Thermal Variability and Fish Species Composition in Buck Creek,
Delaware and Henry Counties, Indiana
Katrina C. Hamilton
Marci A. Kurtz
Advisor: Mark Pyron
c7 /,c.. o I J
-rht'~
....'[\
~\.\
I
'I
.Z J
/) .) ~
l"
. H 3C;
/
Abstract
Thermal conditions of a system may reveal a lot about the stream's attributes and
potential. Water temperature in Buck Creek, Delaware County, Indiana was monitored
throughout the months of June, July and August 2004. Mean monthly temperatures did
not vary among sites (June F=12.79, P ~0.001 ; July F=54.03, p ~0.001; August
F=33 .84, P ~0.001; September F=16.34, P ~0.001) and all sites were significantly
different from each other during the period ofthe study (F=98.5, P~O.OOl). The fish
species composition and temperature data are similar throughout Buck Creek, therefore
indicating that Buck Creek is an ideal location for stocking species of trout for sport
angling purposes.
2
Introduction
Stream temperature is an important environmental factor used in biological
assessments (Cuffney et a11996, Maret et aI1997). Agersborg (1930) found that water
with a constant temperature was more abundant with life than water with a fluctuating
temperature indicating that it may be difficult for fish to survive in areas that are
constantly experiencing drastic changes. Water temperature has a remarkable influence
on fish behavior because it controls their respiration (Agersborg 1930).
The fish community structure is also determined by seasonal changes due to flow
regime in streams (Gormon 1978). In the spring months (March, April, May and June)
fish move in mass numbers (Hall 1972); therefore, fish composition may differ between
seasons depending on the temperature. A study on different ecoregions suggested that
fish populations cluster naturally due to stream temperature (Waite and Carpenter 2000).
The fish diversity in our four microhabitats should relate to the water temperatures as
water temperature is positively correlated with microhabitat diversity (Grossman et al
1998). The temperature trend and fish species composition may be used to help
determine success for stocked fish.
Buck Creek is a first order stream with attributes that instigated speculation of
possible fish species that could be introduced into the system. The stream runs through
Delaware County and Henry County, Indiana and terminates as it flows into the West
Fork of the White River in Yorktown, Indiana. Considering it a viable recreational
resource, local residents have fished Buck Creek for many years. The cool temperatures
have inspired our inquiry into the likelihood of introducing species of trout to Buck Creek
following the research ofKeading (1996) who followed trout utilization of cool water
3
creeks similar to Buck Creek. The addition of lhis species would definitely increase the
recreational appeal of Buck Creek. However, considerations of stocking a new species
into an existing ecosystem should be taken seriously including an evaluation of the
habitat and life strategy compatibility, food availability, and possible implications on
other fish.
This study will evaluate the possibility for trout species introduction by
examining temperature trends in the stream and native fish species. Fish groups at each
location can be determined from bottom type, depth and the current (Gormon 1978).
Relative abundance of these fish groups at any time is a result of unpredictable
environmental fluctuations (Hall 1972). Physical water conditions are not the only
influence on the fish community structure. For example, biological interactions and
overlapping resources (Grossman et al 1998) have an influence on species diversity.
Studies have been done on fish behavior as a result of temperature changes and also on
the influence of habitat structure on stream fish communities. This study was done to
incorporate the two. The objectives are: 1) to determine ifthere is a noticeable
temperature difference between the four study sites of Buck Creek, 2) determine iflhere
is a correlation between temperature and fish species composition, and 3) utilize Ihis
information to examine the potential success of stocking species oftrou!.
4
Methods
Temperature data were collected using HOBO (Tempcon Instrumentation, West
Sussex United Kingdom) digital temperature recorders from June 2004 through
September 2004 to include seasonal variation in the fish assemblage from late spring to
late summer. These devices were placed at five locations in Buck Creek, Delaware and
Henry Counties, Indiana (Figure 1) HOBOs were attached to individual cement cinder
blocks and placed in appropriate locations in the creek where they would be submerged
even at the lowest summer water level. Temperatures were recorded daily at four hour
intervals. The temperature recorders were removed from four of the five locations in
September. One recorder was not recovered.
In late September and early October fish collections were conducted at the same
four locations on Buck Creek. Our sample area was 25 m downstream from the HOBO
and 25 m upstream for a total sampling area of 50 m. One person carried a backpack
electrofishing unit (Smith-Root Inc. type VIII Electrofisher) and two additional people
used nets to retrieve the fish. The fish were placed in buckets until identification at the
end of the sampling period. Fish were sorted by species, total length and weight were
recorded for the smallest and largest representative of each species and the total weight
for the combined sample of species. All fishes were returned to the creek with the
exception of a few sunfish and darters which were identified in the laboratory.
