 

.S .bwmww.

3%.

g»;

n
.

aru—
: s

w

Emma“

is!
.1

m3.

 

a: 3.
11:1 2:11
.2. v , i u.
.
Sill-ad
1‘2 :1: .
3:1.»

Ema

\I utl
23:1): ..
v3». ..
v.21:

.3: u e

‘ e h ‘at Xi

. a .31: .:1 a

_ vi: :4 :JW
1 ‘ .

I

x.
i. ‘ .9. I».
r...

 

*c‘inﬁ.

‘2': .5.
1.7.1“.
F: v. e?!»
n.1ylaaz
‘5‘ 110.1
.1. u it}.
a

..

a“:

. so: Lt
2.: \59.

 

THESH:

 

i (3.} t. -

‘ LIBRARY
Michigan State
University

 

 

 

This is to certify that the
thesis entitled
Fish and Invertebrate Community Composition: A
Comparison of Headwater and Adventitous
Streams

presented by

David A. Thomas

has been accepted towards fulﬁllment
of the requirements for

M.S. degree in FiSh. & Wildl.

L044, W

Major professor

 

 

Date October 24, 2001

 

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

 

 

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

 

DATE DUE DATE DUE DATE DUE
3.0.le .10 ’5 Zzuu

 

U

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDue.p65-p.15

 

FISH AND INVERTEBRATE COMMUNITY COMPOSITION: A COMPARISON OF
HEADWATER AND ADVENTITIOUS STREAMS

By

David A. Thomas

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

2001

ABSTRACT

FISH AND INVERTEBRATE COMMUNITY COMPOSITION: A COMPARISON OF
HEADWATER AND ADVENTITIOUS STREAMS

By

David A. Thomas

The River Continuum Concept (RCC) is an overarching paradigm in stream ecology
that makes predictions regarding the trophic status, ﬁsh community, and invertebrate
community of rivers based on stream order. The RCC, however, ignores the role of
adventitious streams, which are low—ordered tributaries to larger rivers. I examined the
ﬁsh and invertebrate community and habitat of the ﬁfth-order mainstem, two second-
order adventitious tributaries to the mainstem, and three second-order headwater streams
of the Pine River (Alcona County, Michigan) from May through August 2000. Fish
species richness generally increased with increasing stream order and was higher in the
adventitious streams than in the headwater streams. The ﬁsh species composition of
adventitious streams was more similar to the mainstem than to the headwater streams, but
showed greater month-to-month variability than either the mainstem or headwater
streams. Adventitious streams had a preponderance of tolerant ﬁsh and had lower scores
for invertebrate Indices of Biotic Integrity, suggesting that water quality was impaired in
these streams. Habitat conditions in headwater and adventitious streams were similar
except adventitious streams were generally warmer. These results suggest that factors in
addition to stream order, such as stream connectivity and temperature, are important

determinants of stream ﬁsh assemblage.

For my parents, who always encouraged me to learn.

iii

ACKNOWLEDGMENTS

Funding for this project was provided by the Michigan Department of Natural
Resources Fisheries Division and the US. Department of Agriculture’s McIntyre-Stennis
Program. I am especially grateful to my advisor, Dr. Dan Hayes, for his wisdom,
guidance, and patience. I would like to acknowledge and thank my other committee
members, Dr. Tom Coon and Dr. Rich Merritt, for their insight and guidance. I would
also like to thank Dr. Bill Taylor, Dr. John Giesy, and Dr. Larry Fisher for their advice
and encouragement.

A special thanks goes to PhD candidate Brad Thompson, my cohort on this project,
and to Tom Goniea, Jeff Leighton, Jesse Gabbard, Jodie Anderson, and Ed McCoy, the
interns who devoted much of their time and effort in both the ﬁeld and in the lab. I am
also grateful to my labmates, Ed Roseman, J o Latirnore, Kris Lynch, Bryan Burroughs,
Sharon Johnson, Jeremy Price, Sarah Thayer, Ann Krause, J oAnna Lessard, Hope Dodd,
Kurt Newman, and Jessica Mistak, as well as to fellow graduate students Ali Felix, Nikki
Lamp, and Stephanie Pastva for their ﬁiendship, encouragement, and insight. Finally, I

would like to thank my lovely wife, Kim, for all of her love, patience, and support.

iv

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. vi
LIST OF FIGURES ................................................................................ viii
INTRODUCTION .................................................................................... 1
Goals and Objectives ........................................................................ 7
STUDY AREA ........................................................................................ 9
MATERIALS AND METHODS .................................................................. 13
Fish Collection .............................................................................. 13
Macroinvertebrate Collection ............................................................ 14
Habitat Data Collection ................................................................... 14

Data Analysis ............................................................................... 18
RESULTS ............................................................................................ 25
Fish Community Analysis ................................................................ 25
Species Richness .................................................................. 25

Spatial Variability ................................................................. 25

Tolerance ........................................................................... 29

Temporal Variability .............................................................. 34
Macroinvertebrate Community Analysis ................................................ 34
Habitat Analysis ............................................................................ 47
DISCUSSION ....................................................................................... 55
Fish Community ........................................................................... 55
Macroinvertebrate Community ........................................................... 58
Habitat ....................................................................................... 61
CONCLUSIONS .................................................................................... 64
APPENDICES ....................................................................................... 66
LITERATURE CITED ............................................................................. 78

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

Table 10.

Table 11.

Table 12.

Table 13.

LIST OF TABLES

Expected shifts in trophic, physical, and biotic components by stream order
according to the River Continuum Concept (summarized from Cummins 1977;
Vannote et a1. 1980) ..................................................................... 3

Summary of study sites by stream name, stream order, and stream type. . . . . ....12

Level of taxonomic identiﬁcation used in the Procedure 51 (Michigan
Department of Environmental Quality 1997) ....................................... 15

Modiﬁed Wentworth (Cummins 1962; Harrelson et a1. 1994) scale used for
substrate size classiﬁcation ............................................................ l7

Macroinvertebrate metrics used for calculations of Procedure 51 (Michigan
Department of Environmental Quality 1997) ....................................... 22

Fish species present, tolerance values, and number captured in each stream
type ....................................................................................... 26

Average similarity of ﬁsh species by stream type using S¢rensen’s (1948) Q8
index ...................................................................................... 30

Average of similarity of ﬁsh species composition across stream types using
Morisita’s (1959) F index (as modiﬁed by Horn 1966) ........................... 3O

Similarity of ﬁsh species composition in adventitious streams to Pine River
mainstem reference sites using S¢rensen’s (1948) Q8 and Morisita’s (1959) I‘
indices (as modiﬁed by Horn 1966) .................................................. 31

Average of month-to-month similarities within each stream type using
S¢rensen’s (1948) Q8 index .......................................................... 35

Average of month-to-month ﬁsh community similarities within each stream
type using Morisita’s (1959) F index (as modiﬁed by Horn 1966) .............. 35

Macroinvertebrate taxa present and percent composition for each stream type
over all sites and months ............................................................... 36

Mean invertebrate metrics based on Procedure 51 (Michigan Department of
Environmental Quality 1997) scores for each stream type by month. . . . . . . . ....44

vi

Table 14.

Table 15.

Table 16.

Table 17.

Table 18.

Mean monthly and annual values for River Continuum Concept ratios of
photosynthesis to respiration (PR) for each stream type. Values above 0.75
indicate production in excess of respiration (autotrophic condition) ............ 45

Mean monthly and annual values for River Continuum Concept ratios of
coarse particulate organic matter (CPOM) to ﬁne particulate organic matter

(F POM) for each stream type. Values above 0.25 indicate riparian dominated
streams ................................................................................... 45

Mean monthly and annual values for River Continuum Concept ratios of ﬁne
particulate organic matter in transport (TFPOM) to that stored in the benthos
(BFPOM) for each stream type. Values above 0.50 indicate increased
suspended organic material in transport ............................................. 46

Mean monthly and annual values for River Continuum Concept ratios for

channel stability for each stream type. Values above 0.50 indicate that stable
substrates are not limiting ............................................................. 46

Mean (i ISE) of habitat conditions and standard errors for each stream
type ....................................................................................... 48

vii

LIST OF FIGURES

Figure 1. Location of the Pine River in Alcona and Iosco Counties, Michigan ............ 10
Figure 2. Fish species richness by stream (includes additional data from third-, fourth-,
and ﬁfth-order sampling efforts) ...................................................... 28
Figure 3. Number of ﬁsh species in each tolerance category by stream type ............... 32
Figure 4. Percent composition (i approximate 95% conﬁdence intervals) of ﬁsh
abundance in each tolerance category by stream type ............................. 33
Figure 5. Percent composition (i approximate 95% conﬁdence intervals) of invertebrate
ﬁinctional feeding groups by stream type ........................................... 39
Figure 6. Change in composition of macroinvertebrate functional feeding groups over
sampling period ......................................................................... 40
Figure 7. Percent composition (i approximate 95% conﬁdence intervals) of
Ephemeroptera, Plecoptera, and Trichoptera (EPT) by stream type ............. 42
Figure 8. Ratio (1- approximate 95% conﬁdence intervals) of Chironomidae to
Ephemeroptera, Plecoptera, and Trichoptera (EPT) by stream type ............. 42
Figure 9. Frequency of substrate sizes by stream type ........................................ 51
Figure 10. Weekly staff gauge readings for each stream ...................................... 52
Figure 11. Average daily temperature by stream type during the summer months. . . . . ...54

viii

Introduction

Stream order designation (Horton 1945) has been used extensively by ecologists in
recent decades to make predictions concerning a watershed’s physical, biotic, and
metabolic status. Gradations in many physical features can be attributed to increasing
stream order. Naiman (1983) found traits such as watershed area, channel dimensions,
and discharge to be directly proportional to stream order. His study also demonstrated
that stream width could be predicted for a given stream order. A negative correlation
exists between slope gradient and stream order (Platts 1979). Numerous studies have
found signiﬁcant increases in ﬁsh species richness related to increasing stream order
(Sheldon 1968; Platts 1979; Schlosser 1982a; Naiman et al. 1987; Paller 1994; Fairchild
et al. 1998), but Platts (1979) and Fairchild et al. (1998) found that ﬁsh species richness
is typically maximized in fourth or ﬁfth order streams with a subsequent decline in
stream orders of six or more.

The River Continuum Concept (RCC) (Cummins 1977; Minshall 1978; Vannote et al.
1980) hypothesizes that a watershed’s trophic structure follows a predictable gradient
from low-order headwater streams to high-order large rivers, and that downstream
processes are directly inﬂuenced by upstream processes. Some of the expected shifts in
various trophic, physical, and biotic components along this continuum are summarized in
Table 1. Brieﬂy, stream orders 1 through 3 are typically heterotrophic in nature, the
primary carbon source being coarse particulate organic matter (CPOM) received from the

riparian surroundings (V annote et al. 1980). Instream photosynthesis is minimized due to

 

 

 

 

£5: e285 access

Agog 898—30 Eozom 03882 264 >64 02358803 N75 285— owned
Agmv Eovwoﬁm

Goo: e985
AfammV whoaacom 86on

Agog €88on 53325.3 33 88252 28032 02.3582 0% 8:39.22
AXE Somehow

$3 €98on

$3 e835 seam

$.va 298:8 remasoo SE swim at: eﬁoseaom 3 20358:
axe—.8860 53.5580 Nam—35m Quaint; 2.33.5 3.3m 201.5
cusps—«€25— ..mE ouaauoaﬁoh umwﬁohiﬂ one—m via—9:. 835m

.AowE .3 B 03:53 ":3 2:88an 82m 635ng

3850 8:25:00 83% 05 9 95:88 890 Sacha. 3 35:09:00 385 c5 48653 62905 E 3:? @88me A «San.

vegetative shading of these small streams. The CPOM is reduced to ﬁne particulate
organic matter (FPOM) by invertebrate shredders, the second most abundant frmctional
feeding group after collectors (Vannote et al. 1980), as it moves downstream. Stream
orders 4 through 6 shift to an autotrophic state due to an increased amount of sunlight
reaching the stream, resulting in increased primary production. This increase in
periphyton is coupled to an increase in invertebrate grazers. F POM derived from
upstream processes becomes the primary food source of invertebrate collectors, which are
the most abundant functional feeding group in mid-order reaches. Stream orders 7
through 12 return to a heterotrophic state due to increased turbidity resulting from the
large volmne of FPOM and detritus ﬁom upstream sources. The macroinvertebrate
functional feeding groups consist primarily of collectors in these large rivers. The ratio
of annual instrearn photosynthesis to respiration (PR) is largely dependent on the amount
of sunlight reaching the stream bed, a function of vegetative cover/canopy, water depth,
and turbidity.

The RCC, however, is based on an underlying assumption that convergent streams
differ by no more than one order of magnitude. One of the disadvantages of this concept
is that it ignores lower ordered streams emptying into a higher ordered stream (Allan
1995). Gonnan (1986) offered the term adventitious to describe tributaries that ﬂow into
streams differing by 3 or more magnitudes of order (e.g., a 2nd order ﬂowing into a 5th
order). In an amendment to the RCC, Minshall et al. (1985) realized this and suggested
that adventitious tributaries may represent a deviation from the original notion. Losses of
streamside vegetation due to varying land uses in a lower gradient region of the

watershed could alter the amount of CPOM input and sunlight exposure, creating an

autotrophic situation. Therefore, an adventitious stream’s trophic structure would more
closely resemble its proximal midreach stream than headwaters sharing the same order
designation. Grazers and collectors would be expected to comprise the largest proportion
of invertebrate functional feeding groups in this situation. The lower gradient associated
with the proximity of the midreach should cause a corresponding decrease in variability
of the adventitious tributary’s ﬂow regime (Platts 1979).

Other information provided solely by stream order may also be misleading, especially
when a stream receives inﬂow from numerous streams of lower order magnitude. The
resultant increase in discharge and possible temperature effects are not accounted for by
the Horton (1945) classiﬁcation. Minshall et al. (1985) hypothesized that potential
impacts of an adventitious tributary may be determined by its size, riparian land use, and
trophic structure. These factors may alter the ﬁsh community structure. A study by
Osborne & Wiley (1992) on two warmwater watersheds in Illinois found that ﬁsh species
richness was lower in headwater streams compared to adventitious tributaries and found
no signiﬁcant difference in species richness between adventitious tributaries and the
proximal mainstreams. However, headwater streams and adventitious tributaries
exhibited the greatest similarity in species composition (Osborne & Wiley 1992).

