iv ‘13

‘ I [733.3
{3311:}. 'Rl‘filh 21.3
- ' 3!.‘03‘1'HI‘ “:[- I :3:

V1153” - I
:I' .3143! :31, 3: |.

3538353133333 I' it? 44: .34.

I

'2'.".’L:

I

“REF". I R "r; I
_ .11 ‘ .3!
ML”. 'I'3R. " III 133501!th
: m I‘ .-
“it" My" flkat‘gi
RR " ‘-"|-I3 '3
'3‘3"

(3134'... :‘3L L3 4" :'|
3' 03.0.3“ "nth 3

' ‘ . I ,
R ”0“ 'Ril'33:331'i31‘23‘ $13333 I3
[RR W.) ”MIME" $3353“ RE”

549:" ' «II.
t“ #33":
i f" “""1‘ :1-
»; ' . I Wald; 13; 3 Rt
.3 . ~36! ‘-= ‘I'II‘ 115-31 33433 ":3 ’ 3%- I 3
I334! LH 19%;: ‘ ’ M 3m
3:

In"
334;»?
83.

, It" I
BR'HI‘"J‘R§ I

”he

. 3 A
' ,IILlJIp-IIIImfig .. 3 H
“II “‘I ‘

Rib" I I31 . ”
$3113 III LI

II" "

 

IJLI 'IIII‘I': . I .
5'3 2‘1“”? 1R!” .’ ,-. "mg'z' ‘ J ‘
33 I“ ; _ I 3 ‘3'- ‘ I ,- 43.4%, q- u:, _ ;; II' ‘3 in“

333333 ”R '.'I.-> ' ' ' : . h ‘ I 33.; . ‘20.; ‘Lfifhm' . ’3' (fiaj‘giIR'IT : ,

‘ . 53? _ . by]. . 1- . 4 ¢ I g
‘ 1> 337'“ 3 :{1f33R3I W];

II" |_ K

A I i"I

"l,.
,-
C
I
I3
‘

 

’ “13' 013i
‘9‘ «5

.3.” 3

3371133

£31 ME
"3343‘ . W11,“
43333: '1‘}?- "”3 I

- " 5?:
'2‘5‘.“ 4.“. ...
’-_-':"~:.* #3:: . '
“f": . " —3-:*.- ' .
I ckrlzgl.‘ ' .

:4 v.
u _

A": k-
.744: 1....

 

«:24-

;. Illilllii'lllfllIlllllllIllllllll'llllilllllllll

—"— 31293017141569

This is to certify that the

thesis entitled

The Use of Macroinvertebrates as Indicators
of Water Quality for Two Northern Lake Huron
Coastal Wetlands

presented by

Donna Rebecca Kashian

has been accepted towards fulfillment
of the requirements for

Masters degree in Fisheries & Wildlife

n‘; v . '

Major professor

”—

Date 7
g /

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

 

LIBRARY
Michigan State
University

 

 

PLACE IN RETURN Box
to remove this checkout from your record.
To AVOID FINES return on or before date due.

 

DATE DUE

DATE DUE

DATE DUE

 

 

9%” 1999

 

9")
NOV egg

 

 

 

 

 

 

 

“#3105729"-

 

 

JUN b 37260?!

 

 

 

 

 

use c/CIFICIDIhDquS-p.“

 

THE USE OF MACROINVERTEBRATES AS INDICATORS OF WATER QUALITY
FOR TWO NORTHERN LAKE HURON COASTAL WETLANDS

By

Donna R. Kashian

A THESIS

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

MASTER OF SCIENCE
Department of Fisheries and Wildlife

1998

ABSTRACT

THE USE OF MACROWVERTEBRATES AS INDICATORS OF WATER
QUALITY FOR TWO NORTHERN LAKE HURON COASTAL MARSHES

By

Donna R Kashian

The macroinvertebrate fauna of a moderately impacted Northern Lake Huron coastal,
emergent marsh was compared to the fauna of a nearby, relatively pristine, reference marsh.
The impacted marsh received domestic wastewater twice per year from a lagoon system and
was impacted by marina traffic and stormwater runoff fi'om a nearby urban area. The
reference marsh received no such impacts. Community structure of the macroinvertebrates
was determined from sediment and dip-net samples collected from June-September in 1996.
The invertebrate fauna demonstrated moderate impairment at the impacted marsh with fewer
insects present and a greater portion of the fauna existing as Amphipoda and Isopoda. There
was a diverse Ephemeroptera community of 7 species in the reference marsh for all sampling
dates whereas Ephemeroptera were present only in June with 4 species at the impacted marsh.
Trichoptera exhibited lower abundances and species richness at the impacted marsh compared
with the reference marsh. Differences in the macroinvertebrate communities were used to test
38 potential metrics as indicators of water quality. Ten metrics appeared to be good water
quality indicators including the relative abundance of Ephemeroptera, Isopoda, Trichoptera,

predators, filterers and a ratio of grazers to detritivores.

ACKNOWLEDGMENTS

I would like to thank the members of my committee, Dr. Thomas Burton, Dr. Thomas
Coon and Dr. Richard Merritt, for their constructive criticism and helpful comments. I would
especially like to thank my advisor and committee chair, Dr. Burton, for his support and
guidance throughout my graduate studies.

The study could not have been conducted without the valuable field and lab assistance
of Angela Conklin, Brian Keas, Todd Losee, Sam Rifi‘ell, and Mark Scalabrino. Don Urzarski
and Sam Rifi‘ell were particularly helpful with graphics and statistics. Many long hours at the
microscope were spent by Patrick Hudson and Margret Chriscinske of the US. Geological
Survey, who assisted me with insect identification and verification. A special thanks to
Patrick Hudson, who always took time out of his busy schedule to answer any insect
identification questions and who walked me through Chironornidae identification. Thanks to
everyone who assisted me in macroinvertebrate verification and identification including: Dr.
Brian Armitage (T richoptera), Ethan Nedeau (HirudinidaNature Conservancy fore), Dr. Steve
Chordas (Corixidae), Edward Roseman (Zooplankton), Dr. Richard Snider (Collembola), and
Brian Keas (Gastropoda). I gratefully acknowledge the Michigan chapter of the providing
housing. Funding for this project was provided by a fellowship from the U. S. Department of
Defense.

Thanks to my parents who introduced me to the wonders of nature and have always

iii

given me full support. A final thanks to Dan, my husband, who helped me maintain some
remnant of a sense of humor throughout this program and has been a source of constant
encouragement throughout my graduate studies. I most especially appreciate the patience he

has shown and the support he has provided during the writing of this thesis.

iv

TABLE OF CONTENTS

Page
List of Tables vii
List of Figures x

Chapter One: A Comparisons of Two Northern Lake Huron Coastal Marshes l

1. Introduction 1
2. Description of Study Sites 5
3. Methods 9
3.1. Land-use/Land-cover Characterization of Sites 9
3.2. Chemical-Physical Data Collection 9
3.3. Invertebrate Sampling 10
4. Results 14
4.1. Comparisons of Land-use/Land-cover in watersheds 14
4.2. Comparisons of Water Quality Data , 14

4.3. Comparisons of Macroinvertebrate Community Dynamics 22

Aquatic Insects 25
Non-Insect Macroinvertebrates 53
Zooplankton Community 66
Macroinvertebrate Trends Along a Pollution Gradient 68

v.)

Table of Contents (cont'd)

5. Discussion

Chapter Two: Implications for the Development of a Multimetric Index
of Ecological Integrity

1. Introduction
2. Methods
2.1. Metric selection
2.2. Metric testing
3. Results
4. Discussion
Summary & Conclusions
Appendices

Literature Cited

70

76

76

78

78

84

88

94

103

107

113

Table 1.

Table 2.

Table 3.
Table 4.
Table 5.
Table 6:
Table 7.
Table 8.

Table 9.

LIST OF TABLES

Total area and Percent Land-use/Land-cover in Cedarville and
Mackinac bays watershed (MIRIS 197 8).

Aquatic insects collected with standardized dip-net sweep sampling
from Scimus dominated zones (mean number/sample:S.E) and
with cores from sediments in these zones (mean number'm'2_+S.E)
fiom the reference (Mackinac) and Impacted (Cedarville) marshes
in 1996. "‘ was not collected in the samples.

Total number of Chironomidae collected and percent of the total
Chironomidae represented by taxa in the plant associated
community in Mackinac (reference) marsh, Lake Huron, 1996.

Total number of Chironomidae collected and percent of the total
Chironomidae represented by taxa in the plant associated
community in Cedarville (impacted) marsh, Lake Huron, 1996.

Total number of Chironomidae collected per m2 and percent of the
total Chirononridae represented by taxa in the sediment community
in Mackinac (reference) marsh, Lake Huron, 1996.

Total number of Chironomidae collected per m2 and percent of the
total Chironomidae represented by taxa in the sediment community
in Cedarville (impacted) marsh, Lake Huron, 1996.

Total number of Ephemeroptera collected and percent of the
total Ephemeroptera represented by species in the plant associated
community in Mackinac (reference) marsh, Lake Huron, 1996.

Total number of Ephemeroptera and percent of the total
Ephemeroptera represented by species in the sediment of Mackinac
(reference) marsh, Lake Huron, 1996.

Total number of Odonata and percent of the total Odonata
represented by taxa in the plant associated community in Mackinac
(reference) marsh, Lake Huron, 1996.

vii

14

26

34

34

36

36

.38

38

50

Table 10.

Table 11.

Table 12.

Table 13.

Table 14.

Table 15.

Table 16.

Table 17.

Table 18.

Table 19.

List of Tables (cont'd)

Total number of Odonata and percent of the total Odonata
represented by taxa in the plant associated community of Cedarville
(impacted) marsh, Lake Huron, 1996.

Non-insect macroinvertebrates collected with standardized dip-net
sweep sampling from Scimus dominated zones (mean
number/sample:S.E) and with cores from sediments in these zones
(mean number-mfiSE) fiom the reference (Mackinac) and
Impacted (Cedarville) marshes in 1996. * was not collected in the
samples.

Total number of Oligochaeta and percent of the total Oligochaeta
represented by taxa in the plant associated community at Mackinac
(reference) marsh, Lake Huron, 1996.

Total number of Oligochaeta and percent of the total Oligochaeta
represented by taxa in the plant associated community of Cedarville
(impacted) marsh, Lake Huron, 1996.

Total number of Oligochaeta per m2 and percent of the total
Oligochaeta represented by taxa in the sediment in Mackinac
(reference) marsh, Lake Huron, 1996.

Total number of Oligochaeta per m2 and percent of the total
Oligochaeta represented by taxa in the sediment in Cedarville
(impacted) marsh, Lake Huron, 1996.

Relative abundance of Zooplankton collected in dip-net and core
samples, identified near the source of discharge, at Mackinac
(reference) and Cedarville (impacted) marshes, Lake Huron,
September 1996.

Definitions of potential metrics and expected direction of metric
response to increasing perturbation in Northern Lake Huron coastal
marshes (selected and modified fi'om Karr and Kerans 1992).

Relationships between macroinvertebrate functional groups and
ecosystem attributes for which they can serve as analog (modified
after Merritt et a]. 1996).

Potential metrics and their sensitivity values base on Barbour et a1.
1996, Coefficient of variation based on Mackinac values; low =
(CV<.50), high= (CV>.50). From Cedarville and Mackinac
marshes, August 1996. Metrics underlined indicate candidate
metrics.

viii

50

54

57

57

58

58

69

79

83

89

Table 20.

Table 21.
Table A-1.

Table A-2.

Table A-3.

List of Tables (cont'd)

Potential metrics and their sensitivity values base on Barbour et a1.
1996, Coefiicient of variation based on Mackinac values; low =
(CV<.50), high=(CV>.50). From Cedarville and Mackinac
marshes, September 1996. Metrics underlined indicate candidate
metrics.

Candidate metrics for use in Northern Lake Huron coastal marshes.

Operational taxonomic unity and identification references for
invertebrates collected from Cedarville and Mackinac marsh.

Chironomidae present in the plant associated Dip-net samples at
Mackinac (reference) and Cedarville (impacted) marsh, Lake
Huron, June 1996.

Water quality data for Cedarville and Mackinac marsh, Huron,
1996. DFD = distance from discharge in meters" Instrument
failure or samples were lost.

ix

90

95
107

109

110

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.

Figure 8.

Figure 9.

LIST OF FIGURES

Location of 'study sites in Northern Lake Huron Bioreserve,
Cedarville and Mackinac marshes, Lake Huron, Michigan.

Chloride concentration gradient moving away from the sourceof
discharge at Cedarville (impacted) and Mackinac (reference)
marshes, 1996.

Percent saturation of dissolved oxygen moving away from
the source of discharge at Cedarville (impacted) and Mackinac
(reference) marshes, September 1996.

Concentration gradient of soluble reactive phosphorus (SRP)
and inorganic nitrogen at Cedarville (impacted) marsh, September
1996.

Concentration gradient of NH4-N moving away form the source
of discharge at Cedarville (impacted) and Mackinac (reference)
marshes, September 1996.

Concentration gradient of NO3-N at Cedarville (impacted) and
Mackinac (reference) marshes, September 1996.

Community composition of the macroinvertebrate community at
Cedarville (impacted) and Mackinac (reference) marshes, Lake
Huron, 1996. A) Plant associated community (dip-net samples)
and B) Sediment associated community (core samples).

Abundance trends, including standard error, of total
macroinvertebrates and insects at Cedarville (impacted) and
Mackinac (reference) marshes, Lake Huron, 1996. A) Plant
associated community (dip-net samples) and B) Sediment
associated community (core samples). *Mann-Whitney U test:

Significant at 0< = 0.05.

Community composition of the aquatic insect orders at Cedarville
(impacted) and Mackinac (reference) marshes, Lake Huron,
1996. A) Plant associated community (dip-net samples) and B)
Sediment associated community (core samples) *Diptera group
does not included Chironomidae.

l7

17

18

18

23

24

30

Figure 10.

Figure l 1.

Figure 12.

Figure 13.

Figure 14.

Figure 15.

Figure 16.

Figure 17.

List of Figures (cont'd)

Abundance trends, including standard error, of the Chrionomidae
in the plant associated community (dip-net samples) of Cedarville
(impacted) and Mackinac (reference) marshes, Lake Huron,

1996. *Mann-Whitney U test: Significant a 0< = 0.05.

Abundance trends, including standard error, of the Chironomidae
in the sediment community (core samples)of Cedarville
(impacted) and Mackinac (reference) marshes, Lake Huron,

1996. *Mann—Whitney U test: Significant at °< = 0.05.

Abundance trends, including standard error, of Ephemeroptera in
the plant associated community of Mackinac (reference) and
Cedarville (reference) marshes, Lake Huron, 1996. *Mann-

Whitney U test: Significant at 0< = 0.05.

Abundance trends, including standard error, of Ephemeroptera in
the sediment associated community of Mackinac (reference) and
Cedarville (reference) marshes, Lake Huron, 1996. *Mann-

Whitney U test: Significant at °< = 0.05.

Abundance trends, including standard error, of Trichoptera in the
plant associated community of Mackinac (reference) and
Cedarville (impacted) marshes, lake Huron, 1996. *Mann-

Whitney U test: Significant at 0< = 0.05.

Abundance trends, including standard error, of Trichoptera in the
sediment associated community of Mackinac (reference) and
Cedarville (impacted) marshes, Lake Huron, 1996. *Mann-

Whitney U test: Significant at °< = 0.05.

Abundance trends including standard error of the dominant
Trichoptera species in the plant associated community of
Cedarville (impacted) and Mackinac (reference) marshes, Lake

Huron 1996. *Mann-Whitney U test: Significant at °< = 0.05.

Abundance trends including standard error of the dominant
Trichoptera species in the sediment associated community of
Cedarville (impacted) and Mackinac (reference) marshes, Lake

Huron 1996. *Mann-Whitney U test: Significant at °< = 0.05.

32

32

40

40

43

43

45

46

Figure 18.

Figure 19.

Figure 20.

Figure 21.

Figure 22.

Figure 23.

Figure 24.

Figure 25.

List of Figures (cont'd)

Abundance trends, including standard error, of Odonata in the
plant associated community of Mackinac (reference) and
Cedarville (impacted) marshes, Lake Huron, 1996. *Mann-

Whitney U test: Significant at °< = 0.05.

Abundance trends, including standard error, of Odonata in the
sediment associated community of Mackinac (reference) and
Cedarville (impacted) marshes, Lake Huron, 1996. *Mann-

Whitney U test: Significant at °< = 0.05.

Abundance trends, including standard error, of Oligochaeta in the
plant associated community of Mackinac (reference) and
Cedarville (impacted) marsh, Lake Huron, 1996. *Mann-

Whitney U test: Significant at °< = 0.05.

Abundance trends, including standard error, of Oligochaeta in the
sediment community of Mackinac (reference) and Cedarville
(impacted) marsh, Lake Huron, 1996. *Mann-Whitney U test:

Significant at °< = 0.05.

Abundance trends, including standard error, of Gastropoda in the
plant associated community of Cedarville (impacted) and
Mackinac (reference) marshes, Lake Huron, 1996. *Mann-

Whitney U test: Significant at 0< = 0.05.

Abundance trends, including standard error, of Gastropoda in the
sediment associated community of Cedarville (impacted) and
Mackinac (reference) marshes, Lake Huron, 1996. *Mann-

Whitney U test: Significant at °< = 0.05.

Community composition of the Gastropoda community based on
abundance at Cedarville (impacted) and Mackinac (reference)
marshes, Lake Huron, 1996. A) Plant associated community and
B) Sediment associated community.

Abundance trends, including standard error, of Amphipoda in the
plant associated community of Cedarville (impacted) and
Mackinac (reference) marshes, Lake Huron, 1996. *Mann-

Whitney U test: Significant at 0< = 0.05.

xii

49

49

59

59

61

61

62

Figure 26.

Figure 27.

Figure 28.

Figure 29.

Figure 30.

Figure 31.

List of Figures (cont'd)

Abundance trends, including standard error, of Amphipoda in the
sediment associated community of Cedarville (impacted) and
Mackinac (reference) marshes, Lake Huron, 1996. *Mann-

Whitney U test: Significant at (X = 0.05.

Abundance trends, including standard error, of Isopoda in the
plant associated community of Cedarville (impacted) and
Mackinac (reference) marshes, Lake Huron, 1996. *Mann-

Whitney U test: Significant at 0< = 0.05.

Abundance trends, including standard error, of Isopoda in the
sediment associated community of Cedarville (impacted) and
Mackinac (reference) marshes, Lake Huron, 1996. *Mann-

Whitney U test: Significant at 0< = 0.05.

Evaluation of sensitivity of the metrics. Range bars show
maximum and minimum of non-outliers; Solid lines inside the box
are medians; boxes are interquartile ranges (25%ile to 75%ile),
(Taken fiom Barbour et a1. 1996).

A comparison of candidate metrics of long term impact, for use
in Northern Lake Huron Coastal marshes. Metrics determined
from Dip net samples at Cedarville (impacted) and Mackinac
(reference) marshes, Lake Huron, September 1996. Range bars
show maximum and minimum of non-outliers; Solid lines inside
the box are medians; boxes are interquartile ranges (25%ile to
75%ile); dots are outliers. (Taken fi'om Barbour et al. 1996).

*Mann-Whitney U test: Significant at 0< = 0.05.

A comparison of candidate metrics of short term impacts, for use
in Northern Lake Huron Coastal marshes. Metrics determined
from Dip net samples at Cedarville (impacted) and Mackinac
(reference) marshes, Lake Huron, 1996. Range bars show
maximum and minimum of non-out-liers; Solid lines inside the
box are medians; boxes are interquartile ranges (25%ile to
75%ile); dots are outliers. (Taken from Barbour et al. 1996).

*Mann-Whitney U test: Significant at 0< = 0.05.

xiii

64

67

67

87

92

93

CHAPTER ONE

A COMPARISON OF TWO NORTHERN LAKE HURON COASTAL MARSHES

Introduction

The Great Lakes coastal marshes are an integral part of the Great Lakes ecosystem.
They are important as fish spawning grounds (Jude and Pappas 1992) and waterfowl breeding
grounds (Prince and Flegel 1995). Not only do the coastal marshes provide a diverse habitat
for animals and plants but they also act as a sieve and a trap for allochthonous and
autochthonous materials (Gaudet 1974, Wetzel and Allen 1972). Coastal marshes have the
potential of contributing to the total production of the lake ecosystem and regulate, at least
in part, the metabolism of many lakes (Jude and Pappas 1992, Howard-Williams and Lenton
1975, Wetzel and Allen 1972). The Great Lakes contain about 200 species of fish with most
of these species using the coastal marshes during some part of their lives (Whillans 1990, Jude
and Pappas 1992, Brady 1992). Despite the apparent importance of the macroinvertebrates,
comparatively little is known of their distribution and ecology in Great Lakes coastal marshes
or in freshwater marshes in general (Krieger 1992). Historical data reveals that, since the
mid-1800's, Michigan has lost approximately seventy percent of its original coastal marshes
(Prince and Flegel 1995, Jaworski and Raphael 1978). Encroachment upon these ecosystems
through human development has resulted in a subsequent increase of human impact upon

them (Krieger 1992). The need to protect and monitor the health and productivity of these

2

coastal marshes is of great importance in management and recreational planning.
Traditional approaches for collecting water quality data have been based on chemical
monitoring (e. g. APHA 1985). While chemical monitoring of water quality is important,
there are several inherent problems associated with its use. A number of forms of degradation
imposed on aquatic systems are not firndamentally chemical but are instead physical, such as
habitat disturbance. The results of chemical monitoring may also overlook significant
discharge loads that occur between periods of data collection (Rarnm 1988). Rankin et a1.
(1990) found that water chemistry data failed to detect 50% of the impairment in Ohio surface
waters that was detected with integrated biological and chemical monitoring. Karr (1993)
defined biological monitoring as "the use of a biological entity as a detector and its response
as a measure to determine environmental conditions." The use of natural benthic
macroinvertebrate assemblages is one of the best understood and most economical water '
quality monitoring systems, and it can be used to complement chemical monitoring of water
quality (e.g. Platkin et a1. 1989, Rosenberg and Resh 1993, Karr 1993). A major advantage
of this system is that, because benthic macroinvertebrates are continuously exposed and
influenced by the environment, they continuously "monitor" water quality and can reveal
efl‘ects of episodic as well as cumulative pollution and habitat alteration (Olive et a1. 1988,
Plafldn et al. 1989, Barbour et a1. 1996, Fore et a1. 1993). Benthic macroinvertebrates may
reflect long-term water quality conditions and serve as an integrated measure of discharge
effects through changes in community structure and composition caused by these effects
(Ramrn 1988). Macroinvertebrates have proven to be good biomonitoring tools because they
are numerous in almost every aquatic system, are readily collected and identified, and are not

very mobile (Chandler 1970, Gaufin 1973, Roback 1974, Hilsenhoff 1982, Rosenberg and

3

Resh 1993). Macroinvertebrates generally have life cycles of weeks to years in duration, and
this is important in assessing past perturbations of short duration; once a macroinvertebrate
is eliminated from the ecosystem it will not reappear until the next generation (Hilsenhoff
1977)

The use of biological monitoring to evaluate water quality in streams has a long
history beginning with the work of Kolkwitz and Marsson (1908) and Forbes (1913). These
early biological monitoring programs emphasized an organism's tolerance of organic pollution
(Kolkwitz and Marsson 1908, Chutter 1972, Hilsenhofi‘ 1987). Advances in community and
population ecology led to the use of diversity indices (Wilhm and Dorris 1968, Hughes and
Gammon 1987) which acted as measures of species richness and evenness (review in Fausch
et al. 1990). These early biomonitoring ideas began when investigators discovered a number
of biological patterns associated with increased human disturbance within a watershed. Such
patterns included the decline of a number of species (Metcalfe 1989), the disappearance of
a small group of intolerant species (Hilsenhofl‘ 1977), and a decline of trophic specialists while
trophic generalist increase (Metcalfe 1989). It is now widely accepted that pollution of a
stream reduces the number of species in the stream while frequently creating an environment
that is favorable to a few species (Hilsenhoff 1977). Total taxa richness and taxa richness of
intolerant groups such as mayflies, caddisflies, and stoneflies often decline as human
influences increase (Lenat 1988, Ohio EPA 1987, Plaflcin et al. 1989). In contrast, relative
abundance and dominance of Chironomidae and Oligochaeta ofien increase with increasing
human influences (Lenat 1988, Ohio EPA 1987, Ford 1989, Barbour et al. 1996).

