1’.

l v unm’u-uvgu- n1».

If.

“nu-q.- u ‘~«

In

'1.

u.
n

 

 

This is to certify that the
thesis entitled

Evaluation of Habitat Enhancement Structures in Four
Reservoirs of the Au Sable River, Michigan

presented by

Todd Christian Wills

has been accepted towards fulﬁllment
of the requirements for

M.S. Fish. & Wildl.

degree in

 

 

 

Date June 24, 2002

 

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

 

 

LIBRARY

Michigan State

University

 

 

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

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c-JCIRC/DatoDuo.p65-p.15

EVALUATION OF HABITAT ENHANCEMENT STRUCTURES IN FOUR
RESERVOIRS OF THE AU SABLE RIVER, MICHIGAN

By

Todd Christian Wills

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

2002

ABSTRACT

EVALUATION OF HABITAT ENHANCEMENT STRUCTURES IN FOUR
RESERVOIRS OF THE AU SABLE RIVER, MICHIGAN

Todd Christian Wills

Addition of habitat enhancement structure to aquatic systems is a common
practice by ﬁsheries managers hoping to increase spawning success, production, and
angler catch rates of important game species. However, quantitative evaluations of these
efforts are few and typically do not determine the extent to which natural habitat mediates
effects of habitat enhancement structures. I evaluated the effects of two types of habitat
enhancement structure on four functional ﬁsh groups in four reservoirs of the Au Sable
River, Michigan. Using a combination of sampling methods, I compared several
response variables (including relative abundance, nesting, and angler catch rates) between
areas with and without structures, as well as before and after structure placement, across a
gradient of natural habitat conditions. Signiﬁcant treatment effects of half-log habitat
enhancement structures occurred in some cases, but no signiﬁcant treatment effects of
AquaCrib® structures were detected. Smallmouth bass (Micropterus dolomieu)
responded to half-logs, with signiﬁcantly greater relative abundance, nest density, and
nest success in areas with half-logs compared to areas without half-logs. Other functional
groups displayed few signiﬁcant differences in response variables between areas with and
without structure, or before and after structure placement. Habitat effects varied with
reservoir and ﬁmctional group, but generally inﬂuenced response variables more than the

presence of habitat enhancement structure.

ACKNOWLEDGMENTS

I wish to express my deepest appreciation and a sincere thank you to my major
professor, Dr. Mary Bremigan, for her support, encouragement, and friendship. Dr.
Bremigan made herself available at any time to address my questions and concerns while
also giving me the freedom to tailor my research to my personal interests. It was a
pleasure and an honor to pursue my graduate studies under her direction. I also owe a
great deal of thanks to my graduate committee, Dr. Daniel Hayes and Dr. Gary
Mittelbach, for providing encouragement and advice in developing my research, as well
as constructive criticism on an earlier draft of this thesis. Dr. Hayes also provided
statistical advice and support for which I am most grateful. I am indebted to the
numerous graduate students, undergraduate students, and technicians that provided ﬁeld
and laboratory assistance, especially K. Cheruvelil, J. Chiotti, N. Dom, S. Hanson, P.
Hrodey, A. Mahan, S. Makosky, C. Meier, V. Moore, D. Myers, B. Ordell, N. Pfost, G.
Steinhart, M. Wesener, and S. Wills. I would also like to thank H. Jennings and B.
Stuber of the USDA Forest Service and K. Kruger and K. Newman of the Michigan
Department of Natural Resources for logistical support. The Northeast Michigan
Sportsman's Club installed the habitat enhancement structures studied. Funding for this
research was provided by the Michigan Department of Natural Resources and the USDA

Forest Service.

iii

TABLE OF CONTENTS

LIST OF TABLES .............. v
LIST OF FIGURES ................................................................................. vii
INTRODUCTION .................................................................................... 1
METHODS ............................................................................................ 4
RESULTS ............................................................................................ 14
DISCUSSION ....................................................................................... 20
APPENDIX .......................................................................................... 26
LITERATURE CITED ............................................................................. 44

iv

LIST OF TABLES

Table 1: Physical characteristics of the Au Sable River study reservoirs based on data
from Consumer's Power Company, the Federal Energy Regulatory Commission, the
United States Forest Service (Kinney et a1. 1999), and this study. Reservoirs are ordered
from upstream to downstream. Shoreline development factor describes the irregularity of
the shoreline and is expressed as the ratio of the length of the shoreline to the
circumference of a circle of area equal to that of the lake surface, with a minimum value
of 1.0 for a circular lake. Water level ﬂuctuation is the maximum daily change in water
level resulting from hydroelectric dam operation.

Table 2: Experimental design and number of sites per reservoir. "Before/after" indicates
within-reservoir comparison before and after structure placement. "After only" indicates
within-reservoir comparison aﬁer structure placement.

Table 3: Limnological characteristics of the Au Sable River study reservoirs based on
data from Consumer's Power Company, the Federal Energy Regulatory Commission, the
United States Forest Service (Kinney et a1. 1999), and this study. Reservoirs are ordered
from upstream to downstream. Temperature represents the mean summer epilimnetic
temperature (1m depth), June through August of 2000 and 2001. Dissolved oxygen
represents the range of dissolved oxygen levels observed across all depths during
summer. Chlorophyll a, total phosphorous, and total nitrogen are mean summer values
from samples taken in June through August of 2000 and 2001.

Table 4. P-values from mixed-effect analysis of variance modeling the effects of
treatment, macrophyte cover, coarse woody material (CWM) cover, and substrate on
catch-per-effort by functional group. Treatment refers to the presence or absence of half-
log habitat enhancement structures. N refers to the total number of areas sampled.
NS=not signiﬁcant. No signiﬁcant treatment effects were found for gill netting at
AquaCribs® (N=67), angling at half-logs (N=112), or angling at AquaCribs® (N=189).

Table 5. P-values from ﬁxed-effect analysis of variance modeling the effects of reservoir
and treatment on catch-per-effort by functional group. Individual sampling sites were
treated as a blocking factor. NS=not signiﬁcant.

Table 6. P-values from mixed-effect analysis of variance modeling the effects of year,
treatment, and year*treatment interaction on catch-per-effort by functional group in
Alcona Pond. Treatment refers to the presence or absence of half-log habitat
enhancement structures. N refers to the total number of areas sampled. NS=not
signiﬁcant.

Table 7. P-values from mixed-effect analysis of variance modeling the effects of
treatment, macrophyte cover, coarse woody material (CWM) cover, and substrate on
smallmouth bass nest density and nest success. A total of 126 individual nests were
monitored throughout the entire spawning season in 2001. Treatment refers to the
presence or absence of half-log habitat enhancement structures. NS=not signiﬁcant.

vi

LIST OF FIGURES

Figure 1. Location of the Au Sable River, Michigan (top) and the four study reservoirs
(bottom).

Figure 2. Side and top views of a half-log structure (top), and three-dimensional view of
an AquaCrib® structure (bottom).

Figure 3. Percent cover of habitat variables in each of the four study reservoirs.
Reservoirs are presented from upstream (Alcona) to downstream (Foote). Substrate (a) is
the total cover (%) of each of three substrate classes summed across all study sites.
Macrophytes (b) and coarse woody material (c) are mean cover (%, i 1 SE) averaged
across all study sites. Note y-axes differ among substrate, macrophytes, and coarse
woody material.

Figure 4. Mean catch-per-effort (CPE, ﬁsh/km, i 95% CI) of the four functional ﬁsh
groups for treatment and reference areas for snorkeling (top) and electroﬁshing (bottom).
Total and planktivore CPE correspond to the left y-axis; piseivore and smallmouth bass
CPE correspond to the right y-axis. Note y—axes differ among functional groups and
sampling methods.

Figure 5. Mean catch-per-effort (ﬁsh/km, i 95% CI) of the four functional ﬁsh groups
for treatment and reference areas in each reservoir for snorkeling (leﬁ) and electroﬁshing
(right). Reservoirs are presented from upstream (Alcona) to downstream (Foote). Note
y-axes differ among functional groups and sampling methods.

Figure 6. Mean catch-per-effort (ﬁsh/km, i 95% CI) of the four functional ﬁsh groups for
treatment and reference areas and year for snorkeling (left) and electroﬁshing (right)
before and after structure placement in Alcona Pond. Data before structure placement
were collected in 2000; data after structure placement were collected in 2001. Note y-
axes differ among functional groups and sampling methods.

Figure 7. Mean diet composition (grams of prey per gram of ﬁsh, i 95% CI) of the four
prey groups by treatment for smallmouth bass and largemouth bass diets from within-
reservoir comparisons (a, b) and smallmouth bass from before/after comparisons (c, d).
Mean percent composition is averaged across all study sites and reservoirs. Note y-axes
differ among prey groups and species for within-reservoir comparisons, as well as among
prey groups and year for before/after comparisons.

Figure 8. Mean smallmouth bass nest density (nests/ha, top) and nest success (%,
bottom), _+_ 95% CI, in treatment and reference areas from data collected in 2001 in Cooke
and Foote ponds. Nest success was deﬁned as the presence of swim-up fry. Note
difference in y-axes between nest density and nest success.

vii

Figure 9. Mean smallmouth bass nest density (nests/ha, top) and nest success (%,
bottom), i 95% CI, by substrate from data collected in 2001 in Cooke and F oote ponds.
Note difference in y-axes between nest density and nest success.

Figure 10. Relationship between macrophyte cover and smallmouth bass nest density
(nests/ha, top) and nest success (%, bottom) for treatment and reference areas from data
collected in 2001 in Cooke and Foote ponds. Note difference in y-axes between nest
density and nest success.

viii

INTRODUCTION

Fisheries managers have long recognized the potential of habitat enhancement
structures to attract and hold ﬁsh (Brown 1986). The fust state management agency to
document the use of habitat enhancement structures was the Michigan Conservation
Department, which added brush shelters and gravel piles to lakes in the early 1930's
(Hazzard 193 7). Since that time, numerous state and federal agencies and local
organizations have undertaken habitat enhancement projects in both freshwater and
marine systems using a variety of materials (Tugend et al. 2002). These habitat
enhancement structures are added to aquatic systems when natural habitat is perceived to
be lacking or insufficient (Prince et al. 1977), with the goal of providing additional cover
and concentrating ﬁsh, thereby increasing recruitment, survival, growth, and angler catch
rates (Johnson and Stein 1979).