Differences in ﬁsh assemblages of adventitious and headwater streams may also occur
due to abiotic differences. In Osborne & Wiley’s (1992) study, no difference in slope
gradient was found between headwater streams and adventitious tributaries in two
warmwater river systems. This may partially explain the similarity in species
composition. In coldwater river systems, however, decreases in both slope gradient and

hydrological variation, and increases in temperature, are typically associated with

increasing stream order (Cummins 1977; Platts 1979; Vannote 1980). Therefore,
adventitious streams are more likely to drain regions of lower slope gradient than
headwater streams in the same watershed. Because ﬁsh species richness is signiﬁcantly
and negatively correlated with hydrologic variability (Horwitz 1978; Platts 1979; Gorman
1986; Poff & Allan 1995), low gradient adventitious streams should have relatively
higher species richness. Studies have shown that species richness is positively correlated
with stream size and maximized in ﬁfth- or sixth-order streams (Gorman & Karr 1978;
Platts 1979; Beecher et al. 1988; Fairchild et al. 1998). Thus, it is likely that ﬁsh species
richness would be increased in adventitious tributaries relative to headwater streams due
to the proximity of a species-rich midreach.

Differences in ﬁsh assemblages due to abiotic factors may also occur on a temporal
basis. Ibbotson et al. (1994) found ﬁsh species richness in low— to mid-order streams to
vary over time spans as short as one month. Compared to headwater streams,
adventitious streams may exhibit greater temporal dynamics in terms of ﬁsh species
richness and density for several reasons. The greater level of habitat heterogeneity found
in most headwater streams is usually coupled with a greater stability of ﬁsh assemblages
compared to lowland streams (Gorman 1986). Because midreach streams are subject to
the most extreme seasonal temperature ﬂuctuations (Cummins 1977; Vannote 1980),
adventitious tributaries may provide thermal refuges for coldwater ﬁsh when the
mainstream water temperatures approach intolerable extremes. Adventitious tributaries
may also be used seasonally for spawning by midreach-dwelling ﬁsh, thereby increasing
the density and species richness at certain times each year (Gorman 1986; Osborne &

Wiley 1992). A decrease in species richness in adventitious streams would be expected

during low-ﬂow events, such as a drought or a seasonal dry period, which would cause
the ﬁsh to temporarily relocate to the midreach, thereby decreasing species richness and
abundance (Gorman 1986; Osborne & Wiley 1992).

The distribution of ﬁsh taxa classiﬁed as tolerant, intermediate, or intolerant to
environmental degradation may differ in adventitious streams when compared to
headwater streams. Studies have demonstrated that riparian agricultural activities may
degrade habitat through increased sedimentation (Karr 1981; Walser & Bart 1999),
nutrient addition (Cooke et al. 1995), and addition of toxic chemicals associated with
pesticide use (Loehr 1974; Johnson 1986; Cufﬁrey et al. 2000). A study by Walser &
Bart (1999) found that a signiﬁcant positive relationship existed between agricultural use
and sedimentation in mainstem reaches of a watershed due to the lower slope gradient,
but not in the higher gradient headwater reaches. Therefore, low-gradient adventitious
tributaries to a mainstem are also likely to be negatively impacted by agricultural land
use, resulting in decreased numbers of intolerant species. The water quality may
decrease to a level that renders adventitious streams unsuitable for intolerant ﬁsh species,
thereby allowing tolerant and intermediate tolerant ﬁsh to dominate.

Changes in water quality associated with agricultural land use may also be reﬂected in
the macroinvertebrate community composition. Indices of Biotic Integrity (IBI’s) have
been widely used to assess the quality of streams through examination of the presence
and abundance of macroinvertebrates (Kerans & Karr 1994; Whiles et al. 2000) and ﬁsh
(Leonard & Orth 1986; Steedman 1988; Lydy et al. 2000). However, Berkman et al.
(1986) found that macroinvertebrates were more sensitive than ﬁsh as indicators of

habitat perturbation in streams impacted by agricultural practices. Based on these lBI’s,

rapid bioassessment protocols have been developed for use in the Midwestern U.S.
(Michigan Department of Environmental Quality 1997; Barbour et al. 1999). Other IBI’s
include EPT (Ephemeroptera, Plecoptera, and Trichoptera) abundance, and the ratio of
Chironomidae abundance to EPT abundance (Merritt & Cummins 1996a).
Goals and Objectives

The goal of this project is to examine the role of adventitious tributaries in a watershed
and to compare these streams to headwater streams of the same order designation within
the watershed. Associated with this goal, I developed four hypotheses related to the ﬁsh
communities. I ﬁrst hypothesized that ﬁsh species richness is greater in adventitious
tributaries than in headwater tributaries due to the connectivity with the mainstem.
Second, I hypothesized that the ﬁsh species composition in adventitious streams would
show greater similarity to the proximal mainstem than to headwater streams. Third, I
hypothesized that adventitious streams would have a greater abundance of tolerant ﬁsh
species than headwater streams. Finally, I hypothesized that adventitious streams would
exhibit greater temporal ﬂuctuations in ﬁsh species richness during the summer months
than headwater streams. I also developed two hypotheses with respect to the
macroinvertebrate communities. First, I hypothesized that the IBI scores would be lower
for adventitious streams due to negative impacts of agriculture and urbanization. Second,
I hypothesized that the loss of streamside vegetation associated with these land uses
would create a more autotrophic situation in adventitious streams that would be reﬂected
in the macroinvertebrate composition as a decrease in shredders and an increase in

grazers. The speciﬁc objectives of this study were to:

(1)

(2)

(3)

(4)

(5)

(6)

(7)

Characterize and compare the ﬁsh species richness and relative abundance in
three second-order headwater streams, two second-order adventitious streams, and
the ﬁfth-order mainstem of the Pine River in Alcona County, Michigan.

Using similarity indices, compare the ﬁsh species assemblages among these three
stream types.

Determine and compare the proportion of ﬁsh classiﬁed as tolerant, intermediate,
and intolerant between these stream types.

Examine and compare the temporal dynamics of the ﬁsh community composition
in each of these streams from May to August.

Characterize and compare the invertebrate community composition in each of
these stream types for consistency with the River Continuum Concept.

Using Indices of Biotic Integrity (IBI’s), compare the macroinvertebrate
community compositions in each of these three stream types as bioindicators of
stream water quality.

Characterize and compare habitat parameters such as stream width, depth,
discharge, substrate composition, in-stream temperature, and meso-habitat

composition (i.e. rifﬂe, run, pool, etc.) in each of these types.

Study Area

The Pine River, a tributary of the AuSable River, is located in the northeastern region
of Michigan’s lower peninsula (Figure l). The Pine River consists of ﬁve major
branches: the South Branch, West Branch, East Branch, VanEtten Creek, and the Pine
River mainstem. Approximately 95% of this watershed is located within the southeast
region of Alcona County. Shortly after ﬂowing south into Iosco County, the Pine enters
VanEtten Lake, a recreational impoundrnent formed by a small bottom-release dam at the
south end of the lake. The Pine continues its southerly ﬂow until it reaches the AuSable
River, approximately 3 kilometers downstream from VanEtten Dam.

Approximately one-half of the land drained by the Pine River is contained in the
Huron National Forest. Most of the South and West Branches are contained within the
forest boundary. Land use within the Huron National Forest is primarily dedicated to
recreational activities, including camping, hunting, ﬁshing, and hiking. The headwaters
of the South Branch are found in a management area reserved for the Kirtland warbler, a
federally listed endangered bird species, and are protected against development. The East
Branch, VanEtten Creek, and Pine River mainstem lie east of the forest boundary. Here,
the land use is largely for agricultural purposes.

The Pine River was chosen for my research because it is the subject of an ongoing
habitat assessment for salmon and steelhead by myself, doctoral candidate Brad
Thompson, and our advisor Daniel Hayes. The habitat data that we collect will be used
to create a model for predicting the number of j uvenile salmon and steelhead that the Pine

River could support if ﬁsh passage is created at VanEtten Dam.

33¢ 05.—

dmeom—z £03580 8on can «:82 E 82% 8E 05 mo 5:83 A unsur—

       

J

h0>a 0—D¢m=<

10

Stream order was determined from US. Geological Survey maps (scale 1:24,000).
Two second-order adventitious tributaries to the Pine River mainstem were chosen for the
purpose of this research: Hill Creek and an unnamed creek (hereafter referred to as
Unnamed Creek). These were chosen because they are the only second-order
adventitious tributaries to the ﬁfth-order mainstem of the Pine River. Three second-order
non-adventitious headwater tributaries were chosen for this research: McGillis Creek,
McDonald Creek, and VanderCook Creek. The criteria used for selection of these
streams were road access, spatial distribution within the watershed, and connectivity with
streams differing by no more than one order designation.

Two sites separated by a minimum of one hundred meters were chosen for sampling
purposes within each stream (Table 2). Each site consisted of a seventy-ﬁve meter
section of stream, marked at each end to ensure month-to-month sampling consistency.
Site selection for all second-order streams was determined by availability of access. For
each of these streams, only one road crossing was available. Thus, one site was selected
upstream of the road and one site was selected downstream, with the exception of
UnNamed Creek. UnNamed Creek attains second-order status approximately ﬁve meters
upstream from Cruzen Road, the only adequate access point. Thus, both sites were
selected downstream from the road crossing. Sites were selected in the Pine River
mainstem a short distance upstream from the conﬂuence with each of the adventitious

tributaries, Hill Creek and UnNamed Creek.

11

Table 2. Summary of study sites by stream name, stream order, and stream type.

 

Site # Stream Name Stream Order Stream Type
1 McDonald Creek 2 Headwater
2 McDonald Creek 2 Headwater
3 McGillis Creek 2 Headwater
4 McGillis Creek 2 Headwater
5 VanderCook Creek 2 Headwater
6 VanderCook Creek 2 Headwater
7 Hill Creek 2 Adventitious
8 Hill Creek 2 Adventitious
9 UnNamed Creek 2 Adventitious
10 UnNamed Creek 2 Adventitious
11 Pine River 5 Mainstem
12 Pine River 5 Mainstem

12

Materials and Methods

Fish Collection

Fish were sampled at all sites during the ﬁrst full week of May, June, July, and
August of 2000. Blocknets with a 111-inch mesh were placed at the downstream and
upstream end of each site to prevent ﬁsh from migrating into or out of the site during
sampling. A backpack-mounted DC electroﬁshing unit with a single anode probe was
employed to sample ﬁsh in all 2nd order sites. For the larger and deeper 5th order sites, a
barge-mounted DC electroﬁshing rmit with 2 anode probes was employed. We used a
three-pass depletion method to collect ﬁsh ﬁom each site. All passes were initiated at the
downstream end of the site, working in an upstream direction. Captured ﬁsh were held in
a water-ﬁlled bucket or cooler and aerated with a battery-powered unit. After each pass,
all ﬁsh collected were identiﬁed to species, counted, and released downstream of the site.
Uncertain ﬁsh identiﬁcations were retained in a 10% formalin solution for subsequent
veriﬁcation in the lab.

Additional data were included from ﬁsh sampling efforts conducted in 1999 on
several branches of the Pine River. A backpack electroﬁshing unit was used in three 3rd
order sites on the East Branch, a 3rd order site on the South Branch, and a 3rd order site on
VanEtten Creek. The barge-mounted electroﬁshing unit was used to sample ﬁsh on a 4th
order site in the South Branch, 4th order site in the West Branch, and a 5th order site in the

Main Branch.

13

MﬁroinvertelLate Collection

Macroinvertebrates were sampled in late May 2000, mid-August 2000, and early
January 2001 to seasonally represent larval instars of different insect groups. We
followed the protocol described by the Michigan Department of Environmental Quality’s
(1997) Procedure 51. Equal sampling effort was given to all habitat types (i.e. pools,
rifﬂes, and runs). Organisms were collected from silt, sand, gravel, cobble, leaf packs,
submerged vegetation, and woody material by sweeping with a D-frame net, kicking
substrate material upstream of the net, and by hand-picking with forceps. All organisms
captured at each site were placed in a 5-gallon bucket to form a single composite sample.
The composite sample was rinsed through a sieve with a 1-mm mesh and large organic
and inorganic debris fragments were removed. The remaining sample containing the
organisms was then placed in a white enamel counting pan ﬁlled approximately half full
with water. 100 organisms were removed from the sample with forceps and placed in a
95% ethanol solution for subsequent identiﬁcation in the lab. Identiﬁcation of the
organisms was determined at the taxonomic level described by Appendix H of Procedure
51 (Michigan Department of Environmental Quality 1997) and summarized in Table 3.
Organisms were also classiﬁed by ﬁrnctional feeding groups (i.e. shredders, grazers,
predators, and collectors) according to Merritt and Cummins (1996a) and Pennak (1989).
H_abitat Data Collection

Within each site, geomorphic habitat units were classiﬁed as pools, rifﬂes or runs
(Simonson et al. 1993). Habitats classiﬁed as pools were deeper than average, had a slow
water velocity, and an unbroken water surface. Rifﬂe habitats were shallow, had a higher

water velocity, and a turbulent water surface. Habitats classiﬁed as runs had an

14

 

8%: ﬁgs Seam «3232 $8852
68: manage a. 3E0: 38$ e050 =< 9085 88852
63: €885 a. 3E0: S5 288230 5085 80852

8%: sauces coco «86.282 8825:.

Ema: 0353 SEC mooﬁmEU «meagre

8M3: gages awn—U 8:635 «2652‘

33: vacuum 320 seaﬂowzo «2352

Ema: vacuum 8255 290603802

tom: >3— .CaEv—h Susana—33 382.35. no .96: .8qu mun—U 52.25

ASE 3:30 Esoﬁoougcm .«o Eve—Eamon gwmnomé E 056395 05 5 com: 8:85:82 £82553 mo .034 .m 039—.