An analysis of the temperature data was conducted to evaluate the species
composition data. Analysis of variance (ANOVA) and/or Kruskal-Wallis tests were
conducted to determine if temperature significantly differed: I) between the four sites for
the entire duration of the study and, 2) between each site for each full month, and 3) on a
5
site by site basis during the entire study period. Principal components analysis was used
to examine all fish abundance data for all sites simultaneously.
6
RESULTS
Temperature records were available for 87 consecutive days on the length of Buck
Creek. HOBOs recorded temperature and produced 522 values at each of the four
stations.
Temperature data at each site showed similar fluctuations (Figures 2-5). Site 2,
nearest Buck Creek headwaters, and site 3, higher mid section of Buck Creek, showed
fluctuations in temperature that coincided on a daily basis. Sites 4 and 6 (lower middle
Buck Creek and near the intercept with the West Fork of the White River, respectively)
also had variation in temperature patterns similar to sites 2 and 3, but the range of
temperatures was reduced. Temperatures at sites 2 and 3 ranged from 14 - 23 ° C.
However, site 4 temperature ranged from 16 - 20°C. Site 6 also had less variation; the
temperature range was in the range of 14 - 20° C.
When analyzed by individual month, ANOVA indicated significant differences
between at least two of the mean site temperatures (June F=12.79, P :::..0.001 ; July
F=54.03, p :::..0.001 ; August F=33.84, P :::,,0.001; September F= 16.34, P < 0.001). A
Tukey multiple comparisons test was conducted on the data for each month to determine
which sites were significantly different (Table I). Overall, the sites farthest upstream
(sites 2 and 3) were typically similar to each other and the downstream sites (sites 4 and
6) were not similar. In June and July, stream temperature at sites 3 and 6 were not
significantly different.
Though temperatures were similar during some months for sites, the ANOV A test
and Tukey comparisons of pooled sample of total combined data for all of the months
showed that each site was significantly different from the others (F=98.5, P:S;O.OOI).
7
A Kruskal-Wallis test was used to compare the unequal sample sizes of each
month for all sites, to compare the thermal consistency of each site. Table 2 shows the
median and interquartile ranges for the temperatures at sites for June, July, August, and
September. No site was significantly different from itself (Site 2 H = 28.93 DF = 3 P :::
0.001; Site 3 H = 28.47 DF = 3 P ::: 0.001; Site 4 H = 9.13 DF = 3 P ::: 0.028; Site 6 H
=
6.98 DF = 3 P ::: 0.073).
The first two axes ofthe principal components analysis explained more than 88 %
of the total variance in the species abundance data. The first axis accounted for 54.3 %
and the second axis accounted for 33.7 % of total variance. The ordination from the
principal components analysis arranged sites from upstream to downstream on the first
axis (Figure 6), reflecting the pattern of average June and July water temperatures as
these sites (Table 1). Sites on the right side of the first axis had increased abundance of
four species (listed to right of Figure 6), and decreased abundance of four other species
(listed on left of Figure 6) include second axis species. Loadings of the other species were
less than 0.7 and are not further discussed .
8
Discussion
Fluctuating temperatures are characteristic of most naturallotic systems (Sweeney
1976). This thermal diversity has been characterized (Macan 1958;Edinton 1966);
however, overall temperature data for Buck Creek suggest no significant variation over
the months of June through early September. The tests show the overall uniformity in
temperature throughout all of Buck Creek for the summer of2004. Keading (1996)
found mean monthly temperature trends similar among sites in summer months. Those
results from a similar system to Buck Creek provide a reasonable comparison for species
patterns that may potentially be present at Buck Creek. The system did not show
instability as Needham and Jones (1959) suggest that most streams experience. If the
system were to be monitored throughout a cycle of seasons, the stream may be
considered to have less stability. Furthermore, a mUltiple year study may prove even
further instability of stream temperature.
Irregularity in seasonal air temperatures may be a strong factor in the temperature
of Buck Creek. Eaton and Scheller (1996) explain variation in temperature and thermal
consistencies of streams on climate warming. They have found streams in regions of
Indiana to parallel air temperatures in fluctuations and abnormalities.
Thermal conditions of Buck Creek may also affect the composition and diversity
of present fish species. Species such as Cot/us bairdi (mottled sculpin), Semolitus
atromaculatus (creek chub), and Rhinichthys atratulus (blacknose dace) reoccur in most
of the creek. The uniform temperature variation found in the length of Buck Creek
allows fish species to be successful in any portion ofthe stream. The fish and
temperature database matching system (FTDMS) described by Eaton et al. (J 995) proves
9
that much of the composition offish species collected in Buck Creek are cool water
species that may be adversely affected by fluctuating temperatures. The FTDMS
described cold water species C. bairdi may be a good indicator species for other cold
water fishes.
