2 283:2 .n .82 v2 .285 88:22 2282.82 880 82.2285... 22.522
m m W 8:88 82
w m W m m m m m m. 2.2 82. .2 8.02.8 2.82.08 .2 28 .83 88:25 a 8. m
M m m M Du. M. w E

 

 

:0 23:52

 

 

83

Literature cited

Allan J .D. 2004. Landscapes and riverscapes: The influence of land use on stream
ecosystems. Annual Review of Ecology and Evolutionary Systems. 35: 257-284.

Barr DJ. and D.N. Swanston. 1970. Measurement of Creep and Shallow, Slide-Prone Till
Soil. American Journal of Science. 269: 467-480.

Baker T.T., A.C. Werthweimer, 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.

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.

Boulton, A.J., S. Findlay, P. Marmonier, E.H. Stanley, and HM. Valett. 1998. The
functional significance of the hyporheic zone in streams and rivers. Annual Review of
Ecological Systems. 29: 59-81.

Bourque C.P.A. and J .H. Pomeroy. 2001. Effects of forest harvesting on summer stream
temperatures in New Brunswick, Canada: an inter-catchment, multiple-year comparison.
Hydrology of Earth Systems Science. 5: 599-613.

Bowles, D.E., R.A. Short. 1988. Size composition of invertebrate drift and fish predation
in a Texas stream. The Southwest Naturalist. 33(2): 177-184.

Bryant MD. 1980. Evolution of large, organic debris after timber harvest: Maybeso
Creek, 1949 to 1978. United States Department of Agriculture, Forest Service. Pacific
Northwest Forest and Range Experiemntal Station, general technical report PNW-lOl.

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

Chutter, F .M. 1976. Variation in the day-time drift of a natal river. Verh. International
Verein. Limnology. 19: 1728-1735.

Crowder, LB. and WE. Cooper. 1982. Habitat structural complexity and the interaction
between bluegills and their prey. Ecology 63: 1802-1813.

Cushing C.E., K.W. Cummins, and G.W. Minshall. 1995. River and stream ecosystems

of the world. Chapter 2: River and stream ecosystems of Alaska, pp. 9-32. University of
California Press Berkeley and Los Angles, California.

84

Death R.G., M.J. Winterbourn. 1995. Diversity patterns in stream benthic invertebrate
communities: the influence of habitat stability. Ecology 76(5): 1446-1460.

Duncan, W.F.A. and MA. Brusven. 1985. Energy dynamics of three low-order southeast
Alaska streams: Allochthonous Processes. Journal of Freshwater Ecology. 3(2): 223-248.

Dupuis L. and D. Stevenson. 1999. Riparian management and the tailed frog in northern
coastal forests. Forest Ecology and Management. 124: 35-43.

Ehrman T.P. and G.A. Lamberti. 1992. Hydraulic and particulate matter retention in a
3'd-order Indiana stream. Journal of the North American Benthological Society. 11: 341-
349.

Ellis R.J. 1970. Alloperla stonefly nymphs: predators or scavengers on salmon eggs and
alevins? Transactions of the North American Fisheries Society. 4: 677-683.

Field-Dodgson, MS. 1987. The effect of salmon red excavation on stream substrate and a
benthic community of two salmon spawning streams in Canterbury, New Zealand.
Hydrobiologia 154: 3-1 1.

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

Gjerlov C., A.G. Hildrew and J .1. Jones. 2003. Mobility of stream invertebrates in
relation to disturbance and refugia: a test of habitat templet theory. Journal of the North
American Benthological Society. 22(2):207-223.

Gregory S.V., F.J. Swanson, W.A. McKee, and K.W. Cummins. 1991. An ecosystem

perspective of riparian zones: focus on links between land and water. BioScience. 41:
540-551.

Growns 1.0. and J .E. Growns. 2001. Ecological effects of flow regulation on
macroinvertebrate and periphytic diatom assemblages in the Hawkesbury-Nepean River,
Australlia. Regulated Rivers: Research and Management. 17(3): 275-293.

Halupka, K.C., Bryant, M.D., Willson, M.F., and Everest, F .H. 1999. Biological
characteristics and population status of anadromous salmon in Southeast Alaska. General
Technical Report. PNW-GTR-468. US. Department of Agriculture, Forest Service,
Pacific Northwest Research Station, Portland, Oregon.

Harvey B.C., J .L. White and R.J. Nakamoto. 2009. The effect of deposited fine sediment

on summer survival and growth of rainbow trout in riffles of a small stream. North
American Journal of Fisheries Management. 29: 434-440.

85

Hawkins, C.P., Murphy, M.L., Anderson, NH. 1982. Effects of canopy, substrate
composition, and gradient on the structure of macroinvertebrate communities in cascade
range streams of Oregon. The Ecological Society of America, Ecology 63(6): 1840-1856.

