781
E.J. South and W.E. Ensign
22001133 SOUTSoHuEthAeSaTsEteRrnN NNaAtTurUaRliAstLIST 1V2o(4l.) :1728,1 N–7o8. 94
Life History of Campostoma oligolepis (Largescale
Stoneroller) in Urban and Rural Streams
Eric J. South1 and William E. Ensign1,*
Abstract - We conducted a year-long investigation of the effect of watershed urbanization on
the life history of Campostoma oligolepis Hubbs and Greene (Largescale Stoneroller). Lifehistory
characteristics of separate populations of Largescale Stoneroller were compared by
sampling two stream systems differing in urbanization in their upstream catchments. Both
streams are located in the Etowah River drainage basin within the Piedmont ecoregion. We
determined degree of urbanization by estimating the percent area of impervious catchment
surfaces using ArcGIS, and we recorded stream temperatures with continuous-monitor
probes. We sampled each stream system once a month during the spawning period and two
additional months during the remainder of the year and recorded standard lengths, tuberculation,
total weight, and gonadal weight for all retained specimens. Gonadosomatic index
(GSI) values showed reproductively active individuals present in January in the urbanized
system, one month before we found similar individuals in the non-urbanized system.
Comparison of GSI values between systems suggested that reproductive maturity occurs at
a larger size for females in the urbanized system. Comparison of standard lengths of reproductively
active females indicated that growth rates are higher in the urban system.
Introduction
Streams in urbanized landscapes are often degraded. The symptoms that characterize
the ecological degradation of urbanized lotic systems have been termed the
urban-stream syndrome (Meyer et al. 2005, Walsh et al. 2005). One of the primary
causes of the urban-stream syndrome is large amounts of stormwater runoff due
to impervious surfaces and hydraulically efficient drainage systems (Walsh et al.
2005). Urban stormwater runoff alters stream hydrology, morphology, and water
quality, all of which effect fish assemblages. Populations of sensitive fish taxa are
often reduced or extirpated as a result of these habitat changes (Walsh et al. 2005).
In contrast, tolerant species frequently persist in degraded stream environments and
respond with increased abundance or dominance (Walsh et al. 2005).
The purpose of this study was to investigate the effect of catchment urbanization
on the life-history characteristics of Campostoma oligolepis Hubbs
and Greene (Largescale Stoneroller). We compared populations of Largescale
Stoneroller in two stream systems with differing levels of catchment urbanization.
Previous research has shown that other Campostoma (stoneroller) species
persist in urban environments (Waits et al. 2008). Spranza and Stanley (2000)
observed increased juvenile growth rates and extended spawning seasons for
Campostoma anomalum Rafinesque (Central Stoneroller) in non-urban systems
with elevated temperatures and highly variable flow regimes. Urbanized stream
1Department of Biology and Physics, Kennesaw State University, Kennesaw, GA 30144.
*Corresponding author - bensign@kennesaw.edu.
E.J. South and W.E. Ensign
2013 Southeastern Naturalist Vol. 12, No. 4
782
systems typically have a higher percentage of increased impervious surface and a
more open canopy than non-urban systems, which usually leads to warmer water
due to stormwater runoff and increased insolation. Because water temperature and
photoperiodicity influence spawning behavior in fishes (Bond 1996), we expected
to find either a shifted or extended reproductive season for the urban stonerollers.
While a shifted spawning season would likely be a response to changing thermal
regimes, an extended spawning season could potentially be the result of a
bet-hedging reproductive strategy that enhances juvenile survival rates in a variable
environment (Spranza and Stanley 2000). Increased hydrologic variability
(both in terms of intensity and periodicity) is also a common feature of urbanized
streams (Roy et al. 2006). Stonerollers are primarily herbivores (Boschung and
Mayden 2004); therefore, an increase in algal growth due to the warmer water
temperatures and greater light availability may also contribute to increased adult
and young-of-year (YOY) growth rates. Higher YOY growth rates and an earlier
spawning season in the urban system could result in reproductive activity at a
larger size. To test these ideas, we investigated four of the sixteen life-history
characteristics of North American fishes described by Winemiller and Rose
(1992), including length at maturation, duration of spawning season, YOY growth
rates, and adult growth rates. We predicted that the more urbanized stream system
would have a population of stonerollers with a longer spawning season, larger
size at maturation, and higher adult and YOY growth rates.