In the last decade, biological monitoring and indices have been increasingly used to

interpret how similar an assemblage at a site is to the potential assemblage that would have

4

occurred there if the site were undisturbed (Karr 1991, Gerritsen 1995, Rosenberg and Resh
1993). This interpretation relies on use of regional, reference data for relatively undisturbed
sites. The indices currently used are variations of the Index of Biotic Integrity (IBI)
developed by Karr and his colleagues (e. g., Karr 1981, Karr et al. 1996) for fish communities
of streams. The concept was extended to the benthic macroinvertebrate assemblages of
streams by the Ohio Environmental Protection Agency (Ohio EPA 1987), and by Plaflcin et
al. (1989) and Kerans and Karr (1994). Elements of community structure and composition
are the most widely used assessments of stream biological conditions because community
structure is a result of both long-term environmental factors and critical conditions of short
duration (e.g., Hilsenhofl‘ 1977; Karr et al. 1986, Lenat 1988, 1993; Plaflcin et al. 1989).
These monitoring programs and indices are currently used by several states for assessment
and management of streams (Southerland and Stribling 1995, Ohio EPA 1987, Kerans and
Karr 1994, Barbour et al. 1996, Gerritsen 1995).

In comparison to the large body of literature that exists on the role of freshwater
macroinvertebrates in stream communities as they apply to their use as water quality
indicators (Karr 1991), the macroinvertebrate fauna of Great Lakes coastal marshes has
received only limited study (Krieger 1992). As in stream systems, the macroinvertebrate
community structure in coastal marshes may provide a sensitive index to pollution inputs. In
order to monitor the health and productivity of these coastal marshes, it is necessary to first
obtain descriptive data on the distribution and life history of the macroinvertebrate community
and to examine the response of these communities to pollution.

Objectives

The overall goal of this study was to document anthropogenic impacts on the

5

macroinvertebrate fauna of a Great Lakes coastal marsh by comparing it to a nearby reference
marsh which appeared to have been less impacted by human activities. Specific objectives
were to:

1) Compare the diversity and community composition of the macroinvertebrates in

a "reference" Great Lakes coastal marsh with diversity and community composition
of a nearby human impacted marsh.

2) Determine which macroinvertebrates in these coastal marsh were most susceptible
to pollution from a small community such as Cedarville, Michigan.

3) Select and test various metrics for the use in the firture development of an multimetric
index of ecological integrity for use in northern Lake Huron coastal marshes.
Description of Study Sites

This study was conducted in two protected bays along the northern Lake Huron
shoreline at the southeastern shore of Michigan's Upper Peninsula (Figure 1). The study sites
were selected based on the premise that test sites which have similar characteristics in the
absence of water quality impairments would be expected to yield equivalent
macroinvertebrate communities (Barbour 1992). The two sites selected were coastal, or
lacustrine, marshes, that were less than 2 km apart and were part of the Les Cheneaux Islands
marsh complex (Figure 1).

A small coastal marsh immediately adjacent to the town of Cedarville, Michigan at the
northwestern end of Cedarville Bay (45° 59' N latitude, 85° 21‘ W longitude), was selected
as the impacted site. This marsh typically receives discharge from the Cedarville domestic
wastewater treatment lagoon system via Pearson Creek, two times per year in the spring and

the fall. Discharge occurred five times during 1996 due to heavy rainfall causing the

Study Location

Area in detail

 

 

 

 

 

\
\ 65%
’1
\
\
\
\
\
\
\
\
\
\
\
\
\
\
w E \

 

M-1 34

 

 
      

Cedarville Bay

    
    

La Salie
Island

 
     

  

Marquette Island

Sheppards
Bay

    
   
 

Voight Bay

A 2 Km
pprox. <—>

 

Lake Huron

 

 

Figure 1. Location of study site in Northern Lake Huron Bioreserve, Cedarville and
Mackinac marshes, Lake Huron, Michigan.

 

7

wastewater treatment ponds to fill with excess water. Partial discharge occurred May 21-25
and the remainder of the treatment ponds were discharged June 2—5. Fall discharge occurred
with an initial partial release September 26-30 and again October 18-24. The final discharge
for 1996 occurred November 14-18. The wastewater treatment lagoons consist of three
treatment ponds one of which is aerated to promote decomposition and as an aid in settling
particles out of solution. Sewage effluent is chemically injected with 3,000 gallons of FeCl2
every six months. FeCl2 was initially added as a means of odor control, however it also
reduces phosphorus output from the lagoon through chemical precipitation (Landerville pers.
comm).

Other anthropogenic impacts on Cedarville marsh include the partial filling of the wet-
meadow zone and separation of this wet-meadow zone from the deeper emergent marsh,
urban storrnwater runoff from Cedarville, MI, and impacts associated with marina
maintenance and traffic. A portion of the wet meadow zone along Pearson creek was filled
beginning in 1911 with the construction of a road separating the wet meadow zone from the
deeper portion of the marsh. Filling continued with the addition of a lumber retail store in
1939. The upper end of Cedarville Bay also receives storm-water runoff from Cedarville,
Michigan (Population 2000). In addition, a small marina supporting 57 boat slips and a
heavily used, public boating access ramp was located within 100 meters of the study sites.

A similar sized marsh at the northwestern end of Mackinac Bay (46° 00' N latitude,
85° 25' W longitude) was selected as the reference site. This coastal marsh was
geomorphically similar to Cedarville marsh but received no wastewater discharge, was not
immediately adjacent to a town and had no marina or public-access boating ramp near the site.

There were several houses around this bay but none were closer than 1 km of the sampling

sites.

These two marshes had similar geomorphological characteristics within the marshes
in terms of their location in relation to stream discharge, size and shape of their watersheds
and underlying geology. The two marshes were located on the Niagara escarpment, which
consisted primarily of resistant limestone and dolomite (Albert et a1. 1986). The dominant
land use in both was forest with a mixture of northern hardwood forest and dense stands of
northern white-cedar, balsam fir and spruce along the lakes (Albert et al. 1986, personal
observation). The average growing season for this area is 125 days long, with the growing
season heat sum relatively low at 1860° C days (Albert et a1. 1986). The catchments of both
sites were drained by small first order streams, which were relatively similar in length, width,
and area drained.

Within the marsh areas of both bays, the study sites were located in the emergent plant
zones dominated primarily by Scirpus acutus. Macrophytes grew from late May through July,
and started senescing in August. Dead stems were partially scoured from the marsh by ice
during the winter. Following ice-out, the emergent stand re-grew fiom rhizomes. The
emergent zone at the impacted marsh extended approximately 200-3 00 meters from the shore
at the road into Cedarville Bay. The emergent zone at the reference marsh extended for
several hundred meters from the wet meadow zone out into Mackinac Bay. The density of
S. acutus was fairly consistent between the two marshes, with an overall summer average of
72 stems per 0.25 m2 at the reference marsh, and an average of 74 stearns per 0.25 m2 at the
impacted marsh. However, major differences in the plant communities occurred for the
submersed plants in this zone with greater dominance by Ulricularia spp., Myriophyllum

spp., and Ceratophyllum spp. at the impacted site. Water temperatures were usually

9

homogeneous between the two marshes, but when differences occurred, they never exceeded
2°C. Both sites were protected from direct, harsh wave action fiom Lake Huron by a
network of islands.

Substrate at the sampling sites of Mackinac Marsh appeared to be composed
primarily of loose organic matter (33%), with lesser fractions of clay and sand. Cedarville's
marsh substrate at the sampling site appeared to consist primarily of organic matter (70%),
with lesser fraction of clay (Personal observation).

Methods
Land-use/Land-cover Characterization of Sites

Watersheds for Mackinac and Cedarville marsh were delineated, digitized and overlaid
on to 1978 Michigan Resource Information System (MIRIS) Land-use/Land-coverages of
Mackinac county. These coverages were used to determine percent Land-use/Land-cover in
both Mackinac's and Cedarville's watersheds.
Chemical-Physical Data Collection

Environmental data on physical and chemical data were collected at each of the nine
sampling sites in each marsh, on each invertebrate sampling date to characterize the water
within the marsh. Water depth was measured at each site to the nearest 0.5 cm. Vegetation
type and density was determined by counting the number of emergent stems taken at a
random distance and direction from the sampling point in a 0.25 m2 plot. Random direction
(0° to 360° in increments of 10°) and random distance (1 m to 5 m in increments of 0.5 m)
were obtained from a table of randomly generated numbers. Dissolved oxygen (YSI model
51B oxygen meter), pH (Altec monitor 11 meter), and conductivity (YSI model 31

conductivity bridge) were determined in situ using appropriate, calibrated probes. Water and

10

air temperatures were measured just above the sediment surface to the nearest : 01°C using
a hand-held mercury thermometer.

For each date at each site, water samples were collected at one-half the depth of the
water column in acid washed, deionized water rinsed opaque plastic bottles. All samples were
immediately placed on ice and transported to the laboratory within three hours. Alkalinity
(APHA 1985) and turbidity (HACH model 2100A turbidirneter) were determined in the field
laboratory. Sub-samples were filtered through 0.45am millipore filters, frozen at ~20° C, and
later sent to Michigan State University's Soils Testing Laboratory for analysis of Cl, NH4-N,
NOz-N, NO3-N and SRP. All samples were tested at a level of detection of 0.01 mg/L except
CL which was tested at a level of detection of 1.0 mg/L. Results were then used to examine
differences in water quality between the two sites, and to examine if a gradient in littoral
water quality existed as a factor of distance from discharge.

Invertebrate Sampling

A fixed transect was established within the Scirpus acutus dominated plant
communities of each marsh. The transects ran parallel to the shoreline 55 meters into the
emergent plant community in each marsh, beginning 10 meters west of the mouths of the
streams. Sampling sites were located every 50 meters for a total of nine stations (within each
marsh) resulting in the last station being 410 meters west of the stream mouths. This design
was used to minimize habitat variability for such factors as depth, water temperature, and
vegetation density, while at the same time establishing a gradient moving away from the
source of discharge. All transects were placed within a predominantly monotypic S. acutus
zone to minimize the possible confounding effects of varying vegetation types.

Macroinvertebrates were sampled monthly fi'om June through September in 1996

11

beginning two weeks after S. acutus stems extended above the water surface. Sampling
occurred on June 24-25, July 24-25, August 18-19 and September 28-29. On each date,
macroinvertebrates were collected using a sediment sample collected with a coring device and
a water column/vegetation sample taken with a standard number of sweeps with a D-framed
dip net. The sediment and plant associated samples were taken at a random distance and
direction from each of the nine fixed sampling station points along the transect. For each
sample taken at each point along the transect, a random direction (0° to 360° in increments
of 10°) and a random distance (1 m to 5 m in increments of 0.5 m) were obtained fi'om a table
of randomly generated numbers. Care was taken never to sample the same direction twice
from any particular fixed point, nor was any sample taken within one meter of the transect
itself.

Vegetation/water column samples consisted of two sets of standardized triplicate
sweeps with 0.3-m wide D-frame dip nets. The first set of sweeps were made with a l-mm
mesh net which was used to prevent net clogging so more active animals could be captured.
The second set of sweeps was made with a 250 um mesh net in order to capture smaller
macroinvertebrates. Samples consisted of 3 sweeps at the surface of the water column, 3
sweeps in the center of the water column and 3 sweeps along the sediment surface at the
bottom of the water column. Each sweep covered 0.15 m2 of substrate (net width of 0.3 m
x 0.5 m length of pass); therefore, the total composite sample was taken fi'om an area of
approximately 0.90 m2.

In the field, all samples were washed through a 250nm sieve, and then transferred to
1 liter large mouth bottles. The macroinvertebrates along with all debris left in the sieve after

washing, were preserved in 95% ethanol containing 100 mg.l" of Rose Bengal dye with

12

enough 95% ethanol added to the jar to produce a concentration of about 70% when mixed
with the water in the debris (Mason and Yevich 1967). The dye stained the protein in the
sample red which helped to increase the accuracy of picking individual specimens from the
debris in the laboratory. Afier retuming to the laboratory, sample jars completely filled with
macroinvertebrates and debris were drained and rinsed again through a 250um screen. The
ethanol was then replaced with 70% ethanol. Invertebrates were picked fi'om the debris and
sorted with the use of a 10x dissecting microscope.

Organisms were classified to the lowest Operational Taxonomic Unit (OTU'S)
possible throughout the study. The OTU's and the references used for identification, are
shown in Table A-1. Every effort was made to sort invertebrates to species level, but some
taxa that were taxonomically dificult (e. g. Chironomidae) or that were not properly preserved
(e.g. Nematoda) were grouped at higher taxonomic levels. Chironomidae larvae were
identified to genus or species group for the June water column samples at the site nearest the
source of discharge in both the impacted and reference wetland (Table A-2), otherwise they
were sorted to sub-family or tribe. Zooplankton were identified to the species level at the site
nearest the source of discharge in both the sediment and water column samples of the
impacted marsh and the reference marsh in June and September. Specimens were also sent
to taxonornists whenever possible for verification or identification of species present as a
means of most accurately reflecting biodiversity for each of the marsh areas.

Benthic invertebrates were sampled using a corer made of a 1.22 m long, 5 cm
(internal diameter) Plexiglas tube capped with a 4.5 cm rubber stopper prior to pulling the
corer out of the sediment. Each core included 15 cm of sediment plus the water column

above it. Two cores comprised each sediment sample. No attempt was made to separate

l3

macroinvertebrates in the water column from those residing in the sediments. All benthic
samples were preserved with the same method used for vegetation/water column samples.
In the laboratory, each sediment sample was divided into sixths before it was picked, using
a sub-sampling device similar to that described by Waters (1969). To insure that an adequate
number of organisms was obtained to characterize the sample, sub-samples were picked until
at least 50 organisms were found, when possible. Once picking of a sub-sample was started,
it was sorted completely.

The numbers of the invertebrates in each sediment sample were converted to number

per m2 based on the following formula:

Area: 1tr7=1t(0.025m)2 x 2 cores per sample

=3.93 x 10’3 m2

Macroinvertebrates were classified into biotic categories describing trophic status
(omrrivore, detritivore, herbivore, carnivore, scavenger), functional-feeding group, and habitat
using information from Merritt and Cummins (1996) to compare macroinvertebrate
communities between sites.

Macroinvertebrate communities that demonstrated a response to impact were then
selected for detemiination of whether a gradient existed within the impacted marsh with
distance from discharge. This was accomplished by plotting the selected taxa with increasing
distance from discharge with a linear regression. If no gradient was detected, the selected
taxa from the three sites closest to discharge were pooled, and compared to the same taxa

pooled from the three sites furthest fiom discharge, using a nonparametric Mann-Whitney U

14

test for two independent samples, to determine if a difference was present.
RESULTS
Comparisons of Land-use/Land-cover in watersheds

The reference and impacted watersheds were both dominated by forest cover (Table
1). Although land-use was primarily forested for both the watersheds, Cedarville Bay's
(impacted) watershed had a higher percentage of urban and agricultural land than did
Mackinac Bay's (reference) watershed, and lower percentage of hardwood forest and

wetlands (Table 1).

Table 1. Total area and Percent Land-use/Land-cover in Cedarville and Mackinac bay's
watershed (NflRIS 1978).

 

MACKINAC CEDARVILLE
Percent
Coniferous forest 39 41
Hardwood forest 39 34
Urban 5 12
Agricultural 1 4
Wetland 7 4
Other 9 5
Total area of watershed 1.31 km2 1.70 km2

 

Comparisons of Water Quality Data

Greatest difi‘erences in water quality between the impacted marsh (Cedarville) and the

15

reference marsh (Mackinac) occurred during times of active discharge via Pearson creek from
the wastewater treatment lagoon in September (Table A—3). The impacted marsh was
characterized by higher levels of Cl, NIL-N, NO3-N, NOz-N, SRP, conductivity, turbidity,
alkalinity and lower levels of dissolved oxygen when compared to the reference marsh during
June and September but not during July and August (Table A-3). These differences in water
quality during the times of active discharge from the wastewater treatment lagoon into the
creek in June and September were greatest near the mouth of Pearson creek , and decreased
to levels similar to those in the reference marsh between 210 and 260 meters from the creek
mouth (Figures 2-6). The relatively limited area of the marsh with detectable impacts was
also documented in detail by Grant (1994). The well-mixed nature of the marsh, may
contribute to a decreasing impact with distance from discharge. June samples indicated very
low chemical impact at the time of sampling, approximately two weeks after the actual
discharge event, suggesting that the marsh had returned to levels near background within two
weeks after the sewage discharge ceased.

Elevated Cl concentrations occurred during or immediately following discharge fi'om
the wastewater treatment lagoon at the impacted marsh in June and September (Figure 2).
These elevated Cl concentrations decreased to mean background levels at 260 m from the
mouth of the creek, indicating rapid dilution and mixing within the marsh water. However,
Cl concentrations were similar in the two marshes on other sampling dates, with
concentrations ranging between 12-21 mg Cl/l (Table A-3). Cl concentrations in natural
freshwater systems average 8.3 mg Cl/l (Wetzel 1983). The slightly elevated Cl levels of 12-

21 mg C1/l may indicate presence of evaporite deposits within the underlying bedrock of this

 

 

 

 

 

 

 

 

June
804
60‘
404
a
20«
' .0 e :13 e e . e
a Dan 0
0 I a r r
S
U
m
E
August
801
60-
40-
20" D an no [I D
e U 0 . 9 ' e e e
o I I I I
0 60 160 260 360
>

 

Distance from discharge in meters

D Impacted marsh

16

July

 

 

 

 

September

 

 

 

 

I I I I

0 60 160 260 360

)
Distance from discharge in meters

 

' Reference marsh

Figure 2. Chloride concentration gradient moving away from the source of
discharge at Cedarville (impacted) and Mackinac (reference) marshes, 1996.

17

 

 

 

 

 

a 80 _
i‘ r: r:
O :r o I:
'3 70 T t: o
z 0
o
3 so
b - . . o
‘5 . o .
E
g :r
s D
to 40 _
3‘ :r

0 10 60 110 160 210 260 310 360 410

>
Distance from discharge in meters
D impacted marsh . Reference marsh

Figure 3. Percent saturation of dissolved oxygen moving away from the source of
discharge at Cedarville (impacted) and Mackinac (reference) marshes, September 1996.

 

 

 

 

 

1.5 - D
_ 1.0 _
a
E
:1
1:]
0.5 _
1:1
0 1:1
. o
0'0 I r r f g Q Q ? T
0 10 60 110 160 210 260 310 360 410
P
Distance from discharge in meters
9 SRP c] inorganic nitrogen

Figure 4. Concentration gradient of soluble reactive phosphorous (SRP) and inorganic
nitrogen at Cedarville (impacted) marsh, September 1996.

18

 

 

 

 

 

0.5 r:
:r
0.4 - [:1
§ 0.3 ‘
f 0
Z 02 _ :1
a . r:
E
.
0.1 ‘ 0
D a :1 D o
o . o o
0.0 ‘
I I I I
0 60 160 260 360
’
Distance from discharge in meters
D Impacted marsh . Reference marsh

Figure 5. Concentration gradient of NH4-N moving away form the source of discharge at
Cedarville (impacted) and Mackinac (reference) marshes, September 1996.

 

 

 

 

 

0.4
D
0.3 " I:
D
E
u“ 0.2 '-
o D
Z
a
E 0.1 -
O D O O . O
0.0 -
-0.1 T I I I
0 60 160 260 360
>
Distance from discharge in meters
D impacted marsh 0 Reference marsh

Figure 6. Concentration gradient of NO3-N at Cedarville (impacted) and Mackinac
(reference) marshes, September 1996.

19

catchment. Road salt from local roads seems unlikely to be the cause of the elevated Cl levels
because impacts fiom road salt would be expected to decline over the course of the summer
but did not in this study (Figure 2). Cl concentrations remained relatively constant in
concentration and distribution within the reference marsh throughout the sampling season
(Figure 2).

Turbidity levels were low at both the impacted and reference marsh never exceeding
5 NTU‘s (Table A-3). The highest turbidity levels at the impacted marsh occurred during June
and September, perhaps indicating slight elevations in turbidity associated with wastewater
discharge, but trends were inconclusive for other sampling dates. June turbidity levels at the
impacted marsh were highest at the sites nearest discharge and decreased with distance from
discharge. Turbidity at neither site was ever high enough to cause major differences in biota
(EPA 1986), especially given the variance from <1 to 5 NTU's for both marshes.

Dissolved oxygen ranged from 61% saturation to supersaturated values up to 114%
saturation from June through August, at the impacted and the reference marsh with no
consistent differences between the sites (Table A-3). Dissolved oxygen concentrations were
substantially below saturation during September for both the reference and impacted marshes.
Plant senescence leading to a decrease in primary productivity and an increase in respiration
is the most likely explanation for the undersaturation at both sites (Table A-3). Dissolved
oxygen saturation was significantly lower at the impacted site compared to the reference site
during wastewater discharge in September. Dissolved oxygen levels in the impacted marsh
fell to levels below 5 mg/l at the sites less than 210 meters from discharge (Figure 3). The

U. S. Environmental Protection Agency (1986) suggested dissolved oxygen levels below 5

20

mg/l are incompatible with maintenance of diverse aerobic aquatic communities. The low
dissolved oxygen levels may be an important factor for many species of aquatic life in the
areas near the source of discharge at the impacted marsh. If dissolved oxygen levels dropped
to low levels following the June discharge event, they had recovered by the time of sampling
two weeks after discharge had ceased (Table A-3). Low dissolved oxygen levels may have
not occurred in June due to discharge coinciding with periods of high primary productivity.
Dissolved oxygen was monitored during mid-day only, so periods of low dissolved oxygen
that may have occurred at night would not have been detected.

During wastewater discharge in September, the two limiting nutrients, Soluble
reactive phosphorus (SRP) and inorganic N, were elevated for only 160 meters from the
source of discharge along the transect (Figure 4). SRP was always below limits of detection
(0.01 mg P/l) in the reference marsh. SRP was at or less than limits of detection in June, July
and August at the impacted marsh (Table A-3). From September 26 to 30, the Clark
township wastewater treatment lagoon discharged sewage eflluent containing a daily average
of 0.96 mg PA of total phosphorus approximately 2.39 km upstream from the impacted marsh.
SRP was detected at a maximum concentration of 0.21 mg P/l nearest the source of discharge
on September 29, 1996 and returned to non-detectable concentrations at 210 m from
discharge (Figure 4). Aqua-Tema Labs performed a water quality survey at the impacted
marsh in 1994 and found that only 64% of the P discharged from the wastewater lagoon
actually entered the marsh (Grant 1994). This suggests that much of the P loss can be
attributed to rapid assimilation by the stream biota, chemical precipitation, and sorption onto

sediment particles. Even so, SRP exceeded 100g PA from the stream mouth to 160 meters

21

along the marsh transect, and levels exceeding long P/l are known to stimulate plant grth
in many lakes (Wetzel 1983).

NIL-N levels generally ranged from 0.03 to 0.07 mgN/l at the reference and impacted
marshes fi'om June through August (Table A-3). The highest concentration of NIL-N
occurred in September at the impacted marsh, near the source of discharge, reaching
concentrations of 0.48 mg NI-L-N/l (Figure 5). NIL-N returned to background levels of 0.06
mg NH4-N/l at 210 meters from discharge at the impacted marsh in September (Figure 5).
NH4-N concentrations appeared to be relatively constant in both concentration and
distribution at the reference marsh (Table A—3).

NO3-N was only detected at the impacted marsh during June and September and was
below the limits of detection of 0.01 mg NO3-N/l in July and August (Table A-3). June N03-
N concentrations were low, at levels no greater than 0.03 mg NO3-N/l. However, during
September NOs-N was detected at levels as high as 0.32 mg NO3-N/l near the source of
discharge and decreased in concentration with distance from discharge until no longer
detected at distances greater than 210 meters from discharge (Figure 6). NO3-N was always
below the 0.01 mg NO3-N/l limit of detection in the reference marsh expect during
September. Concentrations in September were never greater than 0.04 mg NO3-N/l, or an
order of magnitude lower than values detected near the mouth of the stream during
wastewater discharge.

NOz-N was only detected (0.01 mg NOz-N/l = limit of detection) at the impacted
marsh during September, at the three sites nearest the source of discharge at levels no greater

than 0.03 mg NOz-N/l (Table A-3). NOz-N was never detected at the reference marsh.

22

Both marshes tended to be characterized by pH values from 7.0 to 9.5 and usually
in the 8.0 to 9.0 range. There were no consistent trends in pH at the impacted marsh even
during wastewater discharge in September, except for somewhat more variable pH values
(Table A-3). Conductivity and alkalinity were generally lower at the impacted marsh
compared to the reference marsh, except in September when wastewater discharge was
actively occurring at the impacted marsh (September) (Table A-3).

Comparison of Macroinvertebrate Community Dynamics

The plant and sediment associated macroinvertebrate communities within the Scirpus
acutus zone of the reference marsh were dominated numerically by aquatic insects for all
sampling periods (Figure 7). Aquatic insects were less dominant at the impacted site than
they were at the reference site for both the sediment and plant associated communities (Figure
7), and this was particularly true for the sediment community (Figure 7B). Sediment
associated aquatic insects represented 50 to 80% of the macroinvertebrate abundance of the
reference marsh, but only represented 20 to 46% of the sediment community at the impacted
marsh (Figure 7B). Likewise, aquatic insects represented 50 to 63% of the macroinvertebrate
community of the plant associated community of the reference marsh, and only 28 to 50% of
the plant associated community of the impacted marsh (Figure 7A). .

