AN ABSTRACT OF THE THESIS OF
Michelle L. LaRue for the degree of Master of Science in Fisheries Science
presented on July 24, 2001.
Title: Characterization of Stream Fish Assemblages and Land Use Associations
within a Southern Ohio National Forest.
Abstract approved:
Redacted for Privacy
Li
Redacted for Privacy
Hiram W. Li
Seasonally, in 1998 and 1999, I examined spatial and temporal variation in
fish assemblages of agricultural, forested, and acid mine drainage tributaries within
the Wayne National Forest (WNF) in southern Ohio. Land use and natural
disturbance explained patterns in stream fish assemblages. Creek chub and green
sunfish were present in all land use types. Generally, with the exception of creek
chub and green sunfish, species most abundant in one land use type occurred
infrequently in the other land use types sampled. For example, redbelly dace and
blacknose dace dominated forested assemblages but rarely occurred in mining or
agricultural assemblages.
Agricultural sites consisted of higher order streams, located at lower
elevations with reduced canopy cover. Forested sites included intermittent streams
associated with higher elevation, low stream order, and high canopy cover. Acidic
conditions characterized mining sites, which otherwise remained physically similar
to forested sites. Stream order, elevation, and canopy cover explained the majority
of the variance in assemblage structure within 1998, and pH was also important. In
1999, water quality, specifically dissolved oxygen, and seasonal variation became
important. Assemblages changed following drought in 1999. Forested assemblages
remained most similar following drought, while agricultural assemblages exhibited
less similarity (i.e., greater variability). These results suggest that large-scale reach
characteristics and chemical signals related to land use are important to fish
assemblage structure, but in times of environmental fluctuation, water chemistry of
other site-specific variables may be of even greater importance due to physiological
tolerances and limitations of fishes.
Characterization of Stream Fish Assemblages and Land Use Associations
within a Southern Ohio National Forest
by
Michelle L. LaRue
A Thesis Submitted
to
Oregon State University
In Partial Fulfillment of
the requirements for the
degree of
Master of Science
Presented July 24, 2001
Commencement June 2002
Master of Science thesis of Michelle L. LaRue presented on July 24. 2001.
APPROVED:
Redacted for Privacy
,
reprèienting Fisheries Science
Redacted for Privacy
Professor, representing Fisheries Science
Redacted for Privacy
Head of Department
and Wildlife
Redacted for Privacy
Dean of Grâduafe School
I understand that my thesis will become part of the permanent collection of Oregon
State University libraries. My signature below authorizes release of my thesis to
any reader upon re est.
/
Redacted for Privacy
Michelle L. LaRue, Author
ACKNOWLEDGEMENTS
I offer my most sincere gratitude to my mentor, Deanna Stouder, for long term
encouragement and friendship. I would like to thank Judy Li and Hiram Li for their
wonderful assistance and for serving as my advisors at Oregon State University. I also
thank Cliff Pereira for his role on my committee. I thank the staff at the Wayne
National Forest for their friendship, support, and expertise, especially Rebecca Ewing
and Ralph Miller.
Field assistants are essential to any project: I thank Josh Harper for his help
and patience, and for his creative field accessories. I also thank A. Butterworth for her
assistance. In addition, I'd like to recognize D. Stouder, R. Miller, S. Micucci, and R.
Miller for being cheerful volunteers. I thank T. Cavendar at the Ohio Museum of
Biological Diversity, Ohio State University for his help with specimen identification.
I also thank M. White, F. McCormick, R. Ewing, and J. Grow for knowledgeable
conversations regarding the local geography, fauna, and histomy. Funding was
provided by the Ohio Cooperative Fish and Wildlife Research Unit and the Wayne
National Forest (CCS-09- 14-98-03), and from the USDA Forest Service, Pacific
Northwest Research Station, Aquatic and Land Interaction Program. I'd also like to
thank Hecla Water Association, Oregon Cooperative Fish and Wildlife Research Unit,
and USGS for field equipment loans.
There are many individuals who have served as mentors and friends, providing
me with confidence in my blossoming career; thank you so much to all of you who
have ever had, and continue to have, faith in me. Much love to my family who all
believe I am crazy, but believe nonetheless.
TABLE OF CONTENTS
Page
INTRODUCTION ............................................................................... 1
STUDYAREA ..................................................................................4
METHODS ......................................................................................9
SiteSelection ........................................................................... 9
Habitat Characteristics ................................................................ 14
Fishes .................................................................................. 16
DataAnalysis .......................................................................... 16
RESULTS ......................................................................................20
1998 and Land Use Patterns ........................................................ 20
Habitat Characteristics .......................................................... 20
Assemblage Characteristics .................................................... 25
Integrating Assemblage Composition with Habitat Characteristics...... 29
1999 and Examination of Drought .................................................. 31
Habitat Characteristics .......................................................... 31
Assemblage Characteristics .................................................... 36
Integrating Assemblage Composition with Habitat Characteristics...... 37
Response of Fish Assemblages Following Drought ..................... 41
Spatial and Temporal Variation in Fish Assemblages..............................45
DISCUSSION ................................................................................... 52
BIBLIOGRAPHY .............................................................................. 60
APPENDIX ...................................................................................... 68
LIST OF FIGURES
Figure
Page
Schematic map of the state of Ohio highlighting the location of the
Wayne National Forest ................................................................. 5
2
Monthly precipitation received in south central Ohio during
1998and 1999 ........................................................................... 7
3
Schematic of a typical stream reach (50m following thaiweg), depicting
the start and end of the study site ................................................... 15
4
Comparison of means for stream temperature (a), dissolved oxygen (b),
pH (c), conductivity (d), turbidity (e), and canopy (1) for 1998 and 1999
bylanduseandseason ................................................................ 21
5
Substrate composition for 1998 (a) and 1999 (b) by land use and season. . . .23
6
Temporal patterns in number of species captured for all sites .................. 27
7
Temporal patterns in catch per unit effort (CPUE) for all sites ................. 28
8 NMS ordination of 1998 fish assemblages. Overlaid on the ordinations
are joint plots of environmental variables (a) and individual fish
species (b) strongly correlated (r2> 0.5) with the ordination axes ............... 30
9 NMS ordination of 1999 fish assemblages. Overlaid on the ordinations
are joint plots of environmental variables strongly correlated (r2 > 0.5)
with ordination axes one and two (a) and one and three (b)..................... 38
10 NMS ordination of 1999 fish assemblages. Overlaid on the ordinations
are joint plots of individual fish species strongly correlated (r2 > 0.5)
with ordination axes one and two (a), and one and three (b) .................... 39
11 Composition of fish assemblages following drought in autumn 1999 .......... 42
12 HIerarchical agglomerative cluster analysis of autumn 1998 and 1999
fish assemblages for agricultural and forested sites ............................... 44
LIST OF FIGURES (Continued)
Figure
Page
13 Comparison of autumn 1998 and autumn 1999 fish assemblages
in forested and agricultural sites for number of species captured (a)
and catch per unit effort (CPUE)(b) .....................................................46
14 NMS ordination, based on fish assemblage data, projected onto
axes one and two (a), and onto axes one and three (b), for
1998 (filled symbols) and 1999 (empty symbols) .................................... 50
LIST OF TABLES
Table
Page
1
Sampling locations and landscape characteristics within Wayne
National Forest, Ironton District, Ohio ............................................. 10
2
Seasonal samples collected at each site ........................................... 13
3
Percent total catch of the most common fishes
( 1% of total) in 1998 and 1999 by land use ..................................... 26
4
Correlations between axes coordinates from 1998 NMS ordination
and species abundance and physical parameters .................................. 32
5
Percent catch of all fishes captured annually and for entire study ............. 33
6
Correlations between axes coordinates from 1999 NMS ordination
and species abundance and physical parameters.................................. 40
7 Morisita's Index values for assemblage comparisons between
autumn 1998 and autumn 1999 within a site (a) and for autumn 1998
and autumn 1999 assemblage comparisons within agricultural and
forested land uses (b) .................................................................. 43
8
Composition of autumn 1998 and autumn 1999 fish assemblages
in forested and agricultural sites ................................................... 47
9
Correlations between axes coordinates from 1998 and 1999 NMS
ordination and species abundance and physical parameters .................... 51
CHARACTERIZATION OF STREAM FISH ASSEMBLAGES AND LAND USE
ASSOCIATIONS WIT}IIN A SOUTHERN OHIO NATIONAL FOREST
INTRODUCTION
Fish assemblages may reflect patterns of land use and natural disturbances,
thus it is important to understand which environmental characteristics, from
landscape level to microhabitat, are strongly associated with assemblage structure.
Streams are hierarchically organized systems consisting of multiple spatiotemporal
scales (Frissell et
al.
1986) with physical and biological connections linking
streams to their landscapes (THynes 1975, Gregory et
al.
1991, Stanford and Ward
1992, Montgomery and Buffington 1999); therefore landscape alterations
potentially influence streams at different spatial scales. Headwater streams are
particularly sensitive to, and often the most directly impacted by, landscape
alteration Karr and Schlosser 1978, Beschta and Platts 1986). Low order streams
are also important because they comprise the greatest proportion of streams in a
watershed. Natural disturbances (e.g., drought or floods) vary in duration,
frequency, and timing, so impacts may also be hierarchically scaled.
Land use activities may alter stream fish communities (Schlosser 1991).
For example, changes in fish assemblages have been documented with increased
agricultural activity (Smith 1971, Trautman 1981, Menzel et
al.
1984) and water
quality degradation (TKarr 1991). In North America, 33% to 75% of aquatic
2
organisms are rare or extinct (Master 1990, Naiman et al. 1995), including 34% of
fish species (Doppelt 1993); habitat alteration is a contributing cause in 73% of fish
species extinction (Miller et al. 1989). Natural disturbances may also influence
fish assemblages by temporarily altering stream environments.
The primary objective of this study was to describe patterns in fish
assemblage composition within forested, agricultural, and acid mine drainage
tributaries of the Wayne National Forest in south central Ohio. Ohio has a legacy
of landscape alteration and modification, including the extraction of natural
resources. The state was once 95% forested, but much of the land has been
converted for agricultural and residential purposes. The first land purchase to
create a national forest in Ohio was in 1934, but the Wayne National Forest (WNF)
did not become official until 1951. Stewardship and management of watersheds
are priorities of national forests and a current goal of WNF is the development of
an ecological aquatic classification system and determination of ecological
potential of the aquatic resources. The land within WNF is a fragmented
landscape, containing multiple land uses, and both private and public lands.
Current land use practices include timber harvest, agriculture, oil and gas
development, and commercial coal, clay and limestone mining. While the USDA
Forest Service is not currently using timber harvest as a management tool on WNF
lands, it maintains an active recreation program which includes hiking, mountain
biking, off road vehicles (ORV), and horse trails (R. Ewing, USDA Forest Service,
personal communication). Agricultural land use in the watersheds is limited to
3
private lands where channel modification is a common practice. Pastures, hay
fields and row crops generally occur along mainstem reaches and higher order
tributaries, although some pasture land is found on ridges and sideslopes where the
first order streams originate. Intennittent and first order reaches primarily occur in
forested areas, while mining areas exist throughout the watersheds.
I used descriptive patterns in fish assemblage composition within forested,
agricultural, and acid mine drainage tributaries of the Wayne National Forest to ask
the following questions: 1) Do fish assemblages vary over time? 2) Are there
associations between fish assemblage composition and physical habitat variables?
3) Do fish assemblages within different land use types respond differentially to
drought? and 4) Based on land use and temporal variability, what recommendations
can be made regarding classification and restoration of these tributary streams?
4
STUDY AREA
The Wayne National Forest (WNF) covers over 68,394 ha in unglaciated
south central Ohio within the Western Allegheny Plateau physiographic region
(Figure 1). This region is unglaciated and hilly (Hammond
1964)
and consists
mainly of forested patches interspersed with agricultural land (Anderson
1970)
with very few urbanized areas. As a consequence of the hilly terrain, agricultural
lands are found primarily in the valley lowlands. When nomadic native Indians
inhabited the area the land was relatively undisturbed. Later, native people
established seasonal camps and villages, and cultivated river valleys and terraces.
In the late
1700's
colonists came to Ohio and began to settle along the Ohio River
and other major waterways; the hills of WNF were some of the last areas to be
settled in Ohio. Land was initially cleared for agricultural purposes. Commercial
lumbering peaked in
1849;
less than 100 years later most marketable trees had been
harvested. Southern Ohio was also an active producer of oil, coal, and iron ore.
The furnaces used to smelt iron ore required vast amounts of timber for charcoal
fuel, and by the early
1900's,
most furnaces became non-functional, partly due to
the lack of wood. Given these many landscape changes, few undisturbed streams
or areas exist in the region today.