Study Area
We collected Largescale Stonerollers from two stream systems located in the
Etowah River drainage basin within the Piedmont ecoregion. The first collection
site was an urban second-order tributary of Noonday Creek located in Cobb County,
GA, approximately 10 km north of Atlanta. This stream was characterized by riffles
and pools with a cobble substrate heavily embedded with silt and sand. Exposed
bedrock substrate was present in some runs.
The second site included two second-order confluent streams that were tributaries
of Pumpkinvine Creek. The streams were located in a forested area of Paulding
County, GA, approximately 30 km west of Noonday Creek. Collecting from the
upper Pumpkinvine system presented some challenges during the latter part of this
study: we discovered that the stream we initially chose for the rural system, Hulseytown
Branch, had no above-ground flow in August, with surface-water limited to
isolated pools. Consequently, we substituted a second stream, Brushy Branch (confluent
with Hulseytown Branch), for the remaining three sampling collections. The
last collection from Brushy Branch, in January, yielded a small sample consisting
of only ten juveniles and four adults. An obstruction at a downstream culvert which
had formed since our penultimate collection, in November, markedly altered habitat
at the Brushy Branch site, eliminating the majority of the erosional habitat units
where we had captured stonerollers in the previous two samples. Both streams have
riffle and pool morphology similar to Noonday Creek, but the substrate is primarily
cobble, pebble, and gravel with little embeddedness. We determined the level
783
E.J. South and W.E. Ensign
2013 Southeastern Naturalist Vol. 12, No. 4
of urbanization in the watersheds above the study areas by evaluating impervious
surface from the 2006 National Land Cover database (Fry et al. 2011) using Arc-
GIS. We identified Noonday Creek (watershed area approximately 9.8 km2) as the
more urbanized stream system, with impervious surface covering of 26.3% of the
total area. We classified the upper Pumpkinvine streams (watershed area approximately
7.0 km2) as the less impacted stream system, with only 3.4% of the total
surface area impervious.
Methods
We used a backpack electroshocker (Smith-Root Model LR-24) and dip nets to
make monthly collections of specimens from both stream systems from February
to June 2011, and once each in August 2011, November 2011, and January 2012.
We anesthetized all fish with tricaine methanesulfonate, measured them with dial
calipers to obtain standard length (SL) and total length to the nearest 0.1 mm, and
assessed each one for the presence or absence of tuberculation. We preserved approximately
20 fish per stream from each collection for further study in the lab;
all other individuals were allowed to recover before release. We initially preserved
retained specimens in 95% ethanol (ETOH) and subsequently transferred them
to 70% ETOH. We used some of the collected fish for a concurrent study on age
and growth using otoliths, so we used ETOH rather than fixing the specimens in a
10% formaldehyde solution because formaldehyde decalcifies otoliths and would
have rendered the specimens useless for the aging study (Cailliet et al. 1986). Of a
total of 396 captured fish, we retained 323 for laboratory examination. In addition
to the fish collections, we also measured water temperature at each site on each
collection date. We secured temperature loggers in each stream to record hourly
water temperatures from May through August. Although this period was outside the
spawning season, we used the temperature logger data to assess the relative difference
in temperature regimes between the two streams.
Once in the lab, we placed each fish in water for 10 minutes to allow partial tissue
rehydration before processing. We then blotted each fish dry and recorded the
total weight to the nearest milligram. We used an opti-visor and digital dial calipers
to measure SL and total length to the nearest 0.1 millimeter. SL measurements were
taken by removing the lateral musculature at the caudal peduncle and exposing the
hypural plate. We removed the gonads, blotted them dry, and weighed them to the
nearest milligram. We recorded the presence or absence of tuberculation for males.
We examined the gonads of all individuals with a dissecting scope to ensure correct
gender assignment, and estimated gonadosomatic index (GSI) by dividing the
gonadal tissue weight by the total weight of each individual, and multiplying this
value by 100.
We used GSI values to categorize individuals into one of two levels of reproductive
activity. We assumed that individuals preparing for reproductive
activity would undergo a rapid increase in GSI. In contrast, we expected that
non-reproductive individuals (juveniles and post-spawn fish) would make limited
investments of energy into the production and maintenance of reproductive
E.J. South and W.E. Ensign
2013 Southeastern Naturalist Vol. 12, No. 4
784
biomass, resulting in a relatively stable GSI. Thus, we identified a baseline GSI
and classified individuals with GSI values that showed evidence of rapid increase
above the baseline as reproductively active (RA), and individuals with GSI values
near the baseline as reproductively inactive (RI).