Macroinvertebrates of both the plant and sediment associated communities tended
to reach peak abundances in September with numbers tending to increase as the season
progressed from June through September (Figure 8). Macroinvertebrates reached their peak
mean density of 35,000-m2 in the sediment of the reference marsh during September (Figure

8B), approximately 80% of which were aquatic insects (Figure 7B). The macroinvertebrate

23

A)
Plant Associated Community

1 00%

80%

60%

40%

20%

 

 

3 0%
5 June July Aug. Sept. June July Aug. Sept.
2 Reference marsh Impacted marsh
3
m
E 8
g ) Sediment Associated Community
0. I
100% I
i
80% :
I
I
60% I
I
I
40% :
I
20% i
i
l
0% June July Aug. Sept. June July Aug. Sept.
Reference marsh impacted marsh
m Nematode - isopoda I!!! Oligochaeta Gastropoda

 

= Aquatic Mites m Amphipoda i=1 Aquatic Insects other

Figure 7. Community composition of the macroinvertebrate community atCedarville
(Impacted) and Mackinac (reference) marshes, Lake Huron, 1996. A) Plant associated
community (dip-net samples) and B) Sediment associated community (core samples).

24

 

 

 

 

 

 

 

 

A) Plant Associated Community
1,600 ' |
0)
a I
E i
(I!
In I
a. . l
3 1,200 * -— I
3
In I
8 I
§ 800 ' I
w |
'0
= |
3
n 1
‘° I
g 400 r l
O
2 I I
I I
0 . ' I I
June July Aug. Sept. June July Aug. Sept.
Total Macroinvertebrates Aquatic Insects
B) Sediment Associated Community
50,000“ I
l
" l
'§ 40.0001 :
3 - I
8 I '
u, 30,000 I
E r I
B I
°' 1 I
r: 20,000
to I
o
2 1
10,0001 |
I
I
0

 

 

 

 

 

   

 

 

June July Aug. Sept.
Total Macroinvertebrates Aquatic Insects
- Impacted marsh [:| Reference marsh

Figure 8. Abundance trends, including standard error, of total macroinvertebrates and insects at
Cedarville (impacted) and Mackinac (reference) marshes, Lake Huron, 1996. A) Plant associated
community (dip-net samples) and B) Sediment associated community (core samples). *Mann-Whitney
U test: Significant at or = 0.05.

25

community in the sediment at the impacted marsh reached a similar peak mean density in
September, of 39,836-m‘2; butonly 35% of these macroinvertebrates were aquatic insects
(Figures 8B and 7B). Non-insect macroinvertebrates were more abundant than were insects
at the impacted site (Figures 7 and 8). Amphipoda, Oligochaeta, Isopoda, Gastropoda,
Nematoda, and Hydracarina were important members of the non-insect, invertebrate fauna.
Each major group of macroinvertebrates will be discussed separately below.

Aquatic Insecta:

Plant associated aquatic insects reached their greatest mean abundance of
approximately 700 individuals per sweep net sample in September at the reference marsh
(Figure 8A). Plant associated aquatic insects at the impacted marsh reached a lower peak
mean abundance of 400 individuals per sample in August but these differences between the
reference and impacted sites were significant (p< 0.05) only during August (Figure 8A).
Aquatic insects reached a peak abundance in September in both the impacted and reference
marsh sediment (Figure 8). Sediment associated aquatic insects at the reference marsh
reached a peak density of 27,000~m‘2 compared with a peak density of 18,660°m‘2 at the
impacted marsh (Figure 8B). These difi‘erences were not statistically significant (p< 0.05) for
any sampling date (Figure 8B). .

Aquatic insects associated with the plant community in the reference marsh were
represented by 73 taxa from 37 families (Table 2). Thirty-four taxa of insects, representing
17 families, were collected from the sediment core samples in the reference marsh (Table 2).
The Chironomidae, Ephemeroptera, and Trichoptera were dominant insect groups within the

plant and sediment associated communities at the reference marsh (Figure 9), and Odonata

26

Table 2. Aquatic insects collected with standardized dip-net sweep sampling from Scirpus dominated zones
(mean number/sample18E) and with cores from sediments in these zones (mean number-mf18E) from the
reference (Mackinac) and Impacted (Cedarville) marshes in 1996. * was not collected in the samples.

 

 

 

 

Mackinac Cedarville
TAXON ’ ,
l Dip-Net I Core samples I Dip-Net I Core samples
Mean #/sample1$.E. Mean #-m‘_2_18.E. iMean #/sample18.E. Mean #-m'2_+S.E.
COLEOPTERA
Chrysomelidae
Donacr'a app. .141.10 21.211212] .141.10 35.341.21.21
Nehaemom‘a spp. .34120 * .06106 "‘
Curculionidae 031.03 * * *
Dytiscidae
Agabus spp. .03103 * * *
Elmidae
Duiraphr'a app. .06106 "' * *
Gyrinidae
Dineutus app. 081.08 * .03103 "‘
Gyn'nus spp. . 141.08 * 061.03 *
Haliplidae
Halplus spp. 031.03 "‘ .19108
Peltafytes app. "‘ * . 191.05
Ptilidae .03103 * *
COLLEMBOLA
Isotornidae
Isotomurus tricolor * * 031.03 *
Sminthuridae
Entomobrya nivalr's * * * .06106
Pseudobourletiella spinata .03103 * .1 11.08 *
DIPTERA
Ceratopogonidae
Bezzr'a/Palpomyia spp 421.31 1484317243 1471.71 21 .2112] .21
Culicoides spp. .171.13 * * 212112121
Probezzr’a app. .061. 06 "' * *
Serromyia app. * 70717.07 * *
Sphaeromias spp. * 21 .2112121 "' *
Chironomidae"
Chironominae
Chironomini 750813275 4134.861178324 1 183313099 5605001244500
Tanytarsini 389711432 69268126560 27.2511 1.47 657157682
Orthocladiinae
Corynoneura spp. 4081.96 1060216391 18. 1415.54 989612448
Others 256711365 49477121298 550611187 41702115502
Tanypodinae 386412466 2325421138945 21.87141 1 132900167381
Empididae . 141.07 134.30153 .40 081.05 "'
Ephydridae
Hydrellia spp. * * 021.02 *

27

 

 

 

 

Table 2 (cont'd)
Mackinac Cedarville
TAXON .
I Dip-Net 1 Core samples ] Dip-Net 1 Core samples
Mean #lsarnple15.E.' Mean #-mf18.E. Mean #lsample13.E. Mean #-m‘_215.E.
Sciomyzidae * * 031.03 *
Stratiomyidae
Odontomyia app. .1 11.1 1 * .03103 *
Tabanidae
Chrysops spp. * 424114241 .06106 141411414
Haematopota spp. 031.03 * * *
Hybomr'tra spp. 281.24 * 081.08 *
EPHEMEROPTERA
Baetidae
Callibaetisfen-ugineus 3712.3 1272315475 2.171.80
Procleon vr'rr'docularr's 421.29 * *
Caenidae
Caem's amica 249512184 1229861106079 .1711 1 424114241
Caem's latipennis 8.811581 770.43165946 081.08 "‘
Caem's youngi 170314.91 1887.1913984 * *
Ephernerellidae
Eurylophellafimeralis 1031.63 70717.07 .031.03 *
Epherneridae
Hexagem’a limbata 581.58 21205118486 "' *
HEMIPTERA
Belostornatidae
Lethocerus spp. .03103 21.2112121 * *
Corixidae
H esperocon'xa kenm'cotti "' " .021. 02 "'
Sigara transfiturata .20120 "' * *
Sigara variabillis 1001.82 * * *
Palmacorixa buendi .l 11.08 21 .21121 .21 021.02 *
T richocorixa sexcr'ncta * * .06103 *
Others 611.33 7.071707 1.061.46 21 .21121 .21
Hebridridae
Merragata spp. .03103 * * *
Nepidae
Ranatra app. .03103 * 031.03 *
Mesoveliidae
Mesovelia mulsamr’ .08105 * .58145 21 .21121 .21
Gerridae
Gerri: comatus .03103 * .141. 10 *
T repobates app. * "‘ 031.03
LEPIDOPTERA
Noctuidae
Bellura app. 031.03 * 061.06
Pyralidae
Acentria spp. 431.27 * 2061.24 212112121
Munroessa/Synclita/Neocataclysta spp. "' "' 14411.29 *
Parapoynx app. 2.10194 7.071707 .521.24 21.21121 .21

28

 

 

 

 

 

Table 2 (cont'd)
Mackinac Cedarville l
TAXON ’
I Dip-Net Core samples I Dip-Net Core samples I
Mean #/sample1$.E. ’ Mean #-m‘_’1S.E. Mean #/sample1$.E. Mean #-m::s.E.
NEUROPTERA
Sisyra spp. * "‘ 031.03 *
ODONATA
Ashnidae
Anaxjunius 081.08 "' '1'
Aeshna eremita * * .1 11.08
Aeshna app. * "' 331.33
Coenagrionidae
Enallagama boreale 361.32 * * *
Enallagma ebrium/hagem' 43914.17 1767119189 6.421542 20498110602
Enallagma geminatem 061.03 "‘ .1 11.04 *
Enallagma vernale 251.16 "' * *
Ischnura venicalis 304711565 21 .21121 .21 19.2811 1.10 14. 14114. 14
Others 1.701120 * 17011.70 *
Corduliidae
Epitheca app. .42125 "' .39135 *
Cordulia shurlefji 1.081.79 * 251.43 *
Gomphidae
Arigomphus comutus .03103 * * *
Libellulidae
C elithemis app. 061.06 * *
others .l71.l7 * .56135
Lestidae
Lestes app. .06106 * 031.03 *
TRICHOPTERA
Hydropsychidae
C eralopsyche app. 031.03 .141. 14 *
Cheumatopsyche campyla 031.03 * *
Hydroptilidae
Agraylea multipuntata "‘ 21 .211.21 4.441335 353412121
Hydroptr’la spp. * " 3 6713.67 "'
Orthotriclrr'a spp. "‘ 70717.07 " "'
Oxyeihira spp. 9.091505 282711999 63913.8 *
Leptoceridae
C eraclea app. .141. 14 * * *
Mystacides interjecta 2.3011 .50 63.6214060 031.03 *
Mystacr'des sepulchralis 8.141327 12016112016 1.59177 353413534
Necropsyche spp. 221.12 * . 141.07 *
Oecetr's cinerascens 1.1411 .1 * 1,361.79 141411414
Oecetr's inconspicua .03103 "‘ "' *
Oecetr's ostem' * 424014241 "‘ *
Oecetis persimilis .03103 * "‘ *
Oeceris app. 8.751298 12723112723 1,811.89 21.2112121
Triaenodes aba .70143 "‘ . 171.1 1 "'

29

 

 

 

Table 2 (cont'd)
Mackinac l Cedarville
TAXON ’
I Dip-Net Core samples I Dip-Net [ Core samples I
Mean #lsample18.E. Mean #-m'2_+S.E. Mean #/sarnple18.E. Mean #-m"_+S.E.
Limnephilinae
Lilmnephilus spp. .1711 1 * *
Nemotaulr'us hostilus * "‘ 081.08 *
Molannidae
Molanna hyphena .28128 * *
Molanna spp. .14108 56.5511999 *
Phryaneidae
Agrypni improba .56137 * 361.36 *
Fabria app. 031.03 * 251.25 *
Phryganea cinera 141111087 424114241 1451109 212112121
Polycentropodidae
C emotina app. "' " .03103
Polycentropus spp. 1471.81 1343019819 2.511.33 1625716361

Aphids "‘ 212112121 36713.67 *

30

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A)
Plant Associated Community
100% 1
I
80% 1
i
I
60% 1
l
40% 1
I
I
20% I
or 1
o
c: 1
r0
2
{33‘ June July Aug. Sept. June July Aug. Sept.
‘5 Reference marsh Impacted marsh
E 3) Sediment Associated Community
100%'- Q '
[J l l 51335351315353; |
[I I! II I
r
R Q s :
b s :
60%J \ \ I
k I
|
40%1 :
I
20%- 1
l
l
0% 5
June July Aug. Sept. June July Aug. Sept.

Reference marsh Impacted marsh

Ephemeroptera Diptera" fl 0016091818 E Lepidoptera
[:1 Chironomidae fl Trichoptera , Odonata - Hemiptera

 

Figure 9. Community composition of the aquatic insect orders at Cedarville (impacted)
and Mackinac (reference) marsh, Lake Huron. 1996. A) Plant associated community
(dip-net samples) and B) Sediment associated community (core samples). * Diptera
group does not include Chironomidae

31

were also important constituents of the plant associated insects (Figure 9A).

The plant associated community at the impacted marsh was represented by 64 taxa
of aquatic insects, from 33 families (Table 2). Chironomidae made up 73-93% of the plant
associated aquatic insects in the impacted marsh (Figure 9), as compared to 50-60% of plant
associated insects in the reference marsh (Figure 9). The remainder of the plant associated
community at the impacted marsh was primarily represented by Trichoptera and Odonata,
although neither group ever represented more than 15% of the aquatic insects (Figure 9).
Ephemeroptera never comprised more than 2% of the plant associated aquatic insect
community in the impacted marsh even though they represented more than 10% of the fauna
on every sampling date for the reference marsh (Figure 9A). The impacted marsh sediment
community was represented by 22 species of aquatic insects, belonging to 12 families (Table
2). Chironomidae was the only dominant group of aquatic insects in the sediment at the
impacted marsh, representing between 70 to 90% of aquatic insects (Figure 9B).
Ephemeroptera comprised 20% or more of insects collected from the reference marsh
sediments on every sampling date, but were completely absent from impacted marsh
sediments in July, August and September and were only present in low numbers in June
(Figure 9B).

Chironomidae abundance trends were comparable for the sediment and plant
associated communities in the reference and impacted marshes with no consistent trends in
abundance and no significant differences (p<0.05) occurring between the two marshes
(Figures 10 and 11). Thus, differences in percent composition of insect communities for the

two sites (Figure 9) were the result of reductions in Ephemeroptera, Trichoptera and other

32

 

 

 

 

 

   

 

 

 

 

 

500

g
G.
E

'0 400 -
a.
8
a

3. 300 -
8
C
Q

'3 200 -
3
.0
a
5

100 -
‘2’

0—

June July August September
[:21 Reference marsh - Impacted marsh

Figure 10. Abundance trends, including standard error, of the Chironomidae in the plant
associated community (dip-net samples) of Cedarville (impacted) and Mackinac (reference)
marshes, Lake Huron, 1996. 'Mann-Whitney U test: Significant at a = 0.05.

 

 

20,000 -
1% 16,000 -
Q
E
3 12,000 -
8
C
W
E
a 8,000 -
D
to
C
3 4 000 ~
2 r

 

 

 

 

 

 

 

June July August September
E Reference marsh - Impacted marsh

Figure 11. Abundance trends, including standard error, of the Chironomidae in the
sediment community (core samples) of Cedarville (impacted) and Mackinac (reference) marshes,
Lake Huron, 1996. “Mann-Whitney U test: Significant at a = 0.05.

33

insects rather than from reductions in Chironomidae abundances (Figures 10 and l 1).
Chironomidae larvae were ofien found inside the stems of senescent S. acutus,
Utricularia spp., and Nuphar variegata, with densities inside the plants increasing as
decomposition progressed towards late summer. On actively growing Nuphar variegata,
Chironomidae larvae were found on the outside of the stems or underneath the leaves, with
the periphyton community. Some of these larvae were attached to the stems and leaves by
fixed cases, while others were free living. Chironomidae larvae were seldom found on
actively growing S. acutus. Free living Chironomidae included Corynoneura spp.,
Tanypodinae, and several of the Chironomini and Tanytarsini taxa. Most of the other

Orthocladiinae and some of the Chironomini and Tanytarsini, constructed fixed cases.
Even though there were no consistent, significant (p< 0.05) difi'erences between total
abundance of Chironomidae larvae in the two marshes (Figures 10 and 11), the taxonomic
composition of the Chironomidae communities did differ between the two marshes (Tables
3 and 4). Chironomini was consistently the most abundant Chironomidae at the impacted
marsh in both the plant and sediment community (Tables 3 and 4). The reference marsh
sediment was also dominated by Chironomini, however the plant associated Chironomini was
only the dominant Chironomidae in June and September at the reference marsh (Tables 3 and
4). During July, the most abundant Chironomidae in the plant associated community of the
reference marsh was the Tanytarsini comprising 40% of that population (Table 3). The other
Orthocladiinae dominated the Chironomidae community in August at the reference marsh
making up 50% of the Chironomidae community (Table 3). The fies-living Corynoneura spp.

never represented over 5% of the Chironomidae community in the plant associated

Table 3. Total number of Chironomidae collected and percent of the total Chironomidae represented
by taxa in the plant associated community in Mackinac (reference) marsh, Lake Huron, 1996.

 

 

 

Taxa
June July August September

Chironomini 55 % 30 % 29 % 50 %
Orthocladiinae

Corynoneura spp. 3 % 5 % 3 % 1 %

others 6 % 13 % 49 % 4 %
Tanypodinae 9 % 12 % 14 % 33 %
Tanytarsini 27 % 40 % 5 % 12%
Total number 828 1,148 1,219 3,103

 

Table 4. Total number of Chironomidae collected and percent of the total Chironomidae
represented by taxa in the plant associated community in Cedarville (impacted) marsh,
Lake Huron, 1996.

 

 

 

Date

Taxa June July August September
Chironomini 42 % 47% 59 % 44 %
Orthocladiinae

Corynoneura spp. 21% 6 % 4 % 6 %

others 22% 35 % 12 % 30 %
Tanypodinae 6 % 9 % 8 % 9 %
Tanytarsini 9 % 3 % 17 % 11%
Total number 1,457 199 3,076 2,634

 

35

community of the reference marsh, but was frequently collected, with an average of four
individuals per sample (Table 3). Corynoneura spp. comprised 21% of the Chrionomidae
community during June at the impacted site and then fell to levels similar to those exhibited
at the reference marsh of about 5% or less of the community composition (Table 4).
Chironomidae reached a peak mean abundance of 341 individuals per sample in September
in the plant associated community of the reference marsh, and a peak abundance of 341
individuals per sample in August at the impacted marsh (Figure 10). The Chironomidae were
identified to the lowest taxonomic level possible in June (Table A-2). On this date, 28 taxa
were identified. The reference marsh had 19 identifiable taxa compared with 13 at the
impacted marsh (Table A-2). Only one Tanypodinae, Ablabemyia peleenis, was present in
the impacted marsh samples, compared with five species collected at the reference marsh
(Table A-2).

Chironomidae was the dominant group of aquatic insects in the sediment community,
comprising over 50% of the aquatic insects at the reference marsh, and over 90% at the
impacted marsh (Figure 9B). The sediment associated Chironomidae reached a maximum
mean density in both marshes in September, with no significant (p<0.05) difference between
the mean density of 17,331'm'2 at the impacted marsh, and the 16,000-m‘2 at the reference
marsh (Figure 11). Larvae of the tribe Chironomini was the dominant Chironomidae in the
sediment community of both the reference and impacted marsh (Tables 5 and 6). The
predacious, Tanypodinae was also a major constituent of the Chironomidae community of the
reference marsh, representing nearly 40% of the Chrionomidae in September (Table 5).

Tanypodinae in the sediment samples of the impacted marsh had similar abundances to the

ll

36

Table 5. Total number of Chironomidae collected per m? and percent of the total Chironomidae
represented by taxa in the sediment community in Mackinac (reference) marsh, Lake Huron, 1996.

 

 

 

Taxa Date
June July August September

Chironomini 62 % 54 % 33 % 57 %
Orthocladiinae

Corynoneura spp. 0 % 7 % 3 % 0 %
others 6 % 12 % 21 % 1 %
Tanypodinae 11 % 27 % 24 % 39 %
Tanytarsini 21 % 0 % 19 % 3 %
Total number 150,:szrz/m2 eases/m2 47,5821m2 148,601/m2

 

Table 6. Total number of Chironomidae collected per in2 and percent of the total Chironomidae
represented by taxa in the sediment community in Cedarville (impacted) marsh, Lake Huron, 1996.

 

 

 

T axa Data
June July August September

Chironomini 60 % 63 % 66 % 75 %
Orthocladiinae

Corynoneura spp. 3 % 1 % 3 % 0 %
others 22 % 3 % 11% 1 %
Tanypodinae 12 % 29 % 2 % 17 %
Tanytarsini 3 % 4 % 18 % 7 %

Total number 30,535/m2 55,726Im2 49,618Im2 155,9801m2

 

37

reference marsh in June and July (Tables 5 and 6). However, abundance of Tanypodinae
dropped at the impacted marsh from 29% of Chironomidae in July, to 2% of the
Chironomidae population in August, and remained at abundances much less than exhibited
at the reference marsh in September (Tables 5 and 6). The Orthocladiinae, C orynoneura spp.,
was the least abundant Chironomidae larvae in the sediment samples at both the reference and
impacted marshes, representing less than 10% of the total Chironomidae (Tables 5 and 6).

The only other dipteran families present in the plant associated and sediment samples
of the marshes were uncommon to rare especially in the plant associated samples (Figure 9
and Table 2). Empididae and Ceratopogonidae were collected in the sediment community of
the reference marsh (Table 2). The Ceratopogonidae was represented by three genera;
Bezzia/Palpomyia spp. , Serromyia spp., and Sphaermias spp. in the sediment community of
the reference marsh. Ceratopogonidae occurred in much larger abundances in the sediment
community of the reference marsh compared with the plant associated community (Table 2).
The most abundant of these species was Bezzia/Palponnlia spp. which had an average density
of 148 larvae m‘zin the sediment community of the reference marsh and 21 larvae-m‘2 in the
sediment of the impacted marsh (Table 2). The only Diptera, besides Chironomidae, collected
in the sediment of the impacted marsh was Bezzia/Palpomyia spp. and C ulicoides spp.
members of the family Ceratopogonidae, and Chrysops spp. (Tabanidae) (Table 2).

There were 7 species of mayflies in four families collected from the reference marsh,
but only 4 species from 3 families were collected from the impacted marsh (Tables 7 and 8).
Ephemeroptera was a major component of the sediment and plant associated

macroinvertebrate communities of the reference marsh, comprising from 10 to more than 30%

Table 7. Total number of Ephemeroptera collected and percent of the total Ephemeroptera
represented by species in the plant associated community in Mackinac (reference) marsh, Lake

 

 

 

Huron, 1 996.
Taxa Date
June July August September

Baetidae

Callibaetis ferrugineus 10 % 7 % 4 % 7 %

Procleon viridoculan's 3 % 6 % 0 % 0 %
Caenidae

Caenis amica 30 % 13 % 6 % 58 %

Caenis Iafipennis 21 % 14 % 8 % 17 %

Caenis youngi 35 % 60 % 71 % 16 %
Ephemerellidae

Eurylohella funeralis <1 % 0 % 4 % 2 °/o
Ephemeridae

Hexagenia Iimbata 0 % 0 % 7 % 0 %
Total number/dip-net sample 15/sample 21Isamp|e 35Isample 155/sample

 

Table 8. Total number of Ephemeroptera collected per m2 and percent of the total
Ephemeroptera represented by species in the sediment of Mackinac (reference) marsh,

Lake Huron 1996.

 

 

 

Date
Taxa June July August September

Baetidae

Callibaetis ferrugineus 12 % 0 % 3 %
Caenidae

Caenis amica 11 % 6 % 5 % 47 %

Caenis Iatipennis 4 % 18 % 0 % 29 %

Caenis youngi 81 % 64 % 75 % 20 %
Ephemerellidae

E uronheI/a funeralis 0 % 0 % 0 % <1 %
Ephemeridae

Hexagenia Iimbata 0 % 0 % 20 % <1 %
Total number 20,610/m2 12,723Im2 34,097/m2 84,988lm2

 

39

percent of the aquatic insect communities (Figure 9). In contrast, mayflies were only minor
components of the impacted marsh fauna and never represented more than 4% of the plant
or sediment associated aquatic insect communities (Figure 9). Ephemeroptera abundance was
significantly higher (p<0.05) in the sediment and plant associated communities at the reference
marsh than was Ephemeroptera abundance in the impacted marsh throughout the sampling
season (Figures 12 and 13). Ephemeroptera in the plant and sediment associated communities
reached peak abundances in September with numbers generally increasing as the season
progressed fiom June through September (Figures 12 and 13). Ephemeroptera increased
fiom 15 per dip-net sample in June to their peak density of 155 per dip-net sample in the plant
associated community in September. Density also increased in the sediment associated
community at the reference marsh to 84,988-m’2 in September fiom a low of 12,723-m'2 in
July (Tables 7 and 8). In contrast, no Ephemeroptera were collected in July, August, and
September. in the sediment of the impacted marsh. Ephemeroptera reached a mean peak
abundance of only 4 individuals per sample in the plant associated community of the impacted
marsh during July (Figure 12). The predominant Ephemeroptera family in the plant and
sediment associated communities of the reference marsh was the family Caenidae with 3
species present comprising fi'om 80-96% of total mayfly numbers collected (Tables 7 and 8).
An average of 51 Caenidae per dip-net sample was collected over the entire season from the
reference site as compared to 0.25 Caenidae per sample for the impacted site (Table 2).
Likewise, a seasonal average of 3,887 Caenidae-m‘2 were collected from sediments of the
reference site as compared to 42 for the impacted site.