This study was conducted in two of the major drainage basins located
within the WNF, Symmes Creek and Pine Creek; both are tributaries to the Ohio
River. Symmes Creek has a drainage area of 921.3
km2
(Krolczyk
1954)
and is
A Forested sites
added in 1999
Figure 1. Schematic map of the state of Ohio highlighting the location of the Wayne National Forest (three
districts in stippled locations). Locations of sample sites within the Wayne National Forest are indicated.
located in Jackson, Gallia, and Lawrence counties. The mean gradient is low at 0.6
rn/km (Krolczyk 1954). Previous surveys in Symmes Creek (Holeski etal. 1992,
1993 and 1995) reported approximately 44 different fish species including the
eastern sand darter (Ammocryptapellucida), a federally monitored species in Ohio.
Pine Creek is located west of Symmes Creek and drains 478.4 km2 with an average
gradient of 1.6 ni/km (Krolczyk 1954). The watershed is located in Scioto and
Lawrence counties and flows through the western edge of the WNF. In the late
1970's 21 species of fish were collected in the Pine Creek basin (Barnes and
Canine 1978).
South central Ohio normally experiences the most precipitation in spiing
and early summer followed by a dry autumn. In 1998, the average precipitation for
Ohio was 102.44 centimeters (107% of normal; 7.Olcentimeters above normal) and
precipitation in south central Ohio was also above normal (107%) with an average
of 112.57 centimeters (7.70 centimeters above normal) (Ohio Department of
Natural Resources (ODNR) 1998). However, in 1999 precipitation was
consistently below normal for most of the year (Figure 2) and 1999 was ranked as
the 1 8 driest year in 117 years of record (ODNR 1999). Overall precipitation
during the 1999 calendar year was below normal for all of Ohio (89% of nonnal
with average precipitation of 85.06 centimeters; 10.41 centimeters below normal)
and for south central Ohio (87% of normal with average precipitation of 90.73
centimeters; 14.15 centimeters below normal). The Palmer Drought Severity Index
rated precipitation levels in south central Ohio during 1998 (+7.70 centimeters) as a
7
25
-.
20
1998
Regional Average
Departure
from Normal *
15
10
E
5
0
-5
25
1999
Regional Average
20
Departure
*
from Normal
15
10
U
5
0
-5
-10
Jan
Feb
Mar Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dcc
Figure 2. Monthly precipitation received in south central Ohio
during 1998 and 1999. *Depflje from normal is based on 50 year
record (1931-198 1).
8
vely moist spell; the region was rated as an extreme drought in 1999
centimeters) (ODNR
In April
1999,
1998
and
(-14.15
1999).
south central Ohio began experiencing mild drought
conditions (ODNR 1999); flows were considered below normal for this region, but
instream water volumes were not noticeably affected. Severe drought conditions
were reached by June
1999,
and in July 1999 stream flows were deficient in the
southern half of Ohio. Sixty-six counties were declared agricultural disaster areas
including Gallia, Jackson, Lawrence, and Scioto counties (ODNR 1999). Severe
drought conditions continued into autumn
1999.
The south central region had
greater than normal precipitation for August 1999 (Figure 2) largely due to showers
and thunderstorms occurring on August 24-25 (ODNR 1999). After September
1999,
south central Ohio received above normal precipitation for the remainder of
the year.
METHODS
Site Selection
Six tributaries in the Symmes Creek basin and six tributaries in the Pine
Creek basin within the Ironton district of the Wayne National Forest were sampled
seasonally (i.e., spring, summer, autumn) in
1998
and 1999 to assess fish
assemblage structure (Figure 1, Table 1). Four sites (three in Symmes Creek and
one in Pine Creek) were added in spring
1999
to include larger order forested sites
for direct comparison among land use sites of similar stream order. Study reaches
varied in stream order and were chosen to represent three common land uses of the
region; forested areas, agricultural croplands, and acid mine drainage (Table 1).
Land use categories were determined from a land cover inventory map for the state
(ODNR 1994) and verified in the field. Forested and agricultural reaches were
selected so that land use upstream and in adjacent areas was consistent with land
use at the study reach. Acid mine drainage streams were directly influenced by
point-source pollution from abandoned coal mines. Acid mine drainage occurs
when sulfur and suffitic materials are oxidized by air, water, and bacteria into
sulfuric acid and ferrous sulfate (Warner
1971),
and under neutral pH conditions,
ferric hydroxide will precipate onto the stream bottom (Letterman and Mitsch
1978).
Sampling reaches were chosen to be representative of each land use
category and located away from bridges, major tributaries, agricultural drains,
pastures, and feed lots. Habitat information and fish were sampled concurrently
10
Table 1. Sampling locations and landscape characteristics within Wayne National
Forest, Tronton Distict, Ohio. Land use codes: A = Agricultural, F = Forested, M
= Mining. A indicates sites where cross-section data were not collected for Rosgen
Stream Classification. * indicates sites added in 1999. M.a.s.l = meters above sea
level.
Table 1.
LATTITUDE
&
A
DRAINAGE
BASIN
Pine
COUNTY
Scioto
Turkeyfoot Run
A
Pine
Scioto
Camp Creek
A
Symmes
Gallia
Long Creek
A
Symmes
Lawrence
Cooney Branch
F
Pine
Scioto
Howard Run*
F
Pine
Lawrence
Wolcott Hollow
F
Pine
Lawrence
Asbuxy Creek
F
Symmes
Lawrence
Bakers Fork*
F
Symmes
Lawrence
Jenkins Ridge
F
Symmes
Gallia
Meyers Hollow
F
Symnies
Gallia
STREAM
NAME
Sugarcamp Hollow
LAND
USE
RockyFork
F
Symmes
Gallia
Brushy Fork
M
Pine
Lawrence
Negro Creek
M
Pine
Lawrence
Sand Fork
M
Symmes
Lawrence
Webster Creek
M
Symmes
Lawrence
LONGfl1Th)E
38°48.07N
82°42.21W
38°43.83N
82°44.58W
38°45.59N
82°27.49W
38°39.26N
82°28. lOW
38°41.57N
82°44.86W
38°44.09N
82°41 .72W
38°42.57N
82°42.04W
38°44.60N
82°34.40W
38°44.02N
82°31.32W
38°43.3 iN
82°26.45W
38°4 1 .05N
82°24.41W
38°48.96N
82°26.59W
38°46.88N
82°38.72W
38°44.56N
82°37.29W
38°37.36N
82°22.48W
38°43. 18N
STREAM
ORDER
SITE
DRAINAGE
AREA (KM2)
ELEVATION
(m.a.s.L)
STREAM
2
11.1
195
0.5
ROSGEN
STREAM
TYPE
Bc
2
12.2
189
0.8
A
1
14.3
189
0.8
G
3
28.1
183
0.5
Bc
Intermittent
2.2
207
4.5
Bc
Intermittent
1.0
224
4.2
Bc
Intermittent
2.3
189
1.6
C/Cc
1
3.4
213
1.8
B
2
5.8
189
0.3
C/Cc
Intermittent
3.2
232
2.7
C/Cc
Intermittent
1.6
219
2.1
B
1
6.6
201
0.1
A
Intermittent
2.4
219
4.2
C/Cc
1
3.4
207
0.8
C/Cc
1
5.8
224
23.0
B
Intermittent
4.2
183
0.3
A
KM
12
within each 50-rn study reach. Spring sampling occurred during April, May, and
June; summer sampling occurred in July and early August; autumn sampling
occurred in very late August and September.
Due to intermittent conditions and drought, not all sites were sampled in
every season (Table 2). At least one site in all land use types exhibited intermittent
conditions. In autumn 1998, four sites (Cooney Branch, Wolcott Hollow,
Turkeyfoot Run, and Webster Creek) were dry due to intermittent conditions
(Table 2); these same sites were also dry in autumn 1999. The remaining forested
sites (Jenkins Ridge and Meyers Hollow) and agricultural sites (Camp Creek, Long
Creek, Sugarcamp Hollow) did not desiccate in autumn 1998; these same sites were
recharged with enough water to allow sampling in autumn 1999 following drought
(Table 2).
During drought in summer 1999, streams in only four sites (Brushy Fork,
Negro Creek, Sand Fork, and Sugarcamp Hollow) contained enough water to allow
sampling; streams within all forested sites, three agricultural sites, and one mining
site were dry in summer 1999 (Table 2). Stream flow was too negligible to be
measured in most of these sites. Autumn showers recharged study sites, filling
them with enough water to allow sampling from August 31 to September 10, 1999.
Prior to the late August showers, four of the six sites sampled contained no water.
Table 2. Seasonal samples collected at each site. Land use codes: A = Agricultural, F = Forested, M = Mining. Letters in
parentheses indicate basin: P = Pine Creek and S = Symmes Creek. * indicates sites added in 1999, / indicates sites
sampled, and - indicates dry sites.
STREAM NAME
Sugarcamp Hollow (P)
Turkeyfoot Run (P)
Camp Creek (S)
Long Creek (S)
Cooney Branch (P)
Wolcott Hollow (P)
Jenkins Ridge (S)
Meyers Hollow (S)
Asbury Creek* (S)
Bakers Fork* (S)
Howard Run* (P)
Rocky Fork* (S)
Brushy Fork (P)
Negro Creek (P)
Sand Fork(S)
Webster Creek (S)
LAND
USE
A
A
A
A
F
F
F
F
F
F
F
F
M
M
M
M
SPRING
SUMMER
AUTUMN
SPRING
SUMMER
AUTUMN
1998
1998
1998
1999
1999
1999
1
1
1
1
1
1
1
1
1
I
1
1
-
(n=4)
(n=4)
(n=3)
(n=4)
/
/
II
I
(n1)
-
(n =4)
(n =4)
(n =2)
I
II
I
I
/
-
(n=4)
(n=4)
(n=3)
/
I
1
-
I
I
/
I
I
II
I
I
I
I
I
(n
I
I
I/
= 8)
(n=4)
1
(n =0)
1
/
I
(n=3)
/
I
-
I
/
I
(n =2)
-
/
-
(n=3)
(n=1)
14
Habitat Characteristics
I evaluated landscape features, stream reach, and site attributes, including
water quality and instream habitat, at each site. Habitat transects were located
perpendicular to stream flow every 5-rn for the entire 50-rn reach, resulting in a
total often transects in 1998. In 1999, I reduced the number of transects from ten
to five, placed at 10-rn intervals along the stream reach. This reduction in total
transects was made to reduce time
and
effort in the field after assessment of the
1998 data indicated that five transects would adequately describe the 50-rn stream
reach. Channel and bank characteristics were measured at each transect with a
combination of transect-line and transect-point methods similar to those described
in Simonson et al. (1994) (Figure 3). For each transect-line, wetted channel width,
bank erosion, and riparian width (up to 10-rn) were measured
and
at each transect-
point, water depths, substrate composition (visually estimated for a .25 x .25 m
quadrat), embeddedness (visually estimated to the nearest 5% for gravel and
cobble), and canopy cover (estimated to the nearest 10% using a spherical forest
densiometer) were measured. Habitats were classified as riffle, run, or pool habitat.
Water quality measurements collected at one location within each reach included
water temperature (°C, YSI Model 80'), dissolved oxygen (mg/i, YSI Meter 80),
ambient conductivity (pS/cm, YSI Meter 80), turbidity (ntu, Hach 2100A meter'),
and pH
(Oakton pH meter'). Discharge (calculated using stream velocity as
measured across one transect per reach with a Pygmy current meter'), air
'The use of trade or firm names in this publication is for reader information and does not imply
endorsement by the U.S. Department of Agriculture of any product or service.
15
Figure 3. Schematic of a typical stream reach (50m following thaiweg), depicting the
start and end of the study site. A detailed view of the transect describing transect lines
and transect points is shown. (Modified after Simonson et al. 1994).
16
temperature (using a handheld thermometer), and gradient (using a surveyor rod
and level) were also measured. Stream order (Strahier), elevation, and drainage
area for each site were determined from 7.5-minute USGS topographic maps.
Thirteen of the 16 study sites were surveyed for Rosgen stream classification
purposes (Rosgen 1994), and an additional 38 sites within WNF were also surveyed
(Appendix I).
Fishes
For each season at each site, fishes were collected from a seine-blocked, 50-
m reach on two to four successive passes with a battery operated backpack
electrofisher and two fishnetters. On one occasion, only one pass was performed
due to equipment failure. Electrofishing proceeded in an upstream direction;
all
habitats were sampled to ensure that a representative sample of the assemblage was
collected, and substrates were disturbed to ensure benthic fishes were dislodged.
Standard length, weight, and species of all individuals were recorded before fishes
were returned to the stream. Any unknown or uncertain specimens were preserved
in 10% buffered formalin for later identification and stored at the Ohio Museum of
Biological Diversity in Columbus, Ohio.