To differentiate RA from RI individuals, we rank-ordered GSI values from lowest
to highest in each gender category for all individuals collected from all sites
and dates. The difference between each successive observation was then calculated
and used as a measure of the relative increase in the investment in reproductive
products. We employed a nonparametric deviance method proposed by Qian et al.
(2003) to identify the threshold value where GSI begins to increase rapidly. In this
technique, we determined the overall deviance of individual values for a response
variable around the group mean. We then created two subsets of data and calculated
the deviance of each of the subsets from their respective mean. The sum of
the subset deviances will always be less than the group deviance, and the threshold
value for the overall data set is defined as the value separating the two subsets that
produces the greatest difference between the overall deviance value and the sum of
the two subset deviance values (Qian et al. 2003). We used an iterative replacement
approach in Excel to determine the threshold value that maximized the difference
in overall and subset deviances.
We used the non-parametric Mann-Whitney U test for all statistical analyses
because biological data sample sizes were small and we could not assume data
normality. We compared average GSI values of RA individuals between the two
stream systems for each monthly collection. We also compared mean SL of RA
individuals between the two stream systems for each month and collectively across
months during which reproductive activity might be expected based on data presented
in other studies of Campostoma spawning periods (February–May; Etnier
and Starnes 1993, Miller 1962). We compared daily water temperature averages for
each stream from 6:00 AM and 6:00 PM readings using a paired t-test. Because this
was an exploratory study in which we wished to elucidate trends rather than make
strong inferences, we chose to set our significance level for al l comparisons at P =
0.10. For the same reason, we did not use corrections for multiple comparisons in
our comparisons of monthly values.
Results
Temperature differences between the two stream systems were highly significant
(AM: P < 0.001, PM: P < 0.001). There was a 1.9 ºC difference in temperature at
6:00 AM and a 2.0 ºC difference at 6:00 PM. Noonday Creek temperatures were
consistently higher across times and dates.
Using the nonparametric deviance method, we identified a threshold value between
0.976 and 0.990 to separate female Largescale Stonerollers. Consequently,
we defined RA females as those individuals with GSI values of 0.990 and above,
and RI females as those individuals with GSI values of 0.976 and below. We defined
RA and RI males by a similar process using the GSI values of 0.192 and above, and
0.175 and below, respectively.
785
E.J. South and W.E. Ensign
2013 Southeastern Naturalist Vol. 12, No. 4
We observed elevated female GSI values from January through May (Fig. 1).
High RA female GSI values through May as well as a precipitous drop in those
values in June indicated that the completion of the spawning season for both stream
systems was concurrent. Additionally, the males collected in June showed low GSI
values and no tuberculation. GSI values for RA females were significantly higher
in Noonday Creek than Pumpkinvine Creek for the month of February (P = 0.088;
Fig. 1); the median Noonday Creek value exceeded the maximum value from
Pumpkinvine Creek. From March through May, there was no difference in either
median value or range (Fig. 1). RA males had significantly higher GSI values in
Pumpkinvine Creek than those in Noonday Creek during March (P = 0.024).
Based on analysis of monthly data, our results indicated significant differences in
SL of RA females between the two stream systems (Fig. 2). RA females were significantly
longer in the urban system in January, February, April, and May (P = 0.034,
0.004, 0.004, and 0.077 for the January, February, April, and May collections, respectively).
When we combined data from February through May, differences in length of
RA females between Noonday and Pumpkinvine creeks were highly significant (P =
less than 0.0001). We found no significant length differences for RA males.
For most monthly collections, we were unable to determine age classes based
on the length-frequency distributions. However, the length-frequency distributions
in June and August were of interest. In June, no YOY were collected in the
Figure 1. Gonadosomatic index (GSI) values for reproductively active female Largescale
Stonerollers collected monthly from Noonday and upper Pumpkinvine stream systems in
2011 and January 2012. Horizontal bars represent median GSI values and vertical lines
represent ranges of GSI values.
E.J. South and W.E. Ensign
2013 Southeastern Naturalist Vol. 12, No. 4
786
Pumpkinvine system. However, a 23-mm individual was collected from Noonday
Creek. The August collection showed well-defined YOY cohorts for both streams
(Fig. 3). The urban system had fifteen YOY ranging from 38.7–56.0 mm SL, and
the rural system had seven YOY ranging from 27.6–39.5 mm SL. All but two
of the YOY in Noonday Creek were larger than the YOY in upper Pumpkinvine
Creek, and the largest YOY individuals from Noonday Creek were equal in size to
the smallest age 1 individuals in Pumpkinvine Creek. The August collection also
showed well-defined second modes at 72.5 mm in Noonday Creek and at 57.5 mm
for the Pumpkinvine system that can be reasonably interpreted a s age 1+ fishes.