The mechanisms by which habitat enhancement structures can produce these
positive effects are varied. For example, habitat enhancement structures have been
proposed to increase recruitment by providing cover for spawning (V ogele and Rainwater
1975, Hoff 1991, Patrick 1996, Hunt 2000), thereby increasing nest density. Ultimately
nest success may beneﬁt if structures provide habitat that allows adults to more
effectively protect their young (Hoff 1991). Structures can also offer refuge from
predation and alter survival by increasing cover (Bohnsack and Sutherland 1985, Johnson
et al. 1988, Moring and Nicholson 1994), providing shade (Helfrnan 1979, Helfman
1981, Johnson and Lynch 1992), and providing sites for orientation and schooling (Klima
and Wickharn 1971, Bohnsack and Sutherland 1985). Accordingly, forage abundance in

the vicinity of structures may be enhanced (Wege and Anderson 1979, Aadland 1982,

Bohnsack and Sutherland 1985, Moring et al. 1989), in turn increasing the abundance,
feeding efficiency, and growth of predators (Wege and Anderson 1979, Bohnsack 1989).
Habitat enhancement structures, particularly in the form of artiﬁcial reefs, have also been
proposed to increase public access by making it easier for anglers to locate ﬁsh, and to
increase angler catch rates by concentrating ﬁsh (Bohnsack 1989).

Numerous studies have demonstrated that habitat enhancement structures succeed
in concentrating ﬁsh in natural lakes (Wilbur I974, Moring and Nicholson 1994) and
reservoirs (Brouha 1974, Paxton and Stevenson 1979, Prince and Maughan 1979a, Brown
1986, Kayle 1986, Moring et al. 1989, Mabbott 1991, Rogers and Bergersen 1999).
However, the extent of their effects varies for several reasons. For example, the degree to
which attraction occurs is often dependent upon the species present (Hubbs and
Esehmeyer 193 8, Rodeheffer 1939, 1945), diel ﬂuctuations in ﬁsh distribution (Moring
and Nicholson 1994), the age of the structure (Wilbur 1978, Moring and Nicholson
1994), and/or the structure's physical attributes. In particular, the average number of
individuals and species attracted increases with structural complexity achieved by
increasing the volume, size, and surface area of habitat enhancement structures
(Wickham et al. 1973, Shulman 1984, Rountree 1989, Graham 1992, Potts and Hulbert
1994). Further, ﬁsh abundance varies with structure interstice size (the space within
structures) in a complex fashion (Brouha 1974, Wege and Anderson 1979, Johnson et al.
1988, Lynch and Johnson 1989, Walters et al. 1991).

Morphometric characteristics of the lakes and reservoirs where habitat
enhancement structures are installed also inﬂuence their effectiveness. Habitat

enhancement structures may be relatively ineffective in systems with topographically

complex bottoms (Pardue and Nielsen 1979) and/or alternative natural habitat (Wilbur
1978, Mitzner 1981, Madejczyk et al. 1998, Rogers and Bergersen 1999). The depth at
which habitat enhancement structures are placed also inﬂuences structure use, although
this is often dependent upon the foraging and habitat preferences of resident species
(Rodeheffer 1945, Kayle 1986, Walters et al. 1991, Johnson and Lynch 1992).
Reservoirs, in particular, often make ideal candidates for habitat enhancement projects.
Reservoir systems often lack natural habitat due to the removal of standing timber prior
to or following ﬂooding, decay of timber, rapid siltation of rocky reefs and other ﬁrm
substrate in littoral areas, and/or lack of aquatic vegetation caused by ﬂuctuating water
levels (Prince and Maughan 1978).

Although federal and state agencies conduct periodic investigations of the effects
of habitat enhancement structures on ﬁsh populations, relatively little information is
available in the primary literature to aid managers in decisions regarding exactly how
much, what kind, or where habitat enhancement structures should be added to a system to
attain desired production goals (Bassett 1994, Rogers and Bergersen 1999). This
uncertainty remains in part because many investigations have been inadequately
documented, and rarely have assessments determined the effects of habitat enhancement
structures on system-wide abundance of ﬁsh. In addition, many assessments have lacked
controls, pre-treatrnent data, and statistical analysis, while most have also failed to
compare the effectiveness of different structure types within systems (Bassett 1994).

By identifying the conditions under which habitat enhancement structures are
most effective, ﬁsheries managers can most successﬁilly and economically use structure

as a tool for increasing reproduction, survival, growth, and angler catch rates of sport

ﬁshes in systems with insufﬁcient natural habitat (Johnson and Stein 1979). Within this
context, I evaluated two types of structure, half—logs and AquaCribs®, in four reservoirs
of the Au Sable River, Michigan. I used a combination of two approaches to do so: (1)
comparisons within reservoirs between similar areas with and without added structure,
and (2) comparisons within a reservoir before and after structure placement. The natural
habitat in these reservoirs varied, allowing me to evaluate the effectiveness of the
structures across a gradient of natural habitat and turbidity. In so doing, I sought to
develop management recommendations for placement of habitat enhancement structures
such that future endeavors are most beneﬁcial and cost-effective.

I hypothesized that areas with habitat enhancement structures would concentrate
ﬁsh, increase nest densities, and increase angler catch rates compared to areas without
structures. I also hypothesized that habitat enhancement structures would inﬂuence
predator-prey interactions. Speciﬁcally, I expected that invertebrates would colonize the
half-logs, and that prey biomass in gut contents from smallmouth bass and largemouth
bass (Micropterus salmoides) would be greater in areas with structures than in areas
without structures. Overall, I expected the intensity of effects to increase with decreasing
natural habitat.

METHODS
Study reservoirs

Consumer's Energy Company created six reservoirs on the Au Sable River
(Figure 1) between 1911 and 1924, four of which were included in this study. In general,
the reservoirs are mesotrophic in nature with short retention times ranging from 1 to 12

days. Two of the reservoirs are operated to simulate run-of-river ﬂow (Alcona and F oote

ponds), while the other two (Loud and Cooke ponds) are peaking operations with daily
water level ﬂuctuations reaching 0.49 and 0.30 m, respectively (Table 1, Kinney et al.
1999). Turbidities are moderate, and decrease from upstream to downstream reservoirs.
The reservoirs contain widespread littoral habitat supporting a moderate amount of
vegetation, which is unstable in some due to daily water level ﬂuctuations.
Experimental approach

All ﬁeld sampling occurred during the summers of 2000 and 2001. The Northeast
Michigan Sportsman's Club placed habitat enhancement structures (half—logs and
AquaCribs®, described below) in Loud, Cooke, and F oote ponds in the fall of 1999. In
2001, I placed structures in Alcona Pond as well as additional structures in Loud Pond.
In Loud, Cooke, and Foote ponds I conducted within reservoir comparisons between
areas with and without structures, only after structure placement, whereas in Alcona Pond
I conducted within reservoir comparisons before and aﬁer structure placement (Table 2).
In all reservoirs, sites were chosen for treatment (structure) and reference (no structure)
pairs, to insure that within pairs, treatment and reference areas were as similar to each
other as possible and that the paired sites represented a variety of habitat conditions
throughout each reservoir. University and agency researchers chose sites for treatment
and reference pairs in Loud, Cooke, and Foote ponds from habitat maps that
distinguished vegetated and non-vegetated areas by dividing the shoreline into sections
and choosing study sites haphazardly. In Alcona Pond I divided the shoreline into 100m
sections and randomly assigned study sites. To ensure that structures in treatment areas
did not affect ﬁsh behavior in reference areas, the reference area was located a minimum

of 100m from the treatment area. Whenever possible, I paired one treatment area to one

reference area. When this was not possible, one reference area was shared between two
treatment areas. This occurred ﬁve times across all four reservoirs. Minimum distance
between two consecutive treatment/reference pairs within reservoirs was 100m, but
typically was at least 300m.

I arranged half-logs, a nesting and cover device constructed of a hardwood slab
(approximately 2.4 m in length, 25 cm in width, and 5 cm thick) and two to three two-
core masonry blocks (20 x 20 x 41cm), perpendicular to the shoreline 5m apart at the 1m
depth contour in groups ranging from 4 to 97. The Northeast Michigan Sportsman's Club
constructed the half-logs by bolting the masonry blocks to each end of one ﬂat side of the
hardwood slab, leaving 5-10cm of the slab overhanging at each end (Figure 2), similar to
the half-log described by Hoff (1991). The club also assembled and installed
AquaCribs® (Figure 2), a box-shaped shelter device (122 x 152 x 122cm) constructed of
corrugated plastic, in clusters varying in number from 3 to 10, suspended 1m above the
bottom over a depth interval of 2.7 to 5.5m. The numbers of half-log and AquaCrib®
sites varied among reservoirs (Table 2).

Natural habitat and Iimnological data

In order to evaluate the effectiveness of half-log structures across different
environmental contexts and compare the similarity of treatment and reference pairs, I
characterized the natural habitat at each treatment and reference pair using the method
described by Hunt (2000). A diver wearing a snorkel and mask made visual observations
of natural habitat across the entire length of the transect over the middle depth contour of
each area. The diver noted and recorded substrate composition (sand, clay, or gravel),

presence and size of coarse woody material, and presence and size of macrophyte

patches. 1 calculated percent cover of coarse woody material and macrophytes by
summing the length of the individual habitat patches along the transect and dividing by
the total length of the transect. Sampling took place once per site during early June of
2001 and early August of 2000 and 2001.