15

intermediate and uniform depth, moderate ﬂow velocity, and an unbroken water surface.
Habitat units were delineated by starting at the downstream endpoint of each site and
measuring in an upstream direction. A measuring tape was used to measure the length of
each habitat unit to the nearest meter. Woody material was visually estimated as a
percentage of total area for each habitat unit.

To characterize the habitat, transects were set up by extending a measuring tape across
the width of the stream at several positions within each site. In all 2"d order sites,
transects were located by extending a measuring tape across the width of the stream at the
downstream endpoint and midpoint of each habitat unit. Additionally, transects were
positioned at the upstream end of each site. To characterize the more homogeneous 5th
order sites, transects were set up at 0, 25, 50, and 75 meters from the downstream
endpoint. Data recorded at each transect included stream width, velocity, substrate,
temperature, and geographic coordinates. Stream width was determined by reading the
measuring tape and recording the width to the nearest 0.1-meter. Stream velocity was
measured with a Marsh-McBimey® Model 2000 ﬂow meter at 20-cm intervals across the
stream width in all 2nd order sites and at l-m intervals in all 5th order sites. Substrate was
characterized by performing a pebble count (Wohnan 1954) using a modiﬁed Wentworth
(Cummins 1962; Harrelson et al. 1994) classiﬁcation (Table 4) across all transects. In all
2nd order sites, 25 substrate particles were measured at each transect, while 50 substrate
particles were measured in the wider 5th order sites. The latitude and longitude of each
transect was determined using a Garmin® GPS 12XL Global Positioning System unit. A
staff gauge was installed in each stream and measured weekly for hydrologic variation.

Bank stability, bank vegetation, and streamside cover in each site were evaluated using

16

Table 4. Modiﬁed Wentworth (Cummins 1962; Harrelson et al. 1994) scale used for

substrate size classiﬁcation.

 

Size Value Description Size Range (mm)
1 Clay <2 (visual)
2 Silt <2 (visual)
3 Sand <2
4 Fine gravel 2-4
5 Medium gravel 4-8
6 Coarse gravel 8-16
7 Small pebble 16-32
8 Medium pebble 32-48
9 Large pebble 48-64
10 Small cobble 64-128
11 Medium cobble 128-192
12 Large cobble 192-256
13 Small boulder 256-512
14 Medium boulder 512-1024
15 Large boulder 1024-2048
16 Very large boulder >2048
17 Organic material Any (visual)

17

metrics described in the Michigan Department of Environmental Quality’s (1997)
Procedure 51.

The temperature regime in each stream was determined by installing Onset® Optic
Stowaway digital temperature recorders approximately 20 m upstream of each site. To
ensure that these units remained submerged, they were attached to concrete reinforcement
bars that were driven into the stream substrate. The temperature recorders were
programmed to record the stream temperature at 2-h intervals from May 2000 until May
2001.

Data Analysis

All data manipulations and statistical tests were performed using SAS® Version 8
(SAS Institute 1999) software. Results were considered signiﬁcant at or (Type I error)
values of 0.05 for all tests. All pairwise comparisons were performed with a Kramer
(1956) modiﬁcation of Tukey’s (1953) studentized range test as recommended by Day
and Quinn (1989) for ecological data with unequal sample sizes.

Fish species richness was compared using a Mixed GLM with stream type as the main
effect of interest, month as a blocking factor, and stream as a random effect. Least
Squares Means (LSM) were used to obtain point estimates for mean species richness by
type and these were compared using a T ukey-Kramer studentized range test.

Fish communities in each stream and stream type were compared using two indices of

similarity, S¢renson’s (1948) OS and Morisita’s (1959) F (as modiﬁed by Horn 1966).

S¢renson’s index of similarity is based on species presence/absence and is calculated as:

18

QS=2C/(A+B), [1]
where QS is the index of similarity, A is the number of species in one stream or stream
type, B is number of species in another stream or stream type, and C is the number of
species common to both streams or stream types. Values for QS may range from 0 to 1
with values of 0 indicating no species overlap and values of 1 indicating identical ﬁsh
species composition.
Horn’s (1966) modiﬁcation of Morisita’s (1959) F is based on relative abundance of

each species and is calculated as:

F = injﬁk_ [2]
Z 9211 + 213211

where F is the index of similarity, pij and pik are the relative abundance of the ith species
in the jth and kth stream or stream type. Values for F may range from 0 to 1 with values
of 0 indicating no species overlap and values of 1 indicating identical proportions of
species composition.

The proportion of ﬁshes classiﬁed as tolerant, intermediate, and intolerant (Michigan
Department of Environmental Quality 1997; Barbour et al. 1999) was calculated for each
site. The vectors of proportions were arcsine-transformed and statistically compared
using Multivariate Analysis of Variance (MAN OVA; SAS Institute 1999) to determine if
the tolerance designations of the ﬁsh community differed between stream types. When
MANOVA indicated a signiﬁcant difference in the vectors, a GLM was used to evaluate
the differences in proportions of individual tolerance categories.

Month-to-month variability in ﬁsh community composition was assessed for each

stream and stream type using S¢renson’s (1948) Q8 and Morisita’s (1959) F (as modiﬁed

19

by Horn 1966) indices of similarity. QS values were calculated as in Equation 1, where
A is the number of ﬁsh species present in one month, B is the number of ﬁsh species
present in another month, and C is the number of ﬁsh species common to both months for
a given stream or stream type. F values were calculated as in Equation 2, where pij and
pik are the relative abundance of the ith species in the jth and kth month.

The proportion of macroinvertebrate functional feeding groups (collectors, grazers,
shredders, and predators)(Merritt & Cummins 1996a; Pennak 1978) was calculated for
each site. These proportions were arcsine-transformed and analyzed using a Mixed
General Linear Model with stream type as the main effect of interest and month as a
blocking factor.

The proportion of Ephemeroptera, Plecoptera, and Trichoptera (EPT) were calculated
for each site and sampling event (month). In addition, the ratio of Chironomidae to EPT
was calculated for each site and month. These data were analyzed using a Mixed General
Linear Model with stream type and month as the main effects of interest and stream as a
random effect.

The macroinvertebrate community was evaluated following the procedures for the
Northern Lakes and Forest (NLF) ecoregion as described in the Update of GLEAS
Procedure 51 Metric Scoring and Interpretation (Michigan Department of Environmental
Quality 1996) for the Procedure 51 (Michigan Department of Environmental Quality
1997). The nine invertebrate metrics that were scored for each site and stream are
summarized in Table 5. For each metric, a score of +1, 0, or —1 was assigned as follows:

+1 = community performing better than average condition for

“excellent” sites in the NLF ecoregion;

20

0 == community performing at or within minus two standard deviations
of the average condition for “excellent” sites in the NLF ecoregion;
-l = community performing at less than minus two standard deviations
of the average condition for “excellent” sites in the NLF ecoregion.
Metric scores were then totaled for each site or stream by sampling month. Because there

are nine metrics, scores could range from —9 to +9. Total scores are interpreted as

follows:
S -5 Poor;
-4 to —l Tending Toward Poor;
0 Neutral, no tendency toward Excellent or Poor;
1 to 4 Tending Toward Excellent;
2 5 Excellent.

Total scores in the range of —4 to +4 are deemed acceptable under Michigan Department
of Environmental Quality (1996) Water Quality Standards.

Invertebrate functional feeding groups were used to index stream ecosystem attributes
according to Merritt & Cummins (1996b). The ratio of scrapers to shredders plus total
collectors was used to indicate the trophic status of a stream. Elevated numbers of
scrapers relative to shredders and collectors indicate increased dietary reliance on
periphyton from primary production, while increased proportions of shredders and
collectors relative to scrapers indicate increased loading of allochthonous coarse
particulate organic matter (CPOM) as the primary food source. In general, values greater
than 0.75 imply an autotrophic state, while values less than 0.75 imply a heterotrophic

state (Merritt & Cummins 1996b).

21

Table 5. Macroinvertebrate metrics used for calculations of Procedure 51 (Michigan
Department of Environmental Quality 1997).
(1) Total Taxa — the total number of taxa identiﬁed according to Table 3;
(2) Mayfly Taxa — the total number of families in the order Ephemeroptera present;
(3) Caddisfly Taxa — the total number of families in the order Trichoptera present;
(4) Stoneﬂy Taxa — the total number of families in the order Plecoptera present;
(5) % Mayﬂy Composition — ratio of individuals in the order Ephemeroptera to total
number of organisms;
(6) % Caddisﬂy Composition — ratio of individuals in the order Trichoptera to the
total number of organisms;
(7) "/o Contribution of the Dominant Species — ratio of individuals in the most
abundant taxon to the total number of organisms;
(8) % Isopods, Snails, & Leeches — ratio of individuals in the order Isopoda and
classes Gastropoda and Hirudinea to the total number of organisms;
(9) % Surface Dependent — ratio of individuals dependent on obtaining oxygen

directly from the atmosphere to the total number of organisms.

22

The ratio of shredders to total collectors was used as a measure of allochthonous
CPOM to ﬁne particulate organic matter (FPOM) (Merritt & Cummins 1996b). Elevated
proportions of shredders relative to collectors (>0.25) indicate a greater association of the
benthic invertebrate community with the riparian system, while elevated proportions of
collectors (<0.25) indicate a benthic invertebrate community reliant on upstream
processing of organic material.

The amount of F POM in transport (TFPOM) relative to that stored in the benthos
(BF POM) was indicated by calculating the ratio of ﬁltering collectors to gathering
collectors (Merritt & Cummins 1996b). Elevated proportions of ﬁltering collectors
relative to gathering collectors (>0.50) may indicate high levels of TFPOM, while
elevated proportions of gathering collectors relative to ﬁltering collectors (<0.50) are
indicative of high levels of BFPOM.

The in-strearn channel stability was indicated by calculating the ratio of scrapers and
ﬁltering collectors to shredders and gathering collectors (Merritt & Cummins 1996b).
Elevated proportions of scrapers and ﬁltering collectors (>0.50) indicate that stable
substrates are not limiting, while elevated proportions of shredders and gathering
collectors (<0.50) indicate low substrate stability.

Proportions of habitat types (i.e. rifﬂes, runs, and pools) were calculated for each site
by dividing the total length of each habitat type by the length of each site (i.e. 75 m). A
general linear model (GLM) was then applied to the length proportions at each site using
stream type (i.e., headwater, adventitious, mainstem) as the main effect of interest and
site as a blocking factor. Least squares means (LSM) were used to obtain point estimates

for parameter values, accounting for imbalance in sample sizes of habitat data.

23

For each site, average stream width, depth, woody material, and velocity were
calculated. A GLM was then applied to these data for comparison using stream type as
the main effect of interest and site as a blocking factor. Least Squares Means were used
to obtain point estimates for each parameter of interest, accounting for imbalance in these
data.

The values from the modiﬁed Wentworth scale (Table 4) for substrate material
correspond to size ranges for inorganic material (values 1 through 16). The
corresponding classiﬁcation for organic material (17), however, is a descriptive term for
substrate consisting of wood, leaves, etc. regardless of size. Therefore, organic substrate
was treated separately using a Generalized Linear Model (GLIM; Nelder & Wedderbum
1972) assuming a binomial (organic vs. inorganic) distribution of error terms with stream
type as the main effect of interest and stream as a blocking factor. The inorganic
substrate composition was then analyzed using a Generalized Linear Model (GLIM;
Nelder & Wedderbum 197 2) assuming a multinomial distribution of error terms with
stream type as the main effect of interest and stream as a blocking factor. Least Squares
Means were used to calculate point estimates and standard errors for inorganic substrate
material.

Mean daily temperatures were calculated for May through August, the period of ﬁsh
sampling, for each site. Using these data, the mean summer temperatures for each stream
were analyzed using a General Linear Model with stream type as the main effect of

interest and site as a blocking factor.

24

Results

ﬂh Community Analysis
Species Richness

A total of 40 species of ﬁsh were captured in the Pine River watershed during the
study (Table 6). We captured 33 ﬁsh species in the mainstem, 23 species in the
adventitious streams, and 13 species in the headwater streams over the four-month study
period. An average of 3.8 species of ﬁsh were present at each site in the headwater
streams, 9.3 at each site in the adventitious streams, and 23.3 in the mainstem sites.
Signiﬁcant differences in ﬁsh species richness were detected between the stream types
(P<0.0001). Using a Tukey’s adjustment, a signiﬁcant difference in species richness was
detected for all pairwise comparisons of stream types (P<0.02).

When data from ﬁve third-order, two fourth-order, and two additional ﬁfth-order sites
were included, the number of species present showed a general increase with increasing
stream order in non-adventitious streams (Figure 2). Furthermore, the species richness in
the adventitious sites is above the mean for the non-adventitious second-order streams in
this trend.

Spatial Variability

S¢rensen’s QS values for all pairwise comparisons of individual streams ranged from
O to 0.75 (Appendix C). Generally, the mainstem was more similar to the adventitious
streams (QSZO.43) than to the headwater streams (QSSO.28)(Appendix C). On average,
adventitious streams had a higher degree of similarity to the mainstem (QS=0.48) than to

the headwater streams (QS=0.28)(Table 7). Morisita’s F values for all pairwise

25

 

o o o 0 0:80.85 00:00:00.3 0.000002 0.50 00083.03
0 a o c 2.0.80.0 0000.00 0800Q0§n~ 305:2 00055.5
3 o. o o E00285 00.009. 00000030000Q 80.... 335$.
: S o o E00883 .8000\ 000$0A<30~>80§ 0005 E05002
N. . o . _ 80.80:. 0000800090 000085802. :0:Em :0206
o. m m M 8080585 5080.000 0.080000% .800Q 30050.":
S o m 3 80808085 000 000080E 000D 020003. 0:05.52
mm mm o o 8060:0085 000000>0~\ 00000. 00.00. 30.0%
mm mm o o E00885 00000.30 50.080003». :0x000w0m 0:05.02
ow o o 9. 80:08.85 00000000.... 00039 00000305 0.005
wv mm c 0 80:00::83 00000000 00~0§x~ 005.5 :08800
.m .m o o 80:02:08.: 0.300%... 08.00035... 000m 0.00%
mm mm c an 50:80... 0.30090 0...:0Q0N £833 :080
00 mm o 0 .5022... 000008 00.000 .0000 .0520
m0 mm o om 80.80:. 0000005500 0050000009 000—005 82>?
0: N— 8 c 0 80:00:08.: 0000300 00.00.00R 00.0%04
03 SN o om 80:08:85 0003000. 05.20% 00:0Q 0200—005
3m 0mm _ o0 80:05.85 SEN... 0.080005% 08:0O 0:52.
m mm m _ 0o. m .0 €0.80... .00.: 00005 30:50:02). 08:00
0mm 2 wmm 0 0:80.85 0.000.000\000.00>~0m 80:... 0005
8mm. com vow SN 20.80:. 003.000 0.330.3wa 0009 0005005
09 oo. :0 3 E0085 00000 00:09 £035 00—002
0mm. .mm m B an E0083. 050.000.0000 05000.00. 00:0 0.00:0
.005. E80302 000.5000: 3033:0200. 00:20.05 050 Z 052.85 0E0: :0EE00

.095 80080 £000 5 0003000 0008:: 0:0 £0300. 00:880. aE008: 0.800% :05 .0 030,—.