The extreme abundance of C. bairdi at each site invites reliable correlation to
other species of similar temperature habitats that could be successful in Buck Creek. C.
bairdi could aid in the determination of which sport-fishing species of trout could
potentially be stocked into Buck Creek. Oncorhynchus mykiss (rainbow trout) and Salmo
Irutla (brown trout) have very similar maximum temperature tolerances to C. bairdi.
High water temperature does not adversely affect 0. mykiss fishing (Taylor 1978).
Taylor (1978) did report that S. trulla catches fall when temperatures rise above 18·C. 0.
mykiss may be a better match for Buck Creek; however, it would be plausible that S.
trulla may also survive in the temperature range of Buck Creek.
To analyze the potential scenario of stocked trout, the research fmdings of
Keading (1996) on tributaries closely related to Buck Creek will be applied to Buck
Creek. The prey species necessary to sustain trout are present and are expected to
increase in June and be most numerous between mid July and early September (Keading
1996). Species of trout have been observed to make forays between water temperatures
resembling Buck Creek to forage (Beitinger and Fitzpatrick 1979; Coutant 1987). Higher
abundance of species will allow trout to grow as they are being fished from the system.
Food availability would not be a concern in trout mortality. The largest concern in
supporting fishable trout for summer months is the thermal variation of Buck Creek.
10
There are no foreseeable implications of stocking trout in Buck Creek.
Introducing a new species into a system is risky; however, adding trout to the creek
during the summer months may be desirable for anglers with little interruption to the
system. Anglers would remove out trout inhibiting the population from expanding and
persisting following the summer months. Native species would be able to spawn and
grow before trout introduction. The fishes could also rejuvenate populations over winter
months. Habitat would not be paralyzed by the trout, but there is potential of the habitat
to be altered by anglers. The increased presence of humans in the system would have to
be monitored with fishing restrictions to maintain the integrity and beauty of Buck Creek.
11
Acknowled2ements
In order to maintain the integrity of our research, we would like to give boundless
thanks to those that have contributed in the process of creating, conducting, and revising
our project. We would like to thank Ball State University and the Ball State University
Honors College for their funding and support. Also, our appreciation is expressed the
Ball State University Department of Biology and Dr. Tom Lauer for allowing us to utilize
their equipment and resources. We would like to thank Jamie K. Lau for her assistance in
the field and in the lab without which we could not have functioned so smoothly. Lastly,
we owe a great deal of gratitude to Dr. Mark Pyron for his contribution of equipment,
time and mentorship.
12
References
Agersborg, H.P.K. 1930. The Influence of Temperature on Fish. Ecology 11: 136-144.
Bettinger, T. L. and L. C. Fitzpatrick. 1979. Physiological and ecological correlates of
preferred temperature in fish. American Zoology 19: 319-329.
Coutant, e. C. 1987. Thermal preference: when does an asset become a liability?
Environmental Biology of Fish 18: 161-172.
Cuffney, T.F., Meador, M.R., Porter, S.D., & Gurtz, M.E. 1997. Distribution offish,
benthic macroinvertebrate, and al.gal. communities in relation to physical and
chemical. conditions, Yakima River Basin, Washington, 1990. U.S. Geological.
Survey Water Resources Investigations Report: 96-4280.
Eaton, J. G. et al. 1995. A field information-based system for estimating fish temperature
tolerances. Fisheries 20: 10-18.
Eaton, J. G. and R. M. Scheller. 1996. Effects of climate warming on fish thermal
habitat in streams of the United States. Limnology and Oceanography 41: 110911 15.
Edington, J. M. 1966. Some observations on stream temperature. Oikos 15: 265-273.
Gorman, O.T. and Karr, J.R. 1978. Habitat structure and stream fish communities.
Ecology 59: 507-515.
Grossman, G.D., Ratajczak, R.E. Jr, Crawford M., & Freeman, M.e. 1998. Assemblage
organization in stream fishes: effects of environmental variation and interspecific
interactions. Ecological Monographs 68: 395-420.
13
Keading, L. R. 1996. Summer use of cool water tributaries of a geothermally heated
stream by rainbow and brown trout, Oncorhyncus rnykiss and So/rno trullo.
American Midland Naturalist 135: 283-292.
Macan, T. T. 1958. The temperature ofasmall stony stream. Hydrobilogia 12: 89-106.