I collected limnological data biweekly from each of the four study reservoirs,
using a YSI model 55 dissolved oxygen meter to derive temperature/dissolved oxygen
proﬁles and a Secchi disk to gauge turbidity. I utilized HOBO® temperature loggers
(two per reservoir, one each at an upstream and downstream location, at 1m depth) to
monitor temperature over the duration of the ﬁeld season, and an integrated tube sampler
to collect water samples from the epilimnion (as determined by temperature/dissolved
oxygen proﬁles) for chlorophyll a, total phosphorous and total nitrogen analysis. All
samples were returned to the laboratory and processed using standard procedures
(Murphy and Riley 1962, Menzel and Corwin 1965, Nusch 1980, Crumpton et al. 1992).
Fish concentration and predator-prey interaction

To evaluate the effectiveness of half-logs in concentrating ﬁsh, I used counts
obtained by visual observation while snorkeling along transects (modiﬁed from Keast
and Harker 1977, Moring and Nicholson 1994, Brown et al. 2000). Divers wearing a
snorkel and mask conducted visual observations on the half-logs one day per week in
each reservoir during the ﬁrst two weeks of each month (J une-August). Six treatment
and reference pairs within the reservoir were randomly selected for observation on each
sampling date. Separate divers swam each treatment and reference area of a pair
simultaneously. At each half-log treatment area containing less than ten half-logs, the

divers conducted three parallel transects to estimate ﬁsh abundance over the entire area

where the half-logs were placed: the ﬁrst transect was parallel to shore at the nearshore
end of the half-logs (1m depth contour), the second at the middle of the half-logs, and the
third at the deep end. At sites containing ten or more half-logs, a random subsample of
ﬁve half-logs was chosen for observation. All nearshore transects began at the 1m depth
contour and proceeded in a downstream direction parallel to the shore, beginning 5m
before the ﬁrst half-log encountered and ending 5m aﬁer the last one. Upon completion
of the ﬁrst transect, the swimmer reversed direction and completed the second transect,
and then the third in an effort to minimize disturbance over the study area. Transects in
each reference area were of the same length and at the same depths as transects at the
corresponding treatment area. Divers marked transect starting and ending points on the
shoreline with surveyor's tape and used a removable buoy system to ensure that all
transect lengths and depths were equal between treatment and reference pairs. Divers
established transect lengths for each area on the ﬁrst sampling date that a site was
randomly chosen for observation. All ﬁsh observed at each transect were identiﬁed to
species and counted. 1 calculated relative abundance (ﬁsh/km) by averaging the three
transect counts at each area and then dividing by the length of the transect.

I also used nighttime electroﬁshing to evaluate the effectiveness of half-logs in
concentrating ﬁsh and their effects on ﬁsh foraging. Field personnel conducted one
transect in the downstream direction with a boat-mounted electroﬁsher (7 amps pulsed
D.C., 120hz) at the 1m depth contour at each of six randomly chosen treatment and
reference pairs once per month per reservoir during the months of June, July, and August.
All ﬁsh captured were identiﬁed and measured to provide estimates of relative abundance

(ﬁsh/km) by species, and all smallmouth bass and largemouth bass were weighed. Field

personnel used gastric lavage (12 volt electric water pump, 60gal/hr) to collect diets from
all smallmouth and largemouth bass greater than 150mm TL. Large diet items such as
crayﬁsh and large ﬁsh that were not completely extracted by gastric lavage were removed
by hand or with the aid of forceps. Aﬁer funneling all smallmouth and largemouth bass
diets into 10.16 x 15.24cm cloth bags, ﬁeld personnel preserved the diets in 95% ethanol.
Smallmouth and largemouth bass less than 150mm total length were sacriﬁced and
preserved in 95% ethanol for stomach dissection.

In order to quantify possible forage items associated with half-logs, ﬁeld
personnel collected macroinvertebrates colonizing the half-log structures from ﬁve half-
logs chosen randomly from each treatment site in Cooke and F oote ponds in August of
2001 by use of an air-vacuum apparatus. The samples were stored in 95% ethanol and
returned to the laboratory for sorting, identiﬁcation, and enumeration. Laboratory
personnel added Rose Bengal to each sample to facilitate sorting, and then identiﬁed all
invertebrates to order. Dry weights were estimated using regressions (see Diet analysis
below). Due to the short amount of time that elapsed between invertebrate sampling and
when the half-logs had been installed in Alcona and Loud ponds, I did not collect
invertebrate data in these two reservoirs.

Visual observation and electroﬁshing were not possible as sampling methods for
AquaCribs® due to the lower light levels at the depths where AquaCrib® were placed.
Accordingly, I employed gill nets (15.24m length, 1.83m depth, 4-panel experimental
mesh, 1.3cm, 1.9cm, 2.5cm, and 3.8cm mesh size) to evaluate the effectiveness of
AquaCribs® in concentrating ﬁsh. I conducted netting once per week at each reservoir

during monthly two-week periods in June, July, and August of 2000 and 2001, using

short sets at two randomly selected treatment and reference pairs in an effort to minimize
catch mortality. In 2000, I set the nets at the depth contour past the last visible cribs (4.6
to 5.5m) in mid-morning and removed them in the late aﬁemoon. Few ﬁsh were caught
during this time period. Therefore, in 2001 I set the nets in late aﬁernoon and removed
them the next morning. All ﬁsh captured were measured and recorded. I made no
comparison of diets between treatment and control areas, as insufﬁcient numbers of
smallmouth bass and largemouth bass were captured to permit such comparisons.
Diet analysis

Laboratory personnel identiﬁed diet items to species for ﬁsh and to order for
invertebrates and crayﬁsh. When necessary, further distinctions were made below order
for invertebrates to provide more accurate dry weight estimates. In order to determine
dry weights using regressions (G. G. Mittelbach, Michigan State University, Kellogg
Biological Station, unpublished data, Wilson and Samelle 2002) laboratory personnel
measured head capsule width for most invertebrates, carapace length for crayﬁsh, total
length for zooplankton, and standard length for ﬁsh using a Numonics GraphicMasterIITM
digitizer tablet connected to a Nikon SMZ-U stereo microscope (1:10 zoom) magniﬁed at
0.75x. Diet items too large to measure using the digitizer were measured to the nearest
millimeter with a ruler. I established ﬁve primary prey groups (crayﬁsh, ﬁsh,
macroinvertebrates, zooplankton, and total of all stomach contents) and determined diet
composition by prey group on a dry weight (g) per gram of ﬁsh basis.
Attraction of nesting fish

I used visual observation (modiﬁed from Hunt 2000) to evaluate the use of half-

logs by nesting smallmouth bass. Sampling took place two to three times per week during

10

May and early June in Cooke and F oote ponds in 2001. Divers wearing snorkel and
mask sampled each treatment and reference pair within the reservoir on each sampling
occasion. Divers swam transects over the middle depth contour at which the half-logs
were located in treatment areas and over the same depth contour in reference areas using
the aforementioned visual observation protocols. Each nest encountered within 2m of the
transect was assigned a unique number and marked using a piece of surveyor's tape
attached to a large metal washer. Divers recorded nest number, substrate, and
presence/absence of fry (metric of nest success).
Angler catch rates

I utilized standardized angling to evaluate the effectiveness of both half-logs and
AquaCribs® in concentrating ﬁsh for angler capture. Once per month per reservoir
during June, July, and August 2000 and once per week per reservoir during June, July,
and August 2001, two anglers ﬁshed six randomly selected treatment and reference pairs
for ten minutes each with randomly assigned artiﬁcial or live baits. The ﬁrst location to
ﬁsh within each pair (treatment or reference) was assigned randomly; individual baits and
lures were randomly assigned to each angler. Anglers measured, weighed, and recorded
all ﬁsh captured to provide estimates of relative abundance (catch/hour).
Statistical analyses

In order to facilitate data analysis and interpretation, I placed the most abundant
game ﬁsh species that were encountered during sampling into four groups: total
(comprised of black crappie Pomoxis nigromaculatis, bluegill Lepomis macrochirus,
largemouth bass, northern pike Esox lucius, pumpkinseed Lepomis gibbosus, rock bass

Amblopites rupestris, smallmouth bass, walleye Stizostedion vitreum, and yellow perch

ll

Percaﬂavescens), smallmouth bass, other piseivores (included largemouth bass, northern
pike, and walleye), and planktivores (consisted of black crappie, bluegill, pumpkinseed,
rock bass, and yellow perch). All data were analyzed using SAS version 8.0 (SAS
Institute Inc. 1999). When appropriate, I transformed the data to meet the necessary
distributional assumptions.

Within-reservoir comparisons. I used data collected from Cooke and Foote ponds
in 2000 and 2001 and data collected from Alcona and Loud ponds in 2001 for within-
reservoir comparisons between treatment and reference areas (Table 2). In order to
extrapolate results beyond the four reservoir systems studied, I used mixed-effect analysis
of variance (ANOVA, SAS Institute 1999). For each functional group, I used mixed-
effect AN OVA to determine if the mean difference in relative abundance derived from
visual observation and electroﬁshing (comparing treatment versus reference areas) varied
predictably as a function of natural habitat. I conducted a similar analysis to determine if
the mean difference in smallmouth bass nest density in treatment versus reference areas
varied predictably as a function of natural habitat. For each functional group, I used
mixed-effect ANOVA to determine if the mean difference in relative abundance derived
from angling and gill netting (comparing treatment versus reference areas) varied
predictably as a function of treatment. I used mixed-effect ANOVA again to determine if
the mean difference in prey dry weight (grams prey per gram of ﬁsh, treatment versus
reference) from gut contents of smallmouth bass and largemouth bass varied predictably
among reservoirs as a function of natural habitat. For all mixed-effect models, I treated
treatment and substrate as ﬁxed effects and reservoir and site (nested within reservoir) as

random effects. Percent cover of coarse woody material (CWM) and macrophytes were

12

treated as covariates. In order to make comparisons among the four reservoirs studied, I
used ﬁxed-effect AN OVA (SAS Institute 1999) to determine if the mean difference in
relative abundance (treatment versus reference) for each functional group derived from
visual observation, electroﬁshing, angling, and gill netting varied predictably among
reservoirs. Separate analyses were conducted for each functional group and sampling
technique combination. Reservoir and treatment were treated as ﬁxed effects, while site
(nested within reservoir) was treated as a blocking factor. Rejection criterion was set at
a=0.05 for both analyses.

