J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist
64
Vol. 12, Special Issue 4
Movement Patterns of the Threatened Blackside Dace,
Chrosomus cumberlandensis, in Two Southeastern
Kentucky Watersheds
Jason E. Detar1,2 and Hayden T. Mattingly1,*
Abstract - Chrosomus cumberlandensis (Blackside Dace) is a threatened stream fish
endemic to the upper Cumberland River drainage in Kentucky and Tennessee. Little
is known about the movement patterns of this species. Acquiring an understanding of
baseline dispersal patterns is necessary to inform management and recovery actions.
We tagged 653 Blackside Dace with visible implant elastomer injections in the Big Lick
Branch and Rock Creek watersheds of southeastern Kentucky to determine the frequency,
spatial extent, directionality, and environmental correlates of dace movements. We recaptured
dace from February 2003 through March 2004 using baited minnow traps. Most
tagged dace (81% in Big Lick Branch and 58% in Rock Creek) were recaptured within the
same 200-m stream reach where tagging occurred. However, several individuals moved
considerable distances from the original tagging site, including the first documented
intertributary movement for this species. Mean (± SD) distances moved upstream in
Big Lick Branch (148 ± 138 m) and Rock Creek (733 ± 1259 m) were not significantly
different from mean distances moved downstream (77 ± 29 m and 314 ± 617 m, respectively).
However, the mean overall distance moved was greater in Rock Creek, a longer
stream than Big Lick Branch. The spatial arrangement of traps in both watersheds likely
produced a distance-weighted bias such that we slightly overestimated the frequency
of short-distance movements and underestimated the frequency of long-distance movements.
Our results for Blackside Dace are consistent with a number of other studies that
found stream fish populations composed of a large sedentary group and a smaller mobile
group. The demonstrated ability of Blackside Dace to move into and between tributaries
will remain vital for long-term population viability, and emphasizes the importance of
maintaining suitable corridors within and among Cumberland River tributary streams.
Introduction
Chrosomus cumberlandensis (Starnes and Starnes) (Blackside Dace) is a small
stream fish of the minnow family, Cyprinidae, endemic to the upper Cumberland
River drainage in southeastern Kentucky and northeastern Tennessee (Eisenhour
and Strange 1998). Anthropogenic activities, including coal mining, road construction,
agriculture, and poor logging techniques, have led to the decline of the
Blackside Dace and resulted in its federal listing as a threatened species (O’Bara
1990, L.B.Starnes and W.C. Starnes 1981, W.C. Starnes and L.B. Starnes 1978,
USFWS 1988). The entire Blackside Dace range lies within a region of rich coal
and timber reserves (Starnes 1981), which has been and continues to be subjected
to the impacts of mining and logging.
1Department of Biology, Box 5063, Tennessee Technological University, Cookeville, TN
38505. 2Current Address - Pennsylvania Fish and Boat Commission, 450 Robinson Lane,
Bellefonte, PA 16823. *Corresponding author - hmattingly@tntech.edu.
Ecology and Conservation of the Threatened Blackside Dace, Chrosomus cumberlandensis
2013 Southeastern Naturalist 12(Special Issue 4):64–81
65
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist Vol. 12, Special Issue 4
Relatively few studies have been conducted on the Blackside Dace, and none
have focused on determining their movement patterns. Movement and dispersal
are key determinants of population structure and function (Skalski and Gilliam
2000) that allow stream fishes to be resilient to habitat alteration and stochastic
events. Additionally, periodic long-range movements may enable small-bodied
species to respond to variation in resources over space and across a variety of
habitats (Freeman 1995). In general, stream fish movements are important for
maintaining population connectivity, reducing the potential for local extinctions,
and predicting the rate of colonization or recolonization of suitable habitats (Albanese
et al. 2009, Larson et al. 2002, Woolnough et al. 2009). Therefore, a basic
understanding of Blackside Dace population dynamics, including movement patterns,
is vital to the prudent management of this threatened species.
As the extraction of coal and timber resources continues within watersheds
containing Blackside Dace populations, it is critical that management agencies
know whether the species is relatively sedentary or mobile. Quantitative
estimates of movement probabilities are needed to better evaluate population
viability under different scenarios of natural or anthropogenic changes in stream
habitat conditions. If individuals exhibit extensive movement, it may be possible
for them to vacate their present stream reaches to avoid mining, logging, or other