The taxonomic composition between the sediment and the plant associated community

40

 

 

 

 

200
g
Q
E

8 150
Q
d)
O
i
8

100
8
C
G
'O
C
3
'8

c 50
CB
0
E

0

June July August September
- Reference marsh 1:1 Impacted marsh

Figure 12. Abundance trends, including standard error, of Ephemeroptera in the plant
associated community of Mackinac (reference) and Cedarville (impacted) marshes, Lake
Huron, 1996. *Mann-Whitney U test: Significant at or. = 0.05.

 

12.000

10,000

8,000

6,000

4,000

Mean abundance per meter2

2.000

 

 

 

June July August September
- Reference marsh 1:1 Impacted marsh

Figure 13. Abundance trends, including standard error, of Ephemeroptera in the sediment
associated community of Mackinac (reference) and Cedarville (impacted) marshes, Lake
Huron, 1996. *Mann-Whitney U test: Significant at or = 0.05.

41

of the reference marsh differed only in the occurrence of one additional Baetidae species in
the plant associated community, Procleon viridocularis. The remainder of the plant and
sediment associated taxa in the reference marsh communities were similar with three species
of Caenidae, Caenis amica, C. latr'pennis, and C. youngi, and one species each in the family
Ephemerellidae, Ephemeridae and Baetidae (Table 2). The most commonly observed
specimens in the plant and sediment associated communities of the reference marsh were
Caenis amica and C. youngi (Tables 7 and 8). C. youngi was the dominant Ephemeroptera
species in the sediment and plant associated communities of the reference marsh June through
August but was never collected from the impacted marsh (Table 2). C. amica was the most
abundant Ephemeroptera in September in both the sediments and plant associated samples
(Tables 7 and 8). The burrowing mayfly Hexagenia limbata was only collected in the plant
associated community in August and was collected in both August and September in the
sediment associated community of the reference marsh (Tables 7 and 8). However, no H.
limbata were collected in either the sediment or the plant associated communities of the
impacted marsh (Table 2).

The only Ephemeroptera species collected in the plant community of the impacted
marsh were two species of Caenidae, C. amica and C. latipennis, and two species of
Baetidae, Callibaetisferrugineus and Eurylophellafimeralr‘s (Table 2). C. ferrugineus was
the most abundant Ephemeroptera at the impacted marsh, yet only an average of 2 individuals
per sample were collected (Table 2). Only one species of Ephemeroptera was collected in the
sediment samples at the impacted marsh, C. amica, and it was only present in June (Table 2).

The Trichoptera was the most diverse group of insects in the plant and sediment

42

associated communities in both marshes with a total of 25 species in 7 families collected from
the two marshes (Table 2). The Trichoptera in the plant associated community of the
reference marsh was represented by a total of 19 species; two species each in the families
Molannidae and Hydropsychidae; one species each in the families Hydroptilidae,
Limnephilinae and Polycentropodidae; three species of Phryganeinae and nine species of
Leptoceridae (Table 2). Sixteen species representing 6 families of Trichoptera were collected
from the plant associated community of the impacted marsh with the family Molannidae being
the only family present in the reference marsh that was not also represented in the impacted
marsh (Table 2). Ten species of Trichoptera were present in the sediment core samples of the
reference marsh; three species of Hydroptilidae, four species of Leptoceridae and one species
each of Molannidae, Phryganeidae and Polycentropodinae (Table 2). Six species of
Trichoptera were present in the sediment community of the impacted marsh, but only
Polycentropus spp. and Agraylea multz’puntata occurred in more than one month of sampling
(Table 2).

Two weeks following discharge in June and in September during discharge fi'om the
wastewater treatment lagoon, the plant associated Trichoptera in the impacted marsh were
significantly (P<0.05) less abundant than Trichoptera at the reference marsh (Figure 14).
Although Trichoptera were consistently more abundant in the sediment community of the
impacted marsh, the only month a significant difi‘erence (P<0.05) occurred was September
when discharge was actively occurring fi'om the wastewater treatment lagoon (Figure 15).
Trichoptera of both the plant and sediment associated communities reached peak abundances

in September at the reference marsh (Figures 14 and 15). Trichoptera reached a maximum

43

 

2 125 -
O.

E a:
G

(D

g. 100 -
0

5

8 75 -
8

C

G

E

a 50 q
.0

(U

C

CO

0

2 25 .

 

 

   

 

 

 

June July August September
- Reference marsh [:1 Impacted marsh

Figure 14. Abundance trends, including standard error, of Trichoptera in the plant associated
community of Mackinac (reference) and Cedarville (impacted) marshes, Lake Huron, 1996.
*Mann-Whitney U test: Significant at or = 0.05.

 

 

 

2,000
~_ 1,500 — *
2
Q)
E
8
8 1,000 e
C
(U
‘U
C
3
.0
as
Fe 500 e
0)
2
T
o J l _‘

 

 

   

 

   

 

 

 

 

June July August September
- Reference marsh 1:3 Impacted marsh

Figure 15. Abundance trends, including standard error, of Trichoptera in the sediment
associated community of Mackinac (reference) and Cedarville (impacted) marshes, Lake Huron,
1996. *Mann-Whitney U test: Significant at a = 0.05.

44

mean abundance of 102 individuals in the plant associated community, and a maximum mean
density of 1,441 -m‘2 in the sediment of the reference marsh (Figures 14 and 15). Trichoptera
also reached a peak mean abundance in September in the sediment community of the impacted
marsh, however only at 565 individuals-m“2 (Figure 15). Plant associated Trichoptera at the
impacted marsh reached a peak mean abundance a month earlier in August, with a mean
abundance of only 31 individuals per sample (Figure 14).

Few similarities existed in the dominant taxa between marshes and between sampling
regimes (Table 2). Phryganea cinera, Oxyethira spp. , 0ecetis spp. , and Mystacides
sepulchralis were the most abundant Trichoptera species in the plant associated community
at the reference marsh and were also among the most common species collected from
sediment samples (Table 2). Although P. cinera represented the majority of the plant
associated Trichoptera population in July and September, it was not collected in June and was
scarce in August sampling (Figure 16). During wastewater discharge in September, P. cinera
were significantly (p<0.05) more abundant in the plant associated samples at the reference
marsh than at the impacted marsh (Figure 16). The sediment community of the reference
marsh was not consistently dominated by any one species of Trichoptera (Figure 17). During
June, only two species were collected in the sediment community of the reference marsh,
Mystacides interjecta and 0ecetis ostem‘, with these species occurring in equal proportions
(Table 2). During July, the Polycentropodidae, Polycentropus spp., was the dominant
Trichoptera species in the sediment community of the reference marsh reaching a peak mean
density at approximately 4OO-m’2 (Figure 17).

The most common Trichoptera in the plant associated community of the impacted

45

60

 

40d

June

20—

 

 

 

 

 

 

 

 

 

60

 

40—

July

20 d ..

 

 

 

 

 

 

 

 

60

 

40-

August

20-

T,
0 , glib.

60

 

Mean abundance per sweep sample

 

 

 

 

 

401

 

September

201 *

. .r. L

P. cinera A. multipuntata Mystacr’des spp. Oeoefis spp.

 

 

 

 

 

 

 

C21 Reference marsh - Impacted marsh

Figure 16. Abundance trends including standard error of the dominant Trichoptera species
in the plant associated community of Cedarville (impacted) and Mackinac (reference) marshes,
Lake Huron 1996. *Mann-Whitney U test: Significant at a = 0.05.

46

 

750

500 ..

June

250 4

 

 

 

 

 

 

 

 

 

 

750 2

500 ..

 

July

250 -

 

 

 

 

 

 

 

750

 

Mean abundance per rn2

500 _,

August

250 ..

 

 

 

 

 

 

750

 

 

500 -

 

 

September

250 a

 

 

 

     

 

 

 

 

 

i

0 _
A. multipuntata Polycentropus spp. Mysfacides spp. Oecetis spp.

[:1 Reference marsh - Impacted marsh

Figure 17. Abundance trends including standard error of the dominant Trichoptera species
in the sediment associated community of Cedarville (impacted) and Mackinac (reference) marshes,
Lake Huron 1996. *Mann-Whitney U test: Significant at or = 0.05.

47

marsh were members of the Hydroptilidae family, but a large number of other species in
several families were also present (Table 2). Agraylea multipunctata was the dominant
Trichoptera during June and July, and was significantly (p<0.05) more abundant in those
months at the impacted marsh than compared with the reference marsh (Figure 16). Yet
during September, A. multipunctata was uncommon in the plant community of the impacted
marsh (Figure 16). The most abundant Trichoptera in the plant associated community of the
impacted marsh in August was Oxyethira spp. Oxyethira spp. comprised nearly 50% of the
Trichoptera community in the plant associated samples during August and dropped to
comprise less than 10% in September at the impacted marsh. September plant associated
samples at the impacted marsh were dominated by Hydroptila spp. which represented nearly
50% of the Trichoptera population that month (Table 2). Hydroptila spp. had not been
collected in previous samples all summer (Table 2). The remainder of the Trichoptera
community in the plant associated community of the impacted marsh tended to be uncommon,
seldom representing more than 5% of the Trichoptera community composition (Table 2).
Trichoptera occurred in very low numbers in the sediment of the impacted marsh (Table 2).
Polycentropus spp. was the most abundant Trichoptera in the sediment of the impacted marsh
averaging 162-m‘2 (Table 2). The remainder of the sediment associated Trichoptera at the
impacted marsh, Wstacides sepulchralis, 0ecetis cinerascens, 0ecetis spp. , and Phryganea
cinera, occurred in very low abundances (Table 2). No significant difference were detected
between the dominant Trichoptera species between the sediment communities of the impacted
and reference marsh (Figure 17).

Odonata represented approximately 15% of the plant associated macroinvertebrates

48

in both marshes with the majority being Coenagrionidae damselflies (Table 2). Odonata
abundance trends were comparable for the sediment and plant associated communities in the
reference and impacted marshes no significant (p<0.05) differences occurred for any sampling
date (Figures 18 and 19). Peak abundances occurred in September with numbers generally
increasing as the season progressed from June through September (Figures 18 and 19).
Odonata were uncommon in the sediment samples of both marshes (Table 2). Only two
species of Odonata, Ischnura verticalis and Enallagma spp. were collected in the sediment
samples of these marshes (Table 2). Enallagma spp. was lacking the characteristics necessary
to identify them to species, either due to the collection of early instars or physically damaged
specimens.

Plant associated Odonata reached their highest abundance in September with 813
individuals collected at the reference marsh, and 548 individuals collected at the impacted
marsh (Tables 9 and 10). The Odonata associated with the plant community of the reference
marsh was represented by 13 taxa compared to 10 taxa for the impacted marsh (Table 2).
The Coenagrionidae, lschnura verticalis, was the most abundant Odonata in the plant
associated community of the both the reference and impacted marshes, representing up to
80% of the Odonata in the reference marsh and up to 89% in the impacted marsh (Table 9).
Also present but generally uncommon in the plant associated community at the reference
marsh were five other Coenagrionidae species, one Lestidae species, and six Anisoptera
species (Table 2). While occurrence of uncommon species difi‘ered between the reference and
impacted marshes, no significant differences (p<0.05)occurred for any of the common

Odonate taxa (Table 2). Sediment associated Odonata were dominated by Enallagma spp.,

49

125

 

100 -

754

Mean abundance per sweep sample

 

 

 

 

 

 

 

June July August September
- Reference marsh 1:1 Impacted marsh

Figure 18. Abundance trends, including standard error, of Odonata in the plant associated
community of Mackinac (reference) and Cedarville (impacted) marshes,
Lake Huron, 1996. *Mann-Whitney U test: Significant at a = 0.05.

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

 

800
T
“k
.. 600 .1
8
8
C
S
g 400 - .
a 1
I!
C
to
§
200 _
0 _
June July August September
- Reference marsh 1:: Impacted marsh

Figure 19. Abundance trends, including standard error, of Odonata in the sediment
associated community of Mackinac (reference) and Cedarville (impacted) marshes, Lake
Huron, 1996. *Mann-Whitney U test: Significant at or = 0.05.

50

Table 9. Total number of Odonata and percent of the total Odonata represented by taxa
in the plant associated community in Mackinac (reference) marsh, Lake Huron, 1996.

 

 

 

Taxa Date
WWI.

Aeshnidae

Anaxjunius 0 % 0 % 0 % <1 %
Coenagrionidae

Enallagama boreale 0 % 0 % <1 % 1 %

Enallagama ebn’um/hageni 31% <1 % <1 % 18 %

Enallagma geminatem 8 % 0 % 0 % <1 %

Enallagma vemale 0 % 0 % 1 % 1%

Ischnura verticalis 46 % 80 % 78 % 78 %

Others 0 % 10 % 11 % 0 %
Corduliidae

Epitheca spp. 0 % 0 % 1 % 1 %

Cordulia shudefii 0 % 6 % 7 % 0 %
Gomphidae

An'gomphus comutus 0 % <1 % 0 % 0 %
Libellulidae

Celithemis spp. 0 % 0 % <1 % 0 %

Others 0 % 4 % 0 % 0 %
Lestidae

Lesles spp. 15 % 0 % 0 % 0 %
Total number 13 151 413 813

 

Table 10. Total number of Odonata and percent of the total Odonata represent by taxa in the
plant associated community of Cedarville (impacted) marsh, Lake Huron, 1996.

 

 

 

 

T axa Date
June July August September

Aeshnidae

Aeshna eremita 0 % 13 % <1 % 1%
Coenagrionidae

Enallagama ebrium/hageni 50 % 7 % 4 % 37 %

Enallagma geminatem 20 % 0 % <1 % <1 %

lschnura verticalis 10 % 0 % 89 % 62 %

Others 0 % 64 % 0 % 0 %
Corduliidae

Epitheca spp. 0 % 0 % 3 % <1%

Cordulia shunefii 10 % 8 % 0 % 0 %
Libellulidae 0 % 8 % 3 % 0 %
Lestidae

Lestes spp. 10 % 0 % 0 % 0 %

___191aLnumbe.r 19. 45 3.87 5.5.8..—

51

which was collected with a mean density of 177 nymphs-m'z’ at the reference marsh and 205
nymphsm‘2 at the impacted marsh (Table 2). I. verticalis was the only other species collected
fi'om the sediment samples in either the impacted or the reference marsh (Table 2), and it only
occurred during September. No Odonata were collected in June sediment samples. No
significant differences (p<0.05) occurred between the impacted and reference marshes for
either species of sediment occurring Odonata (Table 2).

I. verticalis was collected every month in the plant associated samples of the
impacted marsh with the exception of July, where on this date 64% of the Odonata population
was represented by an early instar Coenagrionidae that were lacking characteristics necessary
for proper identification to the generic level (Table 10). The unidentified Coenagrionidae
dominated the Odonata plant associated community at the impacted marsh during July (Table
2). It appeared that I. verticalis hatched sometime between the June and July sampling date,
and the unidentified Coenagrionidae may have been early instar I. verticalis. If so I. verticalis
abundance in the impacted marsh would have been similar to abundance in the reference
marsh on this date as it was in August and September (Tables 9 and 10). Too few Odonata
were collected in June from either site for inter-site comparisons to be meaningful (Tables 9
and 10).

The remainder of the aquatic insect community in the plant and sediment associated
samples occurred in very low numbers and infi'equently in the reference and impacted marsh
(Table 2). Aquatic Coleoptera were represented by nine taxa at the reference marsh, and six
taxa at the impacted marsh (Table 2). Sediment associated aquatic Coleoptera was

represented by just one larval species, Donacia spp., and it occurred in both the impacted and

52

reference marsh (Table 2).

Hemiptera was represented by nine species belonging to six families in the plant
associated community of the reference marsh and 8 species belonging to four families in the
impacted marsh (Table 2). The most abundant Hemipteran family, the Corixidae, included
three species and a group of nymphs lacking the characteristics necessary to identify them
below the family level in the reference marsh and four species in the impacted marsh (Table
2). Sigara transfigurata and Sigara variablilis were two Corixidae collected in the plant
associated samples of the reference marsh that had not previously been recorded fiom this
area. Sediment associated aquatic Hemiptera at the reference marsh consisted of the family
Corixidae and the Belostomatidae, Lethocerus spp. (Table 2). The sediment associated
Hemiptera of the impacted marsh was represented by unidentified Corixidae nymphs, and
Mesovelia mulsanti (Table 2).

The same three species ofLepidoptera were collected in the plant associated samples
at both the impacted and reference marsh in low numbers; BelIura spp. ,' Acentria spp.; and
Parqpqynx spp. (Table 2). Sediment samples from both marshes contained low numbers of
Lepidoptera larvae (Table 2).

Three species of Collembola, Isotomurus tricolor, Entomobrya nivalis and
Pseudobourletiella spinata were occasionally collected in the plant associated samples at the
impacted marsh (Table 2). Only one species of Collembola, Pseudobourletiella spinata, was
collected from the plant associated community of the reference marsh (Table 2). No
Collembola were collected in the sediment samples of either marsh.

The least diverse aquatic insect group in the plant associated community of the

53
impacted marsh was a sponge predator Neuroptera, Sisyra spp. (Table 2). Sisyra spp. was
uncommon in the plant associated samples at the impacted marsh, although many sponges
were observed along the transect (Table 2).
Non-Insect Macroinvertebrates:

A total of 8,619 non-insect macroinvertebrates were collected in the plant associated
community of the reference marsh, distributed among 23 identified taxa (Table 11). There
were 10,985 non-insect invertebrates in 30 taxa collected from the impacted marsh. Thus,
the impacted marsh supported more non-insect taxa than did the reference marsh (Table 11).
The sediment community at the reference marsh was represented by 17 identified taxa, only
one of which, a Lymnaeid snail, Acella haldemam', was not collected in the plant associated
community of the reference marsh (Table 11).

The sediment community of the impacted marsh consisted of 18 taxa with only two
taxa present that were not also collected from sediments of the reference marsh (Table 11).
The Oligochaeta, Amphipoda, Gastropoda and Isopoda were significant components of the
plant and sediment associated non-insect macroinvertebrate communities in both the reference
and impacted marshes.

Non-insect macroinvertebrates made up more than 50% of the invertebrate community
in both the sediments and plant associated communities of the impacted marsh, while aquatic
insects made up more than 50% of the macroinvertebrate communities in the reference marsh
(Figure 7).

Nematoda were over 5 times more common in the sediments of the impacted marsh

than they were in sediments of the reference marsh (Table 11). Nematoda dominated the

54

Table l l. Non-insect macroinvertebrates collected with standardized dip-net sweep sampling from Scirpus
dominated zones (mean number/sample:S.E) and with cores from sediments in these zones (mean number-m'2
:SE) from the reference (Mackinac) and Impacted (Cedarville) marshes in 1996. * was not collected in the

 

 

 

 

 

 

samples.
Mackinac Cedarville I
TAXON 7
l Dip-Net Core samples Dip-Net TCore samples I
Mean #/sampleiS.E. Mean #-m"_+S.E. Mean #/sample-_tS.E. Mean #-m‘2_+S.E.
ANNEIJDA
OLIGOCHAETA
Naididae
Stylarr'a spp. 67.55zl:23.24 353.4 1&14485 655813055 551 321467.78
Others 19.53:t8.51 982.46i265.56 59.36i6.03 1547921744. 1 3
Tubificidae 1.58:86 453.6 li233.95 S. l7:h2.98 374.61i128.20
POLYCHAETA
Manayunla'a speciosa "' "' 08:12.08 "'
HIRUDINAE a 4- 4- ul:
Erpobdellidae 36:36 * * *
Mooreobdella spp. * * . l 121:.08 *
Glossiphoniidae
Alboglossr'phonr'a heteroclr'ta " "' .3 l:l:.21 *
Batrachobdella phalera * * .22:I:.22 "‘
Helobdella stagnalr's "' * .1 1:04 *
T heromyzon spp. 031.03 "' * "'
CRUSTACEA
AMPHIPODA
Crangnox spp. "‘ "' . 1421:. 14 "'
Gammarus spp. 3.06:7] 24739-14526 l36.45:l:51.93 l936.67:l:799.04
Hyallella azreca 36.78il4.51 883.52zt212.63 72.22i27.20 l448.97:t542.41
DECAPODA
Orconecles propinquus "' "' .03:!:.03 "
ISOPODA
Lirceus lineatus ll.64:t10.01 155.50t106.73 43.56i12.4l 720.98i230.99
Racovr'aar’ racovr'tzai 8.64:l:2.5 91 1.79:1:39538 12. l 71:2.74 1533.79i316.17
MOLLUSCA
GASTROPODA
Pulrnonata
Ancylidae
Fern'ssa parallela 22:10 * lO.28:I:7.89 25445122706
Lymnaeidae
Acella haldemam‘ 5021:.30 42.41:1:42.4l .83183 63.62:I:40.6
Fossarr'a spp. .171. 17 "‘
Physidae
Aplexa elongata "‘ " .08105 "
Physa gyrr'na 5.06:1:256 63.63406 6.98:2.46 134.30i107.89
Planorbidae
Gyraulus deflectus 2.45:1:2.41 * .55:t.38 *
Gyralus parvus 40.97t35.05 445.29i111.53 41 .61:I:l 9.55 195787552352
Promenetus exacuous 1.082697 42.41:I:42.4l 6.03:1:2.97 269.86i12723

55

Table l l (cont'd)

 

 

 

 

Mackinac I Cedarville
TAXON
I Dip-Net J Core samples I Dip-Net I Core samples I
Mean #lsample:S.E. Mean #-m"_+S.E. Mean #/sample-_l-S.E. Mean #-m'2_+S.E.
Prosobranchia
Bithyniidae
Bithym'a tentaculata * "' 55:.34 5654:3224
Hydrobiidae

Amnicola Iimosa 7.72:6.43 183.78:46.88 23.1 1:8.61 728.02:213.57

Valvata bicarr’nata * "' 22:16 *
OTHERS
NEMATODA 8.92:2.07 586.66:29l .85 2.59:.55 301 1.03:871.53
PELECYPODA

Sphaeriidae 5.06:2.02 523.05:l 3 l .35 1.97:1.46 7775:6039
ARACHNOIDEA (MITES)

Oribatei 4.72:1.95 424.09:94.48 3.22:.69 1095.56:3 59.37

Hydrocarina 13.1 1:2.48 l84.30:29. 14 10.42:4.6 346.34:71.5
Spider

T etragnatha laboriosa . l 1:.22 "‘ * *

TUBELLARIA

Tricladida . 14:. 10 21.21:21.21 .50:.20

*
Dugesr'a tifl'na * "‘ 1.36:1.25 "'

56

sediment community of the impacted marsh during June, representing 30% of the
macroinvertebrate community, and was a significant component of this community in July and
August. Nematoda were a less significant component of the reference marsh sediment
community (Figure 7B and Table 11).

Oligochaeta was represented by two families, the Naididae and the Tubificidae (Table
11). There were no significant differences (p<0.05) in Oligochaeta abundance in the two
marshes on any sampling date (Figures 20 and 21). Most of the Oligochaeta collected from
the reference and the impacted marsh were members of the family Naididae (Tables 11-15).
The genus Stylaria spp. was easily identified and was separated from other unidentified
Naididae. Stylaria spp. and other Naididae were abundant in the plant associated community
in both marshes, comprising from 88 to almost 100% of the Oligochaeta for all sampling dates
(Tables 11-13). Tubificidae made up more than 5% of the plant associated Oligochaeta only
during August in the impacted marsh (Table 13). Oligochaeta reached a peak mean density
in the plant associated community in July at both marshes (Tables 12 and 13, Figure 20).
While abundance in the sediments were more variable with peaks in June and September
(Tables 14 and 15, Figure 21). During July's peak density for the plant associated community,
1,286 Oligochaeta were collected, at the reference marsh, compared with 2,105 at the
impacted marsh (Tables 12 and 13). Stylaria numbers were comparable for both marshes
(Tables 12 and 13). Thus, differences in overall numbers in July and August occurred
through increases in other species of Naididae in the impacted marsh (Tables 12 and 13).

Sediment associated Oligochaeta abundance at the impacted marsh, was higher in

September during discharge from the wastewater treatment lagoon than it was for the

57

Table 12. Total number of Oligochaeta and percent of the total Oligochaeta represented
by taxa in the plant associated community at Mackinac (reference) marsh, Lake Huron,
1996.