Before and after comparisons. I used data collected from Alcona Pond in 2000
and 2001 for within-reservoir comparisons before and aﬁer structure placement. I used
mixed-effect ANOVA to determine if the mean difference in relative abundance derived
from visual observation and electroﬁshing (comparing treatment versus reference areas)
for each functional group varied according to year and treatment. Insufficient angling
data were collected to permit comparisons before and aﬁer structure placement. I treated
treatment and year as ﬁxed effects, and individual sites as a random effect. I also used
mixed-effect AN OVA to determine if the mean difference in prey biomass (treatment
versus reference) from gut contents of smallmouth bass collected from electroﬁshing
varied according to year and treatment (low catch of largemouth bass prevented analysis).
I treated treatment and year as ﬁxed effects, and individual sites as a random effect.

Rejection criterion was set at a=0.05 for all analyses.

13

RESULTS

Natural habitat and limnology

Substrate composition and macrophyte and CWM cover at study sites varied
among reservoirs. No clear pattern was evident in substrate composition along a gradient
from upstream to downstream reservoirs. The predominant substrate type in the
reservoirs (Figure 3a) was sand (Alcona, Cooke, and F oote ponds) or gravel (Loud Pond).
Clay comprised greater than 30% of total substrate cover in only one reservoir (Cooke
Pond). Macrophyte cover appeared to increase along the upstream-downstream gradient,
with cover less than 20% at sites in the two upstream reservoirs, and cover greater than
60% at sites in the two downstream reservoirs (Figure 3b). CWM cover displayed a
contrasting pattern. CWM cover was highest (~8%) at sites in Alcona Pond, the most
upstream reservoir, and unifome lower (less than 3%) at sites in the three downstream
reservoirs (Figure 3c).

Mean summer epilimnetic temperature was similar among reservoirs (Table 3).
All reservoirs, with the exception of Loud Pond in 2000, displayed thermal stratiﬁcation
by late summer. Chlorophyll a concentration was similar across all study reservoirs,
ranging from ~2.5 ug/L in Loud Pond to ~4 ug/L in Alcona Pond (Table 3). Total
phosphorous concentration ranged from 14.1 ug/L in Foote Pond to 24.3 ug/L in Loud
Pond and displayed no clear pattern along a gradient from upstream to downstream
reservoirs (Table 3). Conversely, total nitrogen concentration increased along a gradient
from upstream to downstream reservoirs, ranging from 192.5 ug/L in Alcona Pond to

524.8 ug/L in Foote Pond (Table 3).

14

Fish concentration

Half-logs within and among reservoirs. Presence of half-logs had a signiﬁcant
positive effect on smallmouth bass abundance as indicated by the mixed-effect models
for both visual observation and electroﬁshing (Table 4, Figure 4). Although no other
signiﬁcant effects of half-logs on CPE were documented, in most cases (with the
exception of piseivore CPE from visual observation samples), point estimates of mean
CPE were higher in treatment areas compared to reference areas (Figure 4). CPE of all
functional groups was higher for electroﬁshing than visual observation, although patterns
of relative abundance among functional groups were similar between the two sampling
methods (Figure 4).

Two functional groups, piseivores and smallmouth bass, demonstrated signiﬁcant
variation with natural habitat (Table 4), although the direction of response to habitat
differed among habitat types and between functional groups. Piscivore CPE from visual
observation varied with both macrophyte and CWM cover. As macrophyte cover
increased, piseivore CPE decreased. Conversely, as CWM cover increased, piseivore
CPE increased. Similarly, smallmouth bass CPE, both from visual observation and
electroﬁshing, increased with CWM cover. In addition, electroﬁshing data indicate the
presence of a signiﬁcant treatment*macrophyte cover interaction for smallmouth bass
CPE. Smallmouth bass CPE at treatment areas containing half-logs decreased with
increasing macrophyte cover, while the opposite trend was present at reference areas.
Hence, the largest difference in smallmouth bass CPE between treatment and reference

areas occurred at sites with low macrophyte cover.

15

All ﬁmctional groups, with the exception of piseivores from visual observation
data, demonstrated signiﬁcant variation among reservoirs (Table 5a and b) as indicated
by the ﬁxed-effect models. In general, CPE from visual observation for all functional
groups increased from upstream to downstream reservoirs. Total, planktivore, and
smallmouth bass visual observation CPE appeared highest in Foote Pond, followed by
Cooke, Alcona, and Loud ponds. Total and planktivore electroﬁshing CPE demonstrated
no clear pattern with respect to the upstream/downstream gradient, although the highest
CPE for both functional groups occurred in Foote Pond. In contrast, piseivore and
smallmouth bass electroﬁshing CPE decreased from upstream to downstream reservoirs.
CPE of these two functional groups was highest in Alcona Pond and decreased in
subsequent downstream reservoirs. Site, as well as the interaction between reservoir and
site, served as signiﬁcant sources of variation within functional groups for both methods
(Table 5a and b), reﬂective of within-reservoir variability.

Treatment did not appear to be a signiﬁcant source of variation in CPE (with the
exception of smallmouth bass) for visual observation. However, for electroﬁshing, total
and piseivore CPE were higher at treatment areas containing half-logs compared to
reference areas in most instances (Figure 5). Among the four reservoirs, CPE at
treatment and reference areas was similar (with the exception of piseivore CPE), as no
signiﬁcant reservoir*treatment interaction occurred for either visual observation or
electroﬁshing (Table 5a and b). Piscivore electroﬁshing CPE displayed signiﬁcant
variability among reservoirs, site, and treatment, as demonstrated by the presence of

signiﬁcant site"treatment and reservoir*site*treatment interactions (Table 5b).

16

Half-logs before/after placement. In Alcona Pond, no signiﬁcant treatrnent*year
interaction occurred for any of the four functional groups (Table 6), indicating that half-
log structures did not signiﬁcantly concentrate ﬁsh after their placement (Figure 6). Year
presented a signiﬁcant source of variability, with piseivore CPE signiﬁcantly higher in
2001 compared to 2000 for visual observation, but total and planktivore CPE
signiﬁcantly higher in 2001 compared to 2000 for electroﬁshing (Table 6). Treatment
was signiﬁcant for smallmouth bass sampled by electroﬁshing across both years (Table
6), with higher CPE in treatment areas containing half-logs compared to reference areas
without half-logs (Figure 6) both before and after structure placement.

AquaCribs® within and among reservoirs. Presence of AquaCribs® had no
signiﬁcant effect on the abundance of any of the four functional groups as represented by
both the mixed-effect (Table 4) and ﬁxed-effect models (Table 5c, d). Reservoir was a
signiﬁcant source of variability for planktivore CPE in gill nets, and a reservoir*site
interaction was present for total and planktivore gill net CPE, further reﬂecting within
and among reservoir variability (Table 5c). Planktivore CPE in gill nets was nearly ten
times higher in F oote Pond (0.29 ﬁsh/hour) compared to Cooke Pond (0.03 ﬁsh/hour).
Signiﬁcant within-reservoir variability Was present for smallmouth bass CPE in gill nets
(Table 5c). For piseivores, the effect of treatment on CPE varied among reservoirs as
indicated by a signiﬁcant reservoir*treatment interaction (Table So). In Cooke Pond the
presence of AquaCribs® had a positive effect on piseivore CPE (0.03 ﬁsh/hour at
treatment areas compared to 0.01 ﬁsh/hour at reference areas). Conversely, in Foote
Pond the presence of AquaCribs® had a negative effect on piseivore CPE (0.02 ﬁsh/hour

at treatment areas compared to 0.05 ﬁsh/hour at reference areas).

17

Invertebrate biomass and diet analysis

Invertebrate samples from half-logs in Cooke and Foote ponds were dominated by
zebra mussels (Dreissena polymorpha), which constituted over 99% of the total
invertebrate biomass present. Amphipods, diptera, ephemeroptera, and trichoptera
species made up less than 1% of the total invertebrate biomass present. Excluding zebra
mussels, amphipods comprised 64% of the total biomass of the four other invertebrate
orders, followed by trichoptera (24%), diptera (6%), and ephemeroptera (5%).

Total prey dry weight (per gram of ﬁsh) in smallmouth bass and largemouth bass
diets did not vary signiﬁcantly between treatment and reference areas, or with natural
habitat, within any of the four reservoirs. For smallmouth bass, invertebrates were the
most abundant diet item («48% of prey weight per gram of ﬁsh), followed by ﬁsh
(~35%), crayﬁsh (~12%), and zooplankton (~5%, Figure 7a). For largemouth bass, ﬁsh
were the most abundant prey item, comprising nearly 70% of the total prey dry weight
per gram of ﬁsh. Invertebrates were the second most abundant prey item for largemouth
bass, constituting over 20% of prey weight, while crayﬁsh were the third most abundant
prey item (Figure 7b). Zooplankton were not found in any of the largemouth bass
stomachs for within-reservoir analysis. In the before/after study, total prey dry weight
per gram of smallmouth bass did not differ signiﬁcantly between treatment and reference
areas aﬁer structure placement in Alcona Pond. Invertebrates were the most abundant
prey item in both years (~45-55% by year), followed by ﬁsh (~35-40% by year), crayﬁsh

(~10-20% by year), and zooplankton (less than 10% by year, Figure 7c, d).