Discussion
As predicted, our results suggested an extended spawning season for Largescale
Stoneroller in the urbanized system. The increased GSI values we observed for
Noonday Creek females in January, and the significant difference in RA female
GSI values between Noonday and Pumpkinvine creeks in February suggested that
females in the Noonday stream were building reproductive biomass earlier than
females in the upper Pumpkinvine streams. The presence of some heightened GSI
values for RA females in both stream systems through the end of May indicates
that the end of the spawning season was synchronous in both stream systems. Although
the median value for Noonday Creek had declined to lower levels by May,
the single highest female GSI value recorded during the study also came from the
Figure 2. SL (mm) of reproductively active female Largescale Stonerollers collected
monthly (February–November 2011 and January 2012) from Noonday and upper Pumpkinvine
stream systems. Horizontal bars represent median lengths, and vertical lines represent
length ranges.
787
E.J. South and W.E. Ensign
2013 Southeastern Naturalist Vol. 12, No. 4
May Noonday Creek sample, indicating reproductively active individuals were still
present in the system. Additionally, the YOY size differences that we detected in
our August collections suggest that spawning season may begin earlier in Noonday
Creek than in Pumpkinvine Creek.
Previous studies have shown that increased water temperatures in freshwater
habitats due to anthropogenic influence initiate early spawning behavior in temperate
fishes (Cooke et al. 2003, Paller and Saul 1996). Similarly, our findings
suggested that elevated water temperatures in an urbanized stream system corresponded
to early spawning activity in a freshwater temperate-zone cyprinid. Peak
spawning activity for Central Stoneroller has been reported to occur when water
temperatures are 12–14 °C (Etnier and Starnes 1993). In this study’s initial February
sampling from the Noonday stream, we recorded a water temperature of 12 °C,
and collected RA female Largescale Stonerollers with high GSI values. Spawning
behavior of Central Stoneroller has been observed to cease when temperatures drop
to 10.6 °C (Miller 1962). In the February sampling in Hulseytown Branch, we collected
RA female Largescale Stoneroller with low GSI values and recorded a water
temperature of 9.5 °C.
Spranza and Stanley (2000) found that Central Stonerollers in a stream environment
with elevated water temperatures extended their spawning activity beyond the
Figure 3. Length-frequency histograms for all Largescale Stonerollers captured in the Noonday
and Pumpkinvine stream systems in August 2011.
E.J. South and W.E. Ensign
2013 Southeastern Naturalist Vol. 12, No. 4
788
spawning season observed for stonerollers in a lower water temperature environment.
Our study suggests that urbanized Largescale Stonerollers invested energy in
reproductive products earlier in the year than the stonerollers in the rural system.
In each of these studies, the entire spawning season for stonerollers in an elevated
water-temperature environment was not shifted or truncated, but extended earlier
or later than the spawning season of the stonerollers in the lower temperature environment.
An extended spawning season may enhance the ability of stonerollers
to persist in thermally enriched environments by producing offspring over a longer
period of time. The extended production of offspring may buffer the effects of stochastic
events and increase the number of reproductively active adults available for
subsequent reproduction, and thereby promote persistence of the population.
We also predicted that Largescale Stoneroller growth rates would be higher
in the urban system than in the rural system. If we assume that females reached
reproductive maturity at the same age in both stream systems, then the significant
length differences we observed in RA females in January, February, April, and May
suggest that females grew faster in the urban system. Furthermore, the difference
in the putative modal values for age 1+ Central Stonerollers in August (Fig. 3) also
indicates higher growth rates in the urban system. Increased size is associated with
increased reproductive output in fishes; therefore, more eggs will likely be produced
by the urban stream females.
There was also evidence that YOY fish in the urban system were larger in August
(Fig. 3). The larger size of YOY individuals collected in August from Noonday
Creek could have been a function of the earlier initiation of spawning in the urban
stream than in the rural stream, or higher growth rates associated with higher water
temperatures and increased food supplies. Regardless of the causal factor, it is clear
that Noonday Creek YOY were larger than Pumpkinvine Creek YOY as the end of
the summer arrived. A number of studies have shown that larger size is associated
with increased overwinter survival (Shuter et al. 1980, Thompson et al. 1991).