 

 

 

Date
Taxa
June July August September

Naididae

Stylan‘a spp. 38 % 94 °/o 83 % 77%

others 59 % 6 % 17 % 18 %
Tubificidae 3 % <1 % < 1% 5 °/o
Total number 676 1,286 488 742

 

Table 13. Total number of Oligochaeta collected and percent of the total Oligochaeta
represented by taxa in the plant associated community of Cedarville (impacted) marsh,
Lake Huron, 1996.

 

 

 

Taxa Date
June July August September
Percent
Naididae
Stylaria spp. 31 °/o 66 % 41 % 37 °/o
others 68 % 32 % 47 % 61 %
Tubificidae 1 °/o 2 % 12 % 2 "/0

Total number 849 2,105 1,01 6 678

58

Table 14. Total number of Oligochaeta per m2 and percent of the total Oligochaeta
represented by taxa in the sediment in Mackinac (reference) marsh, Lake Huron, 1996.

 

 

 

Date
Taxa
June July August September
Naididae
Stylan'a spp. 21 % 26 °/o 14 "/0 15 °/o
others 48 % 51 "/0 86 % 61 %
Tubificidae 31 % 23 % 0 % 24 °/o
Total number 32,061Im2 10,9421m2 5,598Im2 15,821Im2

 

Table 15. Total number of Oligochaeta per m2 and percent of the total Oligochaeta
represented by taxa in the sediment in Cedarville (impacted) marsh, Lake Huron, 1996.

 

 

 

Date
Taxa
June July August September
Naididae
Stylan’a spp. 60 % 13 % 9 °/o 0 %
others 24 % 87 % 60 % 90 %
Tubificidae 16 °/o 0 % 31 % 10 "lo

Total number 29,008Im2 6,107Im2 16,285Im2 37,659Im2

 

59

 

 

 

 

 

 

 

 

     

300
8
E 250
to
In
8 _
o 200
a
8 1501
8
C
m
'8 100 -
3
.0
to
§ 50-
2
0
June July August September
1:] Reference marsh - Impacted marsh

Figure 20. Abundance trends. including standard mm, of Oligochaeta in the plant
associated community of Mackinac (reference) and Cedarville (Impacted) marshes,
Lake Huron, 1996. 'Mann-Whitney U test: Significant at at = 0.05.

 

1

6,000

1

5,000

 

4,000

 

I

3,000

 

L

2,000

Mean abundance per m2

1 ,000

l

         

 

 

 

 

 

June July August September
1:] Reference marsh - Impacted marsh

Figure 21. Abundance trends, including standard error, of Oligochaeta in the sediment
community of Mackinac (reference) and Cedarville (impacted) marshes, Lake Huron, 1996.
‘Mann-Whitney U test: Significant at at = 0.05.

60

reference marsh (Figure 21 and Table 15), but this difl‘erence was not significant (p=0.86).

Oligochaeta densities in the sediment community of the reference marsh were never
significantly difl‘erent fi'om Oligochaeta densities in the impacted marsh (Figure 21). The peak
mean density in the reference marsh occurred in June at 3,562-m‘2, while the mean peak for
the impacted marsh occurred in September at 4,184-m‘2.

The Tubificidae were never abundant within the plant associated samples at the
reference marsh reaching a peak of 4 per dip-net sample or 5% of Oligochaeta in September
(Table 12), Tubificidae peaked at 13.5 per dip-net sample or 12% of the plant associated
Oligochaeta in the impacted marsh in August (Table 13). Tubificidae were more important
components of the sediment associated communities at both sites comprising 31% of the
Oligochaeta sediment community at the reference marsh in June (Table 14) and up to 31%
of the Oligochaeta plant associated community in the impacted marsh in August (Table 15).

Tubificidae are often classified as very pollution tolerant but no consistent differences
occurred between the impacted and reference sites (Tables 14 and 15). During wastewater
discharge in September into the impacted site, for example, Tubificidae densities reached
3,797-m'2 at the reference site and 3,766'm'2 at the impacted site (Tables 14 and 15).

Gastropoda was the most diverse group of non-insects in the sediment and plant
associated communities at both the impacted and reference marsh (Table 11). Gastropoda
associated with the plant and sediment communities reached peak abundances in September
with numbers generally increasing as the season progressed from June through September
(Figures 22 and 23). Gastropoda was dominated by the Planorbidae snail, Gyraulus parvus,

in the plant and sediment community at both the impacted and the reference marsh (Figure

61

 

300

225-

150 —

 

Mean abundance per sweep sample

 

 

 

   

June July August September
a Reference marsh - Impacted marsh

Figure 22. Abundance trends, including standard error, of Gastropoda in the plant associated
community of Cedarville (impacted) and Mackinac (reference) marshes, Lake Huron, 1996.
‘Mann-Whitney U test: Significant at 0. = 0.05.

 

10,000 -

8,000 -

6,000 a

4,000 .-

Mean abundance per meter2

2,000 -

 

 

 

 

June July August September
1: Reference marsh - Impacted marsh

Figure 23. Abundance trends, including standard error, of Gastropoda in the sediment
associated community of Cedarville (impacted) and Mackinac (reference) marshes, Lake Huron,
1996. *Mann-Whitney U test: Significant at a = 0.05.

62

   

Percent Abundance

     

 

A) Plant Associated Community
100% l
I
I
80% I
I
I
I
60% I
I
I
40% I
I
I
20% I
_ I
= |
0% . x m I ,
June July Aug. Sept. June July Aug. Sept.
Reference marsh Impacted marsh
B) Sediment Associated Community
100%
I
- . I I
I
80% I
I
I
60% I
I
I
40% I
I
|
20% |
|
I :
0% ' a m
June July Aug. Sept. June July Aug. Sept.
Reference marsh Impacted marsh
Amnicolalimosa Physa gyn'na
1:21 Other

- Promenefus exacusous m Gyraulus parvus

Figure 24. Community composition of the Gastropoda community based on abundance
at Cedarville (impacted) and Mackinac (reference) marshes. Lake Huron, 1996.
A) Plant associated community and B) Sediment associated community.

63

24). Physa gyrina, and Amm'cola limosa were also dominant Gastropoda in the plant
associated communities at both marshes (Figure 24A). Gastropoda was represented by eight
species in the plant associated community at the reference marsh, and ten species in the
impacted marsh (Table 11). Sediment associated Gastropoda was represented by five taxa
at the reference marsh, and seven taxa at the impacted marsh (Table 11). Gastropoda were
more abundant in both sediment and plant associated communities at the impacted marsh,
with significant difi‘erences (p<0.05) occurring in July and August in the plant associated
community, and August and September in the sediment associated community (Figures 22
and 23).

Gastropoda were present in relatively low numbers in the plant associated community
early in the summer and increased to a peak abundance of approximately 200 snails per dip-
net sample in September at both the reference and impacted marsh (Figures 22 and 23).
Gastropoda in the sediment of the reference marsh, never represented more than 6% of the
macroinvertebrate community but were more abundant and represented up to 20% of the
sediment macroinvertebrate community at the impacted marsh (Figure 7B). G. parvus
dominated the plant associated Gastropoda community of the impacted marsh fi'om July until
September (Figure 24A). Gastropoda collected in the June plant associated samples of the
impacted marsh was more evenly distributed, P. gyrina, A. limosa and G. parvus all
represented around 30% of the Gastropoda community composition (Table 11 and Figure
24A).

Amphipoda were more abundant in the sediment and plant associated communities

at the impacted marsh compared with the reference marsh (Table 11, Figures 25 and 26).

64

 

400

300

200

100

Mean abundance per sweep sample

 

 

 

June July August September
I: Reference marsh - Impacted marsh

Figure 25. Abundance trends, including standard error, of Amphipoda in the plant associated
community of Cedarville (lmpacted) and Mackinac (reference) marshes, Lake Huron, 1996.
'Mann-Whitney U test: Significant at a = 0.05.

 

6,000 ..

4,000

2,000

Mean abundance per meter2

 

 

 

 

 

June July August September
I: Reference marsh - Impacted marsh

Figure 26. Abundance trends, including standard error, of Amphipoda in the sediment
associated community of Cedarville (impacted) and Mackinac (reference) marshes, Lake Huron,
1996. *Mann-Whitney U test: Significant at a = 0.05.

65

Significant differences (p<0.05) between the two marshes occurred in July, August and
September in the plant associated community, and in September in the sediment associated
community (Figures 25 and 26). Amphipoda abundance was relatively low in the plant
associated community of the reference marsh in June and July, and increased in August and
September reaching a maximum mean abundance in September of 78 individuals per sample
(Figure 25). Amphipoda abundance in the plant associated community of the impacted marsh
also reached high abundances in August and September at levels nearly three times greater
than those exhibited at the reference marsh (Figure 25). Amphipoda density in the sediment
community remained relatively stable averaging around 1,400 individuals 111'2 at the reference
marsh, except during June where they were at their lowest mean density of 424m2 (Figure
26). Sediment associated Amphipoda density at the impacted marsh was greater than at the
reference marsh on every sampling date and reached a maximum mean density of
approximately 4,500-m'2 in September, the only date where differences were significant at the
p=0.05 level (Figure 26). The overall dominant Amphipoda in both the sediment and plant
associated community at the impacted marsh was Gammarus spp., while Hyalella azteca was
dominant in the reference marsh (Table 11). Gammarus spp. abundance was 45 fold greater
at the impacted marsh in plant associated samples and 8 fold greater at the impacted marsh
sediment samples compared to the reference marsh (Table 11) and these differences were
significant at the p=0.05 level. H. azteca abundance was also greater at the impacted marsh
compared to the reference marsh but difierences were less dramatic (Table 11) and were not
significant.

Isopoda were much larger components of the plant associated community at the

66

impacted marsh than at the reference marsh with over 2,000 individuals collected, compared
with only 730 individuals collected in the plant associated community of the reference marsh
(Table 11). Even so, differences between the two sites were only significantly different during
August (Figure 27) because of high variance. Isopoda abundance was also higher in the
sediment community of the impacted marsh compared to the reference marsh with an average
density of 1,534-m’2 at the impacted marsh, compared with 912-m‘2 in the sediment
community of the reference marsh but these differences were not significant (p=0.05) (Table
11 and Figures 27 and 28). Two species of Isopoda were collected in both marshes, Lirceus
Iinealus and Racovilzai racovitzai (Table 11). L. lineatus was the dominant Isopoda in the
plant associated community at both marshes (Table 11), while R. racovitzai was the dominate
species in the sediment community of both marshes (Table 11).

The remainder of the taxa occurred in low abundances in the plant and sediment
communities in both marshes (Table 11). Four species of Leeches (Hirudinea) were collected
in the plant associated samples at the impacted site; Mooreobdella spp. , AIbogIossiphonia
heteroclita, Batrachobdella phalera and Helobdella stagnalis (Table 11). Hirudinea were
represented by two families with only one identifiable genera, Reromjyzon spp., in the plant
associated community of the reference marsh (Table 11). Orb-weaving spiders T etragnatha
Iaboriosa commonly found near water, were collected at the reference marsh (Table 11).
Other less common members of the sediment macroinvertebrate community included the
mites (Oribatei and Hydracarina), Sphaeriidae and Tricladida (Table 11).

Zooplankton Community:

While sampling gear was not designed specifically for sampling microcrustaceans,

67

 

 

 

 

 

       

 

 

 

 

 

 

g-‘g, 120 - T
E
8 ,_
Q 100 -
a:
o
E 80 -
8 __
8 60 -
c ‘88:: 21‘ ..
‘3 :5.- “A
E . .
:l 40 ‘ 1 ~
'0 a. q:
to
c
8 20
5 ............
0 32823352443445; -,.
Angus September

 

Reference marsh 12:1 Impacted marsh

Figure 27. Abundance trends, including standard error, of Isopoda in the plant
associated community of Cedarville (impacted) and Mackinac (reference) marshes
Lake Huron, 1996. *Mann-Whitney U test: Significant at or = 0.05.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

4,000 - ‘I
To
0) "
5 3,000 - 1'
8. __,
8
8 2,000 -
'D
C
3
n ..
(U
5 1,000 -
Q)
E

0
July
. Reference marsh 12:1 Impacted marsh

 

Figure 28. Abundance trends, including standard error, of Isopoda in the sediment community
of Cedarville (impacted) and Mackinac (reference) marshes Lake Huron, 1996.
‘Mann-Whitney U test: Significant at a = 0.05.

68

many were collected in the dip-net and core samples. Cladocera and Copepoda were
identified to the species level at the site nearest the source of discharge in both the dip-net and
core samples of the impacted and reference marsh in June and September and are included
here (Table 16), but results should be viewed as biased by the sampling gear used.

Cladocera was consistently the most abundant group of zooplankters in the plant
associated community in both marshes (Table 16) with Bosmina longirostrr's being the most
common species collected. A major portion of the sediment associated zooplankton
community at both marshes was composed of Ostracoda and Copepoda. The Cyclopoida
Copepoda Diacyclops thomasi was the most common species collected.
Macroinvertebrate Trends Along a Pollution Gradient:

Ephemeroptera and Trichoptera were selected as pollution sensitive taxa that may
occur in reduced abundances nearest the source of discharge. The abundance of these taxa
at each site was plotted independently along the transect representing a gradient moving away
from the source of discharge. A linear regression showed no consistent pattern in abundance
in response to distance from discharge. Isopoda, Amphipoda, Gastropoda, Oligochaeta, and
Chironomidae were selected as tolerant taxa that may occur in greater abundances nearest the
source of discharge. Once again, the abundance of these taxa were plotted at the sites along
the transect moving away from the source of discharge. No consistent patterns in response
to discharge were apparent. The three sites nearest discharge were then grouped and treated
as replicates and compared to the three sites furthest from discharge. None of the groups
selected demonstrated a significant difi‘erence (p<0.05) in abundance between the sites nearest

discharge compared with the sites filrthest from discharge.

69

Table 16. Relative abundance of Zooplankton collected in dip-net and core samples,
identified near the source of discharge, at Mackinac (reference) and Cedarville (impacted)

marshes, Lake Huron, September 1996.

 

Taxa

Dip-net Samples

Core Samples

 

 

 

 

 

Reference Impacted Reference Impacted
marsh marsh marsh marsh
June Sept. June Sept. June Sept. June Sept.
CLADOCERA
Bosminidae
Bosmr'na longirosm's (OF. Muller) 57% -- 64% 20% -- -- -- --
Chydoridae
Alana guttata (Sars) 2% 15% -- -- -- -- 7% --
C amptocercus rectirosm's (Schodler) 8% 3 0% 4% 19% 67% 1 7% 5% --
Daphm'a pulex (Leydig) 5% -- -- 6% -- -- -- 2%
Diaphanosoma birgel' (Fisher) 1 5% 20% 4% 17% -- 1 7% 12% --
COPEPODA
Cyclopoida
Diacyclops nanus (Sars) -- -- 2% -- -- -- 4% --
Dr‘acyclops thamasr‘ (S. A. Forbes) 10% 25% 1 5% 28% 3 3% 49°/o 53% 86%
Tropocyclops prasinus 1% -- - 1 0% -- 1 7% -- 5%
Mesocyclops edax (S. A. Forbes) 1% -- 2% -- -- -- 2% --
Harpactacoida 1% -- 5% -- -- -- -- 2%
Leptodr'aplomus minutus (Lilljeborg) -- -- 1% -- -- -- -- --
Copepodite nauplii -- 10% 3% -- -- -- 17% 5%

 

7O

Proximity to discharge does not appear to impact abundance of sensitive or tolerant
macroinvertebrates. Impacts on the macroinvertebrate community in response to sewage
effluent afl‘ected the overall marsh community and were not restricted to areas near the source
of discharge.

Discussion

The macroinvertebrate community composition was dominated at both the reference
and the impacted marsh by aquatic insects, Oligochaeta and Gastropoda. The impacted marsh
showed signs of moderate degradation in community structure and composition in both the
plant and sediment associated communities compared to the reference marsh. The aquatic
insect community was a more taxa rich community at the reference marsh than it was at the
impacted site in the plant and sediment associated communities. Aquatic insects represented
a smaller fi‘action of the macroinvertebrate community in the plant and sediment community
of the impacted marsh compared to the reference marsh (Figure 7). The smaller proportion
of aquatic insects at the impacted marsh compared with the reference marsh was the result
of reduced abundance of aquatic insects, and higher abundances of non-insect groups such
as Amphipoda, Oligochaeta, and Isopoda (Figure 7). Sediment associated Gastropoda
(p=0.002) and Nematoda (p=0.005) represented a greater portion of the macroinvertebrate
samples at the impacted site than they did at the reference site, with Nematoda dominating
the macroinvertebrate fauna in June (Figure 7B).

Ephemeroptera, an order of aquatic insects commonly used to monitor water quality
based on their sensitivity to disturbance (Plafkin et a1. 1989), had significantly (p<0.05) lower

taxa richness and were significantly (p<0.05) less abundant in the plant and sediment

71

communities of the impacted marsh compared with the reference marsh (Figures 12 and 13,
Table 2). Caenis spp., a relatively tolerant Ephemeropteran (Hilsenhoff 1977), dominated the
Ephemeropteran sediment and plant community at both marshes; however, it was present in
significantly (p<0.05) reduced numbers in the impacted marsh (Table 2). The pollution
sensitive, Hexagenia Iimbata (Britt 1955), comprised 7% of the Ephemeroptera plant
associated community, and 20% of the sediment community at the reference marsh in August,
and was never collected at the impacted marsh (Table 2). The sediment associated
Ephemeroptera community at the impacted marsh reflected the greatest signs of degradation.
Ephemeroptera were only collected in the June sediment samples at the impacted marsh,
however they were present in relatively large numbers in every month of sampling at the
reference marsh. Several taxa were present throughout the summer in the reference marsh,
but Caenis amica was the only species of Ephemeroptera collected from the sediments at the
impacted marsh (Table 2).

Trichoptera, also a group sensitive to disturbance in lotic systems, had lower taxa
richness and were less abundant in both the sediment (p=0.04) and plant associated (p=0.02)
samples from the impacted marsh than they were in the reference marsh (Figure 9 and Table
2). The dominant Trichoptera species at the impacted marsh was always a member of the
family Hydroptilidae (Table 2); however, the reference marsh was dominated at different
times by species of Leptoceridae and Phryganeidae (Figure 16). All three of these
Trichoptera groups were assigned the same tolerance values for lotic systems by Plafldn et
al. (1989). However, the change in species dominance for these marshes may demonstrate

a more diverse and possibly more dynamic system at the reference marsh. Although

72

Trichoptera was uncommon or scarce in the sediment samples of both the impacted and
reference wetland, Trichoptera taxa richness was significantly (p<0.05) greater at the
reference marsh compared to the impacted marsh. Sediment associated Trichoptera reached
a peak density in both marshes in September, but the density of Trichoptera was almost five
fold greater (p=0.04) at the reference marsh than it was at the impacted marsh with a
maximum density of 3,053 individuals-ma, compared with a maximum density of 763
individualsm‘2 at the impacted marsh (Figure 15).

Several groups of taxa that are characterized as tolerant to disturbance in lotic systems
exhibited similar trends of higher relative abundances and total numbers in response to
disturbance in the plant associated community of these coastal marshes. Chironomidae made
up a larger percentage of the insect community, and were more abundant at the impacted
marsh than at the reference marsh (Figures 9A). The tribe Chironomini, which includes such
pollution tolerant genera as blood worms (Chironomus), was consistently the dominant group
of Chironomidae in the plant associated community of the impacted marsh, but was the
dominant Chironomidae only in June and September at the reference marsh (Tables 3 and 4).
Amphipoda and Isopoda, which are commonly associated with organic pollution loading in
lotic systems (Plafldn et al. 1989), were significantly more (p=0.005) prominent in the plant
associated community at the impacted site than at the reference site (Figure 7A). Amphipoda
abundance was more than five times greater (p=0.004) in the plant community of the
impacted marsh as compared to the reference marsh (Figure 7A).

Although aquatic insects usually dominated the macroinvertebrate community at both

the impacted and reference marsh, the impacted marsh was dominated during July by

73

Oligochaeta (Figure 7A). Oligochaeta were more abundant in the plant associated community
at the impacted marsh than at the reference marsh for every sampling period. Oligochaeta
derive most of their nutrition from bacteria and are able to withstand considerable oxygen
depletion and are fiequently associated with organic pollution loading in streams (Brinkhurst
and Cook 1974).

Significant difi‘erences in the abundance of tolerant taxa were also detected (p<0.05)
in the sediment community between the impacted and reference marshes. Stream
communities with 90% of the aquatic insect community represented by Chironomidae are
generally impaired systems (Hilsenhoff pers. comm). Chironomidae represented nearly 90%
of the aquatic insect community at the impacted marsh compared to approximately 60% at
the reference marsh (Figure 9B). Isopoda, which are assigned a very high pollution tolerance
value in Hilsenhoff's (1988) Family Biotic Index, were also a larger component of the
macroinvertebrate community of the impacted marsh compared with the reference marsh
(Figures 26 and 27). Over the course of the sampling season, nearly twice as many Isopoda
were collected fiom the impacted marsh as were collected from the reference marsh.
Amphipoda was a significantly (p<0.05) larger component of the macroinvertebrate
community within the sediment of the impacted marsh compared to the reference marsh
(Figure 7B). Oligochaeta or "sludge worms" were a major part of the sediment community
at both the impacted and reference marshes. Oligochaeta were more abundant in the sediment
community at the impacted marsh than they were at the reference marsh every month of
sampling with the exception of June (Figure 20). Oligochaeta reached a maximum mean

density of 3,652-m‘2 in the sediment of the reference marsh during June (Figure 20). Kairesalo

74

and Koskimies (1987) found a similar peak in Oligochaeta densities in early June. However,
Oligochaeta in the impacted marsh sediment reached a maximum mean density of 4,184-m'2
in September, the month in which sewage emuent from the wastewater treatment lagoon was
discharged, and these differences were not significant (p=0.30).

Gastropoda were more diverse and abundant in the plant and sediment community of
the impacted marsh than they were in the reference marsh (Table 11). Mason et al. (1970)
found Physa spp., F errissr'a spp., and Promenetus spp. were often found in moderately to
grossly polluted situations. In this study Promenetus exacuous was very common in the
sediment of the impacted marsh, yet was only found at the reference marsh in September and
in low numbers (Figure 24B). Phym spp. and F errr'ssia spp. were also more abundant in the
sediment and plant associated community at the impacted marsh compared to the reference
marsh (Table 11). Berg and Ockelmann (1959) have shown that many Gastropoda can
maintain oxygen consumption, despite a lowering of dissolved oxygen, until a critical
threshold is reached, suggesting that many Gastropoda may be able to withstand brief periods
of low dissolved oxygen.

Smock and Stonebumer (1980), found that macroinvertebrate densities increased with
the onset and progressive senescence of vegetation. The plant associated macroinvertebrate
data from both the reference and impacted marsh supported their suggestion (Figure 8).
Total macroinvertebrate abundances increased throughout the sampling season reaching a
peak abundance in September corresponding with the onset and senescence of S. acutus at
the reference marsh (Figure 8A). However, peak abundance of aquatic insects corresponded

with the onset and senescence of S. acutus only at the reference marsh (Figure 8A). At the

75

impacted marsh, aquatic insects reached peak abundance in August, a month earlier than at
the reference marsh. In September, the reference marsh had nearly twice the number of
aquatic insects (p=0.29) when compared to the plant community of the impacted marsh.
Many of aquatic insects at the reference marsh were Ephemeroptera and Trichoptera, and
these taxa were significantly (p<0.05) less abundant at the impacted site (Figure 9A). The
aquatic insects at the impacted marsh reached maximum abundance before discharge fi'om the
wastewater treatment lagoons occurred in August, whereas they reached maximum

abundance in the reference marsh in September.

CHAPTER TWO

IMPLICATIONS FOR THE DEVELOPMENT OF A MULTIMETRIC INDEX OF
ECOLOGICAL INTEGRITY

Introduction

Early biological monitoring programs consisted of biotic indexes that were primarily
sensitive to organic effluent and sedimentation (Kolkwitz and Marsson 1908 as cited in Kemp
et al. (1967)). The most common approach in early biological monitoring of aquatic systems
involved the ranking of taxa, typically family, genera or species of limited numbers of major
groups of organisms such as invertebrates or fish, where taxa were assigned numerical values
on a scale ranging from pollution intolerant to pollution tolerant (Karr and Chu 1997). An
average pollution tolerance level was then assessed for each sample site (Hilsenhoff 1977,
1982 and 1987, Armitage et al. 1983). These biological indices were limited in their ability
to detect degradation from toxins pollution or from altered physical habitat or flow (Karr and
Chu 1997). This led to the use of toxicity bioassay approaches, which typically examined the
tolerances of only a few species, usually the most tolerant taxa, to a known toxicant. This

bioassay, toxicity approach underestimates the effects of contaminants (Karr and Chu 1997).