Data Analysis
Multivariate analyses were used because environmental variables (and
potentially biological assemblages) operate together and not in isolation (Hawkes et
17
al.
1986, Frenzel and Swanson 1996). Multivariate methods are species dependent
methods and are more sensitive in discriminating spatial and temporal patterns than
univariate or graphical/distributional methods that are species independent
(Warwick and Clarke 1991). Non-metric multidimensional scaling (NMS)
(Kruskal 1964, Mather 1976) is a multivariate ordination method based on an
iterative rank-ordering procedure resulting in an optimal solution, and is well suited
to non-normal data (McCune and Mefford 1999). In particular, NMS has been
used in other aquatic ecological studies to examine community patterns of
phytoplankton (Salmaso 1996), zooplankton (Sprules 1980), macroinvertebrates
(Stephenson etal. 1993, Wright 2000), and fish (McCormick etal.
2000,
Rose
2000, Wright 2000). This method positions sites in species space, so those sites
with similar assemblages are placed closer together than sites with dissimilar ones.
Relationships between assemblages and environmental data were examined using
joint plots (McCune and Mefford 1999).
To examine spatial and temporal variability in fish assemblage structure,
three site-by-species matrices for species catch data were analyzed using NMS
(PC-ORD, version 4.14). Streams in which no fish were captured during sampling
were not included in the multivariate analyses. For each analysis, I used a
Sorensen's (Bray-Curtis coefficient) distance measure at the following settings: 40
runs with original data, 50 Monte Carlo simulations,
0.20
step length, and a 400
iteration maximum. Dimensionality (i.e., number of axes) of the final ordination
configurations was determined by choosing only those dimensions that provided
18
the greatest reduction in stress. Stress is an inverse measure of fit to the data and is
reported for the final ordination on a 0-100 scale (McCune and Mefford 1999).
Final stress for the ordination was less then the final stress in 95% of the
randomized Monte Carlo simulations (i.e., Monte Carlo test p-values <0.05).
Stability of the final solution was reported by PC-OR]) as the standard deviation in
stress over the preceding 15 iterations (McCune and Mefford 1999). All final
ordinations were oriented to the same variable to ease interpretation and
communication of results. Primary, secondary, and tertiary axes were identified
using r2 values for proportion of variance explained by each axis so that the
primary axis (Axis 1) accounts for the greatest proportion of variance, or the
greatest r2 value. Logarithmic transformation (log 10 (x+1)) of the species catch
data was used to include contribution by both common and rare species.
Conductivity values were also log transformed prior to analysis, and matrices were
checked for outliers (i.e., ±2 standard deviations of the mean average distance)
prior to ordination.
To express the linear association between environmental variables or fish
species and the ordination scores for each axis, I used a Pearson product-moment
correlation coefficient (r). The correlation coefficient (r) ranges from +1 to 1 for
perfect positive association to perfect negative association. When the correlation
coefficient equals zero there is no linear association (Sokal and Rohlf 1987). The
coefficient of determination (r2) measures the proportion of variation of one
variable determined by the variation of the other. It is a measure of strength and
19
will range from zero to one (Sokal and Rohlf 1987). I used r2 to rank the relative
importance of correlations with the ordinations and considered r2 values> 0.500 (r
> 0.700) to indicate strong correlation.
I used the mean ± one standard deviation to describe habitat variables (e.g.,
agricultural sites had a mean stream temperature of 20.3 ± 2.6 °C).
To evaluate assemblage changes following drought conditions, I compared
1999 autumn assemblages to the 1998 autumn assemblages in agricultural and
forested sites for catch per unit effort (CPUE), number of species captured, and
species composition. I used Morisita's index (Morisita 1959) and hierarchical
agglomerative cluster analysis, with Euclidean distance measure and Ward's
method, to compare assemblages among sites and land use for autumn 1998 and
autumn 1999. Agglomerative methods sequentially merge objects or groups of
objects with other objects or groups, and hierarchical classifications optimize the
route through which the groups are attained; Ward's method is a space conserving
linkage method (i.e., the original space properties are preserved) compatible with
Eucidean distance measure (MeCune and Mefford 1999). Morisita's index has
been used in other studies of stream fish assemblages (Ross et al. 1985, Matthews
1986, Matthews et al. 1988, Bart 1989, Matthews 1990, Meador and Matthews
1992, Stewart et al. 1992), and it is not affected by diversity or sample size (Wolda
1981). Index values range from zero (no similarity between samples) to slightly
greater than one for identical collections; values
0.40 indicate low similarity (Ross et al. 1985).
0.74 indicate high similarity and
20
RESULTS
1998 and Land Use Patterns
Habitat Characteristics
In 1998, agricultural sites had a mean stream temperature of 20.3 ± 2.6 °C
(Figure 4a), and mean pH of 7.6 ± 0.3 (Figure 4c). Dissolved oxygen values
ranged from 1.75 to 9.40 mg/l (Figure 4b). Agricultural sites had low conductivity
(X = 249.8 ± 128.8 .tS/cm) (Figure 4d) and widely ranging turbidity values (i.e.,
from 4.0 to 24.5 ntu; X = 13.3 ± 14.3 ntu) (Figure 4e). Percent canopy cover was
reduced in agricultural sites (X =22±27) (Figure 4f). Substrate composition
included sand, silt, and gravel (Figure 5a). Agricultural sites ranged in elevation
from 183 to 195 meters above sea level. Agricultural sites were classified as Bc
and Ge channel types; Bc channel types are moderately entrenched with low
gradient and Ge channel types are entrenched with low gradient, sinuosity, and
width/depth ratio (Rosgen 1994) (Appendix I).
Forested sites had a mean stream temperature of 15.8 ± 3.0 °C (Figure 4a),
and a mean pH of 7.3 ± 0.4 (Figure 4c). Dissolved oxygen values ranged from 2.62
to 9.70 mg/I (Figure 4b). Forested sites had low conductivity (X = 148.1 ± 60.1
iS/cm) (Figure 4d) and low turbidity values (X = 3.4 ± 2.8 ntu) (Figure 4e).
Percent canopy cover was high in forested sites (X =90±7) (Figure 4f).
Substrate composition included gravel, sand, cobble, and detritus, although
detectable amounts of detritus were present only in autumn (Figure 5a). Forested
30
a. Stream Temperature
25
.
20
.
15
I
10
5.
12
10
b. Dissolved Oxygen
8
6
4.
2
10
C.PH
I
8
6
4
2
0
n4 n4 n3
n=4 n=4 n2
Agricultural
Forested
n=4 n=4 n=3
Mining
I Spring
n=4 n1 n=3
Agricultural
A Summer
n=8 n=0 n=2
Forested
it4 n3 n=1
Mining
S Autunm
Figure 4. Comparison of means for stream temperature (a), dissolved oxygen (b), pH (c), conductivity (d), turbidity (e),
and canopy (f) for 1998 and 1999 by land use and season. Error bars indicate one standard deviation of the mean.
1998
3000
1999
d. Conductivity
2000
I.
1000
A
K
40
f
U
e. Turbidity
30
20
10
I
100
80
U
I
fCanopy
60
4o
2O
-- ____________________
ir4 n=4 n=3
n=4 n=4 n=2
Agricultural
Forested
n=4 n=4 n3 n4 n=1 n3
Mining
U Spring
Figure 4. Continued
Agricultural
£ Summer
Autumn
n=8 n=0 n=2
n=4 n=3 n=1
Forested
Mining
24
sites ranged in elevation from 189 to 224 meters above sea level. Forested sites
were classified as Bc, C or Cc, and E type channels; C or Cc channel types are
slightly entrenched and meandering with moderate to low gradient and a high
width/depth ratio, and E channel types are moderately sinuous with low
width/depth ratios and low to moderate gradient (Rosgen 1994) (Appendix I).
Mining sites were characterized by acidic conditions, with pH ranging from
2.8 to 6.7 (X = 4.7 ± 1.2) (Figure 4c). Mean stream temperature was 17.9 =E 2.1 °C
(Figure 4a), and dissolved oxygen values ranged from 3.12 to 9.00 mg/I (Figure
4b). Mining sites had variable conductivity (X =811 ± 775 itS/cm) (Figure 4e)
and turbidity (X = 11.5 ± 8.2 ntu) (Figure 4d). Percent canopy cover was high in
mining sites (X =91 ±6) (Figure 4f), and substrate composition included gravel,
sand, silt, precipitate, and detritus (Figure 5a). As in forested sites, detectable
amounts of detritus were present only in autumn. Precipitate only occurred in
mining sites. Mining sites ranged in elevation from 183 to 224 meters above sea
level. Mining sites were classified as C or Cc and E type channels (Appendix I).
All land use types demonstrated a strong seasonal trend in dissolved oxygen
values (Figure 4b). Agricultural sites tended to contain greater amounts of sand
and silt, while forested and mining sites typically had larger proportions of gravel
(Figure 5). Percent canopy cover was most variable in agricultural sites; one site,
Turkeyfoot Run, had several large deciduous trees along its banks while the banks
of the other agricultural sites were dominated mainly by herbaceous vegetation,
bare soil, and/or a few small willows. Also, in autumn 1998, not all sites were
25
sampled due to intermittent conditions (see Table 2); this and/or deciduous leaf
drop may have contributed to the trend toward decreasing canopy cover in autumn.
Assemblage Characteristics
In 1998, agricultural assemblages were dominated by Pimephales notatus,
Notropis buccatus, Semolilus atromaculatus, Luxilus chrysocephalus, and
Etheostoma nigrum which together comprised greater than 75% of the catch for
this land use (Table 3). Pimephales notatus, N buccatus, N. stramineus, Ameiurus
natalis, A. me/as, Esox americanus, Hypentelium nigricans, Lampetra aeyptera,
Lepomis humilis, L. macrochirus, L. megalotis, L. chrysocephalus, Lythrurus
umbratilis, and Micropteruspunctulatus all occurred exclusively in agricultural
sites. Agricultural assemblages tended to contain a greater number of species
(Figure 6) and a higher catch per unit effort (CPTJE) (Figure 7) compared to
assemblages in forested or mining sites. The range was greatest among agricultural
sites for these measures; seven to 19 species were captured in spring 1998, and
CPT.JE ranged from 36 to 642 in autumn 1998.
Forested assemblages were dominated by S. atromaculatus; Phoxinus
erythrogaster, Rhynichthys atratulus, L. cyanellus, andE. fiabellare were also
common in forested sites. Together these five species comprised 98% of the total
catch for forested sites in 1998 (Table 3). Both S. atromaculatus and L. cyanellus
occurred in all land use types, but the two dace species rarely occurred in sites other
than forested ones. Species composition of forested assemblages was similar
(
Table 3. Percent total catch of the most common fishes 1% of total) in 1998 and in 1999 by land use. An asterisk (*)
indicates species contributing <1% of the total catch by land use. Numbers in parentheses indicate total catch. Land use codes:
Ag = Agricultural, For = Forested, and Mm = Mining.
Family
Catostomidae
Centrachidae
Scientific Name
Moxostoma erythrurum
Lepomis cyanellus
Lepomis macrochirus
Lepomis megalotis
Common Name
Golden Redhorse
Green Sunfish
Bluegill Sunfish
Longear Sunfish
Campostoma anomalum Central Stoneroller
Cyprinidae
Luxilus chrysocephalus Striped Shiner
Redfin Shiner
Lythrurus umbratilis
Silveijaw Minnow
Notropis buccatus
Spotfin Shiner
Notropis spilopterus
Notropis stramineus
Sand Shiner
Phoxinus erythrogaster Redbelly Dace
Pimephales notatus
Bluntnose Minnow
Rhynichthys atratulus
Blacknose Dace
Semotilus afromaculatus Creek Chub
Amelurus natalis
Yellow Bullhead
Ictaluridae
Fantail Darter
Etheostomaflabellare
Percidae
Johnny Darter
Etheostoma nigrum
Blackside Darter
Percina maculata
Lampetra
aepyptera
Least Brook Lamprey
Petromyzontidae
1998 % Total Catch
For Mm
Species
Ag
Code (2827) (1963) (29)
MOER
12
14
LECY
1
0
LEMA
1
0
LEME
1
0
CAAN
3
0
0
LUCH
6
*
LYUM
0
2
0
20
0
NOBU
NOSP
0
0
NOST
4
*
17
3
PHER
0
34
0
PINO
13
RHAT
1
0
14
51
76
SEAT
0
0
AMNA
3
5
0
ETFL
2
*
ETNI
5
7
PEMA
LAAE
1999 % Total Catch
Ag
For Mm
(930)
1
(633)
*
2
2
2
0
0
8
3
1
2
0
0
0
0
20
1
5
15
1
0
16
49
45
1
10
5
1
1
*
2
0
(62)
0
40
0
0
0
0
0
0
0
0
0
50
0
0
0
0
27
2
---- Camp
O. Long
-- Sugarcamp
Agricultural
2
2(
-
1
Turkeyfoot
--V
\
14
ic
v
4
S
V
2
2
22
O
Forested
-dv-- Meyer
20
5
0
v Wolcott
18
16
0
4-
14
0)
12
C)
0)
10
Cooney
Jenkins
Asbury
Bakers
Rocky
-- Howard
-
8
0
6
0..