18

Attraction of nesting ﬁsh

Smallmouth bass nest density and nest success (Figure 8) were signiﬁcantly
higher in treatment areas with half-logs compared to reference areas (Table 7). In
addition to treatment, substrate and macrophyte cover were also important factors related
to smallmouth bass nest density and nest success. Mean nest success was highest in areas
with gravel substrate (Figure 9), and mean nest density decreased with increasing percent
macrophyte cover (Figure 10). A signiﬁcant treatment*macrophyte cover interaction was
present for mean smalhnouth bass nest density and nest success (Figure 10). Mean nest
density and nest success were higher at treatment areas compared to reference areas at
low macrophyte cover, but this pattern became less noticeable as macrophyte cover
increased.
Angler catch rates

AquaCribs ® and half-logs. Presence of half-logs or AquaCribs® had no
inﬂuence on angler catch rates for any of the four functional groups. Mixed-effect (Table
4) and ﬁxed-effect (Table 5d) models of mean angling CPE data by functional group at
AquaCribs® and half-logs indicate no signiﬁcant treatment effects for either structure
type. Reservoir was a signiﬁcant source of variability for several functional groups as
indicated from the ﬁxed-effect models (Table 5d). Total angler CPE when ﬁshing at
half-log sites (across treatment and reference areas) was highest in Cooke Pond (0.57
ﬁsh/hour), followed by F oote Pond (0.50 ﬁsh/hour), Loud Pond (0.44 ﬁsh/hour), and
Alcona Pond (0.24 ﬁsh/hour). Angler CPE of smallmouth bass at AquaCrib® sites
(across treatment and reference areas) was higher in Foote Pond (0.06 ﬁsh/hour)

compared to Cooke Pond (0.002 ﬁsh/hour). Site was a source of variability for total and

19

planktivore angler CPE when ﬁshing at both half-log and AquaCribs®, reﬂective of
within-reservoir variability for CPE of both functional groups. Additionally,
reservoir“ site and site*treatrnent interactions existed for smallmouth bass CPE at
AquaCribs®, representing further within-reservoir variability in CPE as well as the
inﬂuence of site-speciﬁc characteristics on treatment effectiveness.

DISCUSSION

I speciﬁcally sought to address the uncertainty surrounding where to add
structures within a system, paying particular respect to the effects of pre-existing natural
habitat conditions on structure use, so that future habitat enhancement efforts are most
beneﬁcial and cost-effective. To do so, I used two statistical models: mixed-effect and
ﬁxed-effect ANOVA. In the mixed-effect models, reservoirs and sites within reservoirs
were treated as random effects so that the results of the analyses could be extrapolated
beyond the four reservoir systems I studied, thus emphasizing a general understanding of
how natural habitat inﬂuences the effects of habitat enhancement structure. However, I
also felt that useful information could be gained by treating the four reservoirs as unique
entities with inherently different characteristics. Hence, I also used ﬁxed-effect AN OVA
models to draw inference speciﬁcally among these four reservoirs and their unique
habitat conditions.

Both statistical models reﬂected similar results, indicating that addition of habitat
enhancement structures had little effect on the four functional groups of ﬁsh I studied.
Each model also demonstrated the importance of pre-existing natural habitat conditions.
Habitat variables such as macrophyte cover, CWM cover, and substrate within study sites

were an important source of variability in CPE for certain functional groups as reﬂected

20

by the mixed-effect models. Accordingly, individual study sites, as well as reservoirs,
each varying in their speciﬁc habitat characteristics, were an important source of
variability in CPE in the ﬁxed-effect models. In other words, natural habitat appeared to
inﬂuence CPE more so than did the presence of habitat enhancement structures. Thus,
before any habitat enhancement project is undertaken, the characteristics of the individual
system of interest, as well as speciﬁc locations within systems, must be taken into
account before structures are placed to ensure their necessity and/or effectiveness.

Using a variety of sampling methods, numerous studies have found that habitat
enhancement structure in ponds, lakes, and reservoirs concentrate ﬁsh (Brouha 1974,
Reeves et al. 1978, Prince and Maughan 1979a, Prince and Maughan 1979b, Pierce and
Hooper 1980, Smith et al. 1981, Aadland 1982, Kayle 1986, Moring et al. 1989, Mabbott
1991, Moring and Nicholson 1994). In particular, several studies have indicated that
black bass (Micropterus sp.) demonstrate an afﬁnity towards structure (Reeves et al.
1978, Prince and Maughan 1979a, Prince and Maughan 1979b, Pierce and Hooper 1980,
Smith et al. 1981, Dufour 1989, Rogers and Bergersen 1999). I found similar results for
smallmouth bass in my study; however, contrary to the results of past research, the two
structure types I studied in four reservoirs had little apparent effect at concentrating other
species. I did ﬁnd signiﬁcant variability in CPE among reservoirs, supporting the notion
that characteristics inherent to the system of study may inﬂuence ﬁsh assemblages more
so than the presence of habitat enhancement structure. While the earliest work conducted
in lentic systems indicated structure effectiveness is inﬂuenced by structure type and
species present (Hubbs and Eschmeyer 193 7, Rodeheffer 1939, Rodeheffer 1945), more

recent research has found variability in structure effectiveness due to inherent

21

characteristics of the system under study. For example, the amount of alternative
complex physical habitat available negatively affected structure use by ﬁsh in research
conducted by Mitzner (1981) and Rogers and Bergersen (1999).

Few studies have examined the effects of structure on ﬁsh diets and growth;
therefore, the amount of information available in the literature to support or refute claims
of increased foraging success and growth is limited. Prince et al. (1976) and Aadland
(1982) found increased periphyton growth on reefs but did not document foraging
patterns or diets of ﬁsh. Pardue and Nielsen (1979) found similar results on wooden
substrates, but no increase in the production of tilapia (Tilapia aurea and T ilapia
mossambica) or bluegill biomass. Dufour (1989) found a correlation between
invertebrates present in smallmouth bass and rock bass stomach contents and
invertebrates colonizing half-logs in a river, while Wege and Anderson (1979) found
greater largemouth bass growth in ponds with structure. Due to the relatively small
amount of structure added per surface acre of reservoir in my study, I made no attempt to
quantify change in growth of any of the species encountered. While data collected from
half-logs indicate that some invertebrates do indeed colonize the structures, I found no
signiﬁcant difference in dry weight of gut contents of largemouth bass and smallmouth
bass between treatment and reference areas. My sampling protocols assumed that ﬁsh
would be actively feeding in the treatment and reference areas during sampling, an
assumption that may not be valid. Although treatment and reference areas were sampled
within minutes of each other, ﬁsh were free to move about each area before sampling.
Accordingly, ﬁsh that were foraging in the vicinity of structures may have moved away

from the sampling area, while ﬁsh that were foraging outside of the vicinity of structures

22

may have moved into the sampling area. Thus, detecting any change in ﬁsh diets would
be difficult.

In addition to congregation at structures throughout the summer, other studies
have demonstrated that black bass display a preference towards structure during the
nesting season. Vogele and Rainwater (1975) found higher numbers of spawning spotted
bass (Micropterus punctulatus) and largemouth bass around brush shelters in an Arkansas
reservoir, but noted that smallmouth bass showed no preference towards structure. In
contrast, Coble (1975) and Miller (1975) noted that smallmouth bass preferred overhead
cover throughout all life stages, including nesting. Accordingly, Hoff (1991) found
dramatic increases in smallmouth bass nest density, nest success, and juvenile abundance
in northern Wisconsin lakes after treatment with half-logs. Patrick (1996) found similar
results for smallmouth bass nest density and nest success in a Tennessee reservoir. Hunt
(2000) found that spawning largemouth bass were attracted to half-log structures, but that
the degree to which this occurred was dependent upon the amount of complex physical
habitat nearby. My results are consistent with those of Hoff (1991), Patrick (1996) and
Hunt (2000). Half-log habitat enhancement structures signiﬁcantly increased smallmouth
bass nest density and nest success in comparison to areas without structure. But, natural
habitat conditions were an important factor in determining nest density and nest success
in areas with structure. As Hunt (2000) discovered for largemouth bass, the amount of
available physical habitat played a key role in determining the nesting habits of
smallmouth bass near half-logs.

Standardized angling has been a widely used method for evaluating the

effectiveness of structure in attracting ﬁsh and increasing angler catch rates (Petit 1972,

23

Brouha 1974, Wilbur 1978, Paxton and Stevenson 1979, Wege and Anderson 1979,
Rogers and Bergersen 1999). Several studies have demonstrated that angler catch rates
are higher in areas with structure compared to areas without structure (Petit 1972, Wilbur
1978, Paxton and Stevenson 1979, Wege and Anderson 1979, Rogers and Bergersen
1999), and catch rates, as similar to concentration in general, are dependent upon
structure type and species present (Paxton and Stevenson 1979, Rogers and Bergersen
1999). In my study, angler catch rates for all functional groups were similar between
areas with and without structure, ﬁrrther supporting the ineffectiveness of structure in
concentrating ﬁsh in my systems of study.

Several studies have noted the effectiveness of different structure types used in
habitat enhancement projects. Hubbs and Eschmeyer (193 7) were the ﬁrst to realize that
different structure types vary in their effectiveness at concentrating ﬁsh. Brouha (1974)
noted that centrarchid species displayed a preference towards reefs constructed of trees
compared to reefs constructed of tires. Similarly, Rogers and Bergersen (1999) noted
that largemouth bass displayed preferences towards varying types of synthetic structures.
Although I did not directly compare the effectiveness of half-logs and AquaCribs®, I feel
that it is important to note that half-logs, the simpler of the two structures I studied
appeared more effective, outperforming the more complex and expensive AquaCribs®.
In addition to being considerably more inexpensive than the AquaCribs®, half-logs are
comprised of materials readily available for purchase at any hardware store or
lumberyard and are simple and efﬁcient to assemble.

In summary, I suspect that the abundance of natural habitat present in the

reservoirs that I studied inﬂuenced the effectiveness of the habitat enhancement

24

structures in concentrating ﬁsh. In each reservoir, abundant natural habitat is present in
the form of standing timber, CWM, and macrophytes. Accordingly, when considering
my study reservoirs as a whole, the amount of habitat enhancement structure added
(compared to the available natural habitat present) is relatively small. I believe that this
contributed to the paucity of signiﬁcant structure effects in my study. In other words, the
amount of structure added (compared to existing natural structure) is insufficient to
attract most ﬁsh and increase concentration. Therefore, ﬁsheries managers must
carefully consider the system as a whole when undertaking any habitat enhancement
effort. Managers can plan ahead and improve the likelihood of success of habitat
enhancement projects by considering the amount of natural habitat available within a
particular system. Additionally, managers may also consider structure placement with
regards to the species present within the system and their life history and habitat needs to
ensure greater chances of success. Smallmouth bass, a species that has demonstrated an
affinity towards overhead cover (Coble 1975, Miller 1975) and has beneﬁted from
structure in the past (Hoff 1991, Patrick 1996), responded positively in my study as well.
Through consideration of such species-speciﬁc habitat requirements, as well as available
natural habitat and cost-effectiveness of various structure types, ﬁsheries managers can

plan beneﬁcial and economical habitat enhancement efforts.