76

77

The need for more comprehensive monitoring systems led to the development of the
multimetric index based on several major groups of organisms and/or physical habitat
characteristics (Karr and Chu 1997).

The development of multimetric indices of biological condition gained favor in 1981
with Karl’s (1981) fish based, Index of Biological Integrity (IBI). This first multimetric index
was designed to include a range of attributes (a measurable component of a biological
system), of fish assemblages for use in monitoring the health of stream systems (Kan 1981
and 1986). Shortly thereafter, Ohio EPA (1988) adopted the concepts involved in the IBI to
develop an invertebrate community index (ICI) evaluation system using benthic invertebrates.
A multimetric index has also been developed to use invertebrates to assess the water quality
for rivers of the Tennessee Valley (Kerans and Karr 1994). Multimetric biological indexes
are currently used in monitoring the health of streams in 42 states, and 6 additional states are
in the process of developing them (Kerr and Chu 1997). Multimetric indices encompass
several attributes of the sampled assemblage, including taxa richness, indicator taxa or guilds,
(tolerant and intolerant groups), health of individual organisms, and assessment of processes
(Kan and Chu 1997). A multimetric index that includes a variety of such metrics integrates
information fiom ecosystem, community, population and individual organism levels (Gerritsen
1995, Karr 1991, Barbour et al. 1995), and it can be expressed in numbers and words.

Changes in the physical, chemical, and biological environment resulting from human
activities alter assemblages. These changes may be changes in species composition, species
richness, or trophic structure, such as a decrease in top carnivores or an increase in

omnivores; or shifts fiom specialist to generalist in food or reproductive habitats, reflecting

78

shifts in food web organization, including energy flow and nutrient cycling (Karr and Chu
1997). Multimetric indices incorporate this information by including metrics such as the
percentage of predators, omnivores, or trophic feeding groups (Karr and Chu 1997). A
multimetric index should be able to demonstrate that an attribute, has a consistent quantitative
change across a range, or gradient of human influence (Karr and Chu 1997). They should
increase or decrease as human influences increases. They should be sensitive to a range of
biological stress. Multimetric biological indices are now well documented as effective in the
assessment of ecological condition in a variety of management settings, with many taxa, and
in diverse geographic regions (Karr and Chu 1997). Multimetric indices are predominantly
used in the assessment of lotic systems, yet may be a usefill tool in the assessment of lentic
or wetland systems. The objectives of this study are to select and test various metrics for use
in the future development of a multimetric index of ecological integrity for use in northern
Lake Huron coastal marshes.
Methods
1. Metric selection

I considered 38 stnrctural or functional measurements of the macroinvertebrate
assemblages that had been used or suggested by other authors as having potential use as
biological metrics (Table 17) (Resh et al. 1995, Kerans et al. 1992, Barbour et al. 1996, Resh
et al. 1995, Kerans and KalT 1994). Potential metrics included elements of macroinvertebrate
community structure and composition, including measures of the trophic and firnctional
composition of the assemblage considered to be indicative of ecological processes (Karr and

Chu 1997).

79

Table 17. Definitions of potential metrics and expected direction of metric response to
increasing perturbation in Northern Lake Huron coastal marshes (selected and modified

from Karr and Kerans 1992).

I Category I Potential Metrics I Hypothesized effect of impact I

Community structure and composition
Richness measures

Ephemeroptera plus Trichoptera Decrease
Number of Crustacea plus Mollusca Decrease
Number of Diptera Decrease
Number of Ephemeroptera Decrease
Number of families Decrease
Number of Trichoptera Decrease
Total abundance Decrease
Total taxa richness Decrease
Enumerations
Proportion of individuals as Amphipoda Decrease
Proportion of individuals as Chironomidae Increase
Proportion of individuals as Chrionomini Increase
Proportion of individuals as Cnistacea plus Mollusca Decrease
Proportion of individuals as Diptera Increase
Proportion of the dominant taxon Increase
Proportion of individuals as Ephemeroptera Decrease
Proportion of individuals as Gastropoda Decrease
Proportion of individuals as Isopoda Increase
Proportion of individuals as Odonata Increase
Proportion of individuals as Oligochaeta Increase
Proportion of individuals as Orthocladiinae Decrease
Proportion of individuals as Tanytarsini Decrease
Proportion of individuals as Trichoptera Decrease
Proportion of individuals as Tubificidae Increase
Proportion of individuals as Sphaeriidae Decrease
Proportion of individuals as Sty/aria spp. Increase
Trophic and Functional composition
Number of scraper plus piercer taxa Decrease
Proportion of individuals as collector-gatherers Increase
Proportion of individuals as filterers Decrease
Proportion of individuals as predators Decrease
Proportion of individuals as scrapers Decrease
Proportion of individuals as shredders Decrease
Ratio of scrapers/collector-filterers Increase
Community diversity and similarity indices
Coefficient of community loss Decrease
Evenness Decrease
J accard coefficient Decrease
Margalef diversity Decrease
Shannon diversity Decrease
Simpson diversity Decrease

80

The first group included elements on macroinvertebrate community structure and
composition consisting of richness measures and enumerations (taken from Kerans et al.
1992) (Table 17). Metrics based on measures of community structure and composition were
selected to provide information on the make-up of the assemblage and the relative
contribution of the population to the total fauna. Richness measures included metrics based
on macroinvertebrate taxa that are generally regarded as sensitive to anthropogenic
disturbances including total abundance. Loss of taxa in sensitive groups is an indication of
perturbation (Wallace et a1. 1996, Barbour et al. 1996). Metrics of sensitive
macroinvertebrate taxa included the total number of Odonata taxa and the total number of
Crustacea plus mollusca taxa (Karr and Chu 1997). Richness metrics considered were also
based on the presence of EPT (Ephemeroptera, Trichoptera and Plecoptera) taxa which have
been found to be universal and applicable in monitoring the health of many stream systems
(Plafldn et al. 1989, Barbour et al. 1996). However this metric was modified to exclude
Plecoptera due to lack-of or low number of the predominantly lotic Plecoptera, within the
northern lake Huron coastal marshes. Resulting metrics were measurements of
Ephemeroptera plus Trichoptera taxa richness, and the number of Trichoptera taxa and
Ephemeroptera taxa independently. Richness metrics were also examined which have been
shown to decrease as human disturbances increases and included measures of total taxa
richness and total number of families (Fore and Karr 1996, Barbour et al. 1996, Resh et al.
1995) (Table 17).

Community structure and composition metrics of relative abundance (enumerations)

are based on the premise that a healthy and stable assemblage will be relatively consistent in

81

the proportional representation of major taxa, though individual abundances may vary in
magnitude. Metrics of relative abundances of taxa that have been shown to decrease in the
presence of perturbation, include percent Ephemeroptera, Trichoptera, Amphipoda, Crustacea
plus Mollusca, Gastropoda, Sphaeriidae, and Tanytarsini (Fore and Karr 1996, Barbour et a1.
1996, and Resh et al. 1995). Metrics considered based on the relative abundances of taxa that
would most likely increase in the presence of perturbation were percent composition of
Oligochaeta, Tubificidae, Srylan'a spp., Odonata, Diptera, Chironomidae, Orthocladiinae,
Chironomidae, and Isopoda (Barbour et a1. 1996). Although, certain individual taxa within
each group may be relatively sensitive to pollution they typically should have no effect on the
overall result of the metric (Barbour et al. 1996). Percentage of the dominant taxon was also
measured which is a simple measure of redundancy. A high level of redundancy corresponds
with a pollution tolerant organism dominating a community resulting in a lower overall
diversity (Plafldn et al. 1989, Barbour et a1. 1996).

Metrics on the trophic and functional composition of the assemblage were selected
based on the premises that trophic metrics are surrogates of complex processes such as
trophic interaction, production, and food source availability (Karr et al. 1986, Cummins et al.
1989, Plaflcin et al. 1989, Barbour et al. 1996) (Table 17). The majority oftrophic metrics
were evaluated as relative abundance (percentage). Metrics considered included relative
abundances of sensitive organisms such as specialized feeders including scrapers, piercers,
filter feeders and shredders which are expected to decrease with increasing disturbance
(Barbour et al. 1996, Wallace et al. 1996). The number of scraper plus piercer taxa was also

included. This metric includes taxa that feed primarily on diatoms (scrapers) and living

82

macrophytes (piercers) and is expected to decrease with increasing perturbation (Barbour et
al. 1996). Generalist consumers (collector-gatherers) have a broader range of acceptable
food materials than do specialist consumers and are, therefore, more likely to be tolerant to
pollution that might alter the availability of certain foods. Therefore, a metric of the relative
abundance of collector-gatherers is expected to increase in the presence of perturbation
(Cummins et al. 1989). The ratio of scrapers to collector-filterers was also considered as a
potential metric to reflect available food resources. The dominance of collector-filterers may
reflect organic enrichment (Resh and Jackson 1993). In addition, four firnctional feeding
group metrics suggested by Merritt et al. (1996) were used to evaluate several ecosystem
attributes (Table 18).

Community diversity and similarity indices were examined as potential metrics, to
reflect the diversity of the aquatic assemblages (Resh et al. 1995). Three diversity indices,

one evenness index, the Coefficient of Community Loss, and the Jaccard Coeficient were

apphed.
(1) Shannon Index H'=-2Pilog2Pi.
(2) Simpson's Index 1-D=1-ZNi(Ni-l)/N(N-1)
(3) Margalefs Index D=S-1/Ln(N)
(4) Evenness Index J'=H‘/Log2S
(6) Coefficient of Community Loss d-a/e
(7) Jaccard Coefficient a/ a+b+c
where
N,= number of individuals of species 1
= the total number of individuals in a sample
1: Ni/N
= total species number
a= number of taxa common to both samples
= number of taxa present in sample B but not A
c= number of taxa present in sample A but not b.
d= total number of taxa present in sample a

e= total number of taxa present in sample b

83

Table 18. Relationships between macroinvertebrate fiinctional groups and ecosystem
attributes for which they can serve as analogs (modified after Merritt et al. 1996).

 

 

 

 

 

ECOSYSTEM METHODS FUNCTIONAL EXPECTED RATIOSl
A'ITRIBUTES GROUP RATIOS
Reference Impacted
Herbivore Shredders (Live Vase.
As a proportion of P/R measurements Plants) + Scrapers
detritivore per unit area on a As a proportion of Heterotrophic Autotrophic
as a surrogate for daily basis Shredders (CPOM <.75 2.75
P/R Detritivores) + Total
Collectors
Suspended
Particulate Organic SPOM Measured Not Enriched Enriched
Matter per Unit Volume and Filtering Collectors in Suspended In
As 0 Proportion of BPOM Measured As a Proportion of Particulate Suspended
Deposited per Unit Area Gathering Collectors Organic Particulate
(Benthic) Matter Organic
Particulate Organic <.SO Matter
Matter 2.50
SPOM/BPOM
Measures Available
Surfaces for Stable
Habitat (Substrate) Attachment Scrapers + Filtering Stable Unstable
Stability (Sediment Coarser Collectors Substrates Substrates
HABITAT than Moved by As a proportion of >60 5.60
STABILITY Maximum Transport Total Shredders +
Velocity, Large Gathering Collectors
Woody Debris,
Rooted Vascular
Hydmphytes)
High ratio of slow Predators Normal Top Sensitive
Top Down Control turnover predators As a proportion of Down Species
TOP DOWN indicates high Total of All Other Predator Afi‘ected
proportion of fast Functional Feeding Control 2.15
turnover Groups <.15

 

' The proposed ratios are based on values calculated from the literature for lotic habitats (Merritt et al. 1996).

 

84

The Shannon Index is a measure of information content, and incorporates both
richness and evenness in its formula (Merritt et al. 1996). Simpson's Index takes into account
both the abundance patterns and the species richness (Cao et al. 1996). The Margalef formula
differs from the other two in that is does not contain an evenness component (Menitt et al.
1996). Jaccard Coefficient of Community similarity measures the degree of similarity in
taxonomic composition between two sites in terms of taxon presence or absence. The
Jaccard Coefiicient discriminates between highly similar sites. Coefficient values, ranging
fiom 0 to 1.0, increase as the degree of similarity with the reference station increase (J accard
1912, Ohio EPA 1987). The Coefficient of Community Loss Index measures the loss of
benthic taxa between a reference site and the site of comparison, it is an index of
compositional dissimilarity, communities are expected to become more dissimilar as stress
increases with values increasing as the degree of dissimilarity with the reference site increases
(Courtemanch and Davies 1987).

2. Metric testing

Macroinvertebrate data from dip-net samples (plant associated communities, chapter
one) were used to test the various metrics and indices for reliability as water quality indicators
between the reference and the impacted marsh. A metric should be capable of distinguishing
differences in anthropogenic impacts among marshes. Therefore, the initial step was to select
metrics based on low within site variability, followed by a determination of the metric's ability
to detect the hypothesized effect of impact, and finally to examine the sensitivity of the
potential metrics. Potential metrics were calculated for two months, August and September.

Months for metric testing were selected based on a greater abundance and diversity of

85

macroinvertebrates, and a higher frequency of late instar larvae, which aid in a more accurate
identification of most taxa. Macrophytes also contained the largest amount of periphyton at
this time, which is a major food source for many macroinvertebrates, including the scrapers.
August samples were used to test whether long-term effects would be detectable (no
wastewater discharge had occurred since June 5). September samples should reflect both
short and long term impacts of wastewater discharge, since wastewater was being discharged
when samples were taken.

The first step was to select metrics that had minimal variability from within-marsh
factors so that differences resulting fiom anthropogenic impacts could be detected. A highly
variable metric within the reference site would not useful because it would have low
discriminatory power between the impaired and the unimpaired sites (Barbour et a1. 1996).
Conversely, a metric with a narrow variance within the reference marsh is potentially useful
for detecting biological change in response to disturbance. This was determined by measuring
within-site variability of the potential metrics. Coefficients of variation for each potential
metric were calculated among the nine sites within each marsh (Karr and Chu 1997). Metrics
were considered to have high variability if their coefficients of variations were greater than
50%, indicating high within-site variability. Chemical analysis showed that impact from
sewage effluent to be greatest at sites less than 260 meters from discharge (at the impacted
marsh). This type of point source disturbance could potentially cause high within site
variability at the impacted marsh. Therefore, only coefficient of variations for the reference
marsh were used as a method of screening potential metrics, because the reference marsh was

not influenced by point source pollution. Metrics were not considered for further

86

consideration as candidate metrics if a high coefficient of variation was detected in either
August or September at the reference marsh.

Following this initial screening and selection, potential metrics were evaluated to
determine if the hypothesized effect of impact was detected by the metric. Metrics were
calculated for each marsh by pooling the macroinvertebrates from all nine sites. Results from
both marshes were than plotted next to each other with box-and-whisker plots (e. g. Figure
29). The positioning of the box-and whisker plots from each metric were compared between
marshes to determine if the metric either agreed with the predictions of the hypothesized
effect of impact, measured no effect of impact, or measured the opposite effect fiom that
predicted based on impacts known to occur in streams. Metrics which detected the
hypothesized efl‘ect of impact in August and September were selected as metrics likely to pick
up long term impacts on the system. Metrics which detected the hypothesized effect of
impact during wastewater discharge in September, but measured no effect of impact in
August were selected as potential metrics of short term impacts on these systems. Metrics
which measured the opposite effect of impact in either month were not considered as
candidate metrics but recommended for further analysis.

The sensitivity of each metric (its ability to discriminate between the reference and the
impacted site) was then judged according to the degree of interquartile overlap in box-and-
whisker plots (Barbour et al. 1996). Metrics were judged to have one of four sensitivity
values: a sensitivity value of 3 if no overlap existed in the interquartile range; a sensitivity of
2 if there was some overlap that did not extend to the medians; a sensitivity of 1 if there was

a moderate overlap of interquartile ranges but at least 1 median was outside the range; and

87

 

 

 

 

 

3 points = No overlap of .1.
interquartile ranges

 

 

 

 

 

 

2 points = Some overlap of interquartile
ranges but both medians are l 1 T

outside the interquartile range overlap.

 

 

 

 

 

 

 

1 points =. Moderate overlap of 1
interquartile ranges but at least
one median is outside the interquartile

 

 

[:1-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

range overlap. I

a. b I
0 points = a. Extensive overlap of .1.
interquatile range or b. both ‘ I‘LI
medians within the overlap. I—I—I I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 29. Evaluation of sensitivity of the metrics. Range bars show maximum and minimum of non-out-
liers; Solid lines inside the box are medians; boxes are interquartile ranges (25%ile to 75%ile), (Taken
from Barbour et al. 1996).

88

a sensitivity of zero if interquartile overlap was considerable, with no discrimination between
reference and impaired sites (Barbour et al. 1996) (Figure 29).

In addition, potential metrics were statistically compared between the two marshes
in order to determine their ability to detect a significant difference between the two sites.
Because metrics were often proportions, they tended not to be normally distributed, so the
normality of the distribution of each metric was tested before using a parametric statistical test
(Kerans and Karr 1994). Ifdistributions were non-nonnal (p<0.05), richness data were log-
transformed, and proportions were arc-sine transformed and the test repeated. If
transformation was unsuccessfirl, a nonparametric Mann-Whitney U test was used for
statistical comparisons between sites on a monthly basis.

Metrics were considered candidate metrics based on their ability to consistently detect
the hypothesized effect of impact between marshes, and consistently have a low coefficient
of variation indicating low within site variability. Statistical comparisons (Mann-Whitney U
test) were not used as a means of metric elimination, but were instead used to support
previously determined sensitivity values. Final selection of metrics was based on the graphical
analysis (Figure 29), because it provided more insight into biology than a simple p-value
could.

Results

From an original suite of 38 metrics, six (16%) metrics met the strict criteria of low
within site variability in the reference marsh, and no overlap of interquartile ranges between
the two marshes for both months (Tables 19 and 20). Thus, these metrics were able to

unambiguously separate the impacted from the reference marsh and are excellent candidate

89

Table 19. Potential metrics and their sensitivity values base on Barbour et al. 1996, Coefficient of variation
based on Mackinac values; low (C V<.50), high (CV>.50). From Cedarville and Mackinac marshes, August
1996. Metrics underlined indicate candidate metrics.

 

Potential Metrics Sensitivity Coefficient of Hypothesized effect of
value Variation im act

Community structure and composition
Richness measures

No. of Cnrstacea + Mollusca taxa 3* High Opposite effect
No. of Diptera taxa 0 Low No effect
No. of Ephemeroptera taxa 3 * High Agrees
No. of Ephemeroptera + Trichoptera taxa 3* High Agrees
No. of families 3* Low Opposite effect
No. of Trichoptera taxa 0 Low No effect
Total abundance 3* High Opposite effect
Total taxa richness 3 Low Opposite effect
Enumeration-Proportion of individuals as:
Amphipoda 3* Low Opposite effect
Chironomidae 0 Low No effect
Chironomini 0 Low No effect
Crustacea + Mollusca 3* Low Opposite effect
Diptera 0 Low No effect
Dominant taxon 0 Low No effect
Ephemeroptera 3* Low Agrees
Gastropoda 3* Low Opposite effect
Isopoda 3* Low Agrees
Odonata 3* Low Opposite effect
Oligochaeta 0 Low No effect
Orthocladiinae 3* Low Agrees
Siylaria 2* Low Opposite effect
Sphaeriidae 3 * Low No effect
Tanytarsini 0 Low No efi°ect
Trichoptera 3* Low Agrees
Tubificidae 0 Low No eflect
Trophic and Functional composition
Collector-gatherers 3* Low Opposite effect
Filterers 1 Low Agrees
Habitat stability 3 * Low Opposite effect
No. of scrapers/collector-filtcrers 0 High No effect
No. of scraper + piercer taxa 3* Low Opposite efiect
Predators 3* Low Agrees
Production/Respiration 3* Low Agrees
Scrapers 3* Low Opposite efl‘ect
Shredders 0 Low No effect
SPOM/BPOM 1 Low No effect
Top down 3* Low Opposite effect
Community diversity and similarity indices
Evenness 3* Low Opposite effect
Margalef diversity 1 Low Agrees
Shannon diversity 3* Low Opposite effect
Simpson diversity 3 * High Agrees

 

*Significant value based on Mann-Whitney U test (P<0.05)

90

Table 20. Potential metrics and their sensitivity values base on Barbour et a1. 1996, Coefficient of variation
based on Mackinac values; low = (CV<.50), high = (C V>.50). From Cedarville and Mackinac marshes,
September 1996. Metrics underlined indicate candidate metrics.

Potential Metrics Sensitivity Coefficient of Hypothesized effect of
value Variation im act

Community structure and composition

 

Richness measures

 

No. of Crustacea + Mollusca taxa 3* High Opposite effect

No. of Diptera taxa 0 Low No effect

No. of Ephemeroptera taxa 3* Low Agrees

No. of Ephemeroptera + Trichoptera taxa 3* Low Agrees

No. of families 0 High No effect

No. of Trichoptera taxa 2 Low Agrees

Total abundance 0 High No effect

Total taxa richness 0 High No effect

Enumerations- Proportion of individuals as:

Amphipoda 3* Low Agrees

Chironomidae 0 Low No effect

Chironomini 1 Low Agrees

Crustacea + Mollusca 3* Low Opposite effect

Diptera 0 Low No eflect

Dominant taxon 0 Low No effect

Ephemeroptera 3* Low Agrees

Gastropoda 0 Low No efi‘ect

Isopoda 3* Low Agrees

Odonata 0 Low No efiect

Oligochaeta 0 Low No effect

Orthocladiinae 3 Low Opposite effect

Stylaria 0 Low No effect

Sphaeriidae 0 Low No effect

Tanytarsini 1 Low Agrees

Trichoptera 3* Low Agrees

Tubificidae 2 Low Opposite effect

Trophlc and Functional composition

Collector-gatherers 2* Low Opposite effect

Filterers 1 Low Agrees

Habitat stability 3 * Low Opposite effect

No. of scrapers/collector—filterers 3 * High Agrees

No. of scraper + piercer taxa 2 High Opposite effect

Predators 3* Low Agrees

Production/Regpiration 3* Low Agrees

Scrapers 3 * Low Opposite eflect

Shredders 0 Low No effect

SPOM/BPOM 0 Low No effect

Top down 3 * Low Opposite effect
Community diversity and similarity indices

Evenness 1 Low Agrees

Margalef diversity 3* Low Opposite effect

Shannon diversity 3 * Low Agrees

Simpson diversity 3 High Agrees

 

*Significant value based on Mann-Whitney U test (P<0.05).

91

metrics for use as indicators of long term impacts on water quality in Northern Lake Huron
coastal marshes (Figure 30). Candidate metrics included: (1) proportions of individuals as
Ephemeroptera, (2) proportions of individuals as Isopoda, (3) proportions of individuals as
Trichoptera (4) proportion of individuals as predators (5) proportion of individuals as filter-
feeders, and (5) a ratio of herbivores (shredders of live plants plus scrapers) to detritivores
(shredders of detritus plus total collectors) as a surrogate for the production to respiration
(P/R) ratio (Tables 19 and 20, Figure 30).

Ephemeroptera and Trichoptera were, as expected (Table 17), sensitive indicators of
water quality with proportions decreasing in the impacted marsh (Figure 30). Isopoda, as
expected (Table 17), increased with impacts reflecting their known tolerance to pollution
(Figure 30). Three fiinctional feeding group based metrics also changed in the direction
expected (Tables 19 and 20, Figure 30). The proportion of fauna as predators decreased
(Figure 30), perhaps reflecting loss of sensitive top predators. The increase in the
herbivore/detritivore ratio at the impacted marsh compared to the reference marsh (Figure 30)
is consistent with an expected increase in primary production due to nutrient enrichment in
the impacted marsh (Table 18). The increase in filtering collectors in the impacted marsh
(Figure 30) is also consistent with the increased suspended particulates expected in the
impacted marsh fi'om either increased plankton production and/or particulate organic input
into the marsh fiom the wastewater lagoons.

Another potentially usefirl metric was the number of Trichoptera taxa (Tables 19 and
20, Figure 31). This metric received a sensitivity value of zero and was classified as showing

no effect based on examination of the box-and-whisker plots and p-values in August.

92

 

 

 

 

 

 

 

 

 

40 ' 30 - ’ -
"‘ 3O - . 4 .
g org 2 I a
3 20 _ . 8' 18 - -
o .c
E _ ‘ ,0 12 '- + 1
g 10 i .E _ d
“J 5 - . °\ i
O\° — O I- '1
0 -
40 30

 

 

 

% Isopoda*
U

 

 

 

- to
# 09
% Predators
8

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5
ti! 3 s
.5 8 r ‘
ii
'3 2 - - 8 5 . .
.i i
9' ’ , ‘1; 4 - -
. 1- T - ° ~
I: E 2 - .
1:
E 0
Reference Impacted Reference Impacted

Figure 30. A comparison of candidate metrics of long term impact, for use in Northern Lake Huron
Coastal marshes. Metrics determined from Dip net samples at Cedarville (impacted) and Mackinac
(reference) marshes, Lake Huron, September 1996. Range bars show maximum and minimum of
non-outliers; Solid lines inside the box are medians; boxes are interquartile ranges (25%ile to 75%ile);
dots are outliers. (Taken from Barbour et al. 1996). *Mann-Whitney U test: Significant at or = 0.05.