26

 

005% mm: mm 5m .5: 580.5.
5 5 o o 805005.85 0.0.0.0500 00.80000: 00.0.5005 .055
5 5 o o 805005.85 000500.500 00.0.0008: 000m 55500508015
5 5 o 0 80500585 0.0000000 00.80000: 00.0.5005 :00500
5 o o 5 80500585 000.00.20.00 0.30.002 .05.5m 050.05m5
5 5 o 0 80500585 00.00.000.080. 0080.005 8.5.580 05005
N 5 o 5 80500585 005005 R00M 0.555 50.580 Z
0 0 o o 58050.5 0000000000. 000005.809 :0550m 05005.50
0 m o 5 5805055 8.0... 80.0% 500.5. :30.m
0 0 o 0 580505. 000300.000. 0.000805 555005
m o o m 80500585 00.000000 0.50002 .0555 55080.5
m o m 0 5805055 00M00.0..0..0. 0.0.0002 5.50 .035
m o 0 5 80500585 05.0.0 00.800005 .0505 0305
m m 5 5 050500.585 0050.0 000800.... 00855555 05005
0 5 m o 805005.85 0.5000000... 00005830: 0005 5.00.5
w m o o 805005.85 0505000500 000.000.3058 050.0505 .036
o a o o 805005.85 00.0.0500 000030.050... 000m 5.0055505m
a w o 5 80500585 0000005 0.000050% 000055.55

5.08. 0 0......

27

.Amtobo mam—95m BEoAEa Ea wagon £55 88m 8% 35323 maﬁa—05v 885 an magnet 860% 5E .N unsure

 

 

EEO Ewobm
m w m N #
”1.]} 1i! s‘J'! -.lLI,|;!1H|}-iltiL|¢ll (131i 1!! _ 1"]? ll ll: 0
o
u m
o
o o m 2 S
. m
o o o m.
o T, m ~ 3
. o m.
-- om m
o
8386 325533 amen—Er 88on o 1 mm
0
380.5. msouucoﬁvméo: ”$252.“ 33on o

r“

.3.

28

comparisons of streams ranged from 0.00 (no similarity) to 0.96 (Appendix C). Table 8
shows that the average Morisita’s F values were higher for comparisons of adventitious

streams to the mainstem (F=0.48) than to headwater streams (F=0.30).

Similarity indices were calculated comparing each adventitious stream to its mainstem
reference site (Table 9). The similarity values for comparison of Unnamed Creek to
mainstem Site #11 were moderately high whether based on species presence/absence
(QS=0.49) or relative abundance (F=0.56). The similarities between Hill Creek and
mainstem Site #12 showed a different trend, however. The similarity of species overlap
was moderately high (QS=0.49) while similarity based on relative abundance was lower
(F=0.25).

Tolerance

The tolerance proportions were similar across all stream types (Figure 3). The
greatest proportion of ﬁsh species present across all stream types were classiﬁed as
intermediate tolerance, with fewer intolerant and tolerant species (F igure 3). When
viewed by relative abundance, however, a high proportion of tolerant ﬁsh dominated the
ﬁsh communities of adventitious streams (Figure 4). A high proportion of intolerant ﬁsh
dominated the ﬁsh communities of headwater streams (Figure 4). The mainstem had a
more even distribution of tolerance classiﬁcations, with high numbers of intermediate
tolerant ﬁsh, and fewer numbers of tolerant and intolerant ﬁsh (Figure 4).

The proportion of intermediate tolerant ﬁsh was higher for the mainstem compared to
headwater streams (P=0.0002) and adventitious streams (P=0.0043). Adventitious
streams had a higher proportion of intermediate tolerant ﬁsh compared to headwater

streams (P=0.0462). Although the differences visually appear to be large (Figure 4), no

29

 

 

 

 

 

% EPT
O)
'_*__.L_____.3i__el 1 _._.l__f
.__J

 

 

 

 

 

 

 

Adventitious Headwater

Mainstem

Figure 7. Percent composition (i approximate 95% conﬁdence intervals) of

Ephemeroptera, Plecoptera, and Trichoptera (EPT) by stream type.

35‘

N
0100

 

I
x . __ A___..v

 

Ratio of Chironomids/EPT
A

.O
07'

 

 

 

...._.___

W

ﬂ

 

 

O

Adventitious Headwater

 

f—_l ,.

 

Mainstem

Figure 8. Ratio (i approximate 95% conﬁdence intervals) of Chironomidae to

Ephemeroptera, Plecoptera, and Trichoptera (EPT) by stream type.

30

Table 9. Similarity of ﬁsh species composition in adventitious streams to Pine River
mainstem reference sites using S¢rensen’s (1948) Q8 and Morisita’s (1959) F indices (as

modiﬁed by Horn 1966).

S rensen’s Morisita’s
Unnamed Creek-Site #11 0.49 0.56
Hill Creek-Site #12 0.49 0.25

31

Tolerant

,
\

Intolerant
I Intermediate

 

   

 

. ...-.

--n

n.-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

aioods J0 .quumN

 

 

 

 

 

Headwater Mainstem

Adventitious

h tolerance category by stream type.

168 1n eac

Figure 3. Number of ﬁsh spec

.093

Sacha an #8ng 3:828 :80 E conga—5a saw mo @3385 3:03:80 $3 Bmﬁieaa ﬂ scavenge... «38$ .v unsur—

Eoumﬁﬂz .8“ston msouu=o>c<

 

 

 

Y:
SN
.8
-9.

 

 

row
row

 

 

uomsodwoo [191:1 %

[on

 

5823. & - om

 

o§te§8£ E L. 1 cm

H§HO~OHﬁH D

 

 

1 OS

33

differences were detected between stream types for the proportions of intolerant species
(P200824) or tolerant species (P201510) due to the high variability within stream types.
Temporal Variability

Average month-to-month QS values were high for all stream types (Table 10). Using
a Tukey’s adjustment, QS values were higher for headwater streams than for adventitious
streams (P=0.007) indicating higher month-to-month variability of ﬁsh species presence
in adventitious streams compared to headwater streams. No differences were detected
between the mainstem and headwater streams (P=0.834) or between the mainstem and
adventitious streams (P=0.157).

Average month-to-month F values were variable, ranging from moderately high in
adventitious streams to very high temporal similarities in headwater streams (Table 11).
Using a Tukey’s adjustment, Morisita’s F values were lower for adventitious streams
than for headwater streams (P<0.0001) and the mainstem (P=0.019), indicating higher
month-to-month variability of ﬁsh abundance in adventitious streams compared to
headwater streams and the mainstem. No signiﬁcant difference was detected between the
mainstem and headwater streams (P=0.565).

Macroinvertebrate Community Analysis

A total of 52 aquatic macroinvertebrate taxa were collected during the study (Table
12). The Pine River mainstem had the highest richness overall, with 34 taxa present.
Little variation in taxa richness existed between the second-order streams, ranging from
25 (VanderCook Creek and Unnamed Creek) to 29 (Hill Creek)(Appendix G). The most

common taxa present in all streams were the families Chironomidae (non-biting midges)

34

 

 

mod :5 end end and mmd Ed 36 Elouwdg

«ed .36 mod wad bod No.0 owd ad 3

mod No.0 cud omd 86 end cod de msouzco>w<
mm owﬁo>< w:<-_3. m:<-:3. 375:. a=<->§ ETNmE 5:652

.602 Eom B 8585

m8 5?: L Ema: MSBtBZ mam: cab 83% :08 £53, mourn—MEG 5:58:80 :2.“ 55:78-5:08 mo om80>< .: uEwH

 

 

cod owd $6 and mud 36 5.0 wwd a2

mod vwd awd mod mwd and $6 and 39%

vod $6 $6 36 mod mod mod 05.0 9833535.
mm odﬁo>< Max‘s—2. ”maﬁa 37:3. w:<.>@2 3332 5%

48?: m0 Ga: 9:35qu mama: 25 8.8.5 :80 £5; moutmzamm 586-8-588 .«o owﬁo>< .3 wish.

35

Table 12. Macroinvertebrate taxa present and percent composition for each stream type
over all sites and months.

 

Phylum Class Family Adventitious Headwater Mainstem
Annelida Hirudinea - 0. 1 0.2 0.3
Annelida Oligochaeta - 0.3 0.6 0.0
Arthropoda Arachnoidea - 0.9 0.2 0.2
Arthropoda Crustacea - 15.3 10.9 7.7
Arthropoda Insecta - 0. 1 0. l 0.0
Arthropoda Insecta Aeshnidae 1 .4 1 .5 0.7
Arthropoda Insecta Athericidae 0.0 0.0 0.7
Arthropoda Insecta Baetidae 0.3 2.7 1 .5
Arthropoda Insecta Baetiscidae 0.0 0.0 1 .5
Arthropoda Insecta Belostomatidae 0.6 0.0 0.0
Arthropoda Insecta Brachycentridae 0.3 9.5 0.3
Arthropoda Insecta Calopterygidae 2.0 0.8 1 .0
Arthropoda Insecta Chironomidae 23.7 1 1.2 13.5
Arthropoda Insecta Cordulegastridae 0.0 1 .4 0.2
Arthropoda Insecta Corixidae 0.8 0.0 2 .2
Arthropoda Insecta Corydalidae 0.3 1 .4 1.2
Arthropoda Insecta Dixidae 3 .3 0.0 0.0
Arthropoda Insecta Dytiscidae 3. 1 0. 1 0.3
Arthropoda Insecta Elmidae 0.7 0.2 3 .7
Arthropoda Insecta Ephemerellidae 7.5 14.3 9.3
Arthropoda Insecta Ephemeridae 0.0 0.0 3 .8
Arthropoda Insecta Gerridae 1.0 1 .6 1 .7
Arthropoda Insecta Glossosomatidae 0.9 3 .7 3 .0
Arthropoda Insecta Gomphidae 0.0 0.3 0.0
Arthropoda Insecta Gyrinidae 0.4 0.2 0.0
Arthropoda Insecta Helicopsychidae 0.0 0.6 0.0
Arthropoda Insecta Heptageniidae 0.0 4.5 21.8
Arthropoda Insecta Hydropsychidae 2 .2 5 .6 7.2
Arthropoda Insecta Isonychiidae 0.0 0.0 2.2
Arthropoda Insecta Leptophlebiidae 2 .3 2.4 0.0
Arthropoda Insecta Libellulidae 1 .2 0.0 0.0
Arthropoda Insecta Limnephilidae 3 .8 6.8 l .0
Arthropoda Insecta Metretopodidae 0. 1 0.0 2.0
Arthropoda Insecta Molanidae 0. 1 0.0 0.0
Arthropoda Insecta Nemouridae 0.0 6. 8 0.0

36

Table 12 (cont.).

Arthropoda Insecta Perlidae 0.3 0.5 2.0
Arthropoda Insecta Perlodidae 0.0 1 .8 1.7
Arthropoda Insecta Philopotamidae 1 .0 1 .0 0.0
Arthropoda Insecta Pleidae 3.0 0.0 0.3
Arthropoda Insecta Polycentropodidae 0.2 0.4 0.0
Arthropoda Insecta Psephenidae 0.0 0.1 0.0
Arthropoda Insecta Pteronarcidae 0.0 0.4 0.5
Arthropoda Insecta Sialidae‘ 0.2 0.7 0.0
Arthropoda Insecta Sirnuliidae 12.5 3.8 4.0
Arthropoda Insecta Stratiomyidae 0.0 0.3 0.0
Arthropoda Insecta Tabanidae 0.1 0.3 0.2
Arthropoda Insecta Taeniopterygidae 0.0 1 .3 1 .0
Arthropoda Insecta Tipulidae 1 .2 1 .2 1 .7
Mollusca Gastropoda Bithyniidae 4.3 0.0 0.0
Mollusca Gastropoda Lymnaeidae 4.5 0.6 1 .2
Mollusca Pelecypoda Unionidae 0.3 0. 1 0.7
Nematomorpha - - 0.0 0. 1 0.0

 

37

n= 1200 1800 600

and Ephemerellidae (mayflies), and the class Crustacea (primarily amphipods and
crayﬁsh).

The composition of macroinvertebrate functional feeding groups varied among stream
types (Figure 5). Collectors were the most abundant group in all three stream types, but
the proportion of collectors were signiﬁcantly higher in adventitious than in headwater
streams (P=0.0145). The proportion of predators varied little between stream types,
ranging from 11.3% in headwater streams to 15.5% in adventitious streams. The
proportion of grazers varied among types and was higher in the mainstem than both
adventitious streams (P=0.0103) and headwater streams (P=0.0053). The proportion of
shredders was highest in headwater streams relative to adventitious streams (P=0.0004)
and the mainstem (P=0.0031).