Heifetz J ., M.L. Murphy and K.V. Koski. 1986. Effects of logging on winter habitat of
juvenile salmonids in Alaskan streams. North American Journal of Fisheries
Management 6: 52-58.

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

Hynes, H.B.N., 1975. Downstream drifi of invertebrates in a river in southern Ghana.
Freshwater Biology. 5: 515-532.

Janetski, D.J., D.T. Chaloner, S.D. Tiegs, and G.A. Lamberti. 2009. Pacific salmon
effects on stream ecosystems: a quantitative synthesis. Oecologia. 159: 583-595.

Jones C.G., J .H. Lawton, and M. Shachak. 1994. Organsims as ecosystem engineers.
Oikos. 69: 373-386.

Jones E.B., G.S. Helfman, J .0. Harper and RV. Bolstad. 1999. Effects of riparian forest
removal on fish assemblages in southern Appalachian streams. Conservation Biology. 13:
1454-1465.

Lancaster J. 2000. Geomorphic scaling of microhabitat patches and their efficacy as
refugia during disturbance. Journal of Animal Ecology. 69: 442-457.

Lancaster, J. and AG. Hildrew. 1993. Flow refugia and the microdistribution of lotic
macroinvertebrates. Journal of the North American Benthological Society 12(4):
385-393.

Lessard J .L. and R.W. Merritt. 2006. Influence of marine-derived nutrients from

spawning salmon on aquatic insect communities in southeast Alaskan streams. Oikos,
113, 334-343.

Lessard J .L., R.W. Merritt and MB. Berg. 2009. Investigating the effect of marine-
derived nutrients from spawning salmon on macroinvertebrate secondary production in
southeast Alaskan streams. Journal of the North American Benthological Society. 28(3):
683-693.

Lobb III MD. and DJ. Orth. 1991. Habitat use by an assemblage of fish in a large warm
water stream. Transactions of the American Fisheries Society. 120: 65-78.

Lyman J. and N.K. Long. 2002. Alaska’s Wild Salmon. Alaska Department of Fish and
Game.

86

McCune, B., J .8. Grace, and D.L. Urban. 2002. Analysis of Ecological Communities.
MjM Software Design.

Merritt R.W., K.W. Cummins and MB. Berg. 2008. An Introduction to the Aquatic
Insects of North America. 4th edn. Kendall/Hunt, Dubuque, IA.

Monaghan K.A. and A.M. Milner. 2009. Effect of Anadromous salmon red construction
on macroinvertebrate communities in a recently formed stream in coastal Alaska. Journal
of the North American Benthological Society. 28: 153-166.

Milner A.M., A.L. Robertson, K.A. Monaghan, A.J. Veal, and EA. Flory. 2008
Colonization and development of an Alaskan stream communityover 28 years.
The Ecological Society of America, research communications.

Minakawa N, R.I. Gara, and J .M. Honea. 2002. Increased individual grth rate and
community biomass of stream insects associated with salmon carcasses. Journal of the
North American Benthological Society. 21(4): 651-659.

Minakawa N. and RI. Gara. 2003. Effects of chum salmon redd excavation on benthic
communities in a stream in the Pacific Northwest. Transactions of the American Fisheries
Society 132: 598-604.

Montgomery DR. and L.H. MacDonald. 2002. Diagnostic approach to stream channel
assessment and monitoring. Journal of the American Water Resources Association. 35: 1-
1 6.

Moore, J .W., D.E. Schindler, M.D. Scheuerell. 2004. Disturbance of freshwater habitats
by anadromous salmon in Alaska. Oecologia 139: 298-308

Moore, J .W. and DE. Shindler. 2008. Biotic disturbance and benthic community
dynamics in salmon-bearing streams. Journal of Animal Ecology 77: 275-284.

Moyle RB. 1976. Some effects of channelization on the fishes and invertebrates of Rush
Creek, Modoc County, Califomia. California Fish and Game. 62: 179-186.

Murphy, M.L. and A.M. Milner. Alaska timber harvest and fish habitat. 1996. In: Milner,
A.M. and M.W. Oswood. (Eds.). Ecological studies 119: Freshwaters of Alaska:
Ecological Synthesis. New York: Springer-Verlag. pp 229-263.

Negishi, J .N., M. Inoue and M. Nunokawa. 2002. Effects of channelisation on stream
habitat in relation to spate and flow refugia for macroinvertebrates in northern
Japan. Freshwater Biology 47: 1515-1529.

Oliver M.J., P. Marmonier, and J .L. Beffy. 1997. Response of invertebrates to lotic
disturbance: is the hyporheic zone a patchy refugium? Freshwater Biology. 37:

87

257-276.

Olsen DA. and CR. Townsend. 2005. Flood effects on invertebrates, sediments and
particulate organic matter in the hyporheic zone of a gravel-bed stream.
Freshwater Biology. 50: 839-853.

Palmer, M.A., A.E. Bely, and K.E. Berg. 1991. Response of invertebrates to lotic
disturbance: a test of the hyporheic refuge hypothesis. Oecologia. 89(2): 182-194.