...
4
2
-0- Negro
22
22
Brushy
Mining
-- Sand
--c' Webster
20
18
16
14
12
10
8
6
---v
4
2
-----V----V
0
SP 98
SU 98
AU 98
SP 99
SU 99
AU 99
Season
Figure 6. Temporal patterns in number of species captured for all sites.
700
--- Camp
600
0 Long
Agricultural
v Sugarcamp
Turkeyfoot
500
I\
/\
400
\
300
\
\
200
\
\o
100
/
700
-- Cooney
600
';;'
O Jenkins
-Y-- Meyer
Forested
-ci
Wolcott
Asbury
-0-- Bakers
500
-U---
400
-- Rocky
0
300
-a-- Howard
0 ...................
200
100
700
Brushy
Mining
-0- Negro
-Y- Sand
-V- Webster
650
40
30
/
Q.
/
w
_______
SP 98
SU 98
AU 98
SP 99
SU 99
AUO9
Season
Figure 7. Temporal patterns in catch per unit effort (CPUE) for all sites.
29
among sites and sites contained between three and nine species (Figure 6). Catch
per unit effort was consistently lower at two sites desiccated in autumn (Cooney
Branch and Wolcott Hollow) than the two sites persistently wetted (Meyers Hollow
and Jenkins Ridge) (Figure 7).
Mining sites contained few to no fish, with few species (Figure 6). The
number of species captured from any site was typically between zero and two
during 1998. Semotilus atromaculatus and L. cyanellus were the dominant species
in mining assemblages, accounting for 90% of the fish catch (Table 3); however,
CPUE was rarely greater than ten (Figure 7). Catch per unit effort fluctuated
minimally during 1998, ranging from zero to 17 in spring.
Integrating Assemblage Composition with Habitat Characteristics
The ordination of 1998 fish assemblages displayed a land use pattern with
three clusters of samples; agricultural sites were located on the left half of the
ordination, forested sites in the mid-upper right, and mining sites on the mid-lower
right (Figure 8) (Stress of 3-dimensional ordination = 8.94; instability = 0.0000 1;
cumulative r2 = 0.9 15). Large-scale reach characteristics (i.e., stream order,
elevation, and canopy cover) explained 58% of the variance, while pH accounted
for 33%. Only pH seemed to distinguish between assemblages in forested and
mining sites (axis two), but stream order, elevation, and canopy cover were the
differing characteristics between agricultural and forested/mining assemblages
(axis one) (Figure 8a). Low pH was associated with mining sites. Higher
30
a.
A
AA
Land Use
o
Agricultural
A
Forested
Mining
O
A
aim
lull
111i
IIJIIJ
U
Axisi
Intermittent
r2=.343
o 1Order
2Order
b.
o
AA
0
0
0
A
Seat
3"Order
A
A
A
Nobu
OPino
Esam
A
A
0
.-
11111
11111
U
Axis!
111111
r=.343
Figure 8. NMS ordination of 1998 fish assemblages. Overlaid on the
ordinations are joint plots of environmental variables (a) and individual fish
species (b) strongly correlated (r2 > 0.5) with the ordination axes. Species
codes consist of first two letters of genus and species (see Table 5). Each
symbol represents one site/date. Final stress of 3-dimensional ordination =
8.94; cumulative r2 = .9 15.
31
elevation, greater percent canopy, and lower stream order were associated with
forested and mining sites. Lower elevation, lower percent canopy, and higher
stream order were associated with agricultural sites.
Species negatively correlated with axis one, including P. notatus, N.
buccatus, and L. chrysocephalus, were found in agricultural assemblages and were
associated with large-scale reach characteristics (i.e., higher stream order, lower
elevation, and less canopy cover) (axis one) (Table 4, Figure Sb).
Esox americanus
was captured only from agricultural sites, in low abundance overall (Table 5), but
this species was also associated with large-scale reach characteristics (Table 4,
Figure Sb).
Semotilus atromaculatus was a ubiquitous species, occurring in high
abundance in all sites regardless of the associated land use (Table 3). However, in
1998
only 22
S. atromaculatus were captured in mining sites compared to 403 in
agricultural sites and 1,003 in forested sites. Semotilus atromaculatus abundance
increased along a pH gradient from mining sites to agricultural and forested sites
(axis two) (Figure 8b,Table 4).
1999 and Examination of Drought
Habitat Characteristics
In 1999, agricultural sites had a mean stream temperature of 16.2 ± 5.4 °C
(Figure 4a), and a mean pH of 8.2 ± 0.6 (Figure 4c). Dissolved oxygen values
ranged from 1.34 to 10.48 mg/I (Figure 4b). Agricultural sites had low
Table 4. Correlations between axes coordinates from 1998 NMS ordination and species abundance and physical
parameters. Correlations greater than r = ± 0.600 are shown, and r2 values> 0.500 indicate strong correlation.
Fish Species
Axis!
r
r2
Pimephales notalus
Notropis buccatus
Esox americanus
Luxilus
chrysocephalus
Lepomis macrochirus
Etheostoma nigrum
Lampetra aepyptera
Ameiurusnatalis
Phoxinus
erythrogaster
-0.832
-0.739
-0.737
0.692
0.546
0.543
-0.718
-0.683
-0.647
-0.628
-0.615
0.515
0.466
0.611
0.373
Axis2
r
Semotilus atromaculatus 0.867
Rhynichthys atratulus
0.680
Axis3
0.752
0.462
0.4 18
Etheostoma nigrum
Pimephales notatus
Ameiurus natalis
Luxilus
chrysocephalus
Notropis buccatus
Esox americanus
r
r2
0.797
0.786
0.709
0.63 6
0.699
0.68 5
0.489
0.469
0.645
0.4 16
0.618
0.503
0.395
0.378
Physical Parameters
Axis!
Stream Order
% Canopy
Elevation
Land Use
r
-0.843
0.797
0.757
0.691
r2
0.711
0.636
0.573
0.478
Axis2
r
pH
0.730
r2
0.532
Axis3
r
% Canopy
-0.82 1
Land Use
Stream Order
Stream Temperature
Elevation
Dissolved Oxygen
-0.740
0.674
0.654
-0.610
-0.600
r2
0.674
0.562
0.455
0.428
0.372
0.360
t'J
33
Table 5. Percent catch of all fishes captured annually and for entire study.
*indicates species contributing <1% of total fish catch. Species codes are the
first two letters of genus and species.
34
Table 5.
%
%
%
1998
1999
Total
Common Name
Species
Code
Catostomus commersoni
Hypentelium nigri cans
Moxostoma eryfhrurum
White Sucker
Northern Hogsucker
Golden Redhorse
CACO
RYNI
MOER
*
*
*
*
0
*
*
1
*
Lepomis cyanellus
Lepomis cyanellusX
Lepomisgibbosus
Lepomis gulosus
Lepomis humilis
Lepomis macrochirus
Lepomis megalotis
Green Sunfish
Green Sunfish Hybrids
Pwnpkinseed
Warmouth
Orangespotted Sunfish
Bluegill
Longear Sunfish
Spotted Bass
LECY
LECX
LEGI
LEGU
LEHU
LEMA
LEME
MIPU
5
3
5
*
*
*
*
0
*
0
*
*
*
0
*
Central Stoneroller
Steelcolor Shiner
Common Carp
Striped Shiner
Rosefin Shiner
Redfln Shiner
Golden Shiner
CAAN
CYWH
CYCA
LUCH
LYAR
LYUM
NOCH
NOBU
NOSP
NOST
Family
Scientific Name
Catastomidae
Catch Catch Catch
Centrachidae
Micropteruspunctulatus
1
1
*
*
1
*
*
*
*
2
0
*
1
*
*
*
0
*
4
6
4
*
0
*
1
Campostoma anomalum
Cyprinella whipplei
Cyprinus carpio
Luxilus chrysocephalus
Lythrurus ardens
Lythrurus umbratilis
Notemigonus chrysoleucas
Notropis buccatus
Notropis spilopterus
Notropis stramineus
Phoxinus erythrogaster
Pimephales notatus
Rhynichthys atratulus
Semotilus atromaculatus
Spotfin Shiner
Sand Shiner
Redbelly Dace
Bluntnose Minnow
Blacknose Dace
Creek Chub
Esox americanus
Grass Pickerel
ESAM
*
0
*
Ameiurusme!as
Ameiurus natalis
Black Bullhead
Yellow Bullhead
AMME
AMNA
*
*
*
2
*
1
Fantail Darter
Johnny Darter
Blackside Darter
ETFL
ETNI
PEMA
3
5
3
3
3
3
*
1
*
Least Brook Lamprey
LAAE
*
1
*
Silveçjaw
FIlER
PINO
RHAT
SEAT
1
*
*
1
*
12
3
9
0
2
7
20
6
30
1
*
*
2
7
8
9
6
48
17
6
34
Esocidae
Ictaluridae
Percidae
Etheostomaflabellare
Etheostoma nigrum
Percina maculata
Petromyzontidae
Lampetra aepyptera
35
conductivity (X = 253.3 ± 213.8 jiS/cm) (Figure 4d) and low turbidity values (X
= 5.11 ± 1.78 ntu) (Figure 4e). Percent canopy cover was low in agricultural sites
(X =3.2±7.1) (Figure 4f), and substrate composition included sand and silt
(Figure 5b).
Forested sites had a mean stream temperature of 16.2 ± 2.0°C (Figure 4a),
and a mean pH of 7.7 ± 0.3 (Figure 4c). Dissolved oxygen values ranged from 1.65
to 8.42 mg/I (Figure 4b). Forested sites had low conductivity (X = 167.9 ± 67.9
j.tS/cm) (Figure 4d) and low turbidity values (X = 3.23 ± 3.78 ntu) (Figure 4e).
Percent canopy cover was high in forested sites (X =66±29) (Figure 4.
Substrate composition included gravel, sand, and silt (Figure 5b).
Mining sites were characterized by acidic conditions, with pH ranging from
5.4 to 6.7 (X = 6.2 ± 0.6) (Figure 4c). Mean stream temperature was 16.8 ± 5.3 °C
(Figure 4a), and dissolved oxygen values ranged from 3.72 to 9.72 mg/i (Figure
4b). Mining sites had variable conductivity (X = 1368 ± 923 .tS/cm) (Figure 4e)
and turbidity (X
11.9 ± 9.1 ntu) (Figure 4d). Percent canopy cover was high in
mining sites (X =60±34) (Figure 4f), and substrate composition included gravel,
sand, and precipitate (Figure Sb).
Spring canopy cover was variable in forested and mining sites and may
have been associated with the timing of deciduous trees leafing out (Figure 4f). Silt
was the dominant substrate in summer at Sugarcamp Hollow, the only agricultural
site sampled in summer 1999 due to drought (Figure Sb). The proportion of gravel
in forested sites seemed to decrease in autumn and therefore the relative proportion
36
of sand and silt increased (Figure 5b), but some forested sites were not sampled in
this season due to drought. Substrate composition in agricultural and forested sites
exhibited a trend toward finer particle size in 1999 (Figure 5). Agricultural sites
exhibited less range in turbidity than in 1998 (Figure 4e), and mining sites were not
as highly acidic as in 1998 (Figure 4c).
Assemblage Characteristics
In 1999, agricultural assemblages were dominated by S. atromaculafus, P.
notatus, L. chrysocephalus, E. nigrum, and N. buccatus, comprising 82% of the
total catch for agricultural sites (Table 3). These five species were also dominant in
1998 agricultural assemblages, although species ranking changed in 1999.
Cyprinella whipplei, L. gulosus, and N. spilopterus were three new species captured
in 1999. As in 1998, agricultural assemblages tended to contain a greater number
and range of species (Figure 6), higher CPUE (Figure 7), and greater range in
CPUE compared to assemblages in forested or mining sites. In Sugarcamp Hollow,
the only persistent agricultural site in 1999, fish species numbers ranged from six to
11, and CPUE fluctuated from 20 to 508.
Forested assemblages in 1999 were similar to 1998 assemblages, except that
L. cyanellus was not a dominant species in 1999 (Table 3). Etheosromaflabellare
was rarely found in sites other than forested ones (Table 3). The number of species
captured in forested assemblages was between two and eight (Figure 6). Catch per
37
unit effort was relatively low in spring and autumn 1999, ranging from 22 to 110,
and was remarkably constant among sites and between seasons (Figure 7).