Needham, P. R. and A. C. Jones. 1959. Flow, temperature, solar radiation, and ice in
relation to activities of fishes in Sagehen Creek, California. Ecology 40:465-474.
Sweeney, B.W. 1976. A diurnally fluctuating thennal system for studying the effect of
temperature on aquatic organisms. Limnology and Oceanography 21: 758-763.
Taylor, A. H. 1978. An analysis of the trout fishing at Eye Brook-a eutrophic reservoir.
Journal of Animal Ecology 47: 407-423 .
14
Figure I . Map of sites on Buck Creek Delaware County, IN. Locations are 'indicated by
numbers.
B
-.,.. 3
Z
~ 4
L ...~
IJ~
6
1
15
Figure 2. Temperature variation at site 2 of Buck Creek, Delaware and Henry Counties,
IN for June through early September.
23s------------------------------------------------,
22
21
G' 20
*
'-'
16
15
14~------------._------------~--------------_r--~
6/1/2004
8/1/2004
7/1/2004
Date
9/1/2004
16
Figure 3. Temperature variation at site 3 of Buck Creek, Delaware and Henry Counties,
IN for June through early September.
23~--------------------------------------------~
22
21
G' 20
.....
*
16
15
14
r------------.-------------.-------------.--~
6/1/2004
8/1/2004
7/1/2004
Date
9/1/2004
17
Figure 4. Temperature variation at site 4, of Buck Creek Delaware and Heruy Counties,
IN for June through early September.
22
20
G'
.....
*
~
...
18
:I
i
16
~
14
12~
____________' -____________- .______________- r__
6/1/2004
7/1/2004
8/1/2004
Date
9/1/2004
~
18
Figure 5. Temperature variation at site 6 of Buck Creek, Delaware and Henry Counties,
IN for June through early September.
24~-----------------------------------------------'
22
G' 20
......
*
...11/
~ 18
~
~
~
16
14
12r--------------r--------------.--------------.--~
6/1/2004
7/1/2004
8/1/2004
Date
9/1/2004
19
Figure 6. Principal component analysis accounts for 88% of total variation of species
abundance with temperature data at Buck Creek sites from June through early August.
'"'"
'"
1
.c
:5
'"
'"
--'
'"
-2
-4
0
E
0.5
C>
~
o
-6
N
U
a..
~
-0.5
o
<=
<=
- 3
-2 ~--------~----------~--------~----------~----------
-1
o
-0.5
Mottled sculpin, Central stoneroller,
Blacknose dace, Orangethroat darter
0.5
pel
0
1
1.5
Rosyface shiner, Golden red horse ,
Logperch, Rockbass, Smallmouth bass
,,•;
;
!
20
Chart I. Species composition for sites on Buck Creek, Delaware and Henry Counties,
IN.
90
80
70
60
II Si!e 2
IISi!e 3
DSi!e4
DSi!e6
50
;
40
30
20
10
0
n
~
.n
n
n
~
Jl
L,
n
n
ll
21
Table 1. Results of Tukey multiple comparisons test comparing average water temperature at each sampling site for June , July, August, and
the combined months for Buck Creek.
Site
Site
Site
Site
2
3
4
6
x
June
SE
18.1
17.6
16.7
17.2
0 .1
0 .1
0.2
0.2
Jul~
N
x
SE
126
126
126
126
19.1
18.6
17.3
17.5
0.1
0.1
0.1
0.1
1m
n
mn
N
186 0
186 P
186 q
186 q
x
18.5
18.0
16.8
17.2
August
SE
0.1
0.1
0.1
0.1
Se~tember
N
186
186
186
186
r
r
s
s
x
19.0
18.5
17.0
17.8
SE
N
0.2
0.2
0.2
0.2
24
24
24
24
x
v
vu
t
tu
18.6
18.1
17.0
17.3
Combined
Months
N
SE
0.1
0.1
0.1
0.1
522
522
522
522
z
Y
x
w
22
Table 2. Results for comparisons of water temperature at sites ovet time using the Kruskal-Wallis test. No comparisons were significantly different.
Site 2
-C
June
July
August
September
18.1
19.4
18.7
19.0
2.7
2.3
2.7
1.3
n
126
186
186
24
-C
17.3
18.7
18.3
18.7
Site 6
Site 4
Site 3
x
x
2.3
1.9
2.3
1.3
n
126
186
186
24
-C
16.8
17.1
17.0
17.1
x
2.4
2.7
2.3
1.5
n
126
186
186
24
·C
17.1
17.5
17.1
17.9
x
3.1
2.3
3.0
2.0
n
126
186
186
24