The higher female GSI values in the Noonday Creek February collection and the
presence of RA females in May suggest an extended spawning season and higher
growth rate for Largescale Stoneroller in urbanized Etowah basin streams. The significantly
longer SL of reproductively active females collected from the Noonday
Creek system provides strong evidence of higher growth rates in the urban setting.
The larger YOY individuals collected in the urban system in the August sample suggest
an increased likelihood of survival overwinter, and may also indicate higher
growth rates in the urban system. Each of these characteristics should contribute to
Largescale Stonerollers’ ability to persist in urbanized systems.
Acknowledgments
We appreciate field assistance provided by Chris Yates and John Bremner and technical
assistance from David Neely. Logistic and financial support from the Department of
Biology and Physics at Kennesaw State University allowed the senior author to conduct an
undergraduate-directed study project that culminated in this manuscript.
789
E.J. South and W.E. Ensign
2013 Southeastern Naturalist Vol. 12, No. 4
Literature Cited
Bond, C.E. 1996. Biology of Fishes, 2nd Edition. Saunders College Publishing, Orlando,
FL. 750 pp.
Boschung, H.T., and R.L. Mayden. 2004. Fishes of Alabama. Smithsonian Institute, Washington,
DC. 736 pp.
Cailliet, G.M., M.S. Love, and A.W. Ebeling. 1986. Fishes: A Field and Laboratory Manual
on their Structure, Identification, and Natural History. Waveland Press, Inc., Long
Grove, IL. 194 pp.
Cooke, S.J. 2003. Nesting activity, parental care behavior, and reproductive success of
Smallmouth Bass, Micropterus dolomieu, in an unstable thermal environment. Journal
of Thermal Biology 28:445–456.
Etnier, D.A., and W.C. Starnes. 1993. The Fishes of Tennessee. University of Tennessee
Press, Knoxville, TN. 681 pp.
Fry, J., G. Xian, S. Jin, J. Dewitz, C. Homer, L. Yang, C. Barnes, N. Herold, and J. Wickham.
2011. Completion of the 2006 national land cover database for the conterminous
United States. Photogrammetric Engineering and Remote Sensing 77:858–864.
Meyer, J.L., M.J. Paul, and W.K. Taulbee. 2005. Stream ecosystem function in urbanizing
landscapes. Journal of the North American Benthological Society 24:602–612.
Miller, R.J. 1962. Reproductive behavior of the stoneroller minnow Campostoma anomalum
pullum. Copeia 1962:407–417.
Paller, M.H., and B.M. Saul. 1996. Effects of temperature gradients resulting from reservoir
discharge on Dorosoma cepedianum spawning in the Savannah River. Environmental
Biology of Fishes 45:151–160.
Qian, S., R.S. King, and C.J. Richardson. 2003. Two statistical methods for the detection
of environmental thresholds. Ecological Modelling 166:87–97.
Roy, A.H., M.C. Freeman, B.J. Freeman, S.J. Wenger, W.E. Ensign, and J.L. Meyer. 2005.
Investigating hydrologic alteration as a mechanism of fish assemblage shifts in urbanizing
streams. Journal of the North American Benthological Society 24:656–678.
Shuter, B.J., J.A. Maclean, F.E J. Fry, and H.A. Regier. 1980. Stochastic simulation of temperature
effects on first-year survival of Smallmouth Bass. Transactions of the American
Fisheries Society 109:1–34.
Spranza, J.J., and E.H. Stanley. 2000. Condition, growth, and reproductive styles of fishes
exposed to different environmental regimes in a prairie drainage. Environmental Biology
of Fishes 59:99–109.
Thompson, J.M., E.P. Bergersen, C.A. Carlson, and L.R. Kaeding. 1991. Role of size,
condition, and lipid content in the overwinter survival of age-0 Colorado Squawfish.
Transactions of the American Fisheries Society 120:346–351.
Waits, E.R., M.J. Bagley, M.J. Blum, F.H. Mccormick, and J.M. Lazorchak. 2008. Sourcesink
dynamics sustain Central Stonerollers (Campostoma anomalum) in a heavily urbanized
catchment. Freshwater Biology 53:2061–2075.
Walsh, C.J., A.H. Roy, J.W. Feminella, P.D. Cottingham, P.M. Groffman, and R.P. Morgan.
2005. The urban stream syndrome: Current knowledge and the search for a cure. Journal
of the North American Benthological Society 24:706–723.
Winemiller, K.O., and K.A. Rose. 1992. Patterns of life-history diversification in North
American fishes: Implications for population regulation. Canadian Journal of Fisheries
and Aquatic Sciences 49:2196–2218.