Semoilus atromaculatus and L. cyanellus were once again the dominant
species in mining assemblages, accounting for 90% of the fish catch (Table 3).
Catch per unit effort fluctuated more in 1999 with as many as 35 fish captured from
Sand Fork in autumn (Figure 7). Species numbers in mining sites remained low,
but increased marginally in one creek, Sand Fork, during autumn (Figure 6).
Integrating Assemblage Composition with Habitat Characteristics
The 1999 ordination of fish assemblages revealed a strong seasonal
influence (Figure 9b); regardless of land use all spring samples formed one cluster,
and a second cluster contained summer and autumn samples (Stress of 3dimensional ordination = 7.73; instability = 0.00001; cumulative r2
0.944). No
trend associated with land use was detected in the ordination. Dissolved oxygen
and stream order were also correlated with the ordination axes (Figure 9a).
Dissolved oxygen explained 40% of the variance; stream order and seasonal
patterns explained 37% and 17% of the variance, respectively (Figures 9a and 9b).
Luxilus chrysocephalus, E. nigrum, P. notatus, N buccatus, L. megalotis,
and L. macrochirus were associated with agricultural sites; L. chrysocephalus was
associated with low dissolved oxygen (axis one) while the other five species were
associated with stream order (axis two) (Figure lOa,, Table 6). Etheostoma
a.
0
0
o
Stream Order
N
2nd Order
0
0
G
A
1
Intermittent
i Order
3'' Order
Dissolved
Oxygen
N
.-
A
Ilifi
AAA
Axis!
AAA
Land Use
o
r2=.400
L
O
b.
Agricultural
Forested
Mining
Sugarcamp
Sand
Camp
Season
0
Long
Sand
Sugarcamp
II
0
Sugarcanip
I
A
AAJenkins
Long
Camp
AMeyers
A
Axis!
AA
A
Dissolve
Oxygen
Season
Spring
Summer
Autumn
r2=.400
Figure 9. NMS ordination of 1999 fish assemblages. Overlaid on the ordinations
are joint plots of environmental variables strongly correlated (r2> 0.5) with
ordination axes one and two (a) and one and three (b). Each symbol represents one
site/date. Only sites sampled more than once per year are identified by name (b).
Final stress of 3-dimensional ordination = 7.73; cumulative r2 = .944.
39
Intermittent
o 1Order
8 2n Order
Etni
N
Plnoj.eL:ma
A
3Order
dIED
A
ALA
Axis!
0
AAA
r2=.400
Land Use
o
b.
O
Agricultural
Forested
Mining
N
Season
p.-
0
Axisi
Spring
Summer
Autumn
r2=.400
Figure 10. NMS ordination of 1999 fish assemblages. Overlaid on the
ordinations are joint plots of individual fish species strongly correlated (r2 > 0.5)
with ordination axes one and two (a), and one and three (b). Species codes
consist of first two letters of genus and species (see Table 5). Final stress of
3-dimensional ordination = 7.73; cumulative r2 = .944.
Table 6. Correlations between axes coordinates from 1999 NMS ordination and species abundance and physical parameters.
Correlations greater than r = ± 0.600 are shown, and r2 values> 0.500 indicate strong correlation.
Fish Species
Axis!
r
Luxilus chrysocephalus
Semotilus
afromaculatus
Pimephales notatus
-0.730
0.534
-0.696
-0.667
0.484
0.445
r2
Axis2
r
Etheostoma nigrum
Pimephales notatus
Notropis buccatus
Lepomis megalotis
Lepomis macrochirus
Phoxinus
erythrogaster
Notropis stramineus
Rhynichthys atratulus
Lampetra aepyptera
0.796
0.773
0.738
0.634
0.598
0.545
0.7 17
0.5 14
0.5 10
0.714
-0.676
0.663
-0.649
0.606
r2
r
Axis3
Etheostomaflabellare
-0.762
Axis3
r
Season
0.753
r2
0.580
0.457
0.440
0.42 1
0.368
Physical Parameters
Axis!
Dissolved Oxygen
Stream Temperature
r
0.782
-0.660
r2
0.612
0.436
Axis2
Stream Order
Mean Stream Width
Elevation
r
0.840
0.695
-0.663
r2
0.706
0.484
0.439
r2
0.567
0
41
flabellare, a dominant fish in forested assemblages, was associated with season
(axis three) (Figure lob, Table 6).
Response offish assemblages following drought
In 1999 fish assemblages from all land use types exhibited seasonal
variation, as spring assemblages differed from autumn assemblages. This seasonal
variation was not apparent in 1998, and one of the biggest differences between the
two years was the occurrence of drought in 1999. Even with a dominant seasonal
pattern in 1999, differences in fish assemblages persisted among the different land
use types (Figure 11). Because forested sites were classified as intermittent and
agricultural sites were perennial streams, I contrasted these sites in autumn 1998
before drought conditions with the same sites in autumn 1999 following drought
conditions.
Overall, fish assemblages were similar within a location between autumn
1998 and autumn 1999 for both land uses, with the exception of Camp Creek
(Table 7a). However, when assemblages were compared by land use, differences
between forested and agricultural sites emerged. Comparison of autumn 1998 and
autumn 1999 samples revealed higher similarity among forested sites than among
agricultural sites; forested sites were grouped together with 90% of the information
remaining, while only 40% of the information remained after agricultural sites were
grouped (Figure 12).
Table 7*. Morisita' s Index values for assemblage comparisons between autumn 1998 and autumn 1999 within a site.
Forested_Sites
Autumn 1998 - Autumn 1999
0.76
0.63
Jenkins
Meyer
Agricultural Sites
Autumn 1998 - Autumn 1999
0.29
0.88
0.83
Camp
Long
Sugarcainp
Table 7b. Morisita's Index values for autumn 1998 and autumn 1999 assemblage comparisons between sites within
agricultural and forested land uses. Sites were compared for assemblage similarity within a season.
Agricultural Sites
Forested Sites
Autumn 1998
Autunin 1999
Jenkins-Meyer
0.51
0.81
Autumn 1998
Autumn 1999
Camp-Long
0.85
0.35
Camp-Sugarcamp
0.82
0.22
Long-Sugarcamp
0.78
0.67
q.j
Distance (Objective Function)
I
3.IE+01
1.6E+01
2.SE-01
I
I
I
I
4.6E+01
I
I
6.2E+01
I
I
Information Remaining (%)
I
50
75
100
I
I
I
I
0
25
I
I
I
Camp 98
Long 98
Sugar 98
Camp 99
Sugar 99
Long 99
Jenkin 98
Jenkin 99
Meyer 99
Meyer 98
Figure 12. Hierarchical agglomerative cluster analysis of autumn 1998 and 1999 fish assemblages for
agricultural and forested sites.
45
Forested assemblages were more similar following drought conditions in
autumn 1999 than in autumn
(Table 7b, Figure
1998
12).
Assemblages in all
forested sites had reduced CPUE and reduced species numbers following drought
in autumn 1999 (Figure 13). Species composition of both forested assemblages in
1999 was
reduced to three dominant species: S. atromaculatus and the two dace
species (Table
8).
These were the same three species that accounted for over 91%
1999 catch in forested sites (Table 3). Less abundant species in
of the
absent from the assemblages in
1999,
1998 were
with the exception of Campostoma
anomalum (Table 8).
Agricultural assemblages in autumn 1998 and autumn 1999 were highly
variable (Figure
12,
Table
8),
and agricultural assemblages were more similar prior
to drought than after (Table 7b). Three new species (N.
and L. gulosus) were
spilopterus, C. whipplei,
captured only in autumn 1999 (Table 8). Rare species (i.e., <
1% of total catch for a given location) in
1998
remained rare, were absent, and/or
were common in 1999 assemblages (Table 8). Catch per unit effort decreased in
two out of three agricultural sites in autumn 1999 (Figure 13b).
Spatial and Temporal Variation in Fish Assemblages
The ordination of combined
1998
and 1999 fish assemblages revealed
strong differences between years, and an association with land use (Figure 14)
(stress of 3-dimensional ordination = 9.90; instability = 0.00001; cumulative r2
0.9 16).
Total catch was highest in
1998
with 4,819 fishes compared to
1,625
=
fishes
18
a.
C'
16
14
I
Jenkins
0 Meyer
XMean
12
10
8
4
OCamp
0 Long
2
A Sugarcamp
6
0
900
* Mean
b.
___________________________________________
0 Jenkins
0 Meyer
Z Mean
o Camp
800
700
I]
A Sugarcamp
Long
* Mean
I
600
500
400
300
200
100
0
A_____________
1998
Agricultural
1999
1998
Forested
1999
Figure 13. Comparison of autumn 1998 and autumn 1999 fish assemblages in forested and agricultural
sites for number of species captured (a) and catch per unit effort (CPUE) (b). Error bars indicate one standard
deviation of the mean.
47
Table 8. Composition of autumn 1998 and autumn 1999 fish assemblages in
forested and agricultural sites. Species are listed in descending order based on
overall percent catch (see Table 5). A "x" indicates >1% of the total catch for a
given location; a "+" indicates 1% of the total catch for a given location. Species
indicated in bold are species captured only in autumn 1999.
+
x
maculata Percina
spiopterus Notropis
x
erythrurm Moxostoma
ardens Lythrurus
megalotis Lepomis
macrochirus Lepomis
x
x
x
x
x
+
x
+
+
+
+
+
x
x
x
gulosus Lepomis
x
+
x
+
+
x
+
+
x
x
x
x
+
+
+
+
x
x
x
x
x
+
x
+
+
+
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
1999
1998
1999
Autumn
1998
1999
Autumn
1998
x
+
x
x
x
x
x
x
x
x
1999
1998
Autumn
1999
Autumn
1998
Autumn
Autumn
Sugarcamp
Long Sites A2ricultural
+
+
Autumn Autumn
Camp
Autumn
+
Meyer
aepyptera Lampetra
americanus Esox
whipplei Cyprinella
commersoni Catastomus
melas Amelurus
umbratilis Lythrurus
anoinalum Cainpostoma
natalis Ameiurus
stramineus Notropis
nigrum Etheostoma
Etheostomaflabellare
chrysocephalus Luxilus
cyanellus Lepomis
atratulus Rhynichthys
erythrogaster Phoxinus
buccatus Notropis
notatus Pimephales
atromaculatus Semotilus
Species
Autumn
Jenkins
Sites Forested
8. Table
49
in 1999. Twenty-nine species were captured in 1998 versus 26 species in 1999.
The overall pattern of land use was somewhat muted by seasonal patterns and lowwater conditions in the second year, but land use was a strong correlate of axis one
(Table 9, Figure 14) and explained 47% of the variance. Fish assemblage patterns
were not distinguishable among basins in the ordination. Pimephales notatus, L.
chrysocephalus, N. buccatus, and E. nigrum were associated with agricultural sites
(Table 9, Figure 14), and together these four species accounted for 58% of the
overall catch for agricultural sites. The two dace species, P. erythrogaster and R.
atratulus, were associated with forested sites (Table 9, Figure 14b), and they
accounted for 31% of the overall catch for forested sites. Semotilus atromaculatus
were the most abundant species in forested sites (5 0%), but they occurred in high
abundance in assemblages of all land use types.
50
a.
°
0
Agricultural
D
o
0
cP
0
Luch
Etni
Forested
.
A
A
A
Pino
L
(#2
A
.
1'
.
aD
U
A
Axis 1
Mining
AA
A
r2.466
b.
.
S
0
r
cPiIio
U
U
0
SD
00
a
U
A
tni
OLuch
A
(#2
LA
Rhat
Axis 1
iPher
A A
A
r2=.466
Figure 14. NMS ordination, based on fish assemblage data, projected onto
axes one and two (a), and onto axes one and three (b), for 1998 (filled
symbols) and 1999 (empty symbols). Overlaid on the ordinations are joint
plots of species strongly correlated (r2> 0.5) with the ordination; species
codes consist of first two letters of genus and species (see Table 5). Each
symbol represents one site/date. Final stress of 3-dimensional ordination =
9.90; cumulative r2 = .9 16.
Table 9. Correlations between axes coordinates from 1998 and 1999 NMS ordination and species abundance and physical
parameters. Correlations greater than r = ± 0.600 are shown, and r2 values> 0.500 indicate strong correlation.
Fish Species
Axis!
r
Pimephales notatus
Luxilus chrysocephalus
Notropis buccatus
Etheosloma nigrum
Ameiurus natalis
Lepomis macrochirus
-0.890
-0.825
-0.821
-0.754
-0.609
-0.606
r2
Axis2
r
r2
0.791
Axis3
Phoxinus erythrogaster
Rhynichthys atratulus
Semotilus atromaculatus
0.68 1
0.673
0.569
r
-0.877
-0.876
-0.690
r2
0.770
0.768
0.476
0.37 1
0.368
Physical Parameters
Axis!