25

APPENDIX

Tables and ﬁgures

26

 

 

 

 

 

and v. 5 mm...” ho Eon mum 300“.

and Ya Ed 0. _. Sod 9.5 3.000

mvd md mud mﬁmd Sm .50..

o 3 Swansea? Nata...” Ev 2.82
E c _ a E Elwin can. :5 3.5
.33 08:. “5:30.300 :33 3.3

.225 53:23. 05.22% .xaEEuo! moat—5 «co—.5535.

.cozmcooo

Emu 058.093; So: 922mm. .92 .395 5 gauge 23v £25me 9: m_ 5:363: .05. Loan; axe.
5.326 m .2 o... be m:_m> 83.55:. m 53> 58:3 9.x. 9: Co 35 9 .960 mean _o 286 m be 8:998:28
m5 9 65.205 9: Co £95. 9: *0 one. 9: mm commoaxo m_ can 95205 65 no 35:62: 9: 39.8%
.203 Eoan_o>ou 8:905 .Emobmcsov 3 E858: Eot 3.38 So meozomom stem 25 new .88?
._m “o >35: 838w “mono“. magnum page: 05 .co_mm_EEoo boﬁSmom Steam. .903“. as. (Ema—coo Logan.
mLosamcoo Ea: Emu co comma 2626mm: >36 52m £28 :< 9: *0 83:28.20 3295. .F min...

27

.Foom coEEam 25m 5 noon... 358:5 38:63. .82. :3 E cocoa $5825 oEowN

 

 

.mﬁbmcm
>_co norm .8 new: 25 room EB.— mamo .m_m>_mcm cocﬂocouon .8 new: room can ooom Ea: Ema.
mm m o o «36 £5 .0 .02
m 2 : 2 82m 8.. co .oz
mm? 83 NSow new 32. Bow «:25an 2325» he as;
F2:0 .22 new
2.5 .23 2:0 85. 2:0 .2? cornuocohom «3222 new anion
Soon. 9.30 25.. ucoo_< c.0333.

 

.EoEoomE 05826 coco comcmano
Eager—LE“? $5065 :25 .93... .EoEooma 932:6 coco can 983 comtmano .6332
-553, $3065 Lot—macaw? c6239 can we? Co .383: new :93“. .3585an .N oBmh

28

 

 

 

Sn as 3.8 E: 00 3.2. gm 38..

8a .1... 0.48 0.0 o.” 00.2. sum 9.80

05 0.. 0.80 0.8 0.0 0.0-0.0 0.0 08..

3m 2 0.8. E. 3 0.0.0.... 40m 88....

.E. 10.... 3.9.: 33......
32.83 502. couegz 0:22.002... 3.9: .3035 .0...

0.2.8080 .880 .88 .50» a __......o.o_..0 82820 23.28:...» 8858....
Sou 0:0 ooou

.0 .0003. £000.... 0...... 0. 000.0. 00.00.00 80.. m0:_0> .0863 :00... 0.0 0000...: 06. 0:0 030.9309... .06.
.0 __>cao.o_..o .00....00 0....00 05000 =0 000.00 0020000 m_0>0_ 00908 0020020 .0 09.0. 0.... 00.00050.
c0908 0020005 .38 0:0 ooom .0 .0303. 5000.5 0:2. .2300 ES 0.30.0080. 0:050:30 .0EE30 :00...
0:. $0000.00. 0.20.0060... 200.3560 0. E0050: Ea... 00.005 0.0 2.3.000”. $030 0.5 0:0 .809

._0 .0 >055: 8.20m 50.0.... 00.05 00.2.: 05 60.00.2550 b20303. 3.00m. 0.000“. 0... $000800 .026n.
9.0.5080 Ea... 0.00 :0 00000 0.6200. .020 .0>_m 0.00m :< 0... .0 00000089050 0050.008... .m 0.00..

29

 

.0>0o

 

0.>200.00.2 000m
0:05.003 02 03.0.0". 02 «08.0". 5002......20
0.52 02 02 02 02 0.3.00...
0.52 02 02 02 02 0.3.2.20... mmuuz
082 02 02 02 02 .53 022.020.0020
000m
0.52 02 008.0... 02 0.08.0... ..2.02...0E0
0.52 02 003.0"... .800". 02 0.2.8... 00muz
0.52 02 02 02 02 0.02.2.0... 200020000
0.52 02 02 02 02 .53 .555
.5. .2...
020000.82. .300 .050 020.0 00505.
2005205 0.2.0000 5.50 03200.00: «222.00.... 320302.... 02:02.5

 

.amﬁz. 000.0035 .0 05.9.0 .o .0. WE 002...... .0 05.9.0

.ﬁmuz. @00..00:0< .0 02.002 _._0 .0. 0200. 0.02. 0.00:0 .20E.00.. .200..._20_0 02 0:005:90 32an 00.02.00
000.0 .0 .0025: _0.0. 00. 0. 0.0.0. 2 00.0.0000 .2080020020 .0000: 00....00 .0 0020000 .0 0020005 0...

0. 0.0.0. .20....00... 000.0 .020_.02:. >0 20t0..00-00.00 20 0.0..0000 020 ..0>00 2230. .0..0.02. >000; 00.000
..0>00 0.>..00.002. ..202..00.. .0 0.00:0 0... 05.0002. 0020_.0> .0 0_0>_020 .00..0-00x.2. 2.0.. 00:_0>-n. .v 0.00...

30

 

 

.cmgmerogw 000m
0.0002000”. 02 02 2.08.0". 5:05.350
02 02 02 02 0.3.8...
02 02 0:00". 02 0.25.50... 3.2
02 02 0.8.0". 02 .53 3205.0...
800
02 02 02 02 5.55.350
02 02 02 02 0.2.8...
02 02 88.0". 008.0". 0.25.50... «:2
02 02 :80". 008.0". .53 002...... 05.05. .0
800
02 02 83.0". 02 535.350
.20E.00.......0>.000m 02 02 02 0.0200...
5.0102000”. 02 02 .080". 0.0.5.50... .02
0.0002000”. 02 02 02 .53 3205...... 05202 .._0 .0
0000
0.0002000”. 02 0.8.0... 88.0... 5:05.350
”—208.00...
cm._wc._o>..m0mm
..c0§00...0._0 88.0". 008.0... 58...". 0.3.8...
02 02 88.0". 38.9.. 225.50... mmuuz
02 83.0.. 0.8.0.... .0802. .53 002.3... 0250205020 .0
000m
5.0.3.2000”. 85.0". 008?. 300.9. 5:05.350
0.0132000”. 02 02 02 0.2.00...
0.0102000”. 02 500.0". 38.9.. 0.35.50... 000.2 50.35030
0.0102000”. 02 58.9.. 88.9.. .53 000.23.. 3:05 .0
5.02055... 0020 0.5 .00.. ..I....:..
«2005290 2025023 3.0 20.2000... 320302.... 0.0.02.0 02:02.00

 

22005290 .02u02 20.00. 020.020 0 00 00.00.. 0.03 00.001053502002202. 000.0 .0202020. >0 to..0..00
20.00 20 .2025005 020 ..0>.000. .0 0.0050 0... 05.0002. 0020..0> .0 0_0>.020 .00..0.00x.. 2.0.. 00:.0>-n. .0 0.00 ...

31

Table 6. P-values from mixed-effect analysis of variance modeling
the effects of year, treatment, and year*treatment interaction on
catch-per-effort by functional group in Alcona Pond. Treatment
refers to the presence or absence of half-log habitat enhancement
structures. N refers to the total number of areas sampled. NS=not

 

 

siiniﬁcant.
Sampling Functional Year Treatment Year'
Method Group Treatment
19121151120.—
Vlsual Total NS NS NS
Observation Planktivore p=0.0009 NS NS
N=92 Piscivore NS NS NS
Smallmouth NS NS NS
Bass
Electroﬂshlng Total p=0.0043 NS NS
N=58 Planktivore NS NS NS
Piscivore p=0.0027 NS NS
Smallmouth NS p=0.0459 NS
Ease

 

 

32

 

.0>oo

 

02500.00...
000530.. 00.0.0". 02 02 0.00.0". 0000000 .002
.050
02.202005.
000530.. 00000". 02 0000.0". 0.000.... 2.0000 .002
3.. 3..
020300.82. .260 .300
2302520 03.3000 .230 022.0200... 50530... 0.50.2

 

..200.._20«_0 .o2uwz 00.0.0.2.0 .20E0020220

.0200; 0o..._0... .0 0020000 .0 82000.0 02. 0.0.0.0. E02500; .300 2. 200000 02.23000
0.220 0... 502000.... 00.9.20... 0.03 0.002 000.205 mm. .o _0.0. .0. 0000000 .002 020 3.0200
.002 0000 0.:0E__02.0 20 0.050000 020 ..0>00 .230. .020.0E >000; 00.000 ..0>00 0.500.002.
..20E.00.. .0 0.0000 02. 05.0002. 002020> .0 0.02020 .00..0.00x_2. 2.0.. 0029.0 .0 0.00 ..

33

 

ﬂloPond -10 FlveChannelePond

Loud Pond

Figure 1. Location of the Au Sable River, Michigan (top) and the four
study reservoirs (bottom).

34

Half-log, Side View
L l

Iii—Di IQEI

 

Half-log, Top View

 

AquaCrib®

 

Figure 2. Side and top views of a half-log structure (top) and three-
dimensional view of an AquaCrib® structure (bottom).

35

{r- Gravel 1
i D Clay

a. Substrate

   

 

b. Macrophytes

Percent Cover
0)
O

-:’/// 41;» /
7' ’42:? y”,
///

 

 

T

15 - c. Coarse Woody Material

 

 

Alcona Loud Cooke Foote

Figure 3. Percent cover of habitat variables in each of the four study
reservoirs. Reservoirs are presented from upstream (Alcona) to
downstream (Foote). Substrate (a) is the total cover (%) of each of three
substrate classes summed across all study sites. Macrophytes (b) and
coarse woody material (c) are mean cover (%, i 1 SE) averaged across
all study sites. Note y-axes differ among substrate, macrophytes, and
coarse woody material.

36

0000.22 0220.200 020 0020.0 322.022..

0202.0 8%.... 0020-.. 0.0 Z .008-.. .00... 0... 0. 020000.200 $5 0000 0.208.380 020 80200.0 ”00.0-.. to.

0.... 0. 020000.200 mm”. 0.03.220... 020 .08... A8020... 02.00.50.200... 020 .00.. 0228.820 .0. 000.0 002020.02
020 2.0.2.8.. .0. 0020.0 00... 3200022.. .20.. 00. .0 :0 $00 H £0.20... MED. tob0..00-00.00 200.). .v 0.2.0.0.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

000.0
0000 5:05.350 00.60.. 0.0.5.5.. 3.0.
0 0
00. . 80
CV v I 8* O
.0...
. 000 o
00 . ﬂ.
.0
. com m
0 0 m...”
- o
N. 00 M
0 - 3 in...
. w
0 . - 00 W
LF I O” M
0. ..
2 - 00.
0.. 00.00.00 D 00.320000 3005 . 00.

 

 

.20....00... I

 

37

0000.02.
02:02.00 020 0000.0 3200022. 020.20 0....0 000.0.» 0.02 A8000. 800002300 0. 320000. 2.00.00:

.20.. 00.20020 0.0 0..0>.000m .900... 0220000020 020 0.0.. 0208.820 .0. ..0>.000. 0000 2. 000.0 3280.0.
020 .20....000 .0. 0000.0 :0... .0200020. .20. 0.0 .0 20 $3 H .2050... to..0..00-..0.00 200$. .m 0.00.0

 

 

002000”.
OwOOKL 0x000 QCOO_< OuOOu— OJOOU “.30...— NC8_<
. O . + . j p . r O
i if.
. .
. 8 1 o
mmﬂm SEOE=OEW . GNP

0000 £:OE_._0Ew . m.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. . . ﬂ .
O
.01 01 . m
. .. 0
3 . m.
.8 .0 m
08200.0 00 80200.0 0.
. . . . . . u"
r r L Jl I. I 0
El . . u
. cow .9. w
.80 H8. W4
0.030.520 . 0.020.090 .00. (w\
recur . .
. 0 . . . I1 I . .
if. .03 [a ”00
. .8.
TS” r
.80... . .30... .00..
00:20.0«D .80. -

 

 

52500... mc_:0...o..00_m 50020000 .0005

 

38

£00508 mam—9:00 0:0 095% 35:05: mecca 8&6 00x0-» 802 .Som E 00000—30

0003 3080003 0:50:06 00.00 0000 68m E 00000200 0003 3080003 00300.50 0.8.002 S0Q .0003 0:00?

5 3060003 0030300 00¢0 000 0.8009 Gamma wEﬁGobBB 0:0 30: mam—0x008 00m 000» 000 00000 000000.000
0:0 30.5005 .5.“ 098% :mm 10003003 .88 0E .3 :0 $3 H £2205 “08.00-003-580 5002 .o 0.5me

80>
080 Sou

000m 5:06:95

 

 

 

 

 

 

 

 

 

‘ L H O
H j‘: mu.-
7 O
. M
. d
9
I...
a
902005 W»
U.
. )
r H:
_ 3 q M H V s
. U.
. V
w
(
0.62255
8:0.80ma r00

 

 

055000.05 mcfmuobomm 50020000 .0005

 

39

00000003800 00000000300 00.0 000> 000 03003 >03 m0080 00 :03 00 00000003800

03000000000503 00.0 0000030 000 300% >003 w00000 00.00000 00x0-> 0002 000900000 000 0000 >030

:0 000000 00w000>0 00 0000003800 00000003 0002 .90 .8 00000003800 033000.000 800m 0000 0808:0080
000 3 .0v 00000003800 00000000000053» 800.0 000:0 0000 00008303 000 0000 00:08:08.0 00.0 80800000
>0 300% >003 000.0 000 .00 :0 0x03 H 80¢ .00 800w 003 >03 .00 08003 0000003800 00:0 0002 .0 00083

000.00.000N

0:80 >20 0:000 >000
.000 50090 2052002.. 80.5.88 0.00 5.0020 2052025

 

 

 

00000000”. D
000050000. 00

 

 

 

F L

.:

ovuz
000m 5008090.. .0 09

 

    

  

a
1
. ~

      

 

 

 

 

 

 

room 0000 5308:0005 .u .on

 

 

 

 

 

 

 

 

 

RFz o0.
mmuz 000m 000m 5=oE=0Em .0 003 5:08:95 .0
8300858 02003303 28.02.58 03:82.55?

"0000 00002 .n ”00:00 800“. 000 60.000 .003 .0000? .0

40

500
450
400
350
300
250
200
1 50
1 00

50

 

l l 1 PL 1 l 1 J

+——1

Nest Density (nests/ha)

l

 

 

 

 

 

100 ‘
90 ‘
80 ‘
7O ‘ T

50‘
40‘
30‘
20‘
1O1

 

Nest Success (%)
O)
O

 

d-

 

 

 

 

 

 

Treatment Reference

Figure 8. Mean smallmouth bass nest density (nests/ha, top) and nest
success (%, bottom), i 95% CI, in treatment and reference areas from
data collected in 2001 in Cooke and Foote ponds. Nest success was
deﬁned as the presence of swim-up fry. Note difference in y-axes
between nest density and nest success.

41

600

§

§

 

Nest Density (nests/ha)
§ §

 

 

 

 

 

 

 

100.
db
0
1?1ooJ
o\
v
8 so.
0 ..
0
0
3 60~
(I)
ﬂ
(0
0 401
Z
20.
o

 

 

 

 

 

Clay Gravel Sand

Figure 9. Mean smallmouth bass nest density (nests/ha, top) and nest
success (%, bottom), i 95% C1, by substrate from data collected in 2001
in Cooke and Foote ponds. Note difference in y-axes between nest
density and nest success.

42

 

1400

 

 

 

 

A A Treatment
‘5 0 R f

c 1200 --- e erence
E O

3 1000

5

g. 800- .

g 600-
8

u 40

(I)

(D
Z 20

 

 

Nest Success (%)

 

 

 

0 T n T j 4A.;
0 20 4o 60 80 100
Macrophyte Cover (%)

Figure 10. Relationship between macrophyte cover and smallmouth
bass nest density (nests/ha, top) and nest success (%, bottom) for
treatment and reference areas from data collected in 2001 in Cooke and
Foote ponds. Note difference in y-axes between nest density and nest
success.

43

LITERATURE CITED

44

LITERATURE CITED

Aadland, L. P. E. 1982. Artiﬁcial reefs as a management tool to improve sport fishing in
North Dakota reservoirs. Master’s Thesis. North Dakota State University, Fargo.

Bassett, C. E. 1994. Use and evaluation of ﬁsh habitat structures in lakes of the eastern
United States by the USDA Forest Service. Fifth International Conference on
Aquatic Habitat Enhancement, Long Beach, CA. Bulletin of Marine Science 55:
l 1 37-1 148.

Bohnsack, J. A. 1989. Are high densities of ﬁshes at artiﬁcial reefs the result of habitat
limitation or behavioral preference? Bulletin of Marine Science 44: 631-645.

Bohnsack, J. A. and D. L. Sutherland. 1985. Artiﬁcial reef research: a review with
recommendations for future priorities. Bulletin of Marine Science 37: 11-39.

Brouha, P. 1974. Evaluation of two reef materials and point or cove locations for
construction of artiﬁcial reefs in Smith Mountain Lake, Virginia. Master’s Thesis.
Virginia Polytechnic Institute and State University, Blacksburg.

Brown, A. B. 1986. Modifying reservoir ﬁsh habitat with artiﬁcial structures. Reservoir
ﬁsheries management: strategies for the 80's. G. E. Hall and M. J. Van Den
Avyle. Bethesda, American Fisheries Society, Southern Division, Reservoir
Committee: 98-102.

Brown, P. J., D. C. Josephson, and C. C. Krueger. 2000. Summer habitat use by
introduced smallmouth bass in an oligotrophic Adirondack lake. Journal of
Freshwater Ecology 15: 135-144.

Coble, D.W. 1975. Smallmouth bass. Black bass biology and management. R. H. Stroud
and H. Clepper. Washington DC, Sport Fishing Institute: 21-22.

Crumpton, W. G., T. M. Isenhart, and P. D. Mitchell. 1992. Nitrate and organic N
analyses with second-derivative spectroscopy. Limnology and Oceanography 37:
907-91 3.

DuF our, J. A. 1989. Evaluation of half-logs as habitat improvement structures for
smallmouth bass Micropterus dolomieui and rock bass Ambloplites rupestris.

Master's Thesis. Michigan State University, East Lansing.

Graham, R. J. 1992. Visually estimating ﬁsh density at artiﬁcial structures in Lake Anna,
Virginia. North American Journal of Fisheries Management 12: 204-212.

45

Hazzard, A. S. 193 7. Results of stream and lake improvement in Michigan. Transactions
of the 2nd North American Wildlife and Natural Resources Conference 1937:
620-624.

Helﬁnan, G. S. 1981. The advantage to ﬁshes of hovering in shade. Copeia: 392-400.

Helﬁnan, G. S. 1979. Fish attraction to ﬂoating objects in lakes. Response of ﬁsh to
habitat structure in standing water. R. A. Stein and D. L. Johnson, North Central
Division American Fisheries Society. Special Publication 6: 49-5 7.

Hoff, M. H. 1991. Effects of increased nesting cover on nesting and reproduction of
smallmouth bass in northern Wisconsin lakes. Proceedings of the First
International Smallmouth Bass Symposium: 39-43.

Hubbs, C. L. and R. W. Eschmeyer. 1937. Providing shelter, food, and spawning
facilities for game ﬁshes of our inland lakes. Transactions of the 2Ind North
American Wildlife and Natural Resources Conference: 613-619.

Hubbs, C. L. and R. W. Eschmeyer. 1938. The improvement of lakes for ﬁshing. A
method of ﬁsh management. Bulletin of the Institute for Fisheries Research 2: 1-
235.

Hunt, J. 2000. Enhancing largemouth bass spawning: behavioral and habitat
considerations. Black Bass 2000: Conservation. Bethesda, American Fisheries
Society. In review.

Johnson, D. L., R. A. Beaumier, and W. E. Lynch, Jr. 1988. Selection of habitat structure
interstice size by bluegills and largemouth bass in ponds. Transactions of the
American Fisheries Society 117: 171-179.

Johnson, D. L. and W. E. Lynch, Jr. 1992. Panﬁsh use of and angler success at evergreen
tree, brush, and stake-bed structures. North American Journal of Fisheries
Management 12: 222-229.

Johnson, D. L. and R. A. Stein. 1979. Response of ﬁsh to habitat structure in standing
water, North Central Division, American Fisheries Society.

Kayle, K. A. 1986. Fish use of different habitat structure locations and conﬁgurations
within a reservoir. Master’s Thesis. The Ohio State University, Columbus.

Keast, A. and J. Harker. 1977. Strip counts as a means of determining densities and
habitat utilization patterns in lake ﬁshes. Environmental Biology of Fishes 12:
1 81-1 88.

Kinney, J ., R. Stuber, and K. Krueger. 1999. Au Sable River Impoundment Assessments,
Michigan Department of Natural Resources and USDA Forest Service.

46

Klima, E. F. and D. A. Wickharn. 1971. Attraction of coastal pelagic ﬁshes with artiﬁcial
structures. Transactions of the American Fisheries Society 100: 86-99.

Lynch, W. E. J. and D. L. Johnson. 1989. Inﬂuences of interstice size, shade, and
predators on the use of artiﬁcial structures by bluegills. North American Journal
of Fisheries Management 9: 219-225.

Mabbott, L. B. 1991. Artiﬁcial habitat for warrnwater ﬁsh in two reservoirs in southern
Idaho. Warmwater Fisheries Symposium 1, Phoenix, AZ, USDA Forest Service,
Rocky Mountain For. and Range Exp. Sta., Gen. Tech. Rep. RM-207: 62-66.

Madejczyk, J. C., N. D. Mundahl, and R. M. Lehtinen. 1998. Fish assemblages of natural
and artiﬁcial habitats within the channel border of the Upper Mississippi River.
American Midland Naturalist 139: 296-310.

Menzel, D. W. and N. Corwin. 1965. The measurement of total phosphorous in seawater
based on the liberation of organically bound fractions by persulfate oxidation.
Limnology and Oceanography 10: 280-282.

Miller, R.J. 1975. Comparative behavior of centrarchid basses. Black bass biology and
management. R. H. Stroud and H. Clepper. Washington DC, Sport Fishing
Institute: 85-94.

Mitzner, L. 1981. Assessment and development of underwater structure to attract and
concentrate ﬁsh, Iowa Conservation Commission, Federal Aid in Fish
Restoration, F-94-R, Study 1, Final Report, Des Moines.

Moring, J. R., M. T. Negus, R. D. McCullough, and S. W. Herke. 1989. Large
concentrations of submerged pulpwood logs as ﬁsh attraction structures in a
reservoir. Bulletin of Marine Science 44: 609-615.

Moring, J. R. and P. H. Nicholson. 1994. Evaluation of three types of artiﬁcial habitats
for ﬁshes in a freshwater pond in Maine, USA. Fifth International Conference on
Aquatic Habitat Enhancement, Long Beach, CA. Bulletin of Marine Science 55:
1149-1159.

Murphy, J. and L. P. Riley. 1962. A modiﬁed single solution method for the
determination of phosphate in natural waters. Analytica Chimica Acta 27: 31-36.

Nusch, E. A. 1980. Comparisons of different methods for chlorophyll and phaepigment

determination. Archiv ﬁir Hydrobiologie Beiheﬁ Ergebnisse der Limnologie 14:
14-36.

47

Pardue, G. B. and L. A. Nielsen. 1979. Invertebrate biomass and ﬁsh production in ponds
with added attachment surface. Response of ﬁsh to habitat structure in standing
water. R. A. Stein and D. L. Johnson, North Central Division American Fisheries
Society. Special Publication 6: 34-37.

Patrick, K. A. 1996. Utilization of half-log structures as spawning sites for smallmouth
bass in Norris Reservoir, Tennessee. The University of Tennessee, Knoxville.

Paxton, K. O. and F. Stevenson. 1979. Inﬂuence of artiﬁcial structure on angler harvest
from Killdeer Reservoir, Ohio. Response of ﬁsh to habitat structure in standing
water. R. A. Stein and D. L. Johnson, North Central Division American Fisheries
Society. Special Publication 6: 70-76.

Petit, C. D. 111 1972. Stake beds as crappie concentrators. Proceedings of the Annual
Conference of Southeastern Game and Fish Commissioners 26: 401-406.

Pierce, B. E. and G. R. Hooper. 1980. Fish standing crop comparisons of tire and brush
ﬁsh attractors in Barkley Lake, Kentucky. Annual Conference of Southeastern
Association of Game and Fish Commissioners 33: 688-691.

Potts, T. A. and A. W. Hulbert. 1994. Structural inﬂuences of artiﬁcial and natural
habitats on ﬁsh aggregations in Onslow Bay, North Carolina. Fiﬁh International
Conference on Aquatic Habitat Enhancement, Long Beach, CA. Bulletin of
Marine Science 55: 609-622.

Prince, E. D., R. J. Strange, and GM. Simmons, Jr. 1976. Preliminary observations on
the productivity of periphyton attached to a freshwater artiﬁcial tire reef. Annual

Conference of Southeastern Association of Game and Fish Commissioners 30:
207-215.

Prince, E. D., O. E. Maughan, and P. Brouha. 1977. How to build a freshwater artiﬁcial
reef. Blacksburg, Sea Grant Division, Virginia Polytechnic Institute and State
University.

Prince, E. D. and O. E. Maughan. 1979a. Attraction of ﬁshes to tire reefs in Smith
Mountain Lake, Virginia. Response of ﬁsh to habitat structure in standing water.
R. A. Stein and D. L. Johnson, North Central Division American Fisheries
Society. Special Publication 6: 19-25.

Prince, E. D. and O. E. Maughan. 1979b. Telemetric observations of largemouth bass
near underwater structures in Smith Mountain Lake, Virginia. Response of ﬁsh to
habitat structure in standing water. R. A. Stein and D. L. Johnson, North Central
Division American Fisheries Society. Special Publication 6: 19-25.

Prince, E. D. and O. E. Maughan. 1978. Freshwater artiﬁcial reefs: biology and
economics. Fisheries 3: 5-9.

48

Reeves, W. C., G. R. Hooper, and B.W. Smith. 1978. Preliminary observations of ﬁsh
attraction to artiﬁcial midwater structures in freshwater. Annual Conference of
Southeastern Association of Game and Fish Commissioners 31: 471-476.

Rodeheffer, I. A. 1939. Experiments in the use of brush shelters by ﬁsh in Michigan
lakes. Papers of the Michigan Academy of Science, Arts, and Letters 24: 183-193.

Rodeheffer, I. A. 1945. Fish populations in and around brush shelters of different sizes
placed at varying depths and distances apart in Douglas Lake, Michigan. Papers
of the Michigan Academy of Science, Arts, and Letters 30: 321-345.

Rogers, K. B. and E. P. Bergersen. 1999. Utility of synthetic structures for concentrating
adult northern pike and largemouth bass. North American Journal of F isherres
Management 19: 1054-1065.

Rountree, R. A. 1989. Association of ﬁshes with ﬁsh aggregation devices: effects of
structure size on ﬁsh abundance. Bulletin of Marine Science 44: 960-972.

SAS Institute Inc. 1999. SAS user's guide: statistics, version 8.0. SAS Institute, Cary,
North Carolina.

Shulman, M. J. 1984. Resource limitation and recruitment patterns in a coral reef ﬁsh
assemblage. Journal of Experimental Marine Biology and Ecology 74: 85-109.

Smith, B. W., G. R. Hooper, and CS. Lawson. 1981. Observation of ﬁsh attraction to
improved artiﬁcial midwater structures in freshwater. Annual Conference of
Southeastern Association of Game and Fish Commissioners 34: 404-409.

Tugend, K. 1., M. S. Allen, and M. Webb. 2002. Use of artiﬁcial habitat structures in US.
Lakes and Reservoirs: a survey from the Southern Division AF S Reservorr
Committee. Fisheries 27: 22-27.

Vogele, L. E. and W. C. Rainwater. 1975. Use of brush shelters as cover by spawning
black basses (Micropterus) in Bull Shoals Reservoir. Transactions of the
American Fisheries Society 104: 264-269.

Walters, D. A., W. E. Lynch, Jr., and D. L. Johnson. 1991. How depth and interstice size
of artiﬁcial structures inﬂuence ﬁsh attraction. North American Journal of
Fisheries Management 1 1: 319-329.

Wege, G. J. and R. 0. Anderson. 1979. Inﬂuence of artiﬁcial structure on largemouth
bass and bluegills in small ponds. Response of ﬁsh to habitat structure in standing
water. R. A. Stein and D. L. Johnson, North Central Division American Fisheries
Society. Special Publication 6: 59-69.

49

Wickham, D. A., J. W. Watson, Jr. and L. H. Ogren. 1973. The efﬁcacy of midwater
artiﬁcial structures for attracting pelagic sport ﬁsh. Transactions of the American
Fisheries Society 102: 563-572.

Wilbur, R. L. 1974. Florida's fresh water ﬁsh attractors. Fishery Bulletin No. 6, Florida
Game and Fresh Water Fish Commission.

Wilbur, R. L. 1978. Two types of ﬁsh attractors compared in Lake Tohopekaliga, Florida.
Transactions of the American Fisheries Society 107: 689-695.

Wilson, A. E. and O. Samelle. 2002. Relationship between zebra mussel biomass and

total phosphorous in European and North American lakes. Archiv fur
Hydrobiologie 153: 339-351 .

50

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

llIllllllllllllllllmllyl