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

Quinn J .M., R.B. Williamson, R.K. Smith, and M.L. Vickers. 1992. Effects of riparian
grazing and channelization on streams in Southland, New Zealand. New Zealand Journal
of Marine and Freshwater Research. 26: 259-273.

Reeves G.H., F.H. Everest and J .R. Sedell. 1993. Diversity of juvenile anadromous
salmonid assemblages in coastal Oregon basins with different levels of timber harvest.
Transactions of the American Fisheries Society. 122(3): 309-317.

Resh V.H., A.V. Brown, A.P. Covich, M.E. Gurtz, H.W. Li, G.W. Wayne, S.R. Reice,
A.L. Sheldon, J .B. Wallace, and RC. Wissmar. 1988. The role of disturbance in
stream ecology. Journal of the North American Benthological Scoiety 7: 433-455.

Ringler, N.H., D.F. Brodowski. 1983. Functional responses of brown trout (Salmo trutta
L.) to invertebrate drift. Journal of Freshwater Ecology. 2(1): 45-57.

Robson B.J. and LA. Barrnuta. 1998. The effect of two scales of habitat architecture on
benthic grazing in a river. Freshwater Biology. 39: 207-220.

Rosario, R.B.D. and V.H. Resh. 2000. Invertebrates in intermittent and perennial streams:
is the hyporheic zone a refuge from drying? Journal of the North American
Benthological Society. 50: 839-853.

Sedell, J .R., G.H. Reeves, F.R. Hauer, J .A. Stanford, C.P. Hawkins. 1990. Role of refugia
in recovery from disturbances: modern fragmented and disconnected river systems.
Environmental Management. 14(5): 71 1-724.

Shaw EA. and J .S. Richardson. 2001. Direct and indirect effects of sediment pulse
duration on stream invertebrate assemblages and rainbow trout (Oncorhynchus mykiss)
growth and survival. Canadian Journal of Fisheries and Aquatic Sciences. 58(11): 2213-
2221.

Schofield K.A., C.M. Pringle and J .L. Meyer. 2004. Effects of increased bedload on

algal- and detrital-based stream food webs: Experimental manipulation of sediment and
macroconsumers. Limnology and Oceanography. 49(4): 900-909.

88

Skinner, W.D. 1985. Night-day drift patterns and the size of larvae of two aquatic insects.
Hydrobiologia. 124: 283-285.

Taniguchi H. and M. Tokeshi. 2004. Effects of habitat complexity on benthic
assemblages in a variable environment. Freshwater Biology. 49: 1164-1178.

Tiegs, S.D., D.T. Chaloner, P. Levi, J. Ruegg, J.L. Tank and G.A. Lamberti. 2008.
Timber harvest transforms ecological roles of salmon in Southeast Alaska rainforest
streams. Ecological Applications. 18(1): 4-1 1.

Tiegs S.D., E.Y. Campbell, P.S. Levi, J. Ruegg, M.E. Benbow, D.T. Chaloner, R.W.
Merritt, J .L. Tank and G.A. Lamberti. 2009. Separating physical disturbance and nutrient
enrichment caused by Pacific salmon in stream ecosystems. Freshwater Biology. 54:
1864-1875.

Toyoshima T., S. Nakano, M. Inoue, Y. Ono, and Y. Kurashige. 1996. Fish population
responses to stream habitat improvement in a concrete-lined channel. Japanese Journal of
Ecology. 46: 9-20.

United States Department of Agriculture, United States Forest Service. 1997. Tongass
land management plan revision. Final Environmental Impact Statement, Region 10
(Alaska Region). US. Department of Agriculture Forest Service, Juneau, Alaska, USA.

Waters, T.F. 1972. The drift of stream insects. Annual Review of Entomology. 17: 253-
272.

Waters, T.F., J .C. Hokenstrom. 1980. Annual production and drift of the stream
amphipod Gammarus pseudolimnaeus in Valley Creek, Minnesota. Limnology
and Oceanography. 25: 700-710.

White RS. and S.T.A. Pickett. 1985. The ecology of natural disturbance and patch
dynamics. Academic Press, Orlando

Williams DD. and H.B.N. Hynes. 1974. The occurrence of benthos deep in the
substratum of a stream. Freshwater Biology. 4: 233-256.

Winterbottom J .H., S.E. Orton, A.G. Hildrew, and J. Lancaster. 1997. Field experiments
on flow refugia in streams. Freshwater Biology. 37: 569-580.

Wood P.J. and RD. Armitage. 1997. Biological effects of fine sediment in the lotic
environoment. Environmental Management. 21: 203-217.

Zweig L.D. and CF. Rabeni. 2001. Biomonitoring for deposited sediment using benthic

invertebrates: a test on 4 Missouri streams. Journal of the North American Benthological
Society. 20(4): 643-657.

89

   

"'1111111111111111s