Land Use
Stream Order
Elevation
%Canopy
r
0.802
-0.671
0.640
0.639
r2
Axis2
r
0.642
Season
0.921
r2
0.848
Axis3
Stream Order
r
0.617
r2
0.380
0.45 1
0.409
0.408
(fl
52
DISCUSSION
Both land use and natural disturbance (i.e., drought) were important in
explaining fish assemblage patterns during a two-year study of tributary streams
within the Wayne National Forest in southern Ohio. With the exception of the
ubiquitous creek chub (S.
atromaculatus)
and green sunfish (L. cyanellus), fish
assemblages within agricultural and forested tributaries were dominated by species
that rarely occurred in the other land use type. Acid mine drainage tributaries
contained few to no fish. Large-scale reach characteristics (i.e., stream order,
elevation, and canopy cover) explained the majority of the variance in fish
assemblage structure within
important. But in
1999,
1998;
water chemistry, specifically pH, was also
large-scale reach characteristics were secondary to water
chemistry (i.e., dissolved oxygen), and seasonal variation was important. These
results suggest that large-scale reach characteristics and chemical signatures related
to land use are important to fish assemblage structure, but in times of
environmental fluctuation, water chemistry or other site-specific variables may also
be of importance due to physiological tolerances and limitations of fishes.
It is common for fish assemblages to be associated with land use
characteristics within a drainage basin. As indicators that land use influences fish
assemblages, percent land use in the watershed was correlated with biotic integrity
in Wisconsin (Wang et
aL 1997),
in a Michigan watershed (Roth et
in Nebraska streams (Frenzel and Swanson
1996).
al. 1996),
In New Zealand, fish
and
53
distribution has been associated with land use, including habitat and elevational
differences (Hanchet 1990, Jowett and Richardson 1996). Riparian deforestation
has also been linked to changes in assemblage structure in southeastern streams
(Jones et al. 1999). Even so, most studies have focused on environmental
degradation rather than patterns in fish assemblage structure within various land
use types. Increased species richness in agricultural sites may be associated with
stream order, a natural gradient (Sheldon 1968, Vannote et al. 1980, Rahel and
Hubert 1991), or with open (i.e., lack ol) canopy, caused by landscape alteration
(Murphy etal. 1981). While trends in stream order and elevation were associated
with fish assemblages in WNF, neither variable could fully describe assemblage
groupings. All sites were categorized as headwater streams (sensu Vannote et al.
1980), and other measurements of stream size, including channel width and depth,
were not strongly correlated in the analyses. Land use often incorporates elements
of natural landscape gradients (e.g., stream order and elevation) with human
impacts (e.g., riparian zone alterations). Therefore, land use and natural gradients
were confounded, and that may explain why stream order and elevation were
associated with the fish assemblages. Given the small difference in stream orders
and minimum elevation change, fish assemblage patterns in small tributary streams
of WNF can be described best by land use patterns.
The distribution of common species in tributaries in WNF is similar to other
studies. For example, P. erythrogaster is commonly found only in headwaters
(Trautman 1981, Stewart et al. 1992) and Hawkes etal. (1986) reported that S.
54
atromaculatus in the Kansas River system preferred intermittent streams. In a midAtlantic classification study, McCormick etal. (2000) also found that headwater
streams were dominated by two ubiquitous species, R. atratulus and S.
atromaculatus, as did Pinder and Morgan (1995) in Maryland Appalachian plateau
streams. Additionally, both S. atromaculatus and R atratulus were more abundant
in smaller streams in Iowa than in larger ones, and although P. notatus is often
depicted as a colonizing species, the species is generally more abundant in larger
streams (Starrett
1950).
The high tolerance of S. atromaculatus to acid mine
drainage conditions was also documented in a Pennsylvania study (Letterman and
Mitsch
1978).
Forested streams within the WNF are often intermittent and prone to annual
desiccation. Such streams may be considered environmentally harsh, containing
assemblages of limited species composition. In Illinois streams, Smith (1971)
included desiccation as a factor responsible for changes in fish populations, and
also in reducing the range of some species. In the Des Moines River, Starrett
(1950)
attributed limited distribution of certain fishes to their inability to cope with
harsh physiochemical and crowded conditions in intermittent streams. Species
common in intermittent streams may exhibit greater tolerances to temperature,
oxygen, and pH extremes (Matthews and Styron
1981)
and high vagility (Poff and
Ward 1989). Wide tolerances to physiochemical extremes, or eulytropy, are
helpful to organisms that remain in isolated pools during desiccation as these
organisms may face suffocation, heat stress, crowding, starvation, and increased
55
predation. In the WNF, intennittent assemblages were dominated by few species,
and dominance patterns changed very little over time. Meador and Matthews
(1992) also observed that intermittent assemblages were temporally consistent.
The consistency of assemblages is likely due to the hardiness of headwater species
like S. atromaculatus (Shelford 1937) and P. erythrogaster (Matthews and Styron
1981) combined with the ability to rapidly recolonize desiccated areas. Phoxinus
oreas is a congener ofF. erythrogaster and occupies similar habitat (Jenkins and
Burkhead 1994).
Assemblages in all sites changed following drought. Agricultural sites
within the WNF were perennial streams and they exhibited greater variability
following drought. Intermittent assemblages in WNF were most similar following
drought, implying that assemblages experiencing desiccation on a regular and
perhaps predictable basis may be better suited to cope with drought than those
assemblages within perennial systems. But not all intermittent streams became
desiccated every year, and in those years additional species, such as L.
cyanellus,
may enter the assemblages of non-desiccated streams; however, these additional
species were eliminated during periods of desiccation. A similar phenomena
occurs in southwestern montane streams where floods eliminate introduced
mosquitofish and allow native topminnows to persist; in streams without floods,
introduced mosquitofish have extirpated the native topminnows (Meffe 1984).
Moyle and Light (1996) also attributed the success or failure of species invasions to
abiotic conditions. Even though short-term changes may be dramatic, faunas
usually recover to nonnal levels, and drought typically does not have long term
effects on streams (Lanmore
et al. 1959,
Bayley and Osborne 1993). A number of
studies have indicated that small stream assemblages are persistent and stable over
time (Moyle and Vondracek 1985,
Matthews etal.
1988,
Ross et al. 1985,
Meffe and Berra
1988)
Minckley and Meffe
1987,
in spite of short term changes
resulting from drought and/or floods.
Some physical habitat differences among land use types did not influence
fish assemblage structure in this study. Substrate in agricultural sites was finer than
in forested and mining sites, yet substrate was not associated with fish assemblages.
However, dominant fishes in agricultural sites prefer sandy creek beds (N
buccatus), or are silt-tolerant (P. notatus and L. chrysocephalus); dominant fishes
in forested sites are usually associated with coarser substrates (Trautman 1981).
Conductivity was greatest in mining sites while forested sites were typically less
turbid, but neither variable was strongly correlated with fish assemblage structure.
In acid mine drainage studies, pH and conductivity are negatively correlated due to
increased leaching under low pH conditions (Dills and Rogers
1974),
but
conductivity may also reflect underlying geology and soils (Frenzel and Swanson
1996). Low turbidity is common in mining streams due to precipitation by acid
effluents (Parsons
1955),
but turbidity also may be increased by human induced
landscape alterations (Karr
1991).
Stream habitats are hierarchically nested from
the watershed or landscape level down to the microhabitat, and each level of habitat
is controlled by characteristics of the higher level, ultimately constrained by
57
watershed characteristics. While patterns in substrate, conductivity, and turbidity
are evident, the larger scale variables revealed by the analyses (e.g., stream order,
elevation, canopy cover) apparently have greater influence on fish assemblage
structure because they help define lower level characteristics.
Classification of tributary streams within the WNF will be more accurate
when both physical and biological components are incorporated. Multiple
environmental variables at multiple scales influenced fish assemblage structure
within the WNF, so channel, or geomorphologic, classification must be tightly
linked with the biota. Classifiers recognizing the importance of both spatial and
temporal variation will gain a better understanding of healthy stream function. A
hierarchical channel classification is strongly recommended because the
hierarchical nature of streams influences fish assemblage structure over multiple
scales (Hawkins et
1999).
al. 1993,
Maxwell et
The approach of Maxwell et
al. 1995,
al. (1995)
Montgomery and Buffington
is appropriate for initial
classification efforts because it incorporates important measurements to
hierarchically map and define characteristics for an overall landscape perspective
rather than opting for a single classification system. Fishes are good indicators of
environmental integrity and can be used to monitor ecosystem health and changes
(Karr
1981, Moyle 1994).
In addition, macroinvertebrates are good indicators
especially where fishes are not abundant (e.g., mining streams), and may be used in
conjunction with fish assemblages. Canopy cover, an important variable
explaining fish assemblage patterns, and riparian zones are known to be important
58
to stream function (Karr and Schiosser
1996);
1978,
Gregory etal.
1991,
Roth etal.
therefore, vegetation surveys can provide additional perspective of the
aquatic-terrestrial linkage.
Land management must incorporate the hierarchical nature of streams into a
holistic landscape perspective. Dale etal.
(2000)
proposed five ecological
principles necessary for land use management: time, species, place, disturbance,
and landscape. These principles recognize the following: 1) ecosystems are
dynamic and potentially exhibit seasonal and annual changes in response to
weather (sensu Odum
1969), 2)
species composition and diversity may change in
response to land use alterations, 3) local environmental conditions may constrain
land use patterns, 4) disturbances (both natural and man-induced) influence
populations, communities, and ecosystems and, 5) site characteristics may be
explained by both local and landscape attributes. In this study I have incorporated
these principles by 1) documenting interannual variation in fish assemblages, 2)
identifying patterns in fish assemblages within agricultural, forested, and mining
land use types, 3) identifying patterns in environmental variables among
agricultural, forested, and mining land use types, 4) examining the response of
stream fish assemblages to drought, and 5) identifying enviromnental
characteristics strongly correlated with fish assemblages. By integrating fish and
physical variables in a multivariate gradient analysis I interpreted fish assemblage
structure in association with environmental correlates. Descriptive results from this
type of multivariate analysis can be used to generate hypotheses for future
experimental studies and also provide a template for future monitoring programs or
restoration activities. For example, in acid mine drainage tributaries water quality
and fish assemblages fluctuated throughout the two years of this study. The
variation in these sites along with similarity in physical habitat to forested sites
suggests that remediation of pollution in acid mine drainage tributaries could allow
recovery of fishes. Because not all portions of a stream network are affected by
landscape alteration, unimpacted segments may serve as reffigia (Sedell
et al.
1990), allowing fishes to recolonize adjacent areas. Recovered mining assemblages
may be predicted to resemble forested assemblages based on the environmental
similarities of the sites. Similarly, lack of canopy cover was a significant descriptor
of agricultural assemblages, and therefore assemblage changes could result from
restoration of the riparian zone along agricultural croplands. Because
anthropogenic landscape changes like agriculture may leave a signature on fish
assemblages for decades, recovery of fishes may lag behind habitat restoration
efforts (Harding et
al.
1998). These descriptive results may also be used to identify
areas of impact, determine effects of historical land use, locate special management
areas, or follow assemblage changes following channel modifications and/or
restoration activities.
BIBLIOGRAPHY
Anderson, J.R. 1970. Major land-uses. Pages 157-160. The national atlas of the
United States. A.C. Gerlach. Washington, DC., U.S. Geological Survey,
U.S. Printing Office.
Barnes, M.D. and R.F. Carline. 1978. A survey of the eastern sand darter,
Ammocryptapellucida (Putnam), and other fishes in Symmes Creek and
Pine Creek, Wayne National Forest, Ohio. A cooperative project of the
Ohio Cooperative Fishery
Research Unit and the U.S. National Forest
Service, Eastern Region. 24 pages.
Bart, H.L. 1989. Fish habitat association in an Ozark stream. Environmental
Biology of Fishes 24(3): 173-186.
Bayley, P.B, and L.L. Osborne. 1993. Natural rehabilitation of stream fish
populations in an Illinois catchment. Freshwater Biology 29: 295-300.
Beschta,, R.L. and W.S. Platts. 1986. Morphological features of small streams:
significance and function. Water Resources Bulletin 22(3): 369-379.
Dale, V.H., S. Brown, R.A. Haeuber, N.T. Hobbs, N. Huntly, R.J. Naiman, WE.
Riebsame, M.G. Turner, and T.J. Valone. 2000. Ecological principles and
guidelines for managing the use of land. Ecological Applications 10(3):
639-670.
Dills, G. and Rogers, D.T. 1974. Macroinvertebrate community structure as an
indicator of acid mine pollution. Environmental Pollution 6: 239-262.