93

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

August September

10
a 4' +
r .- - - 1 -
s I ~— I
3 6 F' i _. '- —— -l
8 _ ._
.C
s 4 - I - - + -
t: + +
z

o

 

 

Figure 31. A comparison of candidate metrics of short term impacts, for use in Northern Lake Huron
Coastal marshes. Metrics determined from Dip net samples at Cedarville (impacted) and Mackinac
(reference) marshes, Lake Huron, 1996. Range bars show maximum and minimum of non-outliers;

Solid lines inside the box are medians; boxes are interquartile ranges (25%i1e to 75%ile); dots are outliers.
(Taken from Barbour et al. 1996). *Mann-Whitney U test: Significant at or = 0.05.

94

However, in September this metric received a sensitivity value of 2 (Tables 19 and 20, Figure
31). Although this metric was inconsistent between months, it did determine differences in
these particular coastal marshes during the period of wastewater discharge. Therefore, it may
be a valuable metric for detection of short term impacts in these Northern Lake Huron coastal
marshes.

Three additional metrics demonstrated potential for use as indicators of water quality
in northern Lake Huron coastal marshes (Table 21). However, these metrics consistently
showed the opposite effects of impact than predicted (Tables 17-20). These included the
proportions of individuals as scrapers, and the ecosystem attribute metrics of Habitat Stability,
and Top Down Control. These metrics may prove usefiil when investigated fiirther but were
not recommended as candidate metrics based on the unpredicted response.

In a direct comparison between sites, the Jaccard Coefficient of Community similarity
mean value was 0.48 in August and 0.50 in September. The Jaccard Coefficient of
Community similarity measures the degree of similarity between the reference site and the
impacted site. A value close to 1 indicates very similar assemblage in terms of taxon present
or absent. The Coefficient of Community loss index measures the loss of benthic taxa
between the reference site and the site of comparison. The mean Coefficient of Community
loss was 0.34 in August and 0.42 in September. The higher value in September indicated a
loss of benthic taxa compared with August values.

Discussion
Water chemistry (Table A-3, Figures 2-6) and Land use (Table 1) differences between

the reference (Mackinac Bay) and the impacted (Cedarville Bay) marshes indicated that

95

Table 21. Candidate metrics for use in Northern Lake Huron coastal marshes.

 

Candidate Metrics of Long Term Impact- Metrics which detected the hypothesized effect of
impact in August and September
Proportion of individuals as Ephemeroptera
Proportion of individuals as T richoptera
Proportion of individuals as Isopoda
Proportion of individuals as F ilterers
Proportion of individuals as Predators
Herbivore/Detritivore as surrogate for P/R

Candidate Metrics of Short Term Impact- Metrics which detected the hypothesized effect of
impact during wastewater discharge in September, but measured no effect of impact in August)

Number of T richoptera taxa

Metrics for Further Analysis- Metrics consistently demonstrating results opposite than predicted in
stream systems)
Proportion of individuals as scrapers
Habitat Stability
Top Down

 

96

impacts on the Cedarville Bay marsh were likely to be moderate. Thus, metrics for detecting
water quality differences in the wetland had to be quite sensitive to detect differences. Of the
38 potential metrics tested, six metrics (16%) consistently demonstrated that differences did
exist in August, two months after the previous release of treated domestic wastewater from
the treatment lagoons, and in September, during wastewater release fiom these lagoons
(Tables 19 and 20, Figures 30). An additional metric (Figure 31) was potentially useful but
differences were not as great as they were for the first six. Three metrics gave consistent
results that were opposite the differences predicted on the basis of metrics designed for
streams (Karr and Kerans 1992) (Table 21). Thus, ten or 26% of metrics tested have
potential as indicators of coastal wetland integrity (Table 21). These metrics will need to be
tested for a variety of coastal wetlands before being widely accepted. The remaining 28
potential metrics derived from indices developed for wadable streams and rivers (Karr and
Kerans 1992, Plafldn et al. 1989, Ohio EPA 1988) did not appear to be useful for Great Lakes
coastal wetlands (Tables 19 and 20). However, some of these metrics may prove to be useful
after testing over a wider array of wetlands. Additional work will be required before any of
these 28 can be definitely removed fiom consideration. Some of the more promising of these
potential metrics are discussed below.

Total taxa richness is considered one of the most useful indicators of water quality,
in lotic systems (Ohio EPA 1987, Kerans and Karr 1995, DeShon 1995, Fore et a1. 1996).
However in this study, total taxa richness was not considered a usefirl indicator of water
quality because it demonstrated high within site variability, and indicated that the impacted

marsh had higher taxa richness than the reference marsh in August (Table 19). Taxa richness

97

was predicted to decline at the impacted marsh. Taxa richness was higher at the impacted
marsh in August due primarily to an increase in the number of Gastropoda and Lepidoptera
species and the occurrence of four Hirudinea species. No leeches were collected at the
reference site (Table 11). The metric of family taxa richness was also higher at the impacted
marsh in August. During September, both the taxa and family richness metrics measured no
efl‘ect between the two marshes. Based on these results, metrics based on total and family
taxa richness of all macroinvertebrates do not appear to be good metrics for use in these
northern Lake Huron coastal marshes. The moderate inputs of organic pollution in these
marshes may be sufficient to increase the food source for some of the moderately tolerant
macroinvertebrates without causing negative impacts, and may result in greater diversity.
However, measures of taxa richness in systems with higher degrees of degradation, or
modification of these metrics to include just taxa richness of insects and insect families may
prove to be valuable metrics in these coastal marshes.

Kerans and Karr (1994), suggested that Ephemeroptera and Trichoptera taxa richness
measures were excellent attributes for detecting water quality impairments in streams of the
Tennessee Valley. Ephemeroptera taxa richness at the reference marsh was significantly
higher (p<0.05) in September and August than it was at the impacted marsh, with a sensitivity
rating of 3 (Tables 19 and 20). However, it was not selected for a candidate metric because
of high within site variability in August (Table 19).

Fore and Karr (1996) suggested that the number of Crustacea plus Mollusca taxa was
a useful measure of calcium-dependent taxa, and it is predicted to decrease with the increase

in disturbance, due to the presence of sensitive Crustacea and Mollusca taxa (Karr and Kerans

98

1992). However in the Northern Lake Huron marshes, Crustacea and Mollusca taxa richness
was significantly higher in the. impacted marsh both months than in the reference marsh
(Tables 19 and 20). Very few sensitive Crustacea or Mollusca were collected in either the
reference or the impacted marshes, and pollution tolerant snails increased in the impacted
marsh. The inability of the metric to detect the hypothesized impact may be due to the lack
of sensitive stream Crustaceans in these coastal marshes and the presence of pollution
tolerant snails. This metric may be a useful indicator of disturbance in these coastal marshes
if additional research confirms that the proportion of Crustacea plus Mollusca increases with
an increase in disturbance because of the pollution tolerant snail response.

In fact, all metrics which included Gastropoda indicated the opposite effect of impact
from those that were predicted (Tables 19 and 20). The increase in abundance and diversity
of Gastropoda in the impacted marsh may have been the result of the increased nutrients
stimulating, periphyton growth which is a major food source for the snails.

Karr and Chu (1997), suggested that, in an index of biotic integrity, metrics should
include measures of sensitive and tolerant organisms. Candidate metrics from this study
include the proportion of individuals of Ephemeroptera and Trichoptera as measures of
sensitive groups, and Isopoda as a measure of tolerant groups (Karr and Chu 1997). .

Chironomidae are fairly tolerant of pollution, and high abundances of Chironomidae
are often used as indicators of degradation (Barbour et al. 1995, Plaflcin et al. 1989). Kerans
and Karr (1994), suggested that Chironomidae variability and the need to identify them to
lower taxonomic units whose pollution tolerances are known may make them of limited use

as indicators of water quality. Metrics based on total abundances and metrics based on

99
Chironomidae abundance of tribes including Orthocladiinae and Tanytarsini generally showed
no efl‘ect in this study (Tables 19 and 20). These results and the difficulties associated with
taxonomic identification agree with Kerans and Karr (1994) suggestion that Chironomidae
may not be practical indicators of water quality.

The metrics based on filnctional feeding groups may provide information not readily
obtained fi'om taxonomic metrics (Plafldn et al. 1989, Barbour et al. 1996, Kerans et al.
1992). Useful measures of trophic and functional composition included measures of the
relative abundance of predators, filterers, and the ratio of herbivores to detritivore, a
surrogate for the primary production to community respiration ratio (P/R). Wallace et al.
(1977), suggested that filter feeders are sensitive to pollution in low-gradient streams. The
metric based on proportion of filter feeders at the impacted marsh as compared to the
reference marsh indicated that they also are sensitive in marshes and may be a useful indicator
of water quality in northern Lake Huron coastal marshes. The metric based on the
herbivore/detritivore ratio is used as a surrogate measure for gross primary production to
community respiration ratio (P/R) and is used to evaluate the balance between autotrophy and
heterotrophy (Merritt et al. 1996). The herbivore/detritivore ratio indicated that the impacted
marsh was an autotrophic system while the reference marsh was a heterotrophic system. The
autotrophic nature of the impacted marsh may be due to increased nutrients levels associated
with the sewage effluent.

Metrics based on proportions of scrapers were predicted to decrease based on stream
literature (Karr and Kerans 1992). However, numbers of scrapers plus piercers and

proportion of individuals as scrapers consistently increased (Tables 19 and 20) rather than

I!

 

'r....a__n‘_:-'.u.' b-14uhk _' 'ral.

100

decreasing as predicted. In streams, these metrics are expected to decrease because of the
preponderance of sensitive Ephemeroptera and Trichoptera grazers (scrapers) in streams. In
northern Lake Huron marshes, the increases in scrapers were primarily the result of
consistently greater Gastropoda abundances and taxa at the impacted marsh. These grazers
are more tolerant of pollution than are the grazers expected to dominate streams. If these
results prove to be consistent across an array of coastal wetlands, this metric might prove to
be an excellent predictor of water quality.

The metric of the relative abundance of predators indicated that predators would
decrease with increased disturbance, and they did (Figure 30). However, one of the measures
of ecosystem attributes considered, predicted that normal top down predator control as the
ratio of predators to all other functional feeding groups should be <0.15 (Merritt et al. 1996).
This metric consistently showed the reference marsh to be >015, and the impacted marsh to
be <0.15, indicating that the impacted marsh had normal top down predator control while the
reference marsh did not. This metric was not selected as a candidate metric as a result. It
may be that the ratio threshold calculated to be 0.15 for South Florida marshes (Merritt et al.
1996) needs to be adjusted for Great Lakes Coastal marshes. Otherwise, the use of the two
predator metrics together could lead to confusion, because difi‘erent conclusions can be drawn
from the two closely related metrics.

The metric of suspended particulate organic matter (SPOM) as a proportion of
deposited particulate organic matter (BPOM) was unable to differentiate between these two
marshes, receiving a sensitivity rating of zero, however information can still be drawn from

this metric. The ratio of SPOM/BPOM indicates that both these marshes are enriched in

101

suspended particulate organic matter. This metric was originally developed for lotic systems.
It may not be appropriate for use as a metric in these northern Lake Huron coastal marshes,
which are expected to be net accumulators of particulate organic matter.

Karr and Chu (1997), suggested that diversity indices are often inconsistent because
they respond erratically to changes in assemblages (Davis 1995, Simon and Lyons 1995).
Difi‘erent diversity indices may, therefore, produce a different rank order of the same series
of sites, making it impossible to compare the sites' biological condition (Kerr and Chu 1997).
This study lends credence to these suggestions. Several problems arose with the use of
diversity indices in this study. The various diversity indices demonstrated conflicting results
within the same site and month (Tables 19 and 20). For example, in September, Shannon
diversity indicated higher diversity at the reference marsh, while Margalef diversity indicated
higher diversity at the impacted marsh (Tables 19 and 20). Different diversity indices are
influenced by both number of taxa and their relative abundances, and some are more sensitive
to rare taxa, others to abundant taxa. Therefore, different indices measure slightly different
aspects of assemblages, and can lead to different results. Another potential problem was that
the same indices were inconsistent from month to month (Tables 19 and 20). For example,
evenness indicated that the invertebrate community was more evenly distributed at the
reference marsh in September during wastewater discharge even though it was more evenly
distributed at the impacted marsh in August. These results are opposite from predictions and
explanations of why this might be so will be needed if these observations hold up across an
array of marshes.

Two direct comparisons between the reference and the impacted sites were calculated

102

using the Jaccard Coefficient and the Coefiicient of Community Loss (Jaccard 1912, Ohio
EPA 1987, Courtemanch and Davies 1987). These coefficients may be valuable in the
assessment of water quality in northern Lake Huron coastal marshes, but values obtained from
just a two marsh comparison provided little insight into ecological condition. It is plausible
that the values obtained from these metrics are normal for this area, but this cannot be

assessed without data from a larger array of sites.

SUMMARY AND CONCLUSIONS

The purpose of this study was to evaluate the environmental quality of two northern
Lake Huron coastal marshes and recommend metrics for use in a multimetric index of
ecological integrity for these and similar marshes. The macroinvertebrate fauna and basic
limnological characteristics were examined to determine if impairment existed. A comparison
of the environmental quality as reflected by the macroinvertebrate fauna suggested that
C edarville (impacted) marsh had been moderately degraded, by sewage lagoon effluent,
storrnwater runoff from Cedarville, and habitat changes related to marina activities and road
construction.

Chemical-physical data showed that water quality in the impacted marsh was
significantly difi‘erent fiom water quality in the reference marsh only when discharge from the
wastewater treatment lagoon was actively occurring. Water quality impacts of Cedarville
Bay marsh included moderate increases in chloride, NH4-N, NO3-N, NOz-N, SRP,
conductivity, turbidity, alkalinity and moderate decreases in levels of dissolved oxygen. These
changes in water quality were greatest near the source of discharge during times of active
discharge from the wastewater treatment lagoons and decreased to background levels by 260
m away from the mouth of the discharge stream. Smaller changes in water quality also

occurred in June, approximately two weeks after a discharge event, suggesting that water

103

104

quality in the marsh returns to background levels within two or three weeks after sewage
discharge.

Although chemical analysis failed to detect any significant impairment in water quality
for the months between discharge events, the macroinvertebrate fauna demonstrated moderate
impairment at the impacted marsh throughout the season. Sensitive taxa such as
Ephemeroptera and Trichoptera had lower abundances and lower taxa richness in both the
sediment and plant community at the impacted marsh than at the reference marsh. For
example, only one species of Ephemeroptera was collected in the sediment at the impacted
marsh, and it was present only in June, while up to six species comprised 10-20% of the
sediment macroinvertebrate fauna at the reference marsh in June, July, August, and
September. Trichoptera exhibited a similar trend with nearly twice as many individuals and
taxa collected fiom the sediments at the reference marsh over the course of the sampling
season as compared to the impacted marsh. The plant associated community of the reference
marsh was dominated by a diverse array of aquatic insects including the sensitive taxa
(Ephemeroptera and Trichoptera). At the impacted marsh, the aquatic insect community was
predominantly Chironomidae (75-90%) with a less diverse array and number of sensitive taxa
being present. The plant and sediment community at the impacted marsh also had higher
abundance of pollution tolerant organisms such as Gastropoda, Isopoda, Amphipoda,
Chironomidae and Oligochaeta and fewer insects than were present in the reference marsh.
Impairment at the impacted marsh appeared to be moderate because it still maintained a fairly
diverse and abundant community, which included sensitive taxa even though these taxa were

reduced in abundance. Differences in the macroinvertebrates community were not detected

105

in response to a gradient from discharge at the impacted marsh. However, further
degradation could potentially result in the elimination of the already reduced sensitive taxa,
and the reduction in abundance and diversity of some of the more moderately pollution
tolerant organisms.

Several potential macroinvertebrate metrics derived from the stream literature were
tested to select metrics that were sensitive enough to detect the moderate impacts on the
invertebrate community impacts. The purpose of this metric selection and testing was to
recommend metrics that would be usefirl in development of an index of ecological integrity
to monitor water quality in these and similar marshes.

Thirty-eight potential metrics tested included those that incorporated elements of the
macroinvertebrate community structure and composition. Six metrics consistently detected
predicted changes in the invertebrate community and are strongly recommended for use in
development of a multimetric index of ecological integrity. An additional four metrics were
identified that show some promise.

The six consistent, strongly recommended candidate inetrics of long term impacts
included: (1) the proportion of individuals as Ephemeroptera, (2) the proportion of
individuals as Trichoptera, (3) the proportion of individuals as Isopoda, (4) the proportion
of individuals as filterers, (5) the proportion of individuals as predators, and (6) the ratio of
herbivores/detritivores as a surrogate for P/R. An additional metric, number of Trichoptera
taxa was less consistent in detecting impacts but is recommended. Three additional metrics
gave consistent results that were opposite fiom those predicted for streams. Ifthis opposite

effect proves to be a consistent efl‘ect, across a wide array of coastal marshes, these three

106

metrics will also be useful in development of an index of ecological integrity. These three
metrics are: (l) the proportion of individuals as scrapers, (2) Habitat stability, and (3) Top
Down.

Development of an index of ecological integrity for biomonitoring through the user
of multimetric indices based on invertebrates appears to be feasible for northern Lake Huron
coastal marshes. Ten potential metrics are suggested fiom this study. These ten metrics need
to be tested and validated across a range, or gradient, of human influence for Great Lake
coastal wetlands before they can be widely adopted or discarded. The proposed metrics may
need to be modified for particular ecoregions. The incorporation of additional metrics based
on attributes of several other species groups (e. g. fish, algae, macrophytes) ofl‘ers an
additional means of strengthening this proposed index of ecological integrity for these
wetlands. Thus, testing and validation of the recommended ten metrics and addition of
metrics based on other taxonomic groups are recommended next steps in development of an

Index of Ecological Integrity for Great Lakes coastal wetlands.

APPENDICES

107

Table A-1. Operational taxonomic unit (OTU) and identification references for
invertebrates collected from Cedarville and Mackinac marsh, Lake Huron, Michigan.

 

 

 

 

INVERTEBRATE OTU‘s REFERENCES

INSECTA

Coleoptera Genus White & Brigham 1996

Collembola“ Species Snider 1967; Christiansen 1996

Diptera Family Teskey 1984

Chironomidae“ Tribe Cofl‘man & Ferrington 1996

Ephemeroptera“ Species Edmunds & Waltz 1996;

Provonsha 1990

Hemiptera“ Genus/species Polhemus 1996

Lepidoptera Genus Lange 1996

Neuroptera Species Evans & Neunzig 1996

Odonata Genus/species Westfall and Tennessen 1996

Trichoptera“ Genus/species Morse & Holzenthal 1996;

Wiggins 1995

ANNELIDA

Hirudinea" Genus/species Pennak 1978

Oligochaeta Family/genus Pennak 1978

Polychaeta Genus Pennak 1978

fichnoideg (Mites) -----------

CRUSTACEA

Amphipoda Species Pennak 1978

Isopoda Species Pennak 1978

Cladocera“ Order/Species Balcer et al. 1984

Copepoda* Balcer et al. 1984
Calanoida/Cyclopoida* Species Balcer et al. 1984
Harpactacoida“ Species Balcer et al. 1984

Ostracoda Subclass Pennak 1978

Decapoda Species

PELECYPODA Genus Pennak 1978

MOLLUSCA

GASTROPODA“ Species Pennak 1978

NEMATODA Phylum Pennak 1978

 

 

108

Table A-1 (cont'd).

INVERTEBRATE OTU's REFERENCES

TURBELLARIA Species Pennak 1978

*Dr. Brian Armitage of Ohio Biological Survey, confirmed the identification of the
Trichoptera. Identification of the Ephemeroptera, Caenis amica, C. youngi and C.
Iatipennis, was based on a reference collection identified by Robert Waltz, Indiana
Department of Natural Resources. Chironomidae larvae were identified to genus or
species groups by Patrick Hudson of the Great Lakes Science Center, Biological
Resources Division, US Geological Survey. Dr . Richard Snider of the Department of
Zoology, Michigan State University confirmed the identification of the Collembola. Ethan
Nedeau, Department of Entomology at Michigan State University identified the Hirudinea.
Edward Rosemen, Department of Fisheries and Wildlife identified the zooplankton
species. Brain Keas of the Department of Zoology at Michigan State University
confirmed the identification of the Gastropoda.

 

109

Table A-2. Chironomidae present in the plant associated Dip-net samples at Mackinac
(reference) and Cedarville (impacted) marsh, Lake Huron, Michigan June 1996.

CHIRONOMIDAE"

Chironominae

C ladotanytarsus spp.
Dicrotendipes spp.

H yporhygma spp.
Microchironomus spp.
Microtendipes spp.
Pagastiella spp.
Paratanytarsus spp.
Paratendipes spp.
Polypedilum simulans group
Polypedilum tritum

T anytarsus spp. ‘

Orthocladiinae

Acricotopus spp.

C orynoneura spp.

C ricotopus bicinctus group.
Cricotopus laricomalis group
C ricotopus spp.

C ricotopus sylvestris group
C ricotopus trifascia
Doncricotopus spp.
Nanocladius spp.
Paracricotopus spp.
Psectrocladius spp.
Urienemanniella similis

Tanypodinae
Ablabesmyia peleensis

Ablabesmyia spp.

C onchapelopia spp.
Procladius spp.
Psectrotanypus spp.

MACKINAC

p—sN—IH

CEDARVILLE

* Identification references: Simpson and Bode 1980, and Wiederholm 1983.

110

Table A-3. Water quality data for Cedarville and Mackinac marsh, Lake Huron, 1996.
DFD = distance from discharge in meters. *Instrument failure or samples were lost.

 

 

 

 

 

 

 

 

Alkalinity (mg CaCOJL)
Mackinac Cedarville
’ J
I DFD J 20-June 24-July l9-Aug. 29-Sept. I 21-June 25-July 18-Aug. 28-Sept]
10 ’ I84 157 200 139 125 110 109 136
60 184 169 206 148 85 100 109 143
110 181 167 204 145 80 90 104 138
160 180 I60 205 146 77 93 79 132
210 134 143 200 149 78 81 59 100
260 127 136 179 150 79 77 61 97
310 123 131 150 132 * 76 89 105
360 125 119 120 121 82 74 83 87
410 127 121 115 114 86 73 51 85
NIL-N (mg NIL)
Mackinac Cedarville
'44,
DFD 20-June 24-July 1 9-Aug. 29-Sept. 21 -June 25-July 1 8-Aug. 28-Sept.
10 .02 .05 .03 .03 .03 .09 .04 .43
6O .01 .04 .02 .02 .06 .08 .07 .48
110 .02 .05 .02 .24 .05 .08 .06 .38
160 .04 .06 <IH .09 .09 .06 .06 .18
210 .01 .03 .04 .03 .09 .07 .05 .06
260 .04 .04 .06 .04 .06 .05 .05 .05
310 .04 .03 .04 .14 .15 .05 .05 .05
360 .03 .04 .02 .03 .06 .05 .04 .07
410 .05 .03 .02 .05 .23 .09 .07 .21
Chloride (mg Cl/L)
hdmfldnac Cedmndfle
DF D 20-June 24-July 1 9-Aug. 29-Sept. 2 l -Jtme 25-July 1 8-Aug. 28-Sept.
10 17 14 14 16 31 4 20 48
60 14 15 l6 18 18 4 18 71
110 15 16 15 I7 7 12 21 81
I60 18 I6 13 17 I3 21 20 28
210 I6 13 15 16 12 16 17 28
260 15 I4 16 21 5 18 21 24
310 15 l3 14 18 5 18 19 22
360 I8 15 14 17 4 11 19 21
410 15 l2 l4 l6 4 15 18 22

 

111

Table A-3 (cont'd)

Conductivity (uS/cm)

Mackinac Cedarville

 

DFD 20-June 24-July l9-Aug. 29-Sept. 2l-June 25-July lS-Aug. 28-Sept.