The proportion of functional feeding groups varied over time (Figure 6). No
interactions between stream type and month were detected for any of the ﬁmction feeding
groups (P200550), implying similar seasonal patterns among stream types. Collector
abundance was highest in adventitious streams over all three sampling events, yet was
only signiﬁcantly higher than headwater streams in January (P=0.0004). Collectors
reached peak abundance in January across all stream types. The proportion of predators
decreased from August 2000 to minimum values in January 2001 for all three stream
types (P500390). The proportion of predators in adventitious streams and the mainstem
was greatest in August while headwater streams peaked in May. Grazer abundance in the
mainstem was higher than in second-order streams over all three sampling events, yet
was signiﬁcantly so only in May (P_<_0.0030). The proportion of shredders was the

highest in headwater streams over all three sampling events. However, signiﬁcant

38

.25 88.9. as 3:8» wanton.“ Queue—Sm cumunotgﬁ .«o @3325 Donovwnoo $3 Buckingham ﬂ :ozmmanoo Beacon .m 0.5mm“—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EBmEmE 832633 msoEEo>c<
, ,2
T cm
\x
.- om
_ ow
J. M %
ﬂ ﬂow 3
_ . m
1 co .m
m.
L. W 7 ON. W...
ﬂocvoﬁm E _
$330 I _ ow
£03on a a
muSooon D ..- om

. 02

39

- - o - -Adventitious + Headwater —9— Mainstem

Collectors

 

 

% Composition

% Composition

 

 

 

60

(It
0

 
  

 

.h
C

“~12. J_1_'..

% Compositior

 

   

t-‘Nw
COCO

'...
......
......
.-
.............
......
.....
.0
41.--

 

‘v—‘F

 

28-May-00 1 1-Aug-00 25-Oct-00 08-Jan-01

Figure 6. Change in composition of macroinvertebrate functional feeding groups over
sampling period.

40

differences were only detected in Fall (P500398) and January (P300350). No
signiﬁcant month-to-month differences in shredder abundance were detected for any
stream type (P200972).

The mean proportion of Ephemeroptera, Plecoptera, and Trichoptera (EPT) varied
among the three stream types over all sampling events (Figure 7). These three
invertebrate orders accounted for more than half of all organisms collected in headwater
streams (62%) and the mainstem (59%), yet accounted for about 19% of the organisms
collected in adventitious streams. The proportion of EPT in adventitious streams was
signiﬁcantly lower than that of headwater streams (P<0.0001) and the mainstem
(P<0.0001). No difference in EPT proportion was detected (P= 0.6299) between
headwater streams and the mainstem. No interactions between stream type and month
were detected (P=0.5295) indicating similar seasonal patterns in EPT abundance. In
addition, no differences in month of sampling were detected (P=0.5715).

The mean ratio of Chironomidae to EPT was highly variable for all stream types
(Figure 8). In adventitious streams, chironomids outnumbered EPT. In headwater
streams and the mainstem, chironomids were much less abundant than EPT. No
interaction between stream type and month were detected (P=0.2760). In addition, no
difference in month of sampling was detected (P=0.3609). Differences in stream type
were signiﬁcant only for comparisons of adventitious and headwater streams (P=0.0065).
The difference between adventitious streams and the mainstem was not signiﬁcant
(P=0.0594), and no difference was detected between headwater streams and the mainstem

(p=0.9959).

41

 

 

‘1
O

 

 

% EPT

50
40 i
30-!

.. I
.l 1

 

 

 

 

 

 

 

 

l
0 + f T
Adventitious Headwater Mainstem

Figure 7. Percent composition (i approximate 95% conﬁdence intervals) of
Ephemeroptera, Plecoptera, and Trichoptera (EPT) by stream type.

 

 

 

 

 

 

 

 

 

 

3.5‘1

!

t— 3]
Pg 1
:2 2.51
a 2i;
.8 l
1.5»
a 11
0.51

Adventitious Headwater Mainstem

Figure 8. Ratio (i approximate 95% conﬁdence intervals) of Chironomidae to
Ephemeroptera, Plecoptera, and Trichoptera (EPT) by stream type.

42

The invertebrate metric scores were variable among the stream types (Table 13).
Mean values for adventitious streams were generally low, indicating a tendency toward
poor water quality (Michigan Department of Environmental Quality 1996). The values
for headwater streams were moderately high, indicating a tendency toward high water
quality. The Pine River mainstem had the highest and most consistent invertebrate metric
scores (5) over all three sampling periods and indicate excellent water quality. Values for
all stream types were within the range deemed acceptable (-4 to +4) for Michigan Water
Quality Standards (Michigan Department of Environmental Quality 1996).

Trophic status (PzR), represented by the ratio of scrapers to shredders and total
collectors, varied seasonally over all stream types (Table 14). In May, all second-order
sites tended toward a heterotrophic condition (<075), while the ﬁfth-order mainstem
indicated an autotrophic condition (>075). August values were variable, with the
mainstem, a headwater stream (McGillis Creek), and an adventitious stream (Unnamed
Creek) tending toward an autotrophic condition and the remaining streams tending
toward a heterotrophic state. Values for January were the lowest, with all streams
tending toward a heterotrophic state.

The ratio of coarse particulate organic matter to ﬁne particulate organic matter
(CPOM/FPOM), represented by the ratio of shredders to total collectors, varied among
stream types (Table 15). Values were generally highest for headwater streams across all
months, indicating increased input of allochthonous material relative to adventitious
streams and the mainstem. The mainstem and adventitious streams were generally low

(<0.25), indicative of systems dependent on ﬁne particulate organic matter.

43

Table 13. Mean invertebrate metrics based on Procedure 51 (Michigan Department of
Environmental Quality 1997) scores for each stream type by month.

 

 

Mm Adventitious Headwater Mainstem
May 0.00 4.33 5
August -1.50 2.67 5
January 050 3.67 5
Annual Average ~0.67 3.56 S

Table 14. Mean monthly and annual values for River Continuum Concept ratios of
photosynthesis to respiration (PR) for each stream type. Values above 0.75 indicate
production in excess of respiration (autotrophic condition).

 

 

 

Month Adventitious Hegd_wr_lter MM
Mav 0.24 0.11 0.86
August 0.36 0.25 0.47
January 0.00 0.09 0.14
Average 0.20 0.15 0.49

Table 15. Mean monthly and annual values for River Continuum Concept ratios of
coarse particulate organic matter (CPOM) to ﬁne particulate organic matter (FPOM) for
each stream type. Values above 0.25 indicate riparian dominated streams.

 

 

 

Month Adventitious Hea_dlater Mainstem
Mav 0.15 0.34 0.04
August 0.04 0.69 0.06
January 0.05 0.68 0.11
Average 0.08 0.57 0.25

45

Table 16. Mean monthly and annual values for River Continuum Concept ratios of ﬁne
particulate organic matter in transport (TF POM) to that stored in the benthos (BFPOM)
for each stream type. Values above 0.50 indicate increased suspended organic material in
transport.

 

 

 

Month Adventitious Headyter Mm
M_av 2.34 0.14 0.38
August 0.18 0.78 0.31
January 1.16 0.38 0.52
Average 1.22 0.46 0.41

Table 17. Mean monthly and annual values for River Continuum Concept ratios for
channel stability for each stream type. Values above 0.50 indicate that stable substrates
are not limiting.

 

 

 

Month Adventitious Hewer Mainstem
my 2.52 0.24 1.69
August 0.72 0.86 1.12
January 0.27 0.33 0.67
Average 1.17 0.48 1 .16

46

The ratio of ﬁne particulate organic matter in transport (TFPOM) to that stored in the
benthos (BFPOM), represented by the ratio of ﬁltering collectors to gathering collectors,
varied among stream types and months (Table 16). The ﬁlth-order mainstem had more
moderate and consistent values seasonally than second-order adventitious and headwater
streams. In May, values for all but one stream were below 0.50, indicative of high loads
of FPOM in transport relative to that stored in the benthos. The value for Unnamed
Creek was approximately 4.50, possibly due to invertebrate community responses to
increased FPOM in suspension from high annual springtime ﬂows. In August and
January, values for all streams were more moderate.

The in-stream channel stability, represented by the ratio of scrapers and ﬁltering
collectors to shredders and gathering collectors, varied among stream types and months
(Table 17). The values for adventitious streams were above the reference point (0.50) for
all months except January, indicating high channel stability. Values for headwater
streams exceeded the reference point only in August only, but on average indicated
moderate channel stability. The mainstem was above the reference point for all months,
indicating high channel stability.

Habitat Analysis

The proportion of habitat types (i.e. rifﬂe, run, pool) varied among stream types
(Table 18). Pools comprised approximately 26% of the length of sites in adventitious
streams, followed by headwater streams (15%) and the mainstem (13%). Rifﬂes
comprised approximately 28% of the length in the mainstem, but made up little of the
proportion of length in adventitious (5%) and headwater (2%) streams. Runs comprised

the largest proportion of length in all stream types with headwater streams containing the

47

cod H 5.2
86 H oﬁo
Nd H Qmm
wmd H 5.:
ﬁe H mm
mév H ﬁwm
mm H mm
02 H 02

E82332

vod H 0.2
Ed H mod
@— H 2.2
mNd H mm
v.m H VAN
v; H Name
3» H m6
was H om

335522

#06 H m.m_

N56 H mod

v; H 3.:
m.~ H vvd
NM H NNN
9N H 9mm
w; H w;
Wm H v.3

3:363:

Gov oSHEomEor—r 558% 582
$5 $83 52
GE 58a :82
as 523 502
560 3003 .x.
5?: 3 SE .x.
5mg: 3 are .x.
523 .3 .82 .x.

.25 885 some Ho.“ Echo 2353.0. can $836.80 55m: mo Ammﬁ HV 582 .3 03.5.

8

4

highest average proportion (84%) followed by adventitious streams (69%) and the
mainstem (59%). The high standard error in the mainstem is accounted for by the fact
that the downstream site was comprised entirely of a single run (Appendix A). Across
stream types, the average lengths of habitat units within a type were not signiﬁcantly
different (P=0.0702).

The proportion of estimated woody material varied between the different stream types
(P=0.0211). The main branch had signiﬁcantly less woody material (P<0.05, Tukey’s
studentized range test) than headwater and adventitious streams, yet no difference was
found between headwater and adventitious streams (P=O.9855, Tukey’s studentized range
test).

Differences in habitat conditions were apparent among stream types. Average width
(P<0.0001, GLM), depth (P<0.0001, GLM), and water velocity (P<0.0001, GLM)
differed between the three stream types. The ﬁfth order mainstem had a signiﬁcantly
higher average width (17.3m; P<0.0001) than the second order headwater (2.4m) and
adventitious (2.1m) streams. However, Tukey’s analysis detected no signiﬁcant
difference between the headwater and adventitious stream widths (P=0.6524). Similarly,
the mainstem had a signiﬁcantly greater average depth (35.9cm; P<0.0001) than the
headwater (14.5cm) and adventitious (13.4cm) streams. No difference in depth was
found between headwater and adventitious streams (P=0.8616). The mainstem had the
highest average velocity (0.16m/s) followed by the headwater (0.09m/s) and adventitious
(0.05m/s) streams. All pairwise comparisons of velocity were found to be signiﬁcant

(P<0.05).

49

Substrate material varied among the three stream types (Figure 9). Organic material
composed approximately 34% of the substrate in adventitious streams, 29% in headwater
streams, but only 5% in the mainstem. Signiﬁcant differences were detected for the
percent composition of organic material between the three stream types (P<0.0001).
When the mainstem sites were excluded from the analysis, adventitious streams were
found to have a greater proportion of organic material (P=0.0086) than headwater
streams. Headwater streams had an average inorganic substrate size of 2.8, composed
primarily of sand (56%) and silt (15%). Adventitious streams had an average inorganic
substrate size of 2.6 and were composed primarily of silt (39%) with lesser amounts of
clay, sand, gravel, and pebbles. The mainstem had an average inorganic substrate size of
3.4 and was composed primarily of sand (71%) with some clay, silt, gravel, pebbles, and
cobble. No difference was detected in mean substrate size between adventitious and
headwater streams (P=0.2206, Tukey’s studentized range test), however these stream
types had smaller mean substrate sizes than the mainstem (P500057, Tukey’s
studentized range test).

The staff gauge readings followed similar trends for all stream types (Figure 10). The
headwater streams were generally the most hydrologically stable, followed by the
adventitious streams, and the mainstem. A series of heavy rainfalls in early June caused
the readings for the mainstem and adventitious streams to ﬂuctuate more dramatically
than the headwater streams. Much of the remainder of the sampling period was dry
resulting in stable staff gauge readings. Another series of showers occurred in early

August, primarily affecting the mainstem stream levels.

50

dab 885. .3 moﬁm cabana mo 353ch .a charm

Mucus: 2.18 236 22$ 238 23$ .25 3:5 EEO
2&5 use»: .35 993 5%: £5 3.80 .532 2a 23m .5 Eu

h L .— y p p I
I I . . . l-liq

. .i- 1..
.. .. .. ,.
u a. \‘
,. n U...
A . A t.-

‘1 l 1

. . »

 

 

 

 

 

 

 

 

 

 

 

 

. %
_ J
_ J
_ a
M .m
M m
8 .M
83ng I
633.com MM
88923 D ow

 

51

.Ewobm :23 com $5va owsaw .cﬁm >38? .3 PEME

on: F BEN 5?? SEN 53F 32%

 

i H - M i M . liliiLIil -.i|| ‘4 o
_

 

naﬁm as: 1T 3
x86 M68825 :6... m
segue: :6: m
M36 £002 lal Mm
M85 3880: III ,. .. .. m.
x020 zooococﬁg + .. ..H (w\

m on

M S

__ 8

_

OS

52

The average temperature during the summer months ranged from 135°C for the
headwater streams to 166°C for the adventitious streams and 167°C for the mainstem.
Using a Tukey’s Studentized Range Test, no signiﬁcant difference was found between
the mainstem and adventitious streams (P=0.6958). However, as a group, the headwater
streams were signiﬁcantly colder than the mainstem (P<0.0001) and adventitious streams
(P<0.0001). Although the headwater streams were cooler on average than the
adventitious streams and mainstem, all three stream types followed similar trends in
temperature ﬂuctuations (Figure 11) indicating that temporal variation in these streams
was largely driven by local weather conditions. McGillis Creek, a headwater stream, was
the warmest stream in the study (mean summer temp=17.4°C), followed by the
adventitious streams Hill Creek (168°C) and Unnamed Creek (167°C), the Pine River
mainstem (15.1°C), and the remaining headwater streams McDonald Creek (125°C), and

VanderCook Creek (11.5°C)(Appendix A).

53

~ OOZ/ 8 U6

+~ OOZ/V/ 6

.2588 8883 05 maize 25 83.8 .3 058382 .33 owaco>< .: Baum..—

 

 

 

EBmEmE 5.1.... Q
.6826on I I

335535. I

 

 

m N. L N 9 K 9

Z m Z V Z V Z V W

H N. N m N M M N /

Z Z Z Z Z Z Z Z Z

0 0 0 O O 0 O 0 0

0 0 O O 0 0 0 0 O

_ LilltllL nlllllbirlinl: {IILII 1|.1IILI...[|IIll N‘

M
7 M.
a... M > 1. :3... M
. <2)“ (>1 .M «MK—1...} Z :7... g
M . f. .72. r I:

b
(D) 33119.19de J. 41190 982mm,;

 

 

 

54

Discussion
Fish Community

As in previous studies relating stream order to ﬁsh species richness (Lotrich 1973;
Gorman & Karr 1978; Platts 1979; Beecher et al. 1988; Fairchild et al. 1998), the non-
adventitious sites sampled in this study clearly showed an increase in species richness
with increasing stream order (Figure 2). The deviation from this trend in adventitious
streams suggests that their connection with the species-rich mainstem and their warmer
stream temperature increases species richness above what could be expected based on
stream order alone. This is further illustrated by the observation that adventitious streams
were more similar on average to the mainstem than were headwater streams, whether
based on species presence/absence (Table 7) or relative abundance (Table 8).
Furthermore, adventitious streams were more similar on average to the mainstem than
they were to headwater streams (Tables 7, 8 and 9). This contradicts the Osborne &
Wiley (1992) study, which found no difference in species composition between
headwater and adventitious streams. However, no differences in SIOpe gradient were
detected in their study, regardless of tributary location within the Illinois watersheds.

Individual streams were generally more similar within stream types than across stream
types. An exception to this, however, was McGillis Creek, which was more similar to
adventitious streams than to its headwater counterparts. Based on S¢renson’s QS values
(Appendix B), McGillis Creek had more species in common with the adventitious
streams (QS=O.34, 0.75) than with the other headwater streams (QS=0.27, 0.00). Based

on Morisita’s F values (Appendix C), it is clear that species common to McGillis Creek

and the adventitious streams are present in high numbers (F=0.96, 0.74), while species

55

common to McGillis Creek and McDonald Creek are present in low numbers (P=0.02).
This suggests that factors beyond stream connectivity are inﬂuencing the composition of
ﬁsh species in headwater streams.

As indicated earlier, McGillis Creek was the warmest of all streams during the study,
while the other two headwater streams were the coldest. The species most frequently
occurring in McGillis Creek, creek chubs (S. atromaculatus), blacknose dace (R.
atratulus), and central mudminnows (U. limi), are species typically associated with
warmwater systems and were also found in high abundance in the adventitious streams.
Conversely, the species most frequently occurring in the other headwater streams, brook
trout (S. fontinalis) and mottled sculpin (C. bairdi), are species typically associated with
coldwater systems and were not captured in McGillis Creek. This reinforces the notion
that temperature regime is an important factor, even more so than stream order, in
determining a stream’s ﬁsh assemblage (Paller 1994; Lyons et a1. 1996). Furthermore,
the species composition in McGillis Creek contrasts with the RCC’s prediction that
headwater streams are dominated by coldwater ﬁsh species.

Based on the distribution of ﬁsh tolerance classiﬁcations (Figure 3), most species
present in each stream type were of intermediate tolerance, with relatively fewer tolerant
and intolerant species present. When viewed by relative abundance (Figure 4), however,
relatively few individuals of intermediate tolerance are present in adventitious and
headwater streams. The majority of individual ﬁsh in adventitious stream populations
were classiﬁed as tolerant, with few intermediate and intolerant individuals present. The
majority of individual ﬁsh in headwater stream populations were classiﬁed as intolerant,

with relatively fewer tolerant and intermediate individuals present. This supports the

56

hypothesis that adventitious streams would have a greater abundance of tolerant ﬁsh
species than headwater streams. In the mainstem, the proportion of individual ﬁsh was
comprised primarily of individuals classiﬁed as intermediate, with fewer tolerant and
intolerant individuals present. Lyons et al. (1996) found an increase in ﬁsh species
richness and a shift from intolerant-dominated to tolerant-dominated ﬁsh assemblages
following environmental degradation of several Wisconsin coldwater streams. The study
suggested that declines in water quality associated with such land uses as agriculture may
increase the average temperature and variability of the temperature, creating conditions
unsuitable to intolerant coldwater ﬁsh species.

Based on month-to-month similarity indices, the ﬁsh species composition of
adventitious streams was more variable in adventitious streams than headwater streams or
the mainstem (Tables 10 and 11). This was consistent with the hypothesis that
adventitious streams would show greater temporal variability compared to headwater
streams. S¢renson’s QS values for the adventitious streams were consistently lower than
headwater streams for all month-to-month comparisons. This suggests that the species
assemblage in adventitious streams generally changed throughout the summer, while the
ﬁsh species assemblage in headwater streams remained more stable.

Morisita’s F values were lower for adventitious streams (P=0.62) compared to
headwater streams (F=0.94), ﬁirther supporting the hypothesis that ﬁsh community
composition in adventitious streams are more variable than headwater streams. This
implies that ﬁsh communities of adventitious streams tend to be more variable in terms of

species presence and their relative abundance across months. Conversely, the ﬁsh

57

communities of headwater streams tend to be more stable in terms of species presence
and relative abundance across months.

The ﬁsh community composition in headwater streams and the mainstem generally ﬁt
the predictions made by the River Continuum Concept (Table 1). Intolerant coldwater
species, such as brook trout and mottled sculpin dominated two of the headwater streams.
McGillis Creek, however, was dominated by tolerant species associated with warmwater
systems. The mainstem was dominated by warmwater ﬁsh species of varying tolerance.
The ﬁsh assemblages of adventitious streams, however, were not dominated by coldwater
species as predicted by the RCC. In general, this study has found that shifts in tolerance
classiﬁcations for ﬁsh may occur based on spatial location within a watershed. I would
therefore propose that consideration be given to amending the RCC, based on further
research, to include the element of ﬁsh tolerance.

Macroinvertemte Community

The invertebrate metric scores for each stream type are consistent with the distribution
of ﬁsh tolerance classiﬁcations. The Procedure 51 (Michigan Department of
Environmental Quality 1997) invertebrate metric scores for adventitious streams were
lower than scores for headwater streams across all sampling events (Table 13) and may
indicate lower water quality in adventitious streams. The proportion of EPT (Figure 7),
an indicator of stream water quality (Menit & Cummins 1996a), was lower for
adventitious streams than for headwater streams. Furthermore, the ratio of chironimids to
EPT (Figure 8) was higher for adventitious streams relative to headwater streams.
Overall, the results of the invertebrate metrics suggest that the water quality in

adventitious streams may be unsuitable for intolerant ﬁsh and invertebrate species, while

58

allowing tolerant species to thrive. Reductions in EPT proportions accompanied by
increases in other, more tolerant, taxa have been documented in other studies of streams
impacted by agricultural practices (Dance & Hynes 1980; Lenat 1984; Lenat & Crawford
1994)

The invertebrate composition in headwater streams and the mainstem also ﬁt the
predictions made by the RCC, while adventitious streams did not. Collectors were the
most abundant ﬁrnctional feeding group across all stream types. As expected, shredders
were the second most abundant group in headwater streams due to increased inputs of
allochthonous organic material (V annote et al. 1980). In the mainstem, grazers were the
second most abundant group due to an increased forage base of periphyton from primary
production. In adventitious streams, however, shredders were the least abundant
functional feeding group, suggesting that riparian input of organic material is limited.
Furthermore, functional feeding groups in adventitious streams were the most
homogeneous of the stream types. Delong & Brusven (1998) found that, in watersheds
heavily impacted by agricultural use, invertebrate communities were relatively
homogeneous, dominated by species tolerant of agricultural non-point source pollution,
and were comprised of few shredders throughout the longitudinal stream continuum.

Various ratios of invertebrate functional feeding groups were used to characterize the
trophic status and amount of coarse particulate organic matter (CPOM) and ﬁne
particulate organic matter (FPOM) for the stream types (Table 14). These ratios give a
general indication of stream ecosystem attributes based on functional feeding group
abundance as a response to food resource availability (Merritt & Cummins 1996b).

Heterotrophic streams, dependent on inputs of allochthonous material, would be expected

59

to have a low ratio of scrapers and shredders to total collectors. The headwater streams
in this study had the lowest average score (0.15) for this ratio. However, the adventitious
streams did not score much higher (0.20). McGillis Creek consistently scored the highest
for headwater streams, indicating a greater tendency toward autotrophy. Unnamed
Creek, which had very little vegetative canopy, consistently scored the highest over all
sampling events for the adventitious streams and even approached the reference point for
autotrophy (0.75) during the summer sampling period. The mainstem sites scored the
highest for this ratio (0.49), indicating that it is more autotrophic in nature. These scores
are in general agreement with the predictions made by the River Continuum Concept
(Table 1).

As expected, the ratio of shredders to total collectors indicated a high loading of
CPOM associated with riparian vegetation in headwater streams (Table 15). Conversely,
the mainstem scored much lower, indicating an increased loading of FPOM from
upstream processes. These results support the RCC’s predictions that headwater
invertebrate communities are dependent on input of allochthonous organic material
(CPOM), while mid-order invertebrate communities are dependent on algae from primary
production and FPOM derived from upstream processing of CPOM (V annote et a1.
1980). However, scores for adventitious streams also indicated a high loading of FPOM.
This may be due to a decrease in vegetative canopy, increased siltation from agricultural
practices or urbanization, or a combination of the two. Minshall (1978) suggested that in
headwater streams of “open” regions, input of allochthonous CPOM might be limited, in
which case autochthonous carbon sources would predominate. Vannote et al. (1980) also

suggested that small tributaries to larger streams might have localized impacts on carbon

60

processing, dependent on the “volume and nature of the inputs.” Lenat & Crawford
(1994) found that suspended sediment yield and invertebrate collector abundance were
increased in agricultural-impacted streams relative to forested headwater sites. The
effects of agriculture practices in this region of the watershed may therefore contribute to
the increased F POM loading in the adventitious streams. This is further supported by the
increased ratio of FPOM in transport (TFPOM) relative to that stored in the benthos
(BFPOM) for adventitious streams (Table 16). Interestingly, McGillis Creek consistently
scored the lowest for headwater streams and was below or near the reference point,
indicating relatively high load of FPOM. This further reinforces the idea that agricultural
practices may lead to inconsistencies with the RCC by increasing the amount of sediment
in streams, while the associated loss of streamside vegetation may lead to a shift from
systems dependent on allochthonous inputs to those dependent on autochthonous energy
sources.

321mg!

Many of the in-stream habitat components were similar between headwater and
adventitious streams. Measures of stream size (e.g., width and depth) and habitat units
were similar. Inorganic substrate and physical structure provided by woody material
were also similar. Therefore, these characteristics do not appear to be responsible for the
differences in ﬁsh community structure.

Detectable differences in habitat attributes included the proportion of organic
substrate, hydrologic stability, stream velocity, and summer stream temperature. The
elevated proportion of organic sediments in adventitious streams may be the result of the

agricultural land use in that region of the watershed. Walser & Bart (1999) found that

61

sedimentation from agricultural practices was increased in mainstem reaches relative to
forested headwater reaches. This may partially explain the greater composition of
tolerant ﬁsh and decreased composition of intolerant ﬁsh in adventitious streams.

Based on stream velocity and staff gauge data, the hydrologic regime of headwater
streams and adventitious were different. Compared to headwater streams, adventitious
streams had slower stream velocities and exhibited greater hydrologic variability in
response to precipitation events. A study by Poff & Allan (1995) found that ﬁsh .
assemblages in hydro gically variable streams were characterized by species associated
with slow velocities and with afﬁnities for low-order streams, including creek chubs (S.
atromaculatus), central mudminnows (U. limi), and blacknose dace (R. atratulus). These
were also the most abundant ﬁsh species that we found in adventitious streams and
McGillis Creek, the slowest of the headwater streams.

On average, the temperature regime of adventitious streams was signiﬁcantly warmer
than that of headwater streams, which may also partially explain the higher mean ﬁsh
species richness in adventitious streams. Interestingly, McGillis Creek was the warmest
of all the streams studied (Appendix A) and had the greatest ﬁsh species richness of the
headwater streams. Studies by Paller (1994) and Lyons et al. (1996) have demonstrated
that warmwater streams exhibit greater ﬁsh species richness than coldwater streams,
which may explain the relatively higher species richness in McGillis Creek compared to
the other headwater streams.

Several factors, including groundwater and riparian vegetative shading, may have
contributed to the cooler temperatures in headwater streams relative to adventitious

streams. Although quantifying the groundwater regime was beyond the scope of this

62

study, it is likely that groundwater upwelling contributed to the cooling effect on
headwater streams because of the steep valley side, in headwater regions, producing a
large head for groundwater inputs. Generally, the headwater sites were shaded by a
denser canopy of vegetation than the adventitious streams, which may have also
contributed to the cooler temperatures in headwater streams. Several studies have found
an increase in stream temperature resulting from removal of streamside vegetative
canopy (Schlosser 1982b; Platts & Nelson 1989; Weaver & Garman 1994; Hetrick et al.
1998). McDonald Creek and VanderCook Creek lie completely within the Huron
National Forest and are largely shaded by a dense canopy of vegetation. While the
McGillis Creek and Hill Creek sites were largely shaded by vegetation, portions of these
streams ﬂow through agricultural land with little or no canopy. In addition, the lands
adjacent to these sites are used primarily for agriculture and are largely devoid of shading
canopy. UnNamed Creek ﬂows through a low-lying swampy area, with little vegetative
canopy and these sites are almost entirely exposed to direct sunlight. The loss of
streamside vegetation may also shift processes in low-order streams to an autotrophic
state typically associated with mid-order streams (Vannote et a1. 1980; Minshall et a1.
1985). Schlosser (1982) documented shifts in invertebrate and ﬁsh communities
resulting from removal of riparian vegetation. In addition, his study found an increase in

invertebrate and ﬁsh biomass attributed to an increase in primary production.

63

Conclusions

Stream order designation alone did not account for differences in ﬁsh community in
this study. Adventitious streams had greater ﬁsh species richness than headwater streams
of the same order. The similarity in ﬁsh assemblage between adventitious streams and
the mainstem lead me to conclude that the increase in species richness is largely due to
the connectivity of these stream types. However, connectivity did not entirely account
for the increase in species richness, as is evidenced by the increased ﬁsh species richness
in the headwater stream McGillis Creek. Habitat conditions were similar among the
headwater streams, with the exception of stream temperature. McGillis Creek was the
warmest stream examined in this study and had the highest species richness among the
headwater streams. Therefore, I conclude that connectivity and stream temperature are
the primary factors accounting for species richness.

The high temporal variability in ﬁsh communities of adventitious streams combined
with their similarity to the mainstem suggest that some ﬁsh species utilize these stream
types interchangeably, possibly for reproductive purposes, food, or thermal refuge.
Fisheries managers should therefore recognize the potential importance of adventitious
streams to ﬁsh communities of mid-order streams.

Adventitious streams also had a greater proportion of tolerant ﬁsh compared to
headwater streams, suggesting that the water quality was lower in adventitious streams.
The increase in tolerant ﬁsh was probably due to several reasons, including agricultural
impacts combined with greater ﬂow instability, stream temperatures, and siltation in

adventitious streams. Fisheries managers should realize that these impacts are typically

associated with removal of riparian vegetation and may therefore be minimized by
maintaining a vegetative buffer zone around these tributaries.

The lower IBI scores for macroinvertebrate communities further support the
conclusion that the water quality is lower in adventitious streams. This may be primarily
due to a high sediment load in transport, as indicated by the high ratio of TFPOM to
BFPOM. The remaining RCC ratios for trophic status and CPOM/F POM indicate that
adventitious streams are more autotrophic in nature compared to headwater streams and
are therefore an exception to the River Continuum Concept (Cummins 1977; Minshall
1978; Vannote et al. 1980).

This study was somewhat limited by time constraints and the low number of
adventitious tributaries present in the Pine River watershed, which resulted in a relatively
small number of streams sampled. Therefore, future research may involve larger
watersheds with more adventitious streams and perhaps comparison across watersheds
within a region. This research may serve to expand the scope of the River Continuum

Concept and improve its practical use in management applications.

65

APPENDICES

66

0.: 9: <3

 

v.: mg.— 9:

 

c.m_ 9.2 mm—

 

Gov oSumcanoH .5885 S82

 

 

RN N m N 3N a N m N 2N a N m N 83> 3% anagram 80:
N3 35 N; 3o 36 86 mod cod 85 $5 $83 as:
a: of q: 2: ed N.N_ «.2 as 3: A83 589 as:

wN 9N NN _.N «N 3 Z N Z as 523 so:

«.8 3a SN _.2 m. N NS 3: Ni 92 3:222 €83 .x.

N3 92 NNN OS 2: 33 NS NS 3w Sm .x.

od 3 ed on 2: ed 3 ed 3 85 .x.

EN SN 0: E S: S S: n: 92 38 .x.
:32 :32 :82

N N N N N N cacao sauem

83398: 532633 Houchaom 09C. Ewobm

x020 V309825 820 £502 #85 22802 :85

e m e N N _ a 35.

62953 8?. :03 Now 283980 35am .< umcaonq<

67

 

Flu:

 

 

 

 

 

2.2 Na a2 3.2 NE NS 3.2 NS Q: Ga 058888. 5865 322
2a 3. N.N 3N nN N.N SN 3 N.N 33> 8% s§§m 50:
2 .o Ed NS 85 86 3o 85 Nod mod 35 28:5 522
EN. ”.9 EN 2; w: 3: of a: 0.2 Ea 58a 932
0.: N.NN n: N.N aN N.N N N M: 95 £23 502

Q oN 3 no 3: 3 N.Nm 9mm in 3:222 383 .x.
Sn 2: n: oN NS o.NN is 33 0% SE .x.
3N ed 98 o.N od 3 S 0: 9o 052 .x.
0.: od EN 3N 5% SN N.NN 3: EN Boa .x.

:82 :32 :32

m m N N N N 520 $25
88882 325:3: msomzﬁozé 09$. 885m
35 8: 820 8825 x86 Em $05
2 2 8 a m N a 35

SEE < 5:83.

68

wmd

:d

36

$6

ﬁnd

53M 85

de

cod

5N6

mud

vﬁo

2.0

cod

who

cod

36

£502 goounoc§>

owd

N.N.o

ﬁnd

36

mmd

298902

9.0

mud

cod

ﬁmd

mmd

Beans:

$6

$6

3.0

mmd

mmd

:5

bid 0:5

£5002

xooocov§>

2.25902

Beams:

:5

.59: m0 62$: Pcomcocem wEm: mﬁwobm 35% $88 5382288 36on gm mo 55:85 .m umuuonnt.

69

dmd

2d

wﬁd

Nvd

:md

33% 0:3

dmd

ddd

Ndd

vnd

odd

Qd

ddd

odd

ddd

mdd

£502 x8985>

Ed

Nod

odd

Nod

wdd

225:0:

Nvd

Ed

ddd

Nod

5d

3:353

de

odd

Ndd

Vdd

Ed

:5

53% 0:5

£5002

xooouocﬁg

295:0:

3:353

Em

.Gca :8: 3 :3:on m3 xowﬁ L and: 9862.82 mﬁm: 2:85 398: 5:60:88 86on :mm .«o bum—«gm .U 56:09?

70

 

Nod dwd vwd and who Ed 5d wwd
mod mod nod awd ood dmd mod Nod
dod do; do; do; do; do; do; do;
mod 5d do; mwd owd cwd owd mud
mod Nod nmd mod omd nod mmd mmd
vod and dud Sd vwd wmd Nwd dud

Hm 0:203: M313 w:<-::_. 37:2. u:<.>m2 37%.:2 5.73%

:35 0:5

£5002

€850§>

2.28:0:

3:853

:5

.x0::_ m0 Amvﬂv 0.:0m:0:em 36: 8:05 :80 55:5 b.5286 52:25:00 :25 55:79-5:02 .9 26:05.3:

71

 

odd :d end dmd and mmd End mod
odd 8d 3d 3d mad wad mod mod
odd mwd 3d 3d 3d and end ~:wd
odd 3d 3d wad add wad mod dad
odd mvd Pd 2 d omd nmd Ed 2d
odd dwd mod mwd owd dad mod owd

mm 0E0}: 8223 52:3. 87:3. w:<.>m2 _:_.#m2 8:333

5:: BE
mzzmxaz
06:90:83
2.08:0:
:08:::D

:5

2008:

80m .3 “00:52:08 03 08:8 L 83: £328.82 8%: 8:05 :80 853» 00:80:86 5838800 :25 5::8-8-5:o2 .H ﬁe:25<

72

 

w o c w o o o :02 0.30308: 20.250833 388.3 535
a o o m o o o :02 30.8830 303882 000m 5:08:08m
a o o w o o H 32 33:0 38:3 83:93::
a o : o o o w 6:: 33303 0.882 9:0 323:5:
a o o o o o 0 EH 3:80: “ﬁgmmﬁi 30:82 08855
S o o S o o o 3:: 3:? 32:80:: so: 38%:
>088“:

2 o o 2 o o o :08: SEQ =0§So§30~ 0.005 8088 Z
N: o o _ o o 2 EH 3030~00b0 §§M~2§02 :0:_:m 5:30
o: o m m m o o :02 2303.08 058,808.: Stan 30:83:
2 o N: o N o m :02 80 032:8“: 000D 30::3: 80::02
mm o o mm o o o :02 020000>0~xu0x0m :80: 323%
mm o o mm o o o 38: 800.2%? Eam~0~=0§m Bxgmwom 805.8 Z
ov o mm o o o m :02 883008 00:30 0.030202 0.005
w: o _ mm o o m :02 33500 .3253 :08:m :08800
a o o S o o o 82 ”88:3 83903:: 8am :8:
mm o m mm o o mm .09 03:02:30 ﬁEoRYN :mugm :005
mm o v mm o o o 38: 238380 050.6% SEQ 688:0
mm o E Q. o o m :3. 80805500 03262030 Suﬁsm 0:55
0: o o N: o o v 82 8:088 3.6a: 53ng
ova o o EN o o Om :02 80309: 3.80% BED 0:10:85
am o 5 0mm _ 0 mm 82 530:0: 2805.0 :5: 5:2
mmm o KN m: mo: : on: .8. .550 5:5 >3:88:32 8800
0mm mom o S o mmm o 38: Eucczoxmzzamiam 88H 0:0on
:K o w: com ovm vm a: BB 0333:: 3:30.53¥ 0009 0005—85
0mm 08 o 2: 0 men 2 6:: :33 38.0 aasom 8502
3:: o SN Km 2: 0 gm EH 33300508: 032080,: :30 0.00:0
730,—. U> 75 m2 U02 :02 D: 00820.5. 08:7: 052.3% 08:: :08800

don—“.575 DHNMBEDHE

H :02 3:80—08: u 38: ”8820:; n EH .8005 :000 8 :08900 8:85: ::.0 .0033 8:828 .8088 860% :03 .m 8:59;.

73

@0de

Nv—iv—dv—dv—i

\Olnlnlnlnﬁ'ﬁ'ﬂ'

th

OOOOOO

OOOOOOO O

3:. mo?
o ﬂ
o ﬂ
0 _
o o
c _
o ﬂ
o v
o m
o v
o o
o o
_ o
_ m
o _

Nwa

OOOOOO

Ov-‘V'V‘AOOO O

moo

OOOOOO

WOOOOOO O

Nmo

v—Ov—‘OOO

OOOOWOF‘ O

.508

we:
32
32
to:
32
BE

:55
:55

3H
BE
:55
32
to:
32

Eﬁzﬂna 338.898:
33853.4 Swish»:
53.535» 358.898:

830:.233 Macks?

“SENBQQEEME MSBEQK
”3.83 Roam

uﬁﬁﬁsu-a
maﬁzkfouzQ

8:5 053%
3&8;qu .438qu
33943:: Macho?
«Swagger: $3982
$.38 QESmomﬁm
aSmE EEQEV
8.23335 azumtawga:

38.60% .525
mmam ﬁsoﬁowbﬁ
oﬁonwom 5200

Hoﬁnm 2808m—

oﬁaﬁo £85

82 5252

:oEBm xoocEU
BOMB gay—m
532m

55% =38%
£6 55
5th ~32
Suﬁsm x85
comm imam

.353 m 55:22?

74

2

OO

00
E
mm
vow
mm
3
mm
cm
mm
on
Sm

o:

3
Q.»
mmv

m—
2

130,—.

QESEooo

NO
Nth

Now‘govvoxoxomxog—«fﬂhom

80>?— 00E

ﬂgmfonoovo

o
ooovonoﬂzoao

v-‘V'
v-‘h

0083.5

o o
_ ﬂ
0 0
ﬂ _ on
m h—
o o
E. ow
m o
c o
3.. o
v o
_ o
o o
v: E.
H g o
m E.
o o
o o
o 2
o o
C o
_ ~
03 no
o o
N a
o w

E: 2000.—08.5» £5002 20:09.2

v—ﬂ
mo_,

hm

ON

an

V3 V3
No—_o8oooov,§f£§omo~m

og‘v—ooov—ogoegomo

mm

o
o
mm

o
_
o
v:
N
ﬂ
o

033880020m
88.50
0338800
033080000020
03800
0358035
03580808m
038:5
80680
002me
83380
033800
0350803800
038808.30
03%?08200
03558505
033080.020m
036325
03325
036532
038802

£5...

80005
80005
80005
80005
8085
8085
80005
80005
3000.5
80005
80005
8085
30005
80005
80005
80005
80005
50005
80005
8000.5
8008—
30005
0003080

0020502

302830
8595:

0.00—U

3802
3885.2
8082.2
3802
3885.2
388:2
3802
8885
388:2
382:5.
3802
3885.2
3802
3885.2
308852
388:2
388:2
3885.2
388:2
388:2
3885.2
3802
3802
3885.2

3:02

3:02

8...»...—

.m£:08 33 00:0 =0 .53 8008 £000 E 308800 8008: 300 80008 880 000508858002 .0 58.0.52

75

coon

0N

mm
mm
ov
om

mvm
E
o~

mm
cm
3
vm
mm—

2
v:
3
Nb
2
o:

Nam

coo

Qﬂccoﬁooﬁeomoomcgo—‘oSctxvo

7.:

coo

NO

mm
mm

W)
Sﬂv—occoogocoowococ

mm

v—l

oco

NO

V3 V3
Nocmgcwaomcﬁomchoo—‘cjcg

O

oco

ccmoh:~cm§vcmcm

v—4
M

cocgccﬁm

coo

OO
cg‘ogcococooov—oommco‘EoV—ac

v—i
V

N
[\

coo

cc§v~°ﬁomcc~o~mcmvcvoc

V3 m
VOF'ON

5.;

038083 «vogue—om
8283A £58580
82855 $8.539

8235
ouﬁEboEoanH
omoEwomh
oaoSEoombm
322385
823m
$2888qu
omoﬁonaomm
086388838
3205
omEESOQEEm
2862qu
0%:qu
0383080 Z
egg—02
82398832
32335::
832—35
82502833
omomﬁoxcoﬂ
omtEQAKOHPAm

8280333

8035
8035
3085
3085
8085
8085
8035
8035
8085
8085
8085
8085
$085
8085
3085
8085
8083
8083
8085
803.5
8085

8085

28883802

83:02
83:22
«82:32

330%

8885.2

«885%

8385.2

3885?

3385.3.

3885.2

3885.2.

«@3852

moon—85%

33038.2

«8885.2

3385.2.

mooaoEE

«@8853

3885.2

3885.2

3885?.

380%

380%

88852

«vouch—ohm

.3808

O amt—3&3

76

Appendix H. Invertebrate metric scores for each stream based on Procedure 51
(Michigan Department of Environmental Quality 1997). Stream names are abbreviated
as follows: McD = McDonald Creek, McG = McGillis Creek, VC = VanderCook Creek,
HC = Hill Creek, UN = Unnamed Creek, Pine = Pine River mainstem.

 

 

 

 

Month Metric McD McG VC HC UN Pine
Jan Total # of Taxa O O O O -1 0
Jan Total # of Mayfly Taxa O 1 -1 -1 0 1
Jan Total # of Caddisfly Taxa O O O O -1 -1
Jan Total # of Stoneﬂy Taxa 1 0 1 O -1 1
Jan % Mayﬂy Composition 0 1 0 O 0 1
Jan % Caddisﬂy Composition 0 O O 0 O 0
Jan % Dominant Taxon 1 1 O 0 -1 1
Jan % Isopods, Snails, & Leeches 1 1 1 1 1 1
Jan % Surface Dependent 1 1 1 1 1 1
Totals 4 5 2 l -2 5
May Total # of Taxa O 1 1 1 1 0
May Total # of Mayﬂy Taxa O 0 0 -1 0 1
May Total # of Caddisﬂy Taxa O O O 1 -1 0
May Total # of Stoneﬂy Taxa O -1 1 -1 -1 1
May % Mayfly Composition 1 1 1 1 O 1
May % Caddisﬂy Composition 1 0 O O O 0
May % Dominant Taxon O 1 O 1 -1 0
May % ISOpods, Snails, & Leeches 1 l 1 O -1 1
May % Surface Dependent 1 1 1 1 O 1
Totals 4 4 5 3 -3 5
Aug Total # of Taxa O O O 1 1 0
Aug Total # of Mayﬂy Taxa O 0 -1 -1 O 1
Aug Total # of Caddisﬂy Taxa O O 0 -1 0
Aug Total # of Stoneﬂy Taxa O -1 1 -1 -1 1
Aug % Mayﬂy Composition 0 l -1 0 -1 1
Aug % Caddisﬂy Composition 1 0 l 0 -1 0
Aug % Dominant Taxon 1 1 1 1 1 1
Aug % Isopods, Snails, & Leeches 1 l 1 1 -1 1
Aug % Surface Dependent O 1 O O -1 0
Totals 3 3 2 1 -4 5

77

Literature Cited

Allan, D. J. 1995. Stream ecology: structure and function of running waters. Kluwer
Academic Publishers, AH Dordrecht, The Netherlands.

Barbour, M. T., J. Gerritsen, B. D. Snyder, and J. B. Stribling. 1999. Rapid
bioassessment protocols for use in streams and wadeable rivers: periphyton,
benthic macroinvertebrates and ﬁsh, second edition. EPA 841-B-99-OOZ. US.
Environmental Protection Agency; Ofﬁce of Water; Washington, DC.

Beecher, H. A., E. R. Dott, and R. F. Femau. 1988. Fish species richness and stream
order in Washington state streams. Environ. Biol. Fish. 22(3): 193-209.

Berkrnan, H. E., C. F. Rabeni, and T. P. Boyle. 1986. Biomonitors of stream quality in
agricultural areas: ﬁsh versus invertebrates. Environ. Manage. 10(3): 413-419.

Cooke, S. E., E. E. Prepas, R. W. Bachmann, J. R. Jones, J. H. Peters, and D. M. Soballe.
1995. Impact of agriculture on stream water quality in the boreal mixed-wood
forest of northern Alberta. Lake Reserv. Manage. 11(2): 130-131.

Cuffney, T. F., M. R. Meador, S. D. Porter, and M. E. Gurtz. 2000. Responses of
physical, chemical, and biological indicators of water quality to a gradient of
agricultural land use in the Yakima River Basin, Washington. Environ. Monit.
Assess. 64(1): 259-270.

Cummins, K. W. 1962. An evaluation of some techniques for the collection and analysis
of benthic samples with special emphasis on lotic waters. Am. Midl. Nat. 67:
477-504.

Cummins, K. W. 1977. From headwater streams to rivers. Am. Biol. Teacher. 39: 305-
312.

Dance, K. W., and H. B. N. Hynes. 1980. Some effects of agricultural land use on
stream insect communities. Envir. Pollut. Series A 22: 19-28.

Day, R. W., and G. P. Quinn. 1989. Comparisons of treatments after an analysis of
variance in ecology. Ecological Monographs. 59(4): 433-463.

Delong, M. D., and M. A. Brusven. 1998. Macroinvertebrate community structure along
the longitudinal gradient of an agriculturally impacted stream. Environ. Manage.
22(3): 445-457.

Fairchild, G. W., R. J. Horwitz, D. A. Nieman, and M. R. Boyer. 1998. Spatial variation

and historical change in ﬁsh communities of the Schuykill River drainage,
southeast Pennsylvania. Am. Midl. Nat. 128: 611-616.

78

Gorman, O. T., and Karr, J. R. 1978. Habitat structure and stream ﬁsh communities.
Ecology 59(3): 507-515.

Gonnan, O. T. 1986. Assemblage organization of stream ﬁshes: the effect of rivers on
adventitious streams. Am. Nat. 128: 611-616.

Harrelson, C. C., C. L. Rawlins, and J. P. Potyondy. 1994. Stream channel reference
sites: an illustrated guide to ﬁeld technique. Gen. Tech. Rep. RM-245. Fort
Collins, CO: US. Department of Agriculture, Forest Service, Rocky Mountain
Forest and Range Experiment Station. 61 p.

Hetrick, N. J ., M. A. Brusven, W. R. Meehan, and T. C. Bjomn. 1998. Changes in solar
input, water temperature, periphyton accumulation, and allochthonous input and
storage after canopy removal along two small salmon streams in southeast Alaska.

Trans. Am. Fish. Soc. 127(6): 859-875.

Horn, H. S. 1966. Measurement of “overlap” in comparative ecological studies. Am.
Nat. 100: 419-424.

Horton, R. E. 1945. Erosional development of streams and their drainage basins. Bull.
Geol. Soc. Am. 56: 275-370.

Horwitz, R. J. 1978. Temporal variability patterns and the distributional patterns of
stream ﬁshes. Ecological Monographs. 48: 307-321.

Ibbotson, A., P. Armitage, W. Beaumont, M. Ladle, and S. Welton. 1994. Spatial and
temporal distribution of ﬁsh in a small lowland stream. Fisheries Management
and Ecology. 1: 143-156.

Johnson, B. T. 1986. Potential impact of selected agricultural chemical contaminants on

a northern prairie wetland: a microcosm evaluation. Envir. Toxicol. Chem. 5:
473-485.

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

Kramer, C. Y. 1956. Extension of multiple range tests to group means with unequal
numbers of replications. Biometrics 12: 307-310.

Kerans, BL, and J. R. Karr. 1994. A benthic index of biotic integrity (B-[BI) for rivers
of the Tennessee Valley. Ecol. Appl. 4(4): 768-785.

Lenat, D. R. 1984. Agriculture and stream water quality: a biological evaluation of
erosion control practices. Envir. Mgmt. 8: 333-344.

79

Lenat, D. R., and J. K. Crawford. 1994. Effects of land use on water quality and aquatic
biota of three North Carolina Piedmont streams. Hydrobiologia. 294: 185-199.

Leonard, P. M., and D. J. Orth. 1986. Application and testing of an index of biotic
integrity in small, coolwater streams. Tran. Am. Fish. Soc. 115(3): 401-414.

Loehr, R. C. 1974. Characteristics and comparative magnitude of nonpoint sources. J.
Wat. Pollut. Cont. Fed. 46: 1849-1872.

Lotrich, V. A. 197 3. Growth, production, and community composition of ﬁshes
inhabiting a ﬁrst-, second-, and third-order stream of eastern Kentucky. Ecol.
Mono. 43: 377-397.

Lydy, M. J ., A. J. Strong, and T. P. Simon. 2000. Development of an index of biotic
integrity for the Little Arkansas River Basin, Kansas. Environ. Contarn. Toxicol.
39(4): 523-530.

Lyons, J ., L. Wang, and T. D. Simonson. 1996. Development and validation of an index
of biotic integrity for coldwater streams in Wisconsin. N. Am. J oum. Fish. Man.
16(2): 241-256.

Merritt, R. W., and K. W. Cummins. 1996a. An introduction to the aquatic insects of
North America, 3rd ed. Kendall/Hunt, Dubuque, LA.

Merritt, R. W. and K. W. Cummins. 1996b. Trophic relationships of macroinvertebrates.
Pages 453-465 in Methods in stream ecology, Hauer, F. R. and G. A. Lamberti,
eds. Academic Press, San Diego, CA.

Michigan Department of Environmental Quality, Surface Water Quality Division.
Revised January 1997. GLEAS Procedure #51: Qualitative biological and habitat
survey protocols for wadable streams and rivers, Great Lakes and Environmental
Assessment Section, Lansing, MI.

Michigan Department of Environmental Quality, Surface Water Quality Division. 1996.
Update of GLEAS Procedure 51 metric scoring and interpretation, Great Lakes
and Environmental Assessment Section, Lansing, MI.

Minshall, G. W. 1978. Autotrophy in stream ecosystems. Bioscience. 28(12): 767-771.

Minshall, G. W., K. W. Cummins, R. C. Petersen, C. E. Cushing, D. A. Bruns, J. R.
Sedell, and R. L. Vannote. 1985. Developments in stream ecosystem theory.
Can. J. Fish. Aquat. Sci. 42: 1045-1055.

Morisita, M. 1959. Measuring interspeciﬁc association and similarity between
communities. Mem. Fac. Sci. Kyushu Univ. Ser. B Biol. 3: 65-80.

80

Nelder, J. A., and R.W. M. Wedderbum. 1972. Generalized Linear Model. Journal of
the Royal Statistical Society, Series A (General). 135(3): 370-384.

Naiman, R. J. 1983. The annual pattern and spatial distribution of aquatic oxygen
metabolism in boreal forest watersheds. Ecological Monographs. 53(1): 73-94.

Naiman, R. J ., J. M. Melillo, M. A. Look, T. E. Ford, and S. R. Reice. 1987.
Longitudinal patterns of ecosystem processes and community structure in a
subarctic river continuum. Ecology. 68(5): 1139-1156.

Osborne, L. L., and M. J. Wiley. 1992. Inﬂuence of tributary spatial position on the
structure of warmwater ﬁsh communities. Can. J. Fish. Aquat. Sci. 49: 671-681.

Paller, M. H. 1994. Relationships between ﬁsh assemblage structure and stream order in
South Carolina coastal plain streams. Trans. Am. Fish. Soc. 123: 150-161.

Pennak, R. W. 1989. Freshwater invertebrates of the United States, 3rd ed. Wiley, New
York, NY.

Platts, W. S. 1979. Relationships among stream order, ﬁsh populations, and aquatic
geomorphology in a Idaho river drainage. Fisheries. 4(2): 5-9.

Platts, W. S., and L. N. Nelson. 1989. Stream canopy and its relationship to salmonid
biomass in the intermountain West. N. Am. J oum. Fish. Man. 9: 446-457.

Poff, N. L., and J. D. Allan. 1995. Functional organization of stream ﬁsh assemblages in
relation to hydrological variability. Ecology. 76(2): 606-627.

SAS Institute 1999 SAS/STAT users guide, Version 8. SAS Institute, Inc. Cary, North
Carolina, USA.

Schlosser, I. J. 1982a. Fish community structure and function along two habitat gradients
in a headwater stream. Ecol. Mono. 52(4): 395-414.

Schlosser, I. J. 1982b. Trophic structure, reproductive success, and grth rate of ﬁshes
in a natural and modiﬁed headwater stream. Can. J. Fish. Aquat. Sci. 39: 968-
978.

Sheldon, A. L. 1968. Species diversity and longitudinal succession in stream ﬁshes.
Ecology. 49(2): 193-198.

Simonson, T. D., J. Lyons, and P. D. Kanehl. 1993. Guidelines for evaluating ﬁsh
habitat in Wisconsin streams. Gen. Tech. Rep. NC-164. St. Paul, MN: US.
Department of Agriculture, Forest Service, North Central Forest Experiment
Station. 36 p.

81

S¢renson, T. 1948. A method of establishing groups of equal amplitude in plant
sociology based on similarity of species content. Det. Kong. Dan. Videnski.
Selak. Biol. Skr. 5: 1-34.

Steedman, R. J. 1988. Modiﬁcation and assessment of an index of biotic integrity to
quantify stream quality in southern Ontario. Can. J. Fish. Aquat. Sci. 45(3): 492-
501.

Tukey, J. W. 1953. Some selected quick and easy methods of statistical analysis.
Transactions of the New York Acadamy of Sciences, Series 2, 16: 88-97.

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

Walser, C. A., and H. L. Bart. 1999. Inﬂuence of agriculture on in-stream habitat and
ﬁsh community structure in Piedmont watersheds of the Chattahoochee River
System. Ecol. Freshwat. Fish. 8(4): 237-246.

Weaver, L. A., and C. G. Garman. 1994. Urbanization of a watershed and historical
changes in a stream ﬁsh assemblage. Trans. Am. Fish. Soc. 123: 162-172.

Whiles, M. R., B. L. Brock, A. C. Franzen, and S. C. Dinsmore II. 2000. Stream
Invertebrate Communities, Water Quality, and Land-Use Patterns in an
Agricultural Drainage Basin of Northeastern Nebraska, USA. Environ. Manage.
26(5): 563-576.

Wolman, M. G. 1954. A method of sampling coarse riverbed material. Trans. Am.
Geophysical Union. 35(6):951-956.

82

 

llllllllﬂﬂlﬂllﬂﬂlﬂlllllllllllll