Doppelt, B., M. Scurlock, C. Frissell and J. Karr. 1993. Entering the watershed.
Washington, D.C., Island Press.
Frenzel, S.A. and R.B. Swanson. 1996. Relations of fish community composition
to environmental variables in streams of central Nebraska, USA.
Environmental Management 20(5): 689-705.
Frissell, CA., W.J. Lisa, C.E. Warren and M.D. Hurley. 1986. A hierarchical
framework for stream classification: Viewing streams in a watershed
context. Envfronmental Management 10(2): 199-214.
Gregory, S.V., F.J. Swanson, WA. McKee and K.W. Cummins. 1991. An
ecosystem perspective of riparian zones. Bioscience 41: 540-551.
61
Hammond, E.H. 1964. Classes of land-surface form in the forty-eight states, USA.
Annals of the Association of American Geographers 54: map supplement 4.
Hanchet, S.M. 1990. Effect of land use on the distribution and abundance of
native fish in tributaries of the Waikato River in the Hakarimata Range,
North Island, New Zealand. New Zealand Journal ofMarine and
Freshwater Research 24:159-171.
Harding, J.S., E.F. Benfield, P.V. Boistad, G.S. Helfman and E.B.D. Jones. 1998.
Stream biodiversity: the ghost of land use past. Proceedings of the National
Academy of Sciences 95: 14843-14847.
Hawkes, C.L., D.L. Miller, and W.G. Layher. 1986. Fish ecoregions of Kansas:
stream fish assemblage patterns and associated environmental correlates.
Environmental Biology of Fishes 17(4): 267-279.
Hawkins, C.P., J.L. Kershner, P.A. Bisson, M.D. Bryant, L.M. Decker, S.V.
Gregory, D.A. McColiougb, C.K. Overton, G.H. Reeves, Ri. Steedman,
and M.K. Young. 1993. A hierarchical approach to classif'ing stream
habitat features. Fisheries 18(6): 3-12.
Holeski, P.M., S. Cook, K. Searis, K. Specht, and P. Savage. 1992. An inventory
of fishes of Symmes Creek - 1991 report. A cooperative project of the
USDA Forest Service, Wayne National Forest and the University of Rio
Grande. 31 pages.
Holeski, P.M., K. Searis, and K. Specht. 1993. An inventory of fishes of Symmes
Creek - 1992 report. A cooperative project of the USDA Forest Service,
Wayne National Forest and the University of Rio Grande. 34 pages.
Holeski, P.M., J. Circle, M. Denny, and B. Johnston. 1995. An inventory of fishes
of Symmes Creek - 1995 report. A cooperative project of the USDA Forest
Service, Wayne National Forest and the University of Rio Grande. 38
pages.
Hynes, H.B.N. 1975. The stream and its valley. Verhandlungen, Internationale
Vereinigung fur Theoretische undAugewandte Limnologie 19: 1-15.
Jenkins, R.E. and N.M. Burkhead 1994. Freshwater fishes of Virginia. Maryland,
American Fisheries Society.
Jones, E.B.D., G.S. Helfinan, J.O. Harper, and P.V. Bolstad. Effects of riparian
forest removal on fish assemblages in southern Appalachian streams.
Conservation Biology 13(6): 1454-1465.
62
Jowett, I.G. and J. Richardson. 1996. Distribution and abundance of freshwater
fish in New Zealand Rivers. New Zealand Journal ofMarine and
Freshwater Research 30: 239-255.
Karr, J.R. 1981. Assessment of biotic integrity using fish communities. Fisheries
6(6): 21-27.
Karr, J.R. 1991. Biological integrity: a long-neglected aspect of water resource
management. Ecological Applications 1(1): 66-84.
Karr, J.R. and I.J. Schiosser. 1978. Water resources and the land-water interface.
Science 201: 229-234.
Krolczyk, J.C. 1954. Gazetteer of Ohio Streams. Columbus, Ohio.
Kruskal, J.B. 1964. Nonmetric multidimensional scaling: a numerical method.
Psychometrika 29:115-129.
Larimore, R.W., W.F. Childers, and C. Heckrotte. 1959. Destruction and reestablishment of stream fish and invertebrates affected by drought.
Transactions of the American Fisheries Society 88: 261-285.
Letterman, R.D. and W.J. Mitsch. 1978. Impact of mine drainage on a mountain
stream in Pennsylvania. Environmental Pollution 17: 53-73.
Master, L. 1990. The imperiled status of North American Aquatic Animals.
Biodiversity Network News (The Nature Conservancy) 3(3): 1-2, 7-8.
Mather, P.M. 1976. Computational methods of multivariate analysis in physical
geography. New York, John Wiley and Sons.
Matthews, W.J. and J.T. Styron. 1981. Tolerance of headwater vs. mainstream
fishes for abrupt physiochemical changes. The American Midland
Naturalist 105(1): 149-158.
Matthews, W.J. 1986. Fish faunal structure in an Ozark stream: Stability,
persistence and a catastrophic flood. Copeia 2: 3 88-397.
Matthews, W.J., R.C. Cashner, and PP. Gelwick. 1988. Stability and persistence
of fish faunas and assemblages in three midwestern streams. Copela 4:
945-955.
63
Matthews, W.J. 1990. Spatial and temporal variation in fishes of riffle habitats: A
comparison of analytical approaches for the Roanoke River. American
Midland Naturalist 124(1): 31-45.
Maxwell, J.R., C.J. Edwards, M.E. Jensen, S.J. Paustian, H. Parrott, and D.M. Hill.
1995. A hierarchical framework of aquatic ecological units in North
America (Nearctic zone). USDA Forest Service North Central Forest
Experimental Station General Technical Report NC-176.
McCormick; F.H., D.V. Peck, and D.P. Larsen. 2000. Comparison of geographic
classification schemes for Mid-Atlantic stream fish assemblages. Journal
of the North American Benthological Society 19(3):385-404.
McCune, B. and M.J. Mefford. 1999. Multivariate analysis of ecological data,
Version 4.14. Gleneden Beach (OR), MJM Software Design.
Meador, M.R. and W.J. Matthews. 1992. Spatial and temporal patterns in fish
assemblage structure of an intermittent Texas stream. American Midland
Naturalist 127: 106-114.
Meffe, G.K. 1984. Effects of abiotic disturbance on coexistence of predator-prey
fish species. Ecology 65: 1525-1534.
Meffe, G.K. and T.M. Berra. 1988. Temporal characteristics of fish assemblage
structure in an Ohio stream. Copeia 3: 684-690.
Menzel, B.W., J.B. Barnum, and L.M. Antosch. 1984. Ecological alterations of
Iowa prairie-agricultural streams. Iowa State Journal of Research 59: 5-3 0.
Miller, R.R., J.D. Williams and J.E. Williams. 1989. Extinctions of North
American fishes during the past century. Fisheries 14(6): 22-3 6.
Minckley, W.L. and Meffe, G.K. 1987. Differential selection by flooding in
stream-fish communities of the arid American Southwest. In W.J.
Matthews and D.C. Hems (eds.), Community and evolutionary ecology of
North American stream fishes. Oklahoma, University of Oklahoma Press.
Montgomery, D.R. and J.M. Buffington. 1999. Channel processes, classflcation,
and response, p. 13-42. In R. Naiman and R. Bilby (eds.), River Ecology
and Management. Springer-Verlag, New York.
Morisita, M. 1959. Measuring of interspecies association and similarity between
communities. Mem. Fac. Sci. Kyrushu Univ., Ser. E Biology 3: 65-80.
Moyle, P.B. 1994. Biodiversity, biomonitoring, and the structure of stream fish
communities. In S.L. Loeb and A. Spacie (eds.), Biological monitoring of
aquatic systems. Florida, Lewis Publishers.
Moyle, P.B. and T. Light. 1996. Fish invasions in California: do abiotic factors
determine success? Ecology 77(6): 1666-1670.
Moyle, P.B. and B. Vondracek. 1985. Persistence and structure of the fish
assemblage in a small California stream. Ecology 66:1-13.
Murphy, M.L., C.P. Hawkins and N.H. Anderson. 1981. Effects of canopy
modification and accumulated sediment on stream communities.
Transactions of the American Fisheries Society 110(4): 469-478.
Naiman, R.J., J.J. Magnuson, D.M. McKnight, and J.A. Stanford. 1995. Th
freshwater imperative: a research agenda. Island Press, Washington, DC.
Odum, E.P. 1969. The strategy of ecosystem development. Science 164: 262-270.
Ohio Department of Natural Resources. 1994. Statewide Land Cover Map.
Ohio Department of Natural Resources. 1998. Monthly Water Inventory Report for
Ohio (January December).
Ohio Department of Natural Resources. 1999. Monthly Water Inventory Report for
Ohio (January December).
Parsons, J. 1955. The effects of acid strip mine pollution in the ecology of a
central Missouri stream. Ph.D. Thesis, Missouri University.
Pinder, M.J. and R.P. Morgan. 1995. Interactions of pH and habitat on cyprinid
distributions in appalachian streams of Maryland. Transactions of the
American Fisheries Society 124: 94-102.
Poff, N.L. nd J.V. Ward. 1989. Implications of streamfiow variability and
predictability for lotic community structure: a regional analysis of
streamfiow patterns. Canadian Journal of Fisheries andAquatic Sciences
46: 1805-1818.
Rahel, F.J. and W.A. Hubert. 1991. Fish assemblages and habitat gradients in a
Rocky mountain-Great Plains stream: biotic zonation and additive patterns
of community change. Transactions of the American Fisheries Society 120:
319-332.
Rose, C.E. 2000. Environmental variables as Predictors of Fish Assemblages in
the Tillamook Basin, Oregon. MS Thesis, Oregon State University,
Corvallis, OR.
Rosgen, D.L. 1994. A classification of natural rivers. Catena 22: 169-199.
Ross, S.T., W.J. Matthews, and A.A. Echelle. 1985. Persistence of stream fish
assemblages: effects of environmental change. The American Naturalist
126(1): 24-40.
Roth, N.E., J.D. Allan, and D.L. Erickson. 1996. Landscape influences on stream
biotic integrity asssessed at multiple spatial scales. Landscape Ecology
11(3): 141-156.
Salmaso, N. 1996. Seasonal variation in the composition and rate of change of the
phytoplankton community in a deep subalpine lake (Lake Garda, Northern
Italy). An application of nonmetnc multidimensional scaling and cluster
analysis. Hydrobiologia 337: 49-68.
Sedell, J.R. G.H. Reeves, F.R. Hauer, J.A. Stanford and C.P. Hawkins. 1990. Role
of refugia in recovery from disturbances: modern fragmented and
disconnected river systems. Environmental Management 14(5): 711-724.
Schiosser, I.J. 1991. Stream fish ecology: a landscape perspective. BioScience
41(10): 704-7 12.
Sheldon, A.L. 1968. Species diversity and longitudinal succession in stream
fishes. Ecology 49(2): 194-198.
Shelford, V.E. 1913. Animal communities in temperate North America. Chicago,
University of Chicago Press.
Simonson, T.D., J. Lyons, and P.D. Kanehl. 1994. Guidelines for evaluating fish
habitat in Wisconsin streams. USDA Forest Service North Central Forest
Experiment Station General Technical Report NC-164.
Smith, P.W. 1971. Illinois streams: a classification based on their fishes and an
analysis of factors responsible for disappearance of native species. Illinois
Natural History Survey Biological Notes 76.
Sokal, R.R. and J.F. Rohlf 1987. Introduction to biostatistics. New York,
Freeman.
Sprules. W.G. 1980. Nonmetric multidimensional scaling analyses of temporal
variation in the structure of limnetic zooplankton communities.
Hydrobiologia 69(1-2): 13 9-146.
Stanford, J.A. and J.V. Ward. 1992. Management of aquatic resources in large
catchments: recognizing interactions between ecosystem connectivity and
environmental disturbance, p. 9 1-124 In R.J. Naiman (ed.), Watershed
Management: Balancing Sustainability and Environmental Change.
Springer-Verlag, New York.
Starrett, W.C. 1950. Distribution of the fishes of Boone County, Iowa, with
special references to the minnows and darters. American Midland
Naturalist 43(1): 112-127.
Stephenson, M., G. Mierle, R.A. Reid, and G.L Mache. 1993. Effects of
experimental and cultural lake acidification on littoral benthic
macroinvertebrate assemblages. Canadian Journal of Fisheries and
Aquatic Sciences 51: 1147-1161.
Stewart, B.G., J.G. Knight, and R.C. Cashner. 1992. Longitudinal distribution and
assemblages of fishes of Byrd's Mill creek, a southern Oklahoma Arbuckle
Mountain stream. The Southwestern Naturalist 37(2): 138-147.
Trautman, M.B. 1981. The Fishes of Ohio, 2'' ed. Columbus, The Ohio State
University Press.