 

 

10 386 333 414 300 288 342 271 399
60 409 351 431 302 225 259 251 411
110 390 348 439 299 168 235 250 425
I60 386 341 430 300 I79 242 237 394
210 305 320 365 305 187 213 531 251
260 305 296 404 310 193 222 622 272
310 306 283 404 822 214 214 510 260
360 302 274 380 270 190 193 215 214
410 286 268 321 259 220 210 212 215
% Dissolved Oxygen
hdackuuu: (hxhnidfle

DFD 20-June 24-Ju1y l9-Aug. 29-Sept. 21 -June 25-July 18-Aug. 28-Sept.
10 106 92 100 59 106 78 90 44
60 94 76 104 58 110 93 102 37
110 95 82 84 56 101 95 91 47
I60 107 90 104 55 112 82 104 50
210 101 91 114 56 112 93 74 69
260 91 90 103 58 103 83 73 76
310 88 103 96 69 104 80 82 72
360 ' 86 102 95 72 105 87 84 76
410 . ‘ 81 102 103 67 97 82 61 72

 

 

Mackinac Cedarville
20-June 24-July 1 9-Aug. 29-Sept. 21 June 25-July I8-Aug. 28-Sept.
IO <01 <01 <01 <01 <01 <01 <01 .30
60 <01 <01 <01 <01 <01 <01 <01 .32
I I0 <01 <01 <01 .03 .02 <01 <01 .26
160 <01 <01 <01 <01 .03 <01 <01 .18
210 <I)I <I)I <:OI <I)I .03 <I)I <I)1 .01
260 <01 <01 <01 <01 <01 <01 <01 <01
310 <01 <01 <01 .02 <01 <01 <01 <01
360 <01 <01 <01 .02 .02 <01 <01 <01

410 <01 <01 <01 .04 .01 <.01 <01 <01

 

112

Table A-3 (cont'd)

 

 

H Mackinac Cedarville

DFD 20-June 24-July 19-Aug. 29-Sept. 2 1 June 25-July l8-Aug. 28-Sept.

‘ 10 r 7.43 7.69 8.46 r- 8.33 7.85 8.00
60 * 7.44 7.69 8.74 * 8.73 * 9.00
I 10 * 7.47 7.48 8.49 * 9.29 7.92 6.60
160 * 7.66 7.65 9.04 * 8.55 7.9 8.00
210 * 7.58 7.84 8.82 * 8.86 8.77 8.60
260 * 7.58 7.73 8.90 * 8.37 6.10 6.60
310 * 7.62 7.73 8.97 * 7.30 7.47 7.60
360 * 7.60 7.73 8.86 * 7.80 7.13 5.80
410 * 6.95 8.03 8.75 * 8.08 7.42 6.70

SRP m P/L Mackinac Cedarville

 

 

  

 

20-June 24-July 1 9-Aug. 29-Sept. 21 June 25-July 18-Aug. 28-Sept.
10 <01 <01 <01 <01 .01 <01 <01 .21
60 <01 <01 <01 <01 <01 <01 <01 .19
110 <01 <01 <01 <01 <01 <01 <01 .13
160 <01 <01 <01 <01 <01 <01 <01 .06
210 <01 <01 <01 <01 <01 <01 <01 <01
260 <01 <01 <01 <01 <01 <01 <01 <01
3 IO <01 <01 <01 <01 .01 <01 <01 <01
360 <01 <01 <01 <01 <01 <01 <01 <01
410 <01 <01 <01 <01 <01 <01 <01 <01
Mackinac Cedarville

 

 

   

 

 

20-June 24-July 1 9-Aug. 29-Sept. 21 June 25-July 1 8-Aug. 28-Sept.
10 1.4 2.4 .64 1.2 4.5 1.8 1.1 4.5
60 1.2 1.0 .56 1.0 5.0 1.6 1.1 1.3
110 1.5 3.1 .64 1.0 4.3 1.0 1.0 1.5
160 2.0 0.9 .65 1.0 4.0 2.2 2.0 1.8
210 3.5 1.0 .70 1.0 3.8 1.2 1.0 1.0
260 2.57 3.9 .46 3.7 3.6 1.1 .95 1.0
310 3.2 2.0 1.0 4.5 4.3 1.7 .76 1.0
360 2.2 2.4 1.0 5.2 3.5 1.0 .83 1.0
410 2.9 3.0 1.0 1.0 3.8 1.0 1.0 1.0
Water temperature (° Celsius)
Mackinac Cedarville
DFD 20-June 24-July 1 9-Aug. 29-Sept 2 1 June 25-July 1 8-Aug. 28-Sept.
10 22 19 23 l l 19 12 20 12
60 23 19 22 12 23 22 22 12
l 10 23 20 22 12 20 22 22 13
160 23 21 22 12 20 22 29 13
210 23 21 23 12 21 22 26 14
260 21 21 23 13 20 23 24 13
310 21 21 23 13 21 22 24 13
360 21 20 23 13 20 22 24 14
410 21 21 23 13 20 23 23 14

 

LITERATURE CITED

Albert, D. A., S. R. Denton, B. V. Barnes. 1986. Regional Landscape Ecosystems of
Michigan. University of Michigan, Ann Arbor.

American Public Health Association (APHA). 1985. Standard Methods for the Examination
of Water and Wastewater, 16th Edition. American Public Health Association,
Washington, D. C.

Armitage, P. D., D. Moss, J. F. Wright, and M. T. Furse. 1983. The performance of a
new biological water quality score system based on macroinvertebrates over a
wide range of unpolluted running-water sites. Water Res. 14(3): 333-347.

Balcer, M. D., N. L. Korda, and S. I. Dodson. 1984. Zooplankton of the Great Lakes, a
guide to the identification and ecology of the common crustacean species.
University of Wisconsin Press, Madison, Wisconsin. 174 pp.

Barbour, M. T., J. B. Stribling, and J. R. Karr. 1995. Multimetric approach for establishing
biocriteria and measuring biological condition. In: W. S. Davis and T. P. Simon
(eds. ), Biological Assessment and Criteria: Tools for Water Resource Planning
andDecision Making. pp. 63-77. Lewis, Boca Raton, FL.

Barbour, M. T., J. L. Plaflcin, B. P. Bradley, C. G. Graves, R. W. Wisseman. 1992.
Evaluation of EPA's rapid bioassessment benthic metrics: Metric redundancy and

variability among reference stream sites. Environmental Toxicology and Chemistry
11: 437-49.

Barbour, M. T., J. Gerritsen, G. E. Griffith, R. Frydenborg, E. McCarron, J. S. White,
M. L. Bastian. 1996. A framework for biological criteria for Florida streams

using benthic macroinvertebrates. J. N. AM Benthol. Soc. 15: 185-211.

Beck, W. M., Jr. 1955. Suggested method for reporting biotic data. Sewage and Industrial
Wastes 27: 1193-7.

113

"my,

 

114

Berg, K., and K. W. Ockelmann. 1959. The respiration of freshwater snails. J. Exp. Biol.
35: 690-708.

Brady, V. J. 1992. The invertebrates of a Great Lakes coastal marsh. M.S. thesis,
Michigan State University, E. Lansing, MI. 132 pp.

Brinkhurst, R O. 1972. The role of sludge worms in eutrophication. U. S. Environmental
Protection Agency, Washington, D. C. Ecol. Res. Ser. EPA-R3-72-004, August
1972.

Brinkhurst, R. O. and D. G. Cook. 1974. Aquatic earthworms (Annelida: Oligochaeta).
In: C. W. Hart and S. L. H. Fuller (eds), Pollution Ecology of Freshwater
Invertebrates. pp. 143-155. Academic Press, New York.

Britt, N. W. 1955. Hexagenia (Ephemeroptera) population recovery in western Lake
Erie following the 1953 catastrophe. Ecology 36: 520-22.

Cao, Y., A. W. Bark, P. W. Williams. 1996. Measuring the responses of macroinvertebrate
communities to water pollution: a comparison of multivariate approaches, biotic and
diversity indices. Hydrobiologia 341: 1-19.

Chandler, J. R 1970. A biological approach to water quality management. Water Pollution
Control 69: 415-21.

Christiansen, K. A., and R. J. Snider. 1996. In: R. W. Merritt and K. W. Cumnrins
(eds), An Introduction to the Aquatic Insects of North America. Kendall/Hunt,
pp. 113-125. Dubuque, Iowa.

Chutter, F. M. 1972. An empirical biotic index of the quality of water in South Afiican
streams and rivers. Wat. Res. 6: 19-30.

Coffrnan, W. P. and., L. C. Ferrington, Jr. , 1996. Chironomidae. In: R. W. Merritt and
K. W. Cummins (eds), An Introduction to the Aquatic Insects of North America.
Kendall/Hunt, pp. 63 5-753. Dubuque, Iowa.

Courtemanch, D. L. and S. P. Davies. 1987. A coefficient of community loss to assess
detrimental change in aquatic communities. Water Research. 21: 217-22.

Cummins, K. W. and M. A. Wilzbach. 1985. Field procedures for analysis of fimctional
feeding groups of stream macroinvertebrates. Pymatuning Lab. Ecol, U. Pittsburgh,

18 pp.

 

115

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

Davis, W. S. 1995. Biological assessment and criteria: Building on the past. In: W. S.
Davis and T. P. Simon (eds), Biological Assessment and Criteria: Tools for
Water Resource Flaming and Decision Making. pp. 15-29. Lewis, Boca Raton,
Florida.

DeShon, J. E. 1995. Development and application of the invertebrate community index (ICI).
In. W. S. Davis and T. P. Simon (eds), Biological Assessment and Criteria: Tools
for Water Resource Planning and Decision Making. pp. 217-244. Lewis, Boca
Raton, Florida.

Edmunds, G. W. Jr., and R. D. Waltz. 1996. Ephemeroptera. In: R. W. Merritt and K.
W. Cummins (eds), An Introduction to the Aquatic Insects of North America.
Kendall/Hunt, pp. 126-163. Dubuque, Iowa.

Environmental Protection Agency. 1986. Quality criteria for water 1986. EPA-440/5-86-001.
U. S. Environmental Protection Agency, Office of Water and Regulations and
Standards. Washington, DC.

Evans, E. D,. and H. H. Neunzig. 1996. Megaloptera and Aquatic Neuroptera. In: R.
W. Merritt and K. W. Cummins (eds), An Introduction to the Aquatic Insects of
North America. Kendall/Hunt, pp. 298-308. Dubuque, Iowa.

Fausch, K. D., J. Lyons, J. R. Karr, and P. L. Angermeier. 1990. Fish communities as
indicators of environmental degradation. Am. Fish. Soc. Symp. 8: 123-144.

Forbes, SA. and R. E. Richardson. 1913. Studies on the biology of the upper Illinois
River. Bulletin of the Illinois State Laboratory of Natural History 9: 48 1-5 74.

Ford, J. 1989. The effects of chemical stress on aquatic species composition and community
structure. In: S.A. Levin, M.A. Harwell, J. R. Kelly, and K. D. Kimball (eds),
Ecotoxicology: Problems and Approaches. pp. 99-114. Springer-Verlag, New York.

Fore, L. S., J. R. Karr, and R. W. Wisseman. 1996. Assessing invertebrate responses to
human activities: Evaluating alternative approaches. J. N. Am. Benthol. Soc. 15:
212-231.

Fore, L. S., J. R. Karr, L. L. Conquest. 1993. Statistical properties of an index of
biological integrity used to evaluate water resources. Can. J. Fish. Aquat. Sci. 51:
1077-087.

 

116

Gaudet, J. J. 1974. The normal role of vegetation in water. In: D. S. Mitchell. (eds),
Aquatic Vegetation and its use and Control. pp 24-37. Unesco, Paris.

Gaufin, A. R. 1973. Use of aquatic invertebrates in the assessment of water quality. In:
J. Cairns, Jr. and K. L. Dickson (eds), Biological Methods for the Assessment of
Water Quality. pp. 96-116. American Society for Testing and Materials Special
Technical Publication 528. American Society for Testing and Materials, Philadelphia,
PA.

Gerritsen, Jeroen. 1995. Additive biological indices for resources management. J. N. Am.
Benthol. Soc. 14(3): 451-57.

Grant, Michael. 1994. Cedarville Bay Water Quality Survey. Aqua-Terra Labs. Indian
River, Michigan.

Hilsenhofi; W. L. 1982. Using a biotic index to evaluate water quality in streams. Technical
Bulletin No. 132. Department of Natural Resources, Madison, Wisconsin.

Hilsenhoff, W. L. 1987. An improved biotic index of organic stream pollution. The
Great Lakes Entomologist. 20: 31-39.

Hilsenhofl‘, W. L. 1977. Use of Arthropods to Evaluate water quality of streams. Technical
Bulletin No. 100. Department of Natural Resources, Madison, Wisconsin.

Howard-Williams, C., and G. M. Lenton. 1975. The role of the littoral zone in the
functioning of a shallow tropical lake ecosystem. Freshwater Biology 5: 445-59.

Hughes, R. M., and J. R. Gammon. 1987. Longitudinal changes in fish assemblages and
water quality in the Willamette River, Oregon. Trans. Am. Fish. Soc. 116: 196-209.

Jaccard, P. 1912. The distribution of flora in the alpine zone. New Phytologist 11: 37-50.

J aworski, Eugene and Raphael, C. N. 1978. Fish, Wildlife, and Recreational Values of
Michigan '3 Coastal Marshes. Twin Cities, The United States Fish and Wildlife
Service, MN.

Jude, DJ. and J. Pappas. 1992. Fish utilization of Great Lakes coastal marshes: current
knowledge and research needs. J. Great Lakes Res. 18(4): 651-72.

Kairesalo, T., and I. Koskimies. 1987. Grazing by Oligochaetes and snails on epiphytes.
Freshwater Biology 17: 317-324.

Karr, J. R. 1990. Biological integrity and the goal of environmental legislation: lessons
for conservation biology. Conservation Biology 4: 244-50.

117

Karr, J. R. 1987. Biological monitoring and environmental assessment: a conceptual
framework. Environmental Management 11: 249-56.

Karr, J. R. 1996. Ecological integrity and ecological health are not the same. In: P.
Schulze (eds), Engineering within Ecological Constraints. pp. 110-113. National
Academy Press, Washington, DC.

Karr, J. R. 1981. Assessment of biotic integrity using fish communities. Fisheries 6(6):
21-27.

Karr, J. R. 1993. Defining and assessing ecological integrity: beyond water quality
Environmental. Toxicology and Chemistry 12: 1521-1531.

Karr, J. R., E. W. Chu. 1997. Biological monitoring and assessment using multimetric

indexes effectively. EPA 23 5-R97-001. University of Washington, Seattle,
Washington.

Karr, J. R, K. D. Fausch, P. L. Angerrneier, P. R. Yant, I. J. Schlosser. 1986. Assessing
Biological Integrity in Running Waters a Method and its Rationale. Special
Publication 5. Illinois Natural History Survey, Illinois.

Karr, J. R. 1991. Biological integrity: a long neglected aspect of water resource
management. Ecological Applications 1: 66-84.

Kerans, B. L., J. R. Karr, and S. A. Ahlstedt. 1992. Aquatic invertebrate assemblages:

spatial and temporal differences among sampling protocols. J. N. Am. Benthol.
Soc. 11: 377-90.

Kerans, B. L, J. R. Karr. 1994. A benthic index of biotic integrity (B-IBI) for rivers of
the Tennessee Valley. Ecological Applications 4(4): 768-85.

Kolkwitz, R., and M. Marsson. 1908. Okologie der pflanzlichen Saprobien. Berichte der
Deusschen Botanischen Gesellschaft 26A: 505-19. (Translated 1967. Ecology
of Plant saprobia. In: L. E. Kemp, W. M. Ingram, and K. M. Mackenthusm (eds),
Biology of Water Pollution. pp. 47-52. Federal Water Pollution Control
Administration, Washington, DC.

Krieger, K. A. 1992. The ecology of invertebrates in Great Lakes coastal marshes:
Current knowledge and research needs. .1. Great Lakes Res. 18(4): 634-50.

Landerville, J. 1997. Clark Township wastewater plant. PO. Box 367. Cedarville,
Michigan, 49719.

118

Lange, W. H. 1996. Aquatic and semiaquatic Lepidoptera. In: R. W. Merritt and K. W.
Cummins (eds), An Introduction to the Aquatic Insects of North America.
Kendall/Hunt, pp. 387-398. Dubuque, Iowa.

Lenat, D. R. 1988. Water quality assessment of streams using a qualitative collection
method for benthic macroinvertebrates. J. N. AM. Benthol. Soc. 7: 222-33.

Lenat, D. R. 1993. A biotic index for the southeastern United States: derivation and list
of tolerance values, with criteria for assigning water-quality ratings. .1. N. AM.
Benthol. Soc. 12(3): 279-90.

Lenat, D. R. 1993. A biotic index for the southeastern United States: Derivation and list
of tolerance values, with criteria for assigning water quality ratings. .1. N. Am.
Benthol. Soc. 12: 279-90.

Mason, WT, and P. P. Yevich. 1967. The use of phloxine B and rose bengal stains to
facilitate sorting benthic samples. Transactions of the American Microscopy
Society 89 (2): 221-23.

Mason, W. T., J. B. Anderson, R. D. Kreis, and W. C. Johnson. 1970. Artificial substrate
sampling of macroinvertebrates in a polluted reach of the Klamath River. Ore. .1.
Water Pollut. Contr. Fed 42(8), Part II, R315-R328.

Merritt, R. W., J. R. Wallace, M. J. Higgins, M. K. Alexander, M. B. Berg, W. T. Morgan,
K. W. Cummins,and B. Vandeneeden. 1996. Procedures for the fimctional analysis
of invertebrate communities of the Kissimmee river-floodplain ecosystem. Florida
Scientist 59: 216-74.

Merritt, R. W., and K. W. Cummins (eds), 1984. An introduction to the aquatic insects
of North America, 2nd edition. Kendall/Hunt Publ. Co., Dubuque, IA. 722 pp.

Metcalfe, J. L. 1989. Biological water quality assessment of running waters based on
macroinvertebrate communities: History and present status in Europe. Environmental
Pollution 60: 101-39.

Morse, J. C., and R. W. Holzenthal. 1996. Trichoptera genera. In: R. W. Merritt and K.
W. Cummins, (eds), An Introduction to the Aquatic Insects of North America,
Kendall/Hunt, Dubuque, Iowa. pp. 350-3 86.

Ohio EPA (Environmental Protection Agency). 1988. Biological Criteria for the Protection
of Aquatic life, volumes 1-3. Ecological Assessment Section, Division of Water
Quality Monitoring and Assessment, Ohio Environmental Protection Agency,
Columbus.

 

119

Ohio EPA (Environmental Protection Agency). 1987. Biological criteria for the protection
of aquatic life. Volume 1-3. Monitoring and Assessment Program, Surface Water
Section, Division of Water Quality, Columbus, Ohio.

Olive, J. H., J. L. Jackson, H. Bass, L. Holland, T. Savisky. 1988. Benthic
macroinvertebrates as indexes of water quality in the Upper Cuyahoga River. OHIO
J. SCI. 88: 91-98.

Pennak, R. W. 1978. Freshwater invertebrates of the United States. John Wiley & Sons,
New York.

Plaflcin, J. L., and M. T. Barbour, K. D. Porter, S. K. Gross, and R. M. Hughes. 1989.
Rapid bioassessment protocols for use in streams and rivers: benthic
macroinvertebrates and fish. EPA/440/4-89-001. Ofice of Water, US
Environmental protection Agency, Washington, DC.

Polhemus, J. T. 1996. Aquatic and semiaquatic Hemiptera. In: R. W. Merritt and K. W.
Cummins (eds), An Introduction to the Aquatic Insects of North America.
Kendall/Hunt, pp. 267-297. Dubuque, Iowa.

Prince, H. H. and C. S. Flegel. 1995. Breeding avifauna of Lake Huron. In: M. Munawar,
T. Edsall, and J. Leach (eds), Ihe Lake Huron Ecosystem: Ecology, Fisheries, and
Management, pp. 247-272. SPB Academic Publishing, Amsterdam, The Netherlands.

Provonsha, A V. 1990. A revision of the Genus Caenis in North America (Ephemeroptera:
Caenidae). Transactions of the American Entomological Society. 116(4): 801-84.

Ramm, A. R. 1988. The community degradation index: A new method for assessing the
deterioration of aquatic habitats. Wat. Res. 22: 293-301.

Rankin, E. T., C. O. Yoder, D. Mishne. 1990. 1990 Ohio Water Resource Inventory.
Ohio Environmental Protection Agency, Division of Water Quality Planning and
Assessment, Ecological Assessment Section, Columbus, Ohio.

Resh, V. H., and J. K. Jackson. 1993. Rapid bioassessment approaches to biomonitoring
using benthic macroinvertebrates. In: D. M. Rosenberg and V. H. Resh (eds),
Freshwater Biomonitoring and benthic macroinvertebrates. pp. 195-23 3. Routledge,
Chapman and Hall, Inc. New York.

Resh, V. H. 1995. Freshwater benthic macroinvertebrates and rapid assessment procedures
for water quality monitoring in developing and newly industrialized countries. In: W.
S. Davis and T. P. Simon (eds), Biological Assessment and Criteria: Tools for
Water Resource Planning and Decision Making. pp. 167-177. Lewis Publishers,
Boca Raton, Florida.

 

120

Roback, S. S. 1974. Insects (Arthropoda: Insecta). In: C. W. Hart, Jr. and S. L. H. Fuller
(eds), Pollution Ecology of Freshwater Invertebrates. pp. 313-76. Academic Press,
New York.

Rosenberg, D. M., and V. H. Resh (eds). 1993. Freshwater biomonitoring and benthic
macroinvertebrates. Routledge, Chapman & Hall, NY. 488 pp.

Simon, T. P., and J. Lyons. 1995. Application of the index of biotic integrity to evaluate
water resource integrity in freshwater ecosystems. In. W. S. Davis and T. P.
Simon (eds), Biological Assessment and Criteria: Tools for Water Resource
Planning and Decision Making. pp. 245-62. Lewis, Boca Raton, Florida.

Simpson, E. H. 1949. Measurement of Diversity. Nature 163: 688.

Simpson, K. W., and R. W. Bode. 1980. Common Larvae of Chironomidae (Diptera)
from New York State Streams and Rivers, with particular reference to the fauna
of artificial substrates. Bulletin No. 43 9, New York State Museum, The State
Education Department, Albany, New York.

Smith, R L. 1990. Student Resource Manual to accompany Ecology and Field Biology. 4th
ed. Harper and Row Publishers, New York.

Smock, L. A., and D. L. Stonebumer. 1980. The response of macroinvertebrates to
aquatic macrophyte decomposition. Oikos 35: 397-403.

Smock, L. A. 1980. Relationship between body size and biomass of aquatic insects.
Freshwater Biology 10: 375-85.

Snider, R. J. 1967. An annotated list of the Collembola (springtales) of Michigan. The
Michigan Entomologist 1: 179-234.

Sourtherland, M. T., and J. B. Stripling. 1995. Status of biological criteria development
and implementation. In: W. S. Davis and T. P. Simon (eds), Biological Assessment
and Criteria Tools for Water Resource Planning and Decision Making. pp. 81-96.
Lewis Publishers, Boca Raton, Florida.

Teskey, H. J. 1984. Aquatic Diptera Part one. Larvae of aquatic Diptera. In: R. W.
Merritt and K. W. Cummins (eds), An Introduction to the Aquatic Insects of
North America. Kendall/Hunt, pp. 448-466. Dubuque, Iowa.

Wallace, J. B., J. R. Webster, and W. R. Woodall. 1977. The role of filter feeders in
flowing waters. Archiv f iir Hydrobiologie 79: 506-32.

 

121

Wallace, J. B., J. W. Grubaugh, and M. R. Whiles. 1996. Biotic indices and stream
ecosystem processes: results fiom an experimental study. Ecological Applications
6: 140-51.

Waters, T. F. 1969. Subsampler for dividing large samples of stream invertebrate drifi.
Limnology and Oceanography l4 (5): 813-15.

Weiderholm, T., ed. 1983. Chironomidae of the Holarctic Region: Keys and Diagnoses.
Part 1. Larvae. Entomological Scandinavica, Supplement No. 19. p...

Westfall, M. J. Jr., and K. J. Tennessen. 1996. Odonata. In. R. W. Merritt and K. W.
Cummins (eds), An Introduction to the Aquatic Insects of North America,
Kendall/Hunt, pp. 164-211. Dubuque, Iowa. f

 

Wetzel, R. G., and H. L. Allen 1972. Functions and interactions of dissolved organic ‘:
matter and the littoral zone in lake metabolism and eutrophication. In: Z. Kajak E.
and A. Hillbricht-Ilkowska (eds), Productivity Problems of Fresh waters. pp.

333-347. Warszawa-Krakow, Poland.

Whillans, T. H. 1990. Assessing threats to fishery values of Great Lakes wetlands. In: J.
Kusler and R. Smarton (eds), Proceedings of an International Symposium: Wetlands
of the Great Lakes, Association of State. pp. 68-71. Wetlands Managers, Inc., New
York.

White, D. S., and W. U. Brigham. 1996. Aquatic Coleoptera. In: R. W. Merritt and K.
W. Cummins (eds), An Introduction to the Aquatic Insects of North America.
Kendall/Hunt, pp. 399-473. Dubuque, Iowa.

Wiggins, G. B. 1995. Larvae of the North American Caddisfly Genera (2nd ed).
University of Toronto Press, Toronto.

"Pg“ ‘

 

 

.aun-

  

nun-unnum-