{V

M

H i \ Willi Mill l»

_ 3 1293 00796 1

Till ll

 

 

 

 

This is to certify that the

thesis entitled

FIRST-YEAR RESPONSES OF A STREAM AND
ITS BROOK TROUT POPULATION TO HINGE-CUTTING
OF RIPARIAN BRUSH

presented by

Mark Muir Ultis

has been accepted towards fulfillment
of the requirements for

Master of Science degree in Fisheries and Wildlife

 

 

/‘ , ./
/////Zi,. /// //..///

Major professor

 

Date February 7, 1986

 

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

ll ill

99

2

I-

i L

 

MSU

 

 

 

RETURNING MATERIALS:
Place in book drop to

 

 

LJBRAfiJES remove this checkout from
mun—n. your record. FINES will
be charged if book is
returned after the date
stamped below.
‘ 4\- .17 1': ”:1. 0 E" f’“
a ' ‘ WM @092
E37?§}'e PM) 9
:41; f f "‘T
0;§'€ s
FEB 1% mg
553‘ 0 2 2005

 

 

FIRST-YEAR RESPONSES OF A STREAM AND
ITS BROOK TROUT POPULATION TO HINGE-CUTTING
OF RIPARIAN BRUSH

By

Mark Muir Ultis

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

1986

ABSTRACT
FIRST-YEAR RESPONSES OF A STREAM AND
ITS BROOK TROUT POPULATION TO HINGE-CUTTING
OF RIPARIAN BRUSH

By

Mark Muir Ultis

Changes in brook trout (Salvelinus fontinalis) abundance and

 

stream channel morphology were investigated after the addition of
overhead bank cover in the summer of 1980 by hinge-cutting riparian
brush along the Salmon Trout River, Marquette County, Michigan. Within
the following year, the trout population declined abruptly over the
entire study area (decreases of 21% in numbers, 38% in biomass).
However, trout abundance within five treated sections in the study
area remained relatively constant (no change in numbers, decrease of
11% in biomass). Changes in stream channel morphology included an

8% decrease in width, 5% increase in depth, 20% increase in water
velocity, and 121% increase in overhead bank cover within the treated
sections. The study suggests that hinge-cutting of riparian brush
may be an economical method of cover creation for streams with suit-

able bank vegetation.

ACKNOWLEDGEMENTS

I would like to express appreciation to Drs. Ray J. White and
William Taylor for their guidance in the completion of this project.
Drs. Niles Kevern and Richard Merritt provided review and editorial

comments.

I would like to thank Guy Fleischer, Kurt Fausch, Greg Curtis,

Chris Bennett, Peter Jacobsen, and Paul Scheer for their assistance

in the field studies.

The Huron Mountain Club provided funds and facilities which made

this study possible.

ii

TABLE OF CONTENTS

Page
LIST OF FIGURES . ........ . ....................................... iv
LIST OF TABLES .................................................. v
INTRODUCTION .................................................... 1
STUDY AREA ........... O O O O 0 ........ O OOOOOOOOOOOOOOOOOOOOOOO O 0 0 O O 0 5
Specific Location of Study Area ............................ 7
Water Quality and Discharge ......... . ............ .......... 7
Bed Materials .............................................. 7
Biota ...................................................... 10
METHODS AND MATERIALS ........................................... 12
Population Estimation ...................................... 12
Calculation of Population Estimates ........................ 14
Habitat Studies ............................................ 16
Drift sampling OOOOOOOOOOOOOOOOO O 0 O O O O O O O O O O 0000000000 O ..... 17
Habitat Alteration ......................................... 17
Statistical Analysis ....................................... 18
RESULTS AND DISCUSSION ............. . ..... . ........... . ..... ..... l9
Trout Population ............................ . .............. 19
Trout Growth ................ . .............................. 24
Habitat .................................................... 24
Streambed Material ..................... . ........ . .......... 30
Comparison of Trout Populations between Treatments
and Controls ....... . ...................................... 36
Invertebrate Drift ................................ . ........ 37
Implications for Management ................ . ............... 41
Conclusions ...... . ......................................... 44
LITERATURE CITED ................................ ................ SO
APPENDICES ............................................... . ...... 53

iii

LIST OF FIGURES

Number
1 The Salmon Trout River ..................................
2 Salmon Trout River study area ...........................
3 Mean stock density and standing crOp estimates in
control and treatment sections over four sampling
periods .................................................
4 Relationship between stock density and standing crop
with length of bank cover, September 1979 ..... . .........
5 Relationship between stock density and standing crop
with length of bank cover, June 1980 ....................
6 Relationship between stock density and standing crop
with length of bank cover, September 1980 ...............
7 Relationship between stock density and standing crop
with length of bank cover, May 1981 .....................

iv

23

-37

38

39

4O

LIST OF TABLES

Number .ngg
1 Stock density and standing crop estimates (all sizes) ..... 20
2 Stock densities and standing crop estimates (< 150 mm) .... 21
3 Stock density and standing crop estimates (> 150 mm) ...... 22
4 Physical characteristics of treatment stations ............ 25
5 Physical characteristics of control stations .............. 26
6 Fall standing crop comparisons ............................ 27
7 Growth of marked year-class brook trout ...... . ............ 28
8 Length of streambank meeting overhead cover criteria ...... 30
9 Substrate composition ................ .. ................... 32

10 Correlation coefficients standing crop .................... 34
11 Correlation coefficients population density ............... 35
12 Significance of changes in trout abundance between years 42
13 Significance of changes in trout abundance between

paired control and treatment sections ..... ................ 43
14 Total drift rates ......................................... 44
A1 Discharge estimates ........... ...... ............ .......... 53
A2 Shocking dates ............................................ 54
A3 Finclip record for wild young of year ......... . ........... 55
A4 Recapture efficiency ...................................... 56
A5 Brook trout population estimates by station in the

Salmon Trout River, September 1979 ........................ 57
A6 Brook trout biomass estimates by station in the

Salmon Trout River, September 1979 ........................ 59

Number

A7

A8

A9

A10

A11

A12

Brook trout population estimates by station in
the Salmon Trout River, June 1980 ........ ................

Brook trout biomass estimates by station in the
Salmon Trout River, June 1980 ............... .............

Brook trout population estimates by station in the
Salmon Trout River, September 1980 ..... ..................

Brook trout biomass estimates by station in the
Salmon Trout River, September 1980 .................. .....

Brook trout population estimates by station in the
Salmon Trout River, May 1981 ........ ...... .. .............

Brook trout biomass estimates by station in the
Salmon Trout River, May 1981 .............................

vi

63

65

67

69

71

INTRODUCTION

Trout habitat management can be divided into four general areas:
habitat protection, restoration, enhancement, and maintenance. The
first, habitat protection, involves preventing activities which damage
streams or their drainage basins. This may entail such practices
as fencing streambanks in pastures to decrease damage from grazing,
preventing channel modifications such as snagging, clearing, and
ditching and controlling sources of pollution, including sedimentation
from agriculture and other human activities. Restoration consists
largely of repairing the effects of damaging activities which were
not prevented. Habitat enhancement is the creation of more suitable
habitat than would naturally occur. This may be in relatively undamaged

streams leading to a "hyperhabitat,"

or may be used in stream sections

lying near the end of natural suitability for salmonids. Habitat

maintenance is the continuing upkeep of previous management endeavors

to prevent a return to the prior conditions. This is most critical

in cases of hyperhabitat where the stream will eventually return to

a less productive state without periodic inputs of energy and materials.
State and federal programs of trout stream improvement had begun

by the 1930's (Hubbs et a1. 1932), however, little evaluation of the

effects on trout populations were possible until accurate methods

of estimating trout abundance by electrofishing were developed in

the 1950's (White 1975). One method of habitat management, used in
both restoration and enhancement, is the construction of current de-
flectors and bank covers. The primary purposes of these devices are
to increase the depth of the channel and to increase the amount of
usable bank cover. The importance of increased channel depth to stream
salmonids is that more living space is available. An increase in
the number of suitable microhabitats in a given stream section may
reduce agonistic behavior and allow higher population densities (Chapman
1966; Allen 1969). Many investigators have demonstrated the impor-
tance of bank cover in regulation of trout abundance. Trout populations
declined following removal of brush cover from a Montana stream
(Boussu 1954). Installation of bank covers and deflectors preceded
increased trout abundance in several studies (Saunders and Smith 1962;
Hale 1969, in White 1973; Hunt 1971; White 1975). The most compre—
hensive study of the effects of stream habitat improvement was done
by Hunt (1971) on Lawrence Creek in Wisconsin. He found that by
increasing the amount of available bank cover, the abundance of age
II + trout increased dramatically within three years of the alterations.
He proposed that the increase was due to greater overwinter survival
afforded by the extra cover.

In 1976-77, the relationship between abundance of instream bank
cover and abundance of brook trout was studied in two sections of
the Salmon Trout River, Marquette County, Michigan (Enk 1977). The
study showed that the variation in abundance of trout in 100 m stations
was primarily due to variation in the amount of instream cover. It
was hypothesised that addition of cover could allow the trout popu-

lation to expand provided other environmental factors were favorable.

The present study was designed to test the effect of rapid cover
creation on trout abundance, and on the physical characteristics
of the stream channel in the Salmon Trout River. The method of
cover creation chosen was "hinge-cutting" of riparian brush. This
involved sawing partially through the stems of streamside brush and
folding the tops over into the water. This method is somewhat similar
to the use of brush bundles for stream habitat improvement by the
Wisconsin Department of Natural Resources but takes less time and
effort to construct (R. J. White, pers. comm.). Hinge-cut cover
usually comes in close proximity to the stream bed, which is thought
to be most desirable as a position choice by trout (Bassett 1978).

Certain manipulations of streamside vegetation are regarded as
useful in habitat management for trout (White and Brynildson 1967).
Removal of trees and high brush to promote growth of grasses and low
brush helps to stabilize banks and provide overhangs. Planting of
trees is usually discouraged, except where stream temperatures are
unusually high (White and Brynildson 1967). Recently, experiments
were carried out to test the effect of complete brush removal on
trout abundance and channel form in a number of Wisconsin streams
(Hunt 1979). While disruption of fish populations due to stream
flow variation confounded the results, it was concluded that removal
of brush led to a larger stock of legal sized trout, owing to im-
proved channel conformation.

The hinge-cutting technique of habitat alteration used in the
present study is new, hopefully combining the beneficial effects of

increasing overhead bank cover, narrowing and deepening of the

channel, and removal of shade-producing brush to allow grasses to

stabilize the banks. General objectivesof this study were:

1)

2)

3)

To determine the effect of rapid creation of cover by
hinge-cutting riparian brush on the abundance of trout

in selected treatment sections on the Salmon Trout River.
It was hypothesized that the trout population would in-
crease in response to cover creation, but that there would
be a time lag in response such that the slopes of the
relationships between trout abundance and cover abundance
would decrease in the first years and then gradually

increase in later years.

To determine the effect of hinge-cutting on channel form,

flow, and substrate composition. It was hypothesized that
the hinge-cutting would create a narrower, deeper channel,
that water velocity would increase, and that fine stream—

bed sediments would be eroded away, leaving more gravel

and rubble exposed at the bed surface.

To evaluate riparian hinge-cutting as a practical habitat

management technique.

STUDY AREA

The Salmon Trout River originates in the southeastern portion
of Huron Mountains in Marquette County, Michigan (Figure 1). The
stream flows northeastward about 20 km, with a gradient of 1.181nlqml
until it enters Lake Superior (Hendrickson et a1. 1973). The total
drainage area is 9,790 hectares. The headwaters and central portion
of the Salmon Trout flows primarily through northern hardwood forest,
while the lower stream is located in a mixed coniferous-hardwood
swamp (Enk 1977).

The river is divided by three major waterfall areas which restrict
the movement of fish, and by two man-made dams. Lower Dam is within
the study area, located about 1 km upstream from Sheet Rock Falls,
and was built for flood control and to provide a sediment trap and
fishing pond. The stop-logs were removed from this dam in the fall
of 1978 to eliminate the impoundment of water in preparation for
this study.

Most of the river is located within the boundaries of the Huron
Mountain Club, and is isolated from public access, however, local
residents and other non-club members are known to fish the Salmon

Trout, after trespassing via such routes as hiking trails and logging

roads.

 

I. SU'IIIOI

     

35040” 1.00!

 

 

 

 

 

Figure l. The Salmon

Trout River.

uvu
/”’\
I ’
MICHIGAN
. 0' .
STUDY
0 5 scfiLE 2 AREA
. was (0
‘tv
0 .5 1 2 «noun-s
=f ' ’
' um: an:
may
6
V
n
«.9 Ch
39"

 

L. SUPERIOR

 

Specific Location of Study Area

 

The study area extended from about 100 m above Sheet Rock
Falls (part of the Lower Falls complex) to the base of Middle Falls,
a total of 2.872 stream km (Figure 2). This portion of the river
lies within sections 13 and 14 of Township 51 North, Range 28 West.
The closest town is Big Bay (population: ca. 250), located about

9 km to the east.

Water Quality and Discharge

 

All river basins in the Upper Peninsula were glaciated, however,
the glacial deposits were thin or absent in the area of the Salmon
Trout. This has contributed to the highly variable streamflow, wide
temperature fluctuations, and great floodflows found on the Salmon
Trout. Mean discharge reported by Hendrickson et al. (1973) was
53 cfs in section 12, T 51 N, R 28 W. The ratio of 10 percent to
90 percent duration discharge was 1.86.

Hendrickson (1973) also reported that the hardness of the stream

was 62 mg/l CaCO and the pH about 7.6. The softness of the water

3
is due to outcroppings of crystalline bedrock in the area, which allow

little mineralization of runoff entering the river.

Bed Materials

 

The predominant bed materials vary within the study area. Below
Lower Dam, gravel and rubble is abundant, intermixed with sand.
Above the dam, sand and silt are predominant up to about station 50.

These stations are in a meadow—like area and were formerly impounded

Figure 2. Salmon Trout River study area. Treatment sections are
shaded. Control sections are unshaded.

 

 

NHddhu
an:

0'
ID i» '“
in in

a “‘\‘\‘\\m‘\\\\\\‘_

(D . .

a
In

Figure 2.

 
  

Shoot
Rock
Falls

 

 

 

10

by Lower Dam. From station 50 to Middle Falls, the amount of gravel

increases and occasional patches of clay are present.

Biota

Aquatic macrophytes are scarce within the study area, presumably
due to frequent flooding and sand substrates. Occasional patches

of Potamageton Sp. were present on silt flats below Lower Dam.

 

Previous work on benthic invertebrates in the Salmon Trout
(Smith 1941) found a standing crop of 7.5 cc/m2 for rubble stream
bed and an average of about 6 cc/m2 for all stream bed types in the
unimpounded sections of the current study area. This was low compared
to other streams cited by Smith and was also attributed to severe
flooding, shifting sand bottom, and poor fertility of the water.

In the spring of 1981, a study was undertaken in conjunction with
this research to investigate the effects of hinge-cutting on macro-
invertebrate abundance and biomass (Dr. R. ierritt, Dept. of
Entomology, MSU). Bottom samples in untreated sections revealed a
biomass of 12.3 mg dry wt/m2 in May and 45.6 mg dry wt/m2 in July.
Differences between untreated and experimental sections following
hinge—cutting are discussed in the results.

Within the study area, brook trout (Salvelinus fontinalis) are
the most abundant fish species in both numbers and biomass as de-
termined by electrofishing. Other species present include slimy

sculpins (Cottus cognatus), creek chubs (Semotilus atromaculatus),

 

and dace (Rhinicthys sp.). The Huron Mountain Club stocks legal

 

size brook trout ( > 200 mm) annually within the study area.

Occasionally, rainbow trout (Salmo gairdneri) have been stocked.

 

11

All stocked brook trout are given fin-clips to differentiate them
from wild trout for population estimates (see Table A3 for specific
fin-clips). Fishing pressure could be considered light, due to the
limited access, and catch and release fly-fishing has been required
for club members between Lower Dam and Middle Falls (about 60 percent

of the study area) since 1975.

METHODS

Twenty-seven stations, each about 100 m long, were selected

for use as experimental and control sections to test the brook trout
population response and changes in channel form and flow due to hinge-
cutting of riparian brush. The study area corresponds to stations
34-60 as designated by Enk (1977). Five experimental sections, each
composed of two stations, were chosen along the study area to provide
maximum spacing between treatments (Figure 2). The remaining sections
were designated control sections and were either two or three stations

in length.

P0pulation Estimates

 

Mark-and-recapture electrofishing was done in spring and fall
to estimate brook trout abundance beginning in the fall of 1979 and
continuing through the spring of 1981. Two shocking runs were made
for each estimate.

The electrofishing unit consisted of a wood and styrofoam boat
which carried a gasoline-powered lOO—600-volt DC "generator," formed
by an AC alternator with rectification to DC. Three handheld positive
electrodes were connected to the generator through spring-loaded
retracting reels mounted on the front of the boat. Each electrode
was a fiberglass handle with a head of stainless steel rod bent into

a diamond-shaped 100p about 30 cm long. The cathode was a sheet of

12

l3

galvanized steel covering the bottom of the boat. A live tub for holding
captured fish was carried on the boat, and nets within the tub separated
fish from different stations.

The electrofishing procedure was composed of one crew member pulling
the boat upstream and shocking the mid-channel, while two other men
covered the area along each stream bank. Often, all three men would
converge on large pieces of cover such as log jams, undercut banks, and
brush piles. Fish drawn to the electrodes were netted and transferred
to the live tubs. After a number of stations were shocked, the live
tub was dropped off, and another crew would process the catch.

During processing, the fish were.anesthetized with tricaine methane
sulfonate (MS—222), measured for length, weighed, and examined for
prior fin—clips. On the first shocking run, all fish captured were
given a temporary clip along the bottom of the caudal fin. After pro-
cessing, the fish were placed in fresh water until revived. then carried
back to the downstream end of the station in which they were captured,
and released. This enabled the fish to redistribute normally for the
recapture run. At least two days were allowed to elapse between
shocking runs to enable the fish to recover and redistribute. On the
second shocking run, fish were examined for first run marking clips, and
recorded as either marked or unmarked. Unmarked fish were measured
for length, and weighed, while recaptures were only measured for length.
Each shocking run required at least a day and one-half of electro-
fishing. When processing the fish of the recapture run, the upper
tip of the caudal fin was clipped to prevent double-countingcfi'fish that

might swim upstream past the shocking team's position overnight.

14

Shocking dates on the Salmon Trout River are given in Table A2
in the appendices. During the fall shocking periods, young of the year
trout captured were given specific permanent fin-clips to facilitate

future recognition of the year-classes for growth studies (Table A3).

Calculation of Population Estimates

 

Trout population estimates were calculated using a modified

Petersen method given in Seber (1973):

)

(m1) (r+u+1)
(r+1)

 

Z
I
p-a

where,

2)
II

estimated population
m = number of marked fish
r = number of recaptures

u = number of unmarked fish in second run

When r/r + u > 0.1 the N is asymptotically normally distributed, and

95 percent confidence intervals are given by:

1.96 H V

2»
+

where,
_ (m+1)(r+u+l)(m-r)(u)
(4+1)2 n+2)

 

When r/r + u > 0.1 and r/m < 0.1, the Poisson approximation is recom-
mended using r as the entering variable into the tables (Chapman 1948).

The upper and lower confidence limits are given by:

)

LC, UC = (m)(r+u)(x1,2

15

where,
x1 = lower limit given by table
x2 = upper limit given by table
LC, UC = lower and upper confidence limits

Estimates were made for different length groups of trout, due to
the size selectivity of electrofishing gear (C00per and Lagler 1956).
Efficiency of capture tends to increase as fish size increases.
Separate estimates were made for fish of each 25 mm length interval
beginning with_: 100 mm, continuing up to > 200 mm. Due to great vari—
ability of recapture rates for the individual IOO-m stations, pro-
bably a result of movement of marked fish across station boundaries,
population estimates for the entire study were calculated for each
size group. These estimates were then prorated back to individual
stations by the proportion of marked and unmarked fish ("new fish")
captured in that station relative to the total number of marked and
unmarked fish in the entire study area. This method of combining the
data and then prorating total estimates is more accurate than indivi-
dual estimates because it allows the use of larger units in the esti-
mations, particularly the number of recaptures, upon which the method
is based (Cooper 1952).

Brook trout biomass estimates were calculated by multiplying the
average weight of the fish in each station by the estimated population
in each size class. Numbers and biomass estimates for each station
were converted to stock density (no/km) and standing crOp (gm/m)

according to actual station lengths.

l6

Habitat Studies

 

Eighteen discharge measurements were made throughout the summer
of 1980 by the transect method, measuring current velocity and water
depth every 0.3 m across the stream channel. A staff gauge was in-
stalled on the upstream face of Lower Dam to allow quick reference to
water levels before habitat measurements. These staff gauge readings
were converted to discharge estimates by regression analysis using
actual discharge measurements. Habitat measurements were made at base-
flow (~O.85 cms) to decrease error caused by water level fluctuations,
and because correlations between trout abundance and cover would be
most meaningful when cover is at a minimum (Cooper and Wesche 1976).

Pre-alteration measurements of channel width, depth, and current
velocity at the thalweg were made at baseflows during June and July,
1980, and post-alteration measurements were made in June, 1981. These
measurements were made on transects Spaced 10 m apart, beginning at
the downstream end of each station. Current velocity was determined
using a Swoffer Model 2000 current meter. Substrate type on each
transect was determined by visual estimate and recorded as percent
composition. Approximate size classes of the substrate types are
as follows: Rock, > 8 cm; Gravel, 0.5 - 8 cm; Sand, < 0.5 cm; Silt,
any fine organic matter.

Measurements of overhead bank cover (submerged or at the water
'surface) in each station were made by determining the length of cover
that was at least 9 cm wide and had at least 15 cm of water beneath it.
These criteria for usable bank cover were adapted from Wesche (1976)

as was done by Enk (1977) in previous work on the Salmon Trout River.

17

It has subsequently been clarified that Wesche's criteria for cover
were somewhat different: at least 9 cm wide and 13 water at least

15 cm deep (R. J. White, pers. comm. 1984). A special gauge was con-
structed to facilitate determination of the length of cover (Enk 1977).
The gauge was inserted in the water beside potential bank cover and

the length of cover Which it fit along the stream bank was recorded.
Individual sections of cover less than 15 cm long were not recorded,
and no attempt was made to measure the width of cover greater than 9 cm
wide. Length of overhead cover is more important than area, as trout
tend to position themselves near the edges of coverts (Gibson and

Keenleyside 1966).

Drift Sampling

 

Drift samples were taken in a number of stations during August,
1980 to determine if there were differences in macroinvertebrate
abundance between control and treatment sections. Drift nets were set
for varying amounts of time, ranging from 4 to 24 hours, over
either sand or gravel substrate. The nets were set in pairs, one at
the lower (downstream) end of a control section, and one at the lower
end of the adjoining downstream treatment section. Each site was sampled
for a total of 24 hours. Current velocity was measured to estimate
the amount of water filtered through the nets. Samples were pre-
served in 70 percent ethanol, and later identified to order, blotted

dry, and weighed to determine biomass of each group represented.

Habitat Alteration

 

After initial measurements of channel form and flow characteristics

were completed, hinge-cutting of riparian brush was begun in the

18

treatment sections. The hinge—cutting involved sawing partially through
the stems of the brush, and folding it into the water at a downstream
angle. The brush remains in place, held by the strip of wood and bark
left uncut. Alders (Alggs sp.) were the most prominent streamside
vegetation, and grew in thick clumps along most of the bank from
stations 41-58. The hinge-cutting was done using small bow saws, and

most of the brush which could reach the stream when felled was cut.

Statistical Analysis

 

Using the estimates for brook trout stock density and standing
crop, and the measurements of overhead bank cover, the relationships
between these variables were analyzed using simple regression techniques.
Correlation coefficients were tested for significance using a t-test
given by Gill (1978). The response of the brook trout population to
the increased cover provided by the hinge-cutting was analyzed using
paired t-tests, comparing stock density and standing crop of the same
sections between years, and comparing stock density and standing crop
of paired treatment and control sections in the same season and year.

Channel form and flow characteristics are presented as percentage
change from before and after habitat alteration. Regression analyses
and paired t-tests were performed using the SPSS Version 8.0 statistical

package available through the MSU Computer Laboratory.

RESULTS AND DISCUSSION

Trout Population

 

Over the four electrofishing periods, fall 1979-spring 1981, the
population density of brook trout declined slightly between seasons,
however, biomass levels staymdfairLyconstant until spring 1981 when a
sharp decline occurred (Tables 1-3, Figure 3). Natural fluctuations III
wild trout populations are common, often caused by weak year-classes due
to unfavorable climatic conditions (severe floods, drought, ice; White
1975). Recruitment of age—0 trout into the stock over the summer accounts
for the larger total numbers present between each spring and the next
fall. Trout of 126-175 mm comprised the greatest portion of the total
biomass in all seasons (Tables A5-A12).

The sections directly above Lower Dam (T2, C3) had lower stock den-
sity and standing crop of brook trout both before and after hinge-cutting,
than those sections below the dam and those farther upstream (Tables 1-3).
These sections were fairly shallow .with relatively large amounts of
sand ‘bottom and relatively little cover. In contrast, those stations
below the dam were also shallow, but had more cover available in the form
of log jams, undercut banks, and instream rubble. The coarser substrate
would also be more favorable for aquatic invertebrate production than

the shifting sand flats found above the dam (Hynes 1970). Farther upstream

l9

20

Table 1. Stock density (no/km) and standing crOp (g/m) estimates for brook trout (sll sizes)
in control snd treatment sections of the Salmon Trout River over four sampling
periods.

 

2 ehsnge I change
in numbers in bionass

Section Fall 1979 8 tin 1980 Fall 1980 Sprin‘ 1981 spring 1980 spring 1980
number inokaiiEZn: inoanEZiZmz (nolknlggln) (no/knzgglm) to spring 1981 to spring 1981

 

 

 

 

 

Cl 1044 12.3 446 15.5 1047 21.9 363 7.9 -19 .49
C2 842 10.5 540 17.0 1835 26.1 685 12.8 +27 -25
C3 610 14.5 608 20.5 442 14.0 199 4.3 -67 -79
C4 1030 17.4 570 12.4 436 9.0 217 6.0 -62 -52
C5 1783 23.8 690 17.5 622 12.2 382 7.1 -45 -59
C6 1578 25.9 745 13.6 1302 18.1 650 10.8 -13 -21

Control

means 1109 16.8 589 16.3 911 16.7 395 7.9 -33 -52
T1 1390 18.3 325 8.1 1535 22.2 480 8.7 +48 +7
T2 498 7.2 224 2.7 565 12.5 206 4.1 -8 +52
73 816 15.6 914 24.1 613 18.2 558 15.6 -39 -35
T4 1043 16.9 506 11.7 528 15.0 661 14.5 +31 +24
TS 1658 24.0 650 16.6 1061 21.7 687 11.9 +6 -28

Treatment

means 1066 16.2 514 12.3 830 17.7 516 10.9 +0 -11

Grand

mesns 1092 16.6 560 14.7 879 17.1 443 9.1 -21 ~38

 

21

Table 2. Stock density (no/km) and standing crop (g/m) estimates for brook trout less than
150 mm long in control and treatment sections of the Salmon Trout River over four
sampling periods.

 

2 change I change
in numbers in biomass
Section Fall 1979 Spring 1980 Fall 1980 Spring 1981 spring 1980 spring 1980

 

 

number (no/km)(g/m) (no/km)(g[m) (no/kngg/m) (no/km)(g1m) to spring 1981 to spring 1981

 

 

 

Cl 895 5.3 321 6.4 814 9.0 305 4.1 -5 -36
C2 730 4.6 409 7.3 1592 14.2 615 7.3 +50 0
C3 468 5.2 385 5.6 332 4.6 155 1.8 -60 -68
C4 857 8.3 463 6.6 358 4.3 156 2.7 -66 -59
C5 1566 14.8 525 8.7 500 6.7 312 3.9 -41 -55
C6 1356 14.3 628 8.3 1189 12.8 553 6.9 -12 -17

Control

means 943 8.3 443 7.0 759 8.2 331 4.2 -25 -40
T1 1203 8.4 273 4.5 1330 11.6 449 6.7 +64 +49
TZ 420 3.5 216 2.9 465 5.2 182 2.7 -16 -7
T3 641 6.7 710 10.2 437 7.0 404 6.8 -43 -33
T4 913 9.7 423 6.5 347 7.0 531 8.5 +26 +31
15 1424 12.6 488 7.4 905 11.6 604 8.4 +24 +14

Treatment

means 908 8.1 416 6.2 668 8.2 433 6.6 +4 +6

Grand

means 929 8.2 432 6.6 724 8.2 371 5.1 -14 -23

 

22

 

 

 

 

 

 

 

 

Table 3. Stock density (no/km) and standing crop (g/m) estimates for brook trout greater
than 150 mm long in control and treatment sections of the Salmon Trout River
over four sampling periods.
1 change I change
in numbers in biomass

Section Fall 1979 Spring 1980 Fall 1980 Sprin 1981 spring 1980 spring 1980
number ,(no/km)(g1m) (no/kaSglm) (no/kmzsglm) (no/kmiZEZmE to spring 1981 to sprigs 1981

C1 149 7.0 124 9.2 232 12.9 58 3.8 ~53 ~59

C2 112 5.9 131 9.7 243 12.0 70 5.4 ~47 ~44

C3 142 9.2 223 14.9 110 9.3 44 2.5 ~80 ~83

C4 173 9.1 107 5.8 78 4.6 61 3.3 ~43 ~43

C5 216 8.9 165 8.8 122 5.5 69 3.2 ~58 ~64

C6 222 11.6 117 5.3 113 5.3 98 3.9 ~16 ~26
Control
means 165 8.4 147 9.1 151 8.4 64 3.7 ~56 ~59

T1 186 9.9 52 3.6 205 10.6 31 2.0 ~40 ~44

T2 78 3.7 8 0.4 100 7.2 23 1.4 +188 +250

T3 174 8.9 204 13.9 177 11.3 153 8.8 ~25 ~37

T4 130 7.2 83 5.2 182 9.8 130 6.0 +57 +15

T5 234 11.4 162 9.2 156 10.1 83 3.5 ~49 ~62
Treatment
means 157 8.1 99 6.3 161 9.7 84 4.3 ~15 ~32
Grand
means 162 8.3 128 8.0 155 8.9 72 3.9 ~44 ~51

 

Figure 3.

23

 

 

 

 

 

 

1200‘
1100- "09 1092
1000-
911
909‘ 979
53001 33°
\
g7001
)-
1.16004 589 560
g 514 516
m5“)!
:1 «a
3,400. 395
8
P300‘
(I)
2001
100-
OJ
C T G C T G C - T G C T G
FALL 1919 seems 1990 FALL 1980 SPRING1981
20
17.7
a 17.1
16216.6 16.3 16.7
15 < ‘ 147
E
\E i
u, ‘ 12.3
3 ! 10.9
51° " 91
0 g 1‘ 19
E
a
E
1;. 5 1 ‘
1
0 1

C T G C T G C T G C T 6
FALL 1979 591111161990 FALL1980 spnmc1991

Mean stock density and standing crop estimates for control
(C) and treatment (T) sections and grand averages (G) for
entire study area over four sampling periods.

24

(sections C4-C6, T3-T5) depth and substrate size increases, and more
cover is available (Tables 4 and 5).

Fall standing crops in the study area are considerably lower than
in other Michigan streams that have been studied (Table 6). The other
streams are in the northern lower peninsula where flow regimes are more
stable than in the upper peninsula, due to sandy soils with high in-
filtration rates and groundwater recharge, and where the concentration
of plant nutrients in the water and in riparian soils is probably greater.
Hunt Creek, the only stream listed besides the Salmon Trout River con-
taining exclusively brook trout, had over twice the average fall
standing crop of that estimated for the Salmon Trout River. The re-
mainder of the streams had mixed-species populations of trout, and their
trout standing crops ranged from about one-and-a—half to over four times
that in the Salmon Trout River. More variable flow, lower hardness,
greater sparcity of overhead cover, and more severe winter conditions

in the Salmon Trout may account for the relatively small standing crop.

Trout Growth

 

Growth in length and weight of marked year-class fish (Table 7)
in the Salmon Trout River were approximately equal to those found for

brook trout in lower peninsula streams (Gowing and Alexander 1980).

Habitat

While there was a general decrease in channel width throughout the
study area, it was twice as great in the treatment sections as in the

controls (Tables 4 and 5). The greater stream width decrease in the

25

Table 4. Physical characteristics of 10 treatment stations* before (1980) and after (1981)
hinge-cutting of riparian brush in the Salmon Trout River, Marquette County,

 

 

 

 

 

 

Michigan.
Avg. Width (m) Avg; Depth (m) Avg.Velocity(cm/s) Bank Cover (m)
Station 1 Z Z 2
number pre post change Eye post change pre post chggge pre post change
37 10.6 10.0 ~5 .588 .523 ~11 26.2 28.7. +9 13.1 25.9 +97
38 10.4 9.40 ~10 .575 .545 ~5 33.2 42.1 +27 14.2 38.1 +168
42 8.50 7.70 ~9 .421 .492 +17 27.7 36.9 +33 2.3 21.5 +834
43 6.98 6.31 ~10 .600 .701 +17 24.7 28.7 +16 13.4 80.6 +501
47 6.24 5.91 ~5 .671 .762 +14 30.2 37.2 +23 31.9 75.7 +137
48 6.48 5.58 ~15 .616 .747 +21 38.1 41.5 +9 18.0 45.9 +155
52 6.93 6.28 ~9 .475 .561 +18 33.5 42.1 +26 14.9 56.1 +277
53 7.16 6.97 ~3 .671 .665 ~1 28.0 31.1 +10 48.8 38.9 ~20
57 6.86 6.78 ~1 .652 .625 ~4 29.0 35.1 +21 30.0 34.3 +14
58 7.90 6.99 ~12 .552 .469 ~15 33.8 43.9 +29 27.4 54.6 +99
Average 7.80 7.20 ~8 .582 .609 +5 30.5 36.7 +20 21.4 47.2 +121

 

‘each station approximately 100 m long. Exact station lengths given in Table 8.

26

 

 

 

 

 

 

 

Table 5. Physical characteristics of 15 control stations* before (1980) and after (1981)
hinge-cutting of riparian brush in the Salmon Trout River, Marquette County,
Michigan.
Avg. Width (m) Avg. Dgpth (m) Avg.velocity unis) Bank Cover (m)
Station 1 z z z
number pre post change pre post change pre post chgpge pre gpost change
34 13.0 12.6 ~3 .747 .682 ~9 20.1 20.7 +3 0.4 2. +550
35 12.3 11.2 ~9 .533 .524 -1 36.6 41.1 +12 11.8 5. ~55
36 12.4 12.2 ~2 .491 .472 -4 27.7 26.5 -4 26.0 21. ~18
39 13.6 13.6 0 .387 .310 ~20 25.9 40.5 +56 16.0 12. ~22
40 14.1 14.1 +0 .658 .541 ~18 43.6 34.1 ~22 15.7 18. +15
41 11.0 10.4 ~5 .558 .565 +1 21.0 25.0 +19 1.0 3. +200
44 6.61 6.26 ~5 .756 .742 ~2 25.6 30.5 +19 24.4 15. ~38
45 6.14 5.91 -4 .786 .797 +1 28.0 30.2 +8 28.4 24. ~14
46 6.10 6.01 -1 .768 .736 ~4 23.2 25.6 +10 23.3 17. ~26
49 6.75 6.26 ~7 .771 .629 ~18 18.9 25.6 +35 18.6 23. +28
50 6.50 6.16 -5 .613 .570 ~7 25.3 26.8 +6 21.2 16. ~23
51 6.75 6.42 ~5 .594 .599 +1 24.1 33.8 +40 18.3 11. ~36
54 7.50 7.34 ~2 .536 .457 ~15 20.1 29.6 +47 12.0 3. ~75
55 7.54 6.82 ~10 .664 .594 ~11 31.1 34.7 +12 22.9 18. ~17
56 8.53 8.71 +2 .881 .665 ~25 17.7 25.3 +43 26.7 13. ~51
Average 9.25 8.93 ~3 .650 .592 ~9 25.9 30.0 +16 17.8 13. ~22

 

*each station approximately 100 m long.

Exact station lengths given in Table 8.

27

Table 6. Estimates of fall trout standing crop (kg/km) in the Salmon
Trout River compared to average fall standing crop in some
northern lower peninsula streams.

 

Fall standing crop (kg/km) Reference

 

 

Stream Year Brook Brown Rainbow Total source*

Salmon Trout 1979 165 ~ ~ 165 1
1980 166 - - 166 1

Mainstream AuSable 1974-77 189 3321 153 3663 2

South Branch
AuSable 1974-77 191 1265 - 1456 2

North Branch

 

AuSable 1961-67 736 1616 ~ 2352 2
Pigeon 1961—64 285 126 ~ 411 2
Hunt Creek 1959-64 339 ~ ~ 339 2
Williamsburg Creek 1975-76 39 974 5 1018 2
a

1 = This study
2 = Gowing and Alexander (1980)

28

Table 7. Average lengths (L), weights (W), and standard deviations of
marked year-class brook trout in the Salmon Trout River,
Marquette County, Michigan. Initial measurements taken on
young-of—the-year when given the permanent finclip.

 

 

 

 

 

 

Fall 1979 Springl980 Fall 1980 Spring 1981

L W L W L W L W

FincliE (mIn) (g) (turn) (a) (turn) (3) (MI) (8)
ALV 86 5 112 18 152 35 154 36
:3 :1 :8 :4 :13 :12 :11 :9

A - ~ - - 87 6 116 16

+4 +1 :18 :10

 

29

treatment sections was due to deposition of silts among the hinge-cuts
along the banks. This deposition of silt and sand was evident over the
spring floods as some hinge-cuts were becoming buried along their outer
margins.

Water depth increased 4.6 percent in the treatments, while decreasing
8.8 percent in the controls (Tables 4 and 5). All stations below Lower
Dam decreased in water depth after hinge-cutting. Movement of sand from
above the dam into sections below was obvious over the winter after
cutting, and much of the sand was caught in the spaces between rocks and
led to the decrease in depth. The five treatment stations (42, 43, 47,

48 and 52) closest above Lower Dam were the ones that had increases in
depth (14-21 percent), as the stream carved a new channel through the
previous pond bed.

Water velocity at the thalweg increased in both treatments and controls,
20 and 15 percent, respectively. The overall narrowing of the stream
channel could account for some of the increase in velocity.

Hinge-cutting immediately increased the amount of cover in treatment
stations by an average of 400 percent (160-2400 percent), but by a year
'later, much of the increase had disappeared, with only an average of 120
percent remaining by July, 1981 (Table 8). During the same year, bank
cover decreased 18 percent in the controls. The decrease in the controls
suggests that 1980-81 was a time of generally decreasing cover and demon-
strates the temporary nature of overhead cover in streams which experience
frequent high water. For example, a large area of sunken logs in
station 40 had supplied a great length of overhead cover in previous

studies and in the beginning of this study. These logs were moved to

3O

 

 

 

 

Table 8. Length of stream bank meeting overhead cover criteria in the Salmon
Trout River, Marquette County, Michigan.
2 change 1 change
Station Station Length of overhead cover (m) June-Aug. June 1980-
number length (m) June 1980 August 1980 June 1981 1980 June 1981
34 105 0.4 - 2.6 ~~ -
35 110 11.8 ~~ 5.3 - ~-
36 95 26.0 ~~ 21.3 ~~ -
37 100 13.1 91.9 25.9 +602 +98
38 92 14.2 105.5 38.1 +643 +468
39 99 16.0 ~~ 12.5 ~~ -
40 80 15.7 - 18.1 - -
41 116 1.0 ~- 3.0 - -
42 119 2.3 58.4 21.5 +2439 +835
43 130 13.4 124.1 80.6 +826 +501
44 105 24.4 ~- 15.2 - ~-
45 110 28.4 ~~ 24.5 - -
46 119 23.3 - 17.2 ~~ -
47 103 31.9 125 75.7 +292 +137
as 98 18.0 114.6 45.9 +53i +155
49 108 18.6 ~~ 23.8 ~~ ~-
50 100 21.2 - 16.3 - ~-
51 108 18.3 - 11.7 - ~-
52 102 14.9 128.6 56.1 +763 +276
53 159 48.8 125.4 38.9 +157 ~20
54 81 12.0 - 3.0 - ' ~-
55 116 22.9 ~~ 18.9 ~~ ~-
56 90 26.7 ~~ 13.0 ~~ ~-
57 130 30.0 93.9 34.3 +213 +14
58 98 27.0 91.9 54.6 +240 +102
59 95 38.5 ~~ 38.5 ~~ -
60 100 29.9 ~~ 29,9 - ~-
Totals for treatments 213.6 1059.3 471.6 +396 +121

Totals for controls 335.1 ~- 274.8 -~ ~18

31

higher ground by spring 1981 flood waters, reducing the amount of avail-
able cover, however total overhead cover within the station increased as

new lengths were created.

Streambed Material

 

Following hinge-cutting, treated sections generally underwent sig-
nificant decreases in amount of streambed sand and concurrent increases
in amount of gravel, rock and silt (Table 9). The most important changes
in streambed materials after hinge-cutting were in the amount of rock
and sand below and above the dam (Table 9). Rock decreased slightly below
the dam (C1, C2, T1) while increasing greatly (68-470 percent) in all
sections above the dam. Changes in amount of sand were opposite of the
changes in rock, with all sections below the dam increasing in sand.

The combined effect of opening the dam in fall 1978, with the last

stop log removed in spring 1980, and of hinge-cutting was to move sand
from above the dam into (and through) the sections below the dam.

The density of hinge-cuts was lower below the dam than above, due to
higher banks and less streamside brush, so the scouring effect may

not have been as great there. Rates of movement of sand into and through
the downstream sections was probably still changing at the end of the
study, and I expect the amount of sand streambed to decrease in later
years.

Silt increased in sections T2~T4 where hinge-cuts were fairly
dense. Silt deposited due to slowing of current under felled brush at
the stream margins. No trends in silt deposition were detected in the

control sections.

32

 

 

 

 

 

 

 

 

 

 

Table 9. Substrate composition (2) in control and treatment sections before and after hinge-
cutting of riparian brush along the Salmon Trout River. Marquette County,
Michigan.
Rock Gravel Sand Silt
Section 2 1
number before after change before after chgpge before after chggge before after change
Control Sections
Cl 59.5 53.8 ~9.6 12.8 8.2 ~36 23.7 32.6 +38 4.1 5.5 +34
C2 53.5 33.9 ~37 2.7 2.9 +7.4 34.9 53.4 +53 8.9 9.8 +10
Doun- 56.7 44.1 ~22 7.9 5.6 ~29 29.1 42.7 +47 6.4 7.6 +19
stream
averages
C3 2.1 6.6 +214 6.4 9.3 +45 81.0 76.5 ~5.6 6.0 4.8 ~20
C4 1.2 6.8 +467 7.6 6.1 ~20 83.6 82.0 ~1.9 7.0 4.4 ~37
CS 1.9 3.2 +68 15.5 18.0 +16 73.1 66.3 ~9.3 7.2 8.0 +11
Upstream 1.7 5.6 +222 9.7 11.0 +14 79.4 75.2 ~5.3 6.7 5.7 ~16
averages
Treatmggt Sections
T1 56.0 41.1 ~27 10.5 7.3 ~30 16.8 42.3 +152 15.3 9.5 ~37
T2 12.3 27.5 +124 2.7 10.1 +274 83.7 48.7 ~42 1.4 13.8 +886
T3 0.0 1 5 +00 4.4 9.0 +105 95.3 76.0 ~20 0.0 12.5 +00
T4 1.1 5.3 +382 16.8 21.7 +29 80.1 62.8 ~22 2.0 9.7 +385
T5 5.6 13.1 +134 52.8 53.9 +2.1 32.6 24.7 ~24 9.1 8.4 ~7.7
Upstream 4.9 12.3 +149 19.1 23.7 +24 72.8 52.7 ~28 3.1 11.1 +253

averages

 

33

I conclude that narrowing of the channel due to hinge-cutting
led to scouring of sand, uncovering of rocks and gravel, and deposition
of silts along the stream margin. The increases in substrate size
in treatment sections can lead to a more diverse invertebrate fauna
(Hynes 1970), and increase the amount of suitable spawning areas for
trout. Hunt (1971) also found that constricting the streamflow led
to considerable scouring away of fine sediments in Lawrence Creek,

Wisconsin, a stream with predominantly sand—silt bottom.

Trout Population - Bank Cover Relationships

 

Correlations between trout abundance and cover density were sig-
nificant for most population parameters prior to hinge-cutting, the
exceptions being those stations below Lower Dam (Tables 10 and 11).
After hinge-cutting in the summer of 1980, correlations declined for
all stations combined and for those stations above Lower Dam. This
was the anticipated immediate result as cover ratings increased abruptly,
while the trout population had insufficient time to adjust to the
changes.

Enk (1977) had also found that stations below Lower Dam had poor
correlations between trout abundance and cover. One possible expla—
nation for the weak correlations is the method used to quantify usable
cover. This study was concerned with bank cover that had 15 cm or more
of water beneath it, which may have been relatively rare and applicable
only to larger fish, and overlooked the rather small-sized but perhaps
numerous bits of cover provided by instream rubble. The stream sections

below Lower Dam are wide and predominantly riffles, so any estimate

Table 10. Correlation coefficients (r) and coefficients of determination (r2) for
trout standing crop (Y) and cover density (x) in the Salmon Trout River.

 

 

 

 

 

Standing

crop Fall 1979 Spring 1980 Fall 1980 Spring 1981
£31m) r rIT' r r5 r rz_ r r
All sizes- .718** .515 .453* .205 .210 .044 .465** .216

all stations

All sizes- .756** .571 .577** .333 .413 .171 .744** .554
above dam

All sizes- .117 .014 .035 .001 .725 .526 .094 .009
below dam

5150mm? .683** .466 .632** .399 .319 .102 .433* .188

all stations

5150- .704** .495 .692** .478 .544* .296 .642** .413
above dam

ilSOmm— .187 .035 .545 .297 .727 .528 .362 .131
below dam

>150mm- .610** .372 .259 .067 .057 .003 .406* .164

all stations

>150m- .622** .438 .401 .161 .167 .028 .678** .460
above dam
>150mm- .043 .002 .245 .060 .570 .325 .242 .059
below dam

 

*indicates significance at 52 level
**indicates significance at 1z level

35

 

 

 

 

 

Table 11. Correlation coefficients (r) and coefficients of determination (r2) for
trout population density (Y) and cover density (x) in the Salmon Trout
River.

S‘°°k Fall 1979 Springgl980 Fall 1980 Spring 1981
density 2 2 ———-2- 2
(no/km) r r r r r r r r

All sizes- .576** .332 .668** .447 .144 .021 .391* .153
all stations

All sizes— .611** .373 .723** .522 .418 .175 .637** .405
above dam

All sizes- .558 .311 .272 .074 .662 .438 .232 .054
below dam

‘5150mm- .S46** .298 .664** .441 .147 .022 .336 .113
all stations

ilSOmnr- .580** .336 .687** .472 .398 .158 .573** .329
above dam

liISOmm- .567 .321 .428 .183 .617 .381 .284 .080
below dam

>150mm- .581** .337 .412* .170 .089 .008 .541** .292
all stations

>150mmr .659** .434 .516* .267 .244 .059 .676** .457
above dam

>150mm— .339 .115 .204 .042 .758* .574 .274 .075
below dam

 

* indicates

significance at the 52 level
**indicates significance at the 12 level

36

of the amount of usable cover was probably underestimated by measuring
only large-sized bank cover. Above Lower Dam up to station 58, the
stream is narrower with primarily sand bottom. Little cover is provided
by instream rubble, leading to better estimates of usable cover by the
methods employed.

In the fall of 1980 correlations for stations below Lower Dam were
stronger as cover ratings had increased quickly (Tables 10 and 11),
possibly compensating for the previous underestimation of usable cover.
However, in the spring of 1981 the correlations had dropped to approx-
imately ttua same levels as the spring of 1980. Sand scoured out of
stations above Lower Dam filled in spaces between rocks which may
have provided some cover. The loss of this cover may have contributed
to the population decline in that part of the study area in the spring
of 1981.

The slopes of the trout abundance - cover density regression lines,
which would be the predicted increases in trout numbers or biomass with
the addition of one meter of overhead cover within the observed cover-
density range, decreased steadily from a high of 3.3 fish and 65.2 gm
in the fall of 1979, to 0.6 fish and 14.4 gm in the spring of 1981
(Figures 4-7). An abrupt drop following hinge-cutting would be expected
as the study area would be "cover-saturated." As the hinge-cut areas
are thinned of excess cover, and the trout population expands to utilize

the available living space, the $10pes will begin to rise.

Comparison of Trout Populations between Treatments and Controls

 

Control sections showed no significant differences in trout popu-

lations between the fall of 1979 and the fall of 1980, but they had

37

 

 

 

 

 

Y '448.8+3.Jx
rz-Jsz
240dr
::
z: e
\
c:
- e
:5 T ‘.
0
:- 1800b
_. e
“3 0
0 e
5 e e ° .
o 1
e
25
e
O s
a > 6
:3 am . .
‘0
4h. .
cf : * : : :
0 200 400 800
OVERHERD BRNK COVER H/KH
y -ssas.s.ss.zs
(20.58.
420001-
I:
x:
\
t:
c:
g zsoom
a:
<3
c:
:z
c:
:z
(I
0— 140004
a:

 

 

 

 

or t t : : : e
0 200 400 800
OVERHERD BRNK COVER M/KH

Figure 4. Relationship between stock density and standing crop
with length of bank cover, September 1979.

38

 

 

 

 

 

 

 

 

 

 

y - 187.3HJ s
r3-.sa1
240m'
1:
r
\
O
z 1.
C 160041-
(D
2
LIJ
D
K
L)
O
.—
(D
0. ¢ ‘fi 3 ‘fi 7 7
0 200 400 000
UVERHERD BRNK COVER H/Kfl
y . ssau on."
r’-.zss
4200011-
t
K
\
I:
O db
0.
O
m
U
C3
2
O
2
G
.—
U)
m t t : : +7 :
0 200 400 800

OVERHERD BRNK COVER H/KH

Figure 5. Relationship between stock density and standing crop
with length of bank cover, June 1980.

39

 

 

 

 

 

 

 

Y . 10.20.04:
. r3 - .022
240°?
2
x: Is
\
O
z
.
1: 18001' . '
"‘ .
(D
2
“J
D
g
g; 1e
5; 1s
.
.
C. i i 7 1‘ L 3
0 200 ‘00 800
OVERHERO BRNK COVER H/KH
y - “season s
r’ - .oss

420001"
2 .
x
\
Z
O 7*
o. . .
:3 2300041 .
a
U
0
z
D
2
c
.—
CD

 

 

 

o foo' foo goo
OVERHEHO BHNK COVER H/Kfl

Figure 6. Relationship between stock density and standing crop
with length of bank cover, September 1980.

 

 

 

4O

 

    

 

 

 

 

   

 

 

 

Y 0288.00.81!
2.
2‘00? f .153
1:
I
\
D
Z 3'
:2 xsomr
E75 e
2
LIJ
o 1
Z
8
1— 800+b .
(D
T .e e
e
O C . . O
o - 2'00 ' (on 600
OVERHERO BRNK COVER H/KH
y .ssss.1.u.ax
fa'JiI
420004-
I:
K
\
1:
cs 1'
23 2900:» O
O:
U
c:
z
O
z e
c
.— uoom» e
03
e
1 e
e 3 e O
.L.’ ' ' '
o r 230 R 4'00 ' . 600

OVERHERD BRNK COVER H/KH

Figure 7. Relationship between stock density and standing crop
with length of bank cover, May 1981.

41

significantly fewer trout in the spring of 1981 than in the spring of
1980 for all parameters with the exception of stock density of fish
less than 150 mm long (Table 12). This was not the pattern for the
treatment sections, as no significant differences were present between
the Spring of 1980 and the spring of 1981. Therefore, the decline

in the trout population experienced between the fall of 1980 and the
spring of 1981 was primarily within the control sections, while popu-
lation levels within the treatment sections remained about the same.
This may have been due to improvements in "space-refuge factors" (in-
creased depth, more hiding cover, and increase food supply) within the
treatment sections (Hunt 1969).

In comparing trout abundance between paired control and treatment
sections, the only significant differences were found in the spring of
1980 when control sections had more fish greater than 150 mm in length
than did their paired treatments (Table 13). After hinge-cutting no

significant differences were detectable.

Invertebrate Drift

 

Total invertebrate drift rates (mg wet wt./l/day) were in all
paired comparisons, higher in the control section than in its immediate
downstream treatment section (Table 14). A possible explanation for
this could be that the newly uncovered hard streambed materials and the
dense lacework of hinge-cut brush in the water along stream edges of
treatment sections may have provided increased substrate for attachment
of invertebrates, and that there was less movement out of such sections

and a greater net rate of attachment or "settlement" of drifting

42

Table 12. Directions and significance levels of changes in trout abun-
dance from fall to fall and spring to spring within control
and treatment sections, based upon t-tests.

 

Fall l979-Fall 1980

 

Sprigg 1980 ~§pring 1981

 

 

 

 

Trout population Control Treatment Control Treatment
variable sections sections sections sections
Stock density (fish/km)
§_150 mm ns 1979>l980 ns ns
(P< .2)
> 150 mm ns ns 1980>l981 ns
(p< .02)
all sizes ns ns 1980>l981 ns
(p< .15)
Standinggcropr(g/m)
:_150 mm ns ns 1980>l981 ns
(p< .05)
> 150 mm ns 1980>l979 1980>l981 ns
(p< .15) (p< .05)
all sizes ns ns 1980>l981 ns

(p < .02)

 

43

Table 13. Differences in trout population statistics between paired
control and treatment sections, based on t-tests.

 

Trout population Fall Spring Fall Spring
variable 1979 1980 1980 1981

Stock density (fish/km)

 

§_150 mm ns ns ns us
> 150 mm ns controls> ns ns
treatments
(P< .1)
all sizes ns ns ns ns

Standing crop (g/m)

 

:_150 mm ns ns ns us
> 150 mm ns controls> ns ns
treatments
(P<=.2)

all sizes ns ns ns ns

 

441

 

 

 

 

 

 

Table 14. Invertebrate drift rates at the downstream ends of each treatment (T) section
compared to the immediate upstream control (C) section over two substrate thes.
Drift rate (mg wet wt/l/day)
Section Substrate Date Ephemeroptera Trichoptera Plccoptera Other Total
compared type (1980) C T C T C T C T C T
C2 vs T1 sand-rock 8~16 .188 .179 .468 .499 .051 .000 1.949 1.841 2.656 2.519
rock 8~16 .006 .012 .050 .078 .000 .066 .752 .243 0.808 0.399
C3 vs T2 sand 8~16 .132 .087 .334 .398 585 .031 1.489 .780 2.540 1.296
rock- 8~17 ~~ .025 ~~ .030 ~~ .007 ~~ .006 ~- 0.1‘2
gravel
rock 8~17 .018 - .118 - .012 - .277 ~~ 0.425 ~-
04 vs T3 sand 8~18 .006 .007 .060 .026 .001 .003 .004 .048 0.107 0.084
sand— 8~18 .011 <.001 .039 .016 .049 .003 .208 .052 0.307 0.071
gravel
C5 vs T4 sand 8~21 .002 .002 .096 .060 .044 .001 .105 .048 0.247 0.111
gravo1- 8~21 .001 -- .010 ~~ .050 ~~ .071 ~~ 0.131 ~-
sand
gravel 8-21 ~~ <.001 -~ .019 -~ .000 -- .021 ~~ 0.040
co vs is sand 8~21 -- .ooo -- <.001 -- .001 -- .010 -- 0.011
sand- 8~22 .022 .005 .031 .045 .001 .007 .031 .005 0.085 0.C62
gravel
rock- 8~22 .011 ~~ .042 - .030 ~~ .169 ~~ 0.252 -
gravel
Means .040 .032 .125 .117 .082 .012 .506 .305 0.756 0.471

 

45

invertebrates into treatment sections than in the control sections which
lacked such abundance of attachment sites for invertebrates. Also,

the brush and leaves in the water in treatment sections were probably

a large source of food, "enticing" invertebrates not to drift out.

Salmonids are primarily drift feeders (White 1967‘ in Hunt 1969),
so availability of food for trout was probably better in the control
sections at the time of drift sampling. It has been suggested that
the level of incoming drift food is a factor in determining a stream
sections' trout carrying capacity (Mason and Chapman 1965; Peterson
1966, in Waters 1969).

Dr. R. Merritt (Dept. of Entomology, MSU) studied samples of
hinge-cut twigs, leaf packs, and bottom sediments on two dates (May
16-17, 1981 and July 29-30, 1981) from stations 38 and 47 to investigate
the effects of the hinge-cutting on macroinvertebrate abundance and
biomass. He found that bottom samples immediately adjacent to or under
hinge-cuttings had a higher biomass of aquatic invertebrates (May:

162.8 mg dry wt/ftz; July: 1180.5 mg dry wt/ftz) than comparable sites

in control sections (May: 132.0 mg dry wt/ftz; July: 490.7 mg dry wt/ftz).
Merritt suggested that the differences in biomass might be due to the
scouring effect of the hinge-cuts, which increase current speed and,
subsequently substrate size. Merritt concluded that the practice of
hinge—cutting created a greater variety of habitats for macroinvertebrate
colonization, by providing twig surface area for attachment, trapping

leaf litter which provides additional sites, and changing bottom

sediments by scouring.

46

Implications for Management

 

Due to the short duration of this study, complete evaluation of
hinge-cutting as an effective management tool was not possible. Several
years are required after habitat enhancement for trout populations to
adjust ecologically to improved conditions. White (1975) states that
at least one year is needed before the onset of a positive population
response and Hunt (1976) found that at least five years are needed for
full response of brook trout to habitat improvement. However, such
changes in channel form and flow as preceded increased trout abundance
in Lawrence Creek, Wisconsin (Hunt 1971), occurred due to hinge-cutting
within the treatment sections of the Salmon Trout River.

In Lawrence Creek, the amounts of pool area and overhead bank cover
were the key environmental factors determining trout survival before
installation of bank covers and deflectors, and following habitat
modification, channel depth, pool area, and protective cover all in-
creased. Resultant population responses included reduced emigration.

a decrease in overwinter mortality, and a stockpiling of age-1+ trout
(Hunt 1971). Hinge-cutting has produced an initial narrowing and
deepening of the channel, and has provided additional overhead cover.
These characteristics have been found to reduce unfavorable effects of
low flows (White 1975). There is also evidence of reduced mortality

or emigration from treated sections, as hinge-cut areas did not experi-
ence a significant decline in trout abundance between the springs of
1980 and 1981, as was found in the control sections. Increased water
velocities produced by constricting the flow have scoured sand and silt
off gravel, now available for trout spawning, and would conceivably

also have increased the abundance of macroinvertebrates.

47

The preferred trout feeding microhabitat includes an area of low
current velocity near to a principal line of drift, all with adjacent
overhead cover (Jenkins 1969). The denser areas of hinge-cuts provide
these characteristics by constricting the flow and concentrating the
drift food supply. Hinge—cut brush can provide visual isolation among
the branches, and a higher current velocity in the channel. Both features
can ultimately serve to decrease agonistic behavior, and allow more
trout to occupy a given area.

Hunt (1979) reported on removal of woody brush from stream sections
in Wisconsin, and hypothesized that favorable changes in channel morpho-
metry (similar to those I have found with hinge-cutting) would result
due to proliferation of aquatic macrOphytes, and establishment of grassy
turf banks. Growth of grasses and sedges help stabilize the banks,
prevents erosion, and forms undercuts which provide hiding cover (White
and Brynildson 1967). Hunt's (1979) study was complicated by decreased
stream discharges, which were believed to be the primary reason that
standing crops of brook trout failed to increase following brush removal.
Year-round discharge estimates for the Salmon Trout were not made, so
correlations with trout standing crop are not possible.

While the trout population had little time to react to any bene-
ficial environmental changes brought on by hinge-cutting, macroinverte-
brates quickly colonized the submerged brush. The prevalent species

were filter-feeding caddisflies of the genera Hydropsyche and

 

Brachycentrus. Merritt (Dept. of Entomology, MSU) also demonstrated

 

a greater abundance of benthic invertebrates in hinge sediments than in

comparable control sites. The blend of an easily available food supply

48

and protective cover should provide an attractive position choice for
stream salmonids.

There is obvious advantage of hinge-cutting over conventional bank
covers and deflectors in ease of construction and economy. Virtually
no materials need be purchased and transported to the stream, and labor
is greatly reduced. Each treatment section was completely hinge-cut
by a two-man crew within a single work day. In actual management, only
those areas along streams should be hinge-cut which have suitable quan-
tities and type of riparian brush and are located in stretches with
inadequate overhead cover or in erosion susceptible zones.

Disadvantages of hinge-cutting include unnatural appearance,
especially initially and when applied over long sections of stream.
However, some of the felled brush gradually becomes covered with sediments,
and grasses and sedges invade (or can be planted) to conceal cuttings
on the bank. Fishing can be both enhanced and hindered by hinge-cutting.
Large amounts of bruSh in the stream increase the likelihood of snagged
and lost lures. However, fly-casting is made easier with the stream-
side brush knocked down, and hinge-cuts provide centralized areas on
which fishing effort can be focused. Huron Mountain Club members had
mixed reactions to the fishability of the treatment sections. Again,
applied use of hinge-cutting would likely not involve the full lengths
of both banks, so the aesthetics and access for fishing would not need

to be altered drastically in actual practice.

Conclusions

 

Hinge-cutting of riparian brush is a simple technique for mani-

pulating stream channel form and flow to benefit trout. The method

49

may be particularly suited to streams in steep terrain, where the force
of floods may limit the use of conventional bank covers. Loss of
hinge-cut covers would not be costly.

Hinge-cutting provides overhead cover from predators and shelter
from adverse currents, increases the availability of food organisms,
and promotes bank stabilization. The suitability and effectiveness of
the method of habitat modification is dependent upon physical and
biological characteristics of each proposed treatment area. Proper site
selection will enhance the desired physical changes, while holding

aesthetic losses to a minimum.

LITERATURE CITED

Allen, K. R. 1969. Limitations on production in salmonid populations
in streams. Pages 3-18_ig T. G. Northcote, ed. Symposium on
Salmon and Trout in Streams. H. R. MacMillan Lectures in
Fisheries, Univ. Brit. Col., Vancouver.

Bassett, C. E. 1978. Effect of cover and social structure on position
choice by brown trout (Salmo trutta) in a stream. M.S.
Thesis, Michigan State University. East Lansing, Michigan.
181 pp.

 

Boussu, M. F. 1954. Relationship between trout populations and
cover on a small stream. J. Wildl. Mgt. 18: 29-239.

Chapman, D. C. 1948. A mathematical study of confidence limits of
salmonid populations calculated by sample tag ratios. Int.
Pac. Salmon Fish. Comm. Bull. 2: 67-85.

1966. Food and space as regulators of salmonid populations
in streams. Amer. Nat. 100: 345-357.

Cooper, E. L. 1952. Rate of exploitation of wild eastern brook
trout and brown trout populations in the Pigeon River,
Otsego County, Michigan. Pap. Mich. Acad. Sci. Arts Lett. 38:
151-162.

Cooper, G. P. and K. F. Lagler. 1956. Appraisal of methods
of fish population study: Part III, the measurement of fish
pepulation size. Trans. N. Am. Wildl. Conf. 21: 281-297.

Cooper, C. 0. and T. A. Wesche. 1976. Stream channel modification
to enhance trout habitat under low flow conditions. Water
Resources Series No. 58. Univ. of Wyoming, Larmaie.

Enk, M. D. 1977. Instream overhead bank cover and trout abundance
in two Michigan streams. M.S. Thesis, Michigan State

Gibson, R. J. and M. H. A. Keenleyside. 1966. Responses to light of
young Atlantic salmon (Salmo salar) and brook trout
(Savelinus fontinalis). J. Fish. Res. Bd. Can. 23: 1007-
1025.

 

 

50

51

Gill, J. L. 1978. Design and analysis of experiments in the animal
and medical sciences. Vol. I. Iowa State Univ. Press,
Ames, Iowa.

Gowing, H. and G. R. Alexander. 1980. Population dynamics of trout
in some streams of the northern lower peninsula of Michigan.
Fisheries Research Report 1877, Mich. Dept. Nat. Res.

Hale, J. G. 1969. An evaluation of trout stream habitat improve-
ment in a north shore tributary of Lake Superior. Minn. Fish.
Invest. 5: 37-50. Cited in R. J. White. 1973.
Stream channel suitability for coldwater fish. Proc. 28th
Ann. Mtg. Soil Conserv. Soc. Am.: 61-79.

Hendrickson, G. E., R. L. Knutilla and C. J. Doonan. 1973. Hydro-
logy and recreation on the cold-water rivers of Michigan's
Upper Peninsula. U.S.G.S. Water Info. Ser. Rep. 4. 39 pp.

Hubbs, C. L., J. R. Greeley and C. M. Tarzwell. 1932. Methods for
the improvement of Michigan trout streams. Bull. Inst. Fish.
Res. No. 1., Univ. of Michigan.

Hunt, R. L. 1969. Effects of habitat alteration on production,
standing crops and yield of brook trout in Lawrence Creek,
Wisconsin. Pages 281-312‘i2 T. G. Northcote, ed. Symposium
on Salmon and Trout in Streams. H. R. MacMillan Lectures in
Fisheries, Univ. Brit. Col., Vancouver.

. 1971. Responses of a brook trout population to habitat
development in Lawrence Creek. Tech. Bull. No. 48, Wis.
Dept. Nat. Res., Madison. 35 pp. -

1976. A long-term evaluation of trout habitat development

and its relation to improving management-related research.
Trans. Am. Fish. Soc. 105(3): 361-365.

. 1979. Removal of woody streambank vegetation to improve
trout habitat. Tech. Bull. No. 115, Wis. Dept. Nat. Res.,
Madison.

Hynes, H. B. N. 1970. The ecology of running waters. Univ. Toronto.
555 pp.

Jenkins, T. M. 1969. Social structure, position choice and micro-
distribution of two trout species (Salmo trutta and Salmo
gairdneri) resident in mountain streams. Anim. Behav. Monogr.
2: 57-123.

 

Mason, J. C. and D. W. Chapman. 1965. Significance of early emer-
gence, environmental rearing capacity, and behavioral ecology
of juvenile coho salmon in stream channels. J. Fish. Res.
Bd. Canada. 22: 173-190.

52

Peterson, G.R. 1966. The relationship of invertebrate drift abundance
to the standing crop of benthic organisms in a small stream.
M.S. Thesis, Univ. Brit. Col., cited in T.F. Waters. 1969.
Invertebrate drift—ecology and significance to stream fishes.
Pages 121—134 in T.G. Northcote, ed. Symposium on Salmon and
Trout in Streams; H.R. MacMillan Lectures in Fisheries, Univ.
Brit. Col., Vancouver.

Saunders, J. W. and M. W. Smith. 1962. Physical alteration of stream
habitat to improve brook trout production. Trans. Am. Fish.

Soc. 91: 185-188.

Seber, G. A. F. 1973. The estimation of animal abundance and related
parameters. Charles Griffon and Co. 506 pp.

Smith, L. L. 1941. A fisheries management program for the waters of
the Huron Mountain Club. Ph.D. Dissertation, Univ. Michigan.
424 pp.

Waters, T. F. 1969. Invertebrate drift-ecology and significance to
stream fishes. Pages 121-134.i3 T.G. Northcote, ed.
Symposium on Salmon and Trout in Streams. H.R. MacMillan
Lectures in Fisheries, Univ. Brit. Col., Vancouver.

Wesche, T. A. 1976. Development and application of a trout cover
rating system for IFN determinations. Pages 224-234‘i3
J. F. Osborn and C. H. Allman, eds. Proceedings of the Symposium
Symposium and Specialty Conference on Instream Flow Needs:
Vol. II. Am. Fish. Soc., Bethesda.

White, D. A. 1967. Trophic dynamics of a wild brook trout stream.
Ph.D. Thesis. Univ. Wis., Madison. 183 pp.

White, R. J. 1973. Stream channel suitability for coldwater fish.
Proc. 28th Ann. Mtg. Soil Conserv. Soc. Am.: 61-79.

. 1975. Trout population responses to streamflow fluctuations
and habitat management in Big Roche-a-Cri Creek, Wisconsin.
Verh. Internet. Verein. Limnol. 19: 2469-2477.

White, R. J. and 0. M. Brynildson. 1967. Guidelines for management
of trout stream habitat in Wisconsin. Tech. Bull. No. 39.
Wis. Dept. Nat. Res., Madison. 65 pp.

APPENDIX

53

Table A1. Discharge estimates (cms) at station 40 (Lower Dam) on
the Salmon Trout River during the summer of 1980.

 

 

Discharge Discharge Discharge

Date (cms) Date (cms) Date (cms)
6 - 27 0.768 7 - 16 0.81 8 - 4 —-p
28 0.76 17 -- 5 1.10

29 -—p 18 —— 6 0.78

30 1.048 19 -- 7 0.75

7 - 1 --p 20 -- 8 0.748
2 0.863 21 —- 9 0.72

3 0.81 22 0.81 10 --P
4 0.77 23 0.78 11 1.31

5 --p 24 0.78 12 1.023

6 0.87a 25 0.77 13 1.26p

7 1.33p 26 0.76 14 1.78

8 1.06a 27 0.76 15 1.16

9 0.86 28 0.75 16 0.93

10 0.78 29 0.76 17 0.86

11 0.77 30 0.72 18 0.83p

12 -- 31 0.74 19 0.86

13 0.71 8 — 1 0.72 20 1.02

14 --p 2 0.71 21 0.89

15 0.99a 3 0.70 22 0.80

23 0.77

 

a
actual discharge measurement

pprecipitation

54

Table A2. Shocking dates for brook trout population estimates in
the Salmon Trout River, Marquette County, Michigan.

 

 

Electrofishing
period Date
Fall 1979 September 22-24, 1979
Spring 1980 June 8~13, 1980
Fall 1980 September 14-17, 1980

Spring 1981 May 16-19, 1981

 

55

 

 

 

 

 

Table A3. Finclip record for wild young-of-the-year brook trout and
for stocked hatchery brook trout in the Salmon Trout River,
Marquette County, Michigan.

Wild Trout
Year Finclip Date
1975 A (below Lower Falls) June-July 1976
ARV (above falls)

1976 LV Fall 1976

1977 RV Fall 1977

1978 none not shocked

1979 ALV Fall 1979

1980 A Fall 1980

Hatchery Trout

1976 LP

1977 RP (possible some A clips)

1978 ~-

1979 2V

1980 LP

Key: A - adipose RP ~ right pectoral

RV - right ventral

LP - left pectoral
LV - left ventral 2V - both ventrals

56

 

 

 

Table A4. Recapture efficiency (%) of brook trout by size class
for four shocking periods on the Salmon Trout River,
Marquette County, Michigan.
Size Shocking period
Class Fall 1979 Spring 1980 Fall 1980 Spring 1981
‘5 100 mm 18% 33% 12% 9%
101-125 mm 32% 45% 19% 23%
126-150 mm 34% 55% 30% 29%
151-175 mm 35% 57% 23% 42%
176-200 mm 32% 53% 39% 35%
3_201 mm 38% 47% 41% 14%

 

57

 

 

 

HNH m m mH Nm N Hm Nm
NON o m NN om mN mm om
qu H 0 NH mN NH Nm mm
qu H o mH mN NH No cm
mNH m N mH He 0H «CH mm
mm m a w mH m we Nm
mm m n w mH n no Hm
omH m w mH mN m oN om
aw H N OH «H mH Ne on
wN m o mH mH 0H Hm we
ow n q 0H wH mH mm Na
ow m q 0H wH mH mm on
Nw w HH w 0 HH mm me
mm H o m HH q 0H «a
mN m m NH mN m wN me
Ne o o N o c He Ne
NN H o o N o mH He
mm H m m N N Nm oq
cNH H m m N «H ow mm
oNH H m m N «H om mm
HqH H HH 0H m N mHH Nm
qu o m NH a m mHH cm
no . H N cH N m «a mm
1 I 1 I I I 1 cm
Hmuoe ooNA GONuoNH mNHIHmH omHuoNH mNHIHoH ooHv Humane
AEEV mmmHo oNHm zuwcoH Hmuoe cowumum

 

.mNmH HonEoumom .ao>Hm psosh :oeHmm ozu :H :oHpmum Np moumEHHmo :oHumHsmom unosp xoonm .m< oHan

58

 

 

Noam as oNH 88m own ssm mmHN
usmmm .H4 -88 -NmN lacs -NHN -NomH Ho Nan

mesa mm as awN ems ANN NmNH mHnuos
MNH o N mH on mH am on

me w m ma mm wH HoH am

one a m mm mm HH omH mm
Hnnoe com. oom-8NH maH-HmH omH-e~H mNH-HoH ooH.w Hanan:

 

AEBV mmmHo muwm :uwmoH

Hnuoe

soaumum

 

H.e.ucouv.m< mHnnH

59

 

 

 

momm 0 0mm omoH oNHH Nam Hon om
«ooH mm o qu mmm me NHq mm
quH mm o qu mmm me NHq cm
onN «He NNH «on New oHN wa mm
NcmH mwN NmN mwN NNm 0N MNm Nm
quH mmN NmN mwN NNm ON mNm Hm
mme omm «on mmq coo moH mmq 0m
mmNH om qu oom omm oNH oNN as
mqu mNN o mmm omN NNH NoH we
ooNH mmN qu ooq qu omH mON Na
ooNH mmN qu ooq qu omH moN on
mme NmoH mmo omN NON moH wMN ma
mNm HoH o moH «0N co we an
mme oeN moH owe ch NHH qu ma
NoN o o mm c we cmH Ne
mmH HHH o 0 ea c mm He
mmNH ma omN can mmH ow «NN on
NHqH me mmN cam oNH «0H can on
NHqH me moN com oNH qu oqm mm
HmHN o HmN qu mNH «N HmN Nm
mqu o mwN owe om HHH mcq on
quH Nm qu oNq on mm oaN mm
1 I 1 I I I I an
Hmuoe ooNv ooNuoNH mNHIHmH omHIoNH mNHuHoH ooHv Hones:
.INEEV mmmHu oNHm cuwcoH Hmuoe nowumum

 

.mNmH nonemunmm .nm>wm ozone :OEHmm mzu cw cowumum NH moumEHumm Ava mmmEoHn unouu xooum .c< mHHmH

60

 

 

 

meemm mmNN mama HNNmH mmaNH HNHs NesoH
-8osmm -ssos -Neaa -mamw -mmHm -Hamm usage Ho Naa
sagas mean Hsmm mquH mNHHH mwmm saws Hanna
.ommH o was mum Nae amH NNN on
swam one wnH ohm 8mm mmm mmm an
exam mom me was was NmH Has an
onN «mm osm Has ssHH so «on Nm
Hnnos ooNv ooN-8aH msH-HmH omH-o~H mNHuHoH oon. pagans
AEEV mmmHo mNHm cuwcmH Hmuoe nOHumum

 

H.n.ocouV .8<.6Hnne

61

 

 

 

OO O c O O ON NH NO
OO N O NH OH NO NH OO
OO N H O OH HN OH mm
me O O N O ON O «O
OO O m m ON ON O mm
Om N m O O ON qH Nm
OO O H O OH HH OH HO
HN O c O OH ON ON Om
Om N O O HH OH OH OO
«O m O O OH ON ON Oq
OOH O O OH ON ON ON Nq
OO O HH O OH OH NH Oq
mm N NH NH O OH O me
OO O N HH ON OH ON «q
Hm O O H H OH OH me
ON O O H O O NH NO
ON N O O O N N Hq
Om O O O OH HN O OO
OO N O O HN ON O OO
NN O O H N N O OO
HO O O O O ON O NO
as N m O OH OH N Om
HO O m O NH ON O mm
me O N O HH OH N «m
Hmuoe OONA OONIONH mNHIHmH OOHIONH ONHIHOH OOHv Hones:
AEEV mmmHo oNHm :uwdoH Hmuoe cowumum

 

.OOOH OGSO .um>Hm usoue coEHmO ecu :H :oHumum >3 maumEHumo coHumHnaoa unouu Hooum .N< oHOmH

62

 

 

 

smmH an HmH sew mam was mam
[maNH -ms -am -NnH Ham -sms -Hmm Ho Nam
Nme MN oHH mHH Hmm Nam eHm annoe
as o H H e_ an oH on
as o o m OH on OH mm
mm m s mH ON Hm oH mm
Hnnoe com. ooN-8NH mNHuHmH omHIONH mNHuHoH oon. pagans
AEEV mmmHU wNHm LumeH HMUOH. SOfiuMum

 

H.e.uaoov .Na 6Hnne

 

 

ONNH O NON OOO OOH ONO OO NO

 

 

OOON OOO OON OOO OHO OOO ONH OO
ONOH OHN OO OOO OOO ONO NOH OO
OOOH O ONN OOO NOH NON OO OO
OOOH OOO NNH OON OOO NOO OO OO
OOHH OOH ONH NOH OHN HOO OO NO
OHO O OO OON NOO OOH ONH HO
HOOH O OON OOH HON HNO OOH OO
OOOH OOH NOO OHO NON ONN NHH OO
OONN NOO OOO OOO OON OOO OON OO
OHON HOO OON OOO OHO ONO ONN NO
as OOOH OHO NON ONH OOO NOH OO OO
,6 OOHN ONH NHO ONO OOH OOH NO OO
ONON OHN NHO OHO OOO HOH ONH OO
OOO O O OO NN OOH OHH OO
ONO O O OO OOH OO ON NO
NNO OON OON OOH OOH OHH OH HO
OHON ONO NOO OOH OOO OHO OO OO
OHNH ONH OOO NOH OOO OOO NN OO
HHO NON O OO NOH ON ON OO
OOO ONN O OOH HON OON OO NO
ONNH OOH OOH OHH OOO OON OH OO
OOOH NOO OOH NOH NOO OON O OO
NONH NHO OOO OOH OON NOH OH OO
Hence OONA OONIONH ONHIHOH OOHIONH ONHIHOH OOHv popes:
AEEV mmmHu mNHm Huwde Hmuoe :OHuMum

 

.OOOH mend .um>HO unoue coEHmO ecu aw coHumum On mmumEHumm AEOV mmmEoHO unouu Hooum .O< oHOmH

64

 

 

 

 

8N888 8H88H HH88 N888 8888 88am 8~H8
-88H88 -NwsO -H888 -N888 -888N -8888 -8N8H Ho N88
888H8 88NN HONN mama 8888 88am 888m anaoH
888H 8 888 888 888 8H8 88 88
8NN1 o o 888 moN 8N8 NmH 88
Hmmm 8N8 88N 888 N88 888 88 88
H8869 com. 888-8NH 8NH-H8H 88Hu8~H 8~H1H8H OOHw panes:
HEEV mmmHu muwm HumdoH Hmuoe coauMum
H.8.ucouO .8< 8Hnne

65

 

 

 

OO O O OH OH OH ON NO
OO O H OH ON O ON OO
ON O H O HN O OO OO
OO O O O HH N OH OO
OO H O ON ON OH ON OO
OO O O OH HH N HH NO
HO H O O O O NN HO
HO H O O O O NN OO
OO O O O HH OH HH OO
OO H O ON ON OH OH OO
NO O O O HH N HH NO
NO O H O HH N ON OO
NO O O O OH N HH OO
OO O O O OH OH N OO
OO O O O OH ON OO OO
OO O O O HH OH NN NO
OO O O N O N OO HO
OON O O ON OH OH OOH OO
NHN H O NN OH ON OOH OO
OOH O H ON OH OO OO OO
NOH O H OH O HN OHH NO
OOH O O ON ON NH NN OO
OOH O O OH OH OH OO OO
OO O H O O NH OO OO
HmHOH OONA OONIONH ONHnHOH OOHIONH ONHIHOH OOHv Hanana
AEEO mmmHo muHm Luwwoa Hmuoy cowumum

 

.OOOH HoOEOHOmO .uo>Hm unoue coEHmO ecu cw COHumuw NO moumafiumm coHumHnaoa uDOHu HoouO .O< macaw

66

 

 

 

 

OOHO OO NN OOO OOO ONO OHOH
IHNNH IHO IOO INNN IOOO IONN IONO Ho NOO
ONON ON NO OOO OHO OOO NNNH Hmuoe
OO H O OH ON OH OO OO
OOH H O OH OO ON HO OO
OOH O O OH OO HO OO OO
:38. 8? 88.88: 8:12 8782 821:: SW 38.5:
AEEO mmmHu mNHm OOOCoH Hmuoe noHumuO
A.O.uc00v .O4 OHOMH

 

67

NNNH o 0 OOO OOO ONH ONN NO

 

 

HOOH O ON HOO OOO OO OOH OO
ONOH OOO OO NHO OHO O OON OO
OOO O O NON OON OO ON OO
HOHN NNH NON OHO OOO OON HNN OO
OOOH OOO OON NOO OON OO OO NO
OHO OO O ONH OOH OO OOH HO
OHO OO O ONH OOH OO OOH OO
NOOH OOO OON ONO OON OON HO OO
OOON NOH OOO ONO OOO NOO OO OO
OOHH NOO O OOH OON OOH OO NO
HHOH OON NO OOH OON OO OOH OO
OOOH OON HON NOH OOO ON OO OO
OHON ONO OOO OOO NOO ONN OO OO
OOHN ONO OHO O OHN OOO OON OO
OOOH O OO OOO OON OHH ONH NO
OON ONN O NOH OO OO ONO HO
ONNO NON ONN HNOH ONO ONN HOO OO
ONOO OO NON OHO OOO NOO NHHH OO
OOON NNO ON OOO OHO OOO ONO OO
OOOH O ON ONO OON OHN OOO NO
OOON ONN OON ONO OOO OHN NOO OO
OOON OOO OON OOO OOO OON OON OO
OOOH OOO OO OHO OO OOH NON OO
Hmooe OONA OONIONH ONHIHOH OOHIONH ONHIHOH OOHv ponenc
HEEV mmmHo wNHm :uwde Hmuoe :oHumum

 

.OOOH nonsmoaom .Ho>HO unoue coEHmO ozu CH :oHumum OO moumEHumm HEOV mmmEoHO usonu xooum 493onOmH

68

 

 

 

HOOHO OONHH OOOO OOOOH OONHH OOOO ONNO
IOOOOO IOOHO IOOON IONOO IOHOO IHOOO IOOOO Ho NOO
OOONO OOOO OOOO ONONH OOOOH OHNO OOON HOOOH
HNOH OOH O NOO HOO ONN OOH OO
HOON ONH O OOO ONN OOO NOO OO
OONO NNO O OOO OOO NHO OOO OO
:38 88? 888: 8:12 88782 821:: SW 885:
AEEV mmmHo mNHm HuwcmH Hmuoe COHumuO

 

HOLE”; .82 833

 

69

 

 

HO O H O OH OO NN NO
OO O O N HH O NN OO
HO N H O O NH NN OO
OH O O O O O O OO
ON N O OH NO N HH OO
OO O O O NN ON NN NO
ON O O O O O O HO
ON O O O O N HH OO
ON N O O O O O OO
OO O O N HH ON HH OO
OO O O OH NH N HH NO
OH O O O O N O OO
HN O O H O O OH OO
ON N O H N N OH OO
OO N O O O OH OH OO
OH O O O O N O NO
O O O O O O O HO
OHH O O O N OO OO OO
HO N H O O OH NO OO
OO O H H O NH HH OO
OO N H H OH ON OH NO
NO O O H OH ON HH OO
NN O O O O OH HH OO
HO N O O N OH HH OO
HmooN OONA OONIONH ONHIHOH OOHIONH ONHIHOH OOHv Hanan:
AEBO mmmHo mnwm :uwcmH Hmuoe cowumum

 

.HOOH Omz .um>HO Ozone :oSHmO ecu :H coHumum NO mmumEHumo coHumHnmoa unouu Hooum.HH< oOOmH

7O

 

 

 

NOOH OO OO OOH OOO NOO OON

IOOO 1O ION IOOH uOOH INON INO Ho NOO
OONH ON OO NOH NON OOO NNO Hmuoe

OO O O N NN OH NN OO

OO O O O O OH NN OO

HN O H OH NN NN OH OO
Hence OONA 88~-8N8 8NHnH8H 88HI8NH 8~H1H8H oon. H8885:
AEEV mmmHo oNHm HuOde Hmuoe COHumuO

 

H.O.8cooO.HH< mHnne

71

 

 

 

OONH O OO OON OON ONO OON NO
NOO O ONH ONN OON OO HOH OO
NNO OON OO ONH o OHN OHN OO
OHO O O OOH OOH OO O OO
NOOH NON O OOO OHO OO OO OO
OONH O OOO OOO OHO OOO OHN NO
NOO O o OOH ONH OO OO HO
ONN O OOH ONN OON ON OO OO
NON OON O ONH OON OO o OO
OONH O OON OHO OON NNO NN OO
HOOH OOO OOO OOO OOO OHH OO . NO
OOO O O OOO ON OO O OO
HOO O O NO OO O HOH OO
OOO NOH NON OO OO OO OO OO
ONN NON O NOH OO OOH ONH OO
OOO O O O ONN NO O NO
O O O O O O O HO
OONN OOO OON OOH OO NON HOO OO
NOO OON OO OHH OO OON OON OO
OOO O OO O OO OON OO OO
OONH OON OO NO OOO NOO NHH NO
OHO O OOH OO HNO OON OO OO
OOO O OON NHH O ONH OO OO
NNO OOO OOH ONH NO OHH OO OO
Hmuoe OONA OONIONH ONHIHOH OOHIONH ONHIHOH OOHv penanc
AEEV mmmHu mNHm :Awwwa Hmuoe :OHumuO

 

.HOOH Om: .nm>Hm uaoue :oEHmO mnu :H :oHumum OH mmumaHumm NEOV mmmEoHn unouu xoouO.HH<.mHOmH

72

 

 

88888. 8888 8888 8888 8888 8888 8888
-88888 -888 -8H8H -8888 -8888 -8888 -888 88 888
88888 8888 8888 8888 8888 8888 8888 88888
8888 8 888 888 888 888 H88 88
888 8 8 888 H88 888 888 88
8888 8 88 HH8 888 888 88H 88
88888 8888 888-888 888-H88 88H-88H 888-888 888w 888888

 

AEEV mmmHo mNHm nuwcoa Hmuoe

coaumHO

 

78.8883 .82 388.8

HICHIGRN STRTE UNIV. LIBRRRIES
IIHI "MI I“ lllWl WW I“! II M! W HIM 1|! "1 W lllHl
31293007961992