Vannote, R.L. G.W. Minshall, K.W. Cummins, J.R. Sedell and C.E. Cushing.
1980. The river continuum concept. Canadian Journal of Fisheries and
Aquatic Sciences 37: 130-137.
Wang, L., J. Lyons, P. Kanehl, and R. Gatti. 1997. Influences of watershed land
use on habitat quality and biotic integrity in Wisconsin streams. Fisheries
22(6): 6-12.
Warner, R.W. 1971. Distribution of biota in a stream polluted by acid minedrainage. The Ohio Journal of Science 71(4): 202-2 15.
Warwick, R.M. and K.R. Clarke. 1991. A comparison of some methods for
analysing changes in benthic community structure. Journal of the Marine
BiologicalAssociation of the United Kingdom 71: 225-244.
Wolda, H. 1981. Similarity indices, sample size and diversity. Oecologia (Berl)
50: 296-302.
67
Wright, K.K. 2000. From Continua to Patches: Longitudinal Patterns in the
Middle Fork of the John Day River, Oregon. Ph.D. Thesis, Oregon State
University, Corvallis, OR.
68
APPENDIX
APPENDIX
I collected cross-section data for the purpose of Rosgen stream
classification (Rosgen 1994) at 13 fish sites and at an additional 38 sites within the
WNF. Bankfull widths ranged from 2.2 to 16.5 meters; entrenchment values
ranged from 1.2 to 72; width/depth ration ranged from 6.6 to 87.9. Values for
gradient were generally low, ranging from 0 to 2%. Sinuosity was atypically low
with values usually from 1.0 to 1.2 with only three exceptions. These low values
are associated with "A" type channels and do not concur with the other delineative
measurements. However, sinuosity carries the least weight of all the delineative
measurements and is one of the most frequently measured incorrectly (Rosgen
1994). By excluding sinuosity, 40 of 51 sites could be classified by entrenchment,
w/d ratio,
and gradient;
three B, nine Bc, 11 C or CC, 13 E, and four F type
channels resulted. Disagreement between entrenchment and w/d ratio values was
the reason 11 sites
could not
be classified. However, classification was successful
with 10 of the remaining 11 sites when the "fudge factors" were applied: two Bc,
one E, and six Gc type channels resulted,
and
one channel could be either a Bc or
Gc channel.
B channel types are moderately entrenched with moderate gradient; Bc
channel types are moderately entrenched with low gradient. C channel types are
slightly
entrenched meandering channels with moderate to low gradient and a high
width/depth ratio; Cc channel types are C channels with a gradient less than .001.
70
E channel types are moderately sinuous with low width/depth ratios and low to
moderate gradient. F channel types are entrenched and meandering with low to
moderate gradient, and Gc channel types are entrenched with low gradient,
sinuosity, and width/depth ratio (Rosgen 1994).
APPENDIX (Continued)
Delineative measurements for Rosgen stream classification at 51 sites within the Wayne National Forest boundary. Dominant
substrate was used at sites where a pebble count was not performed. Lattitude and longitude are also included. S = Symmes Creek
and P = Pine Creek.
Bankfull
D50 or Pebble
Site
Basin Width (m) Entrenchment WID ratio Gradient
Sinuosity Dominant Count? Lattitude Longitu
Asbury/Fish
4.48
3.5 E,C
9.9 A,G,E 0.01 Gc,F,E,Bc,C,Cc 1.2 A
99
S
Gravel
Y
3844.60 8234.40
99/00 BakersFork/Fish
5.79
3.0 E,C
13.6 B,C,F 0.01 Gc,F,E,Bc,C,Cc 1.0 A
S
Gravel
Y
3844.02 8231.32
Brushy-lower/Fish
99
P
11.19
2.5 E,C
87.9 B,C,F
0 Gc,F,E,Bc,C,Cc 1.0 A
Silt
Y
3846.88 8238.72
2.56
99
Brushy-upper
P
1.9 B
13.0 B,C,F 0.01 Gc,F,E,Bc,C,Cc 1.0 A
Gravel
Y
3846.96 8238.56
98
Brushyl (CR34)
P
4.94
5.3
E,C
7.7 A,G,E 0.01 Gc,F,E,Bc,C,Cc 1.2 A
Clay
N
BrushyBuckeyel
5.85
98
2.1
B
24.2 B,C,F S
3842.57 8233.53
N
98
BrushyBuckeye2
3.29
2.6 E,C
S
14.3 B,C,F 0.01 Gc,F,E,Bc,C,Cc 1.0 A
Sand
N
3843.04 8233.58
98
BrushyBuckeye3
7.04
1.2 A,G,F 31.1 B,C,F 0.01 Gc,F,E,Bc,C,Cc 1.1 A
S
Gravel
N
3842.50 8234.10
98
BrushyBuckeye4
3.05
2.8 E,C
S
14.9 B,C,F 0.01 Gc,F,E,Bc,C,Cc 1.1 A
Gravel
3843.26 8234.34
N
98 Buffalol
6.40
1.5 B
S
17.0 B,C,F 0.01 Gc,F,E,Bc,C,Cc 1.1 A
Gravel
N
3845.38 8233.50
98
Buffalo2
5.06
S
1.4 B
10.0 A,G,E
0 Gc,F,E,Bc,C,Cc 1.0 A
Gravel
N
3845.26 8233.45
98
Buffalo3
8.50
7.2 E,C
19.2 B,C,F 0.01 Gc,F,E,Bc,C,Cc 1.0 A
S
Gravel
N
3845.23 8233.40
10.15
99
1.3
A,G,F 15.1 B,C,F
S
0 Gc,F,E,Bc,C,Cc 1.2 A
Sand
Buffalo@Bakers
Y
3843.00 8231.18
Buffalo@Bridge
6.70
99
1.5 B
7.4 A,G,E
S
0 Gc,F,E,Bc,C,Cc 1.1 A
Sand
Y
3844.30 8241.37
99
Camp/Fish
4.81
S
1.4 B
6.8 A,G,E
0 Gc,F,E,Bc,C,Cc 1.0 A
Saud
Y
3845.59 8227.49
Caulley
8.17
99
3.6 E,C
13.3 B,C,F
S
0 Gc,F,E,Bc,C,Cc 1.1 A
Gravel
Y
98
Cooneyl
3.99
P
4.4 E,C
12.0 B,C,F 0.01 Gc,F,B,Bc,C,Cc 2.3 E
Gravel
N
3841.30 8244.53
98
Cooueyl/Fish
P
6.16
1.9 B
18.4 B,C,F 0.01 Ge,F,E,Bc,C,Cc 1.1 A
Gravel
N
3841.40 8244.54
98
Cooney3
P
2.96
1.9 B
8.5 A,G,E 0.02 B,G,Fb,Eb,Cb
1.2 A
Gravel
N
3841.54 8244.37
98
Handley
4.66
1.6 B
9.0 A,G,E 0.01 Gc,F,E,Bc,C,Cc 1.0 A
S
Gravel
N
3842.15 8230.23
5.67
99
Howardl/Fish
P
1.8 B
31.1 B,C,F 0.01 Gc,F,E,Bc,C,Cc 1.3 G,B
Gravel
Y
3844.09 8241.72
99
Howardil
P
3.29
1.9 B
11.2 A,G,E 0.01 Gc,F,E,Bc,C,Cc 1.0 A
Gravel
Y
3844.11 8241.68
Year
-4
I-
APPENDIX (Continued)
Bankful
Year
98
98
Site
Indianl
Indian2
98
Indian3
98
Indian4
98
Indian5
98
Indian6
98/99/00 Jenkins/Fish
98
Johnsl
98
Johns2
98
Little Buffalo!
98
Little Buffalo2
Little Pine
99
99
LongfFish
98/00 Meyer/Fish
99
Negro/Fish
99
Pine above Young's
98
Pine2
98
Pine3
99
Pine522
99
Pine@LyraLickRun
PineYoung'sBr.
99
98
Pine Creek Chapel
SandFork/Fish
99
99
Sugarcamp/Fish
99
Symmes I
99
Symmes II
99
TnbtoBakers
99/00 WolcottlfFish
99
Wolcotifi
Basin Wldth(m) Entrenchment WID ratio
S
S
S
S
S
S
S
S
S
S
S
P
S
S
P
P
P
P
P
P
P
P
S
P
S
S
S
P
P
6.06
6.22
5.55
3.17
4.60
7.04
7.25
10.45
13.53
4.45
2.16
11.46
9.54
3.53
6.77
9.17
11.55
11.25
15.18
11.64
10.00
10.2!
7.71
3.78
9.23
16.49
5.18
5.91
6.10
2.0 B
1.2
1.6
1.4
2.1
A,G,F
B
B
B
2.0 B
3.6 E,C
4.3 E,C
3.1 E,C
4.6 E,C
5.6 E,C
2.6 E,C
1.6 B
3.2 E,C
2.6 E,C
1.5 B
3.6 E,C
2.1 B
2.4 E,C
1.5 B
1.3 A,G,F
4.6 E,C
2.7 E,C
1.4 B
3.0 E,C
1.8 B
1.4 B
5.0 E,C
3.7 E,C
Gradient
17.6 B,C,F
26.1 B,C,F
27.3 B,C,F
17.3 B,C,F
22.0 B,C,F
38.9 B,C,F
20.1 B,C,F
0.02 B,G,Fb,Eb,Cb
0.01 Gc,F,B,Bc,C,Cc
0.01 Gc,F,E,Bc,C,Cc
0.01 Gc,F,E,Bc,C,Cc
0.02 B,G,Fb,Eb,Cb
0.01 Gc,F,E,Bc,C,Cc
0 Gc,F,E,Bc,C,Cc
0 Gc,F,E,Bc,C,Cc
6.9 A,G,E
9.2 A,G,E 0.01 GC,F,E,BC,C,CC
0 Gc,F,E,Bc,C,Cc
9.5 A,G,E
6.6 A,G,E
0 Gc,F,E,Bc,C,Cc
10.2 A.G,E
13.8 B,C,F
7.6 A,G,E
13.9 B,C,F
9.7 A,G,E
10.! A,G,B
9.1 A,G,E
10.5 A,G,E
11.8 A,G,E
12.4 B,C,F
10.5 AG,E
11.7 A,G,E
12.! B,C,F
11.5 A,G,E
11.9 A,G,E
D50 or Pebble
Sinuosity Dominant Count? Lattitude Longitude
1.1
1.0
1.0
1.0
1.1
1.1
1.0
1.1
1.1
1.0
1.2
1.0
1.1
1.2
1.1
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
0 Gc,F,E,Bc,C,Cc
0 Gc,F,E,Bc,C,Cc
0.01 Gc,F,E,Bc,C,Cc
0.01 Gc,F,E,Bc,C,Cc
0 Gc,F,E,Bc,C,Cc 1.!
0 Gc,F,E,Bc,C,Cc 1.!
0 Gc,F,E,Bc,C,Cc 1.3 G,B
0 Gc,F,E,Bc,C,Cc 0 Gc,F,E,Bc,C,Cc 0 Gc,F,E,Bc,C,Cc 0 GC,F,E,BC,C,CC 2.0 E
0 Gc,F,E,Bc,C,Cc 1.1 A
0 Gc,F,E,Bc,C,Cc 1.0 A
0 GC,F,E,BC,C,CC 0 GC,F,E,BC,C,CC 8.7 A,G,E 0.01 GC,F,E,BC,C,CC 1.0 A
16.0 B,C,F 0.01 Gc,F,E,Bc,C,Cc 1.4 G,B
0 Gc,F,E,Bc,C,Cc 1.0 A
28.8 B,C,F
Cobble
Gravel
Gravel
-
Cobble
Gravel
Sand
Sand
Sand
Bedrock
Gravel
Silt
Sand
Sand
Gravel
Sand
Sand
Sand
Saud
Sand
Sand
Sand
Gravel
Silt
Sand
Sand
Gravel
Gravel
Saud
N
N
N
N
N
N
N
N
N
N
N
Y
Y
N
Y
Y
N
N
Y
Y
Y
N
Y
Y
Y
Y
Y
Y
Y
3845.25 8234.11
-
3845.03
3845.36
3845.38
3845.22
3843.02
3842.09
-
-
8234.36
8234.51
8235.09
8233.52
8226.47
8230.10
-
-
3837.97
3839.26
3841.05
3844.56
3845.30
3845.49
3845.19
3840.66
3846.25
3845.42
3846.16
3837.36
3848.09
3848.69
3848.07
3844.02
3842.57
3842.59
-
8242.82
8228.10
8224.41
8237.29
8238.84
8241.06
8239.03
8242.45
8242.49
8239.67
8241.55
8222.48
8242.39
8226.05
8226.59
8231.54
8242.04
8242.27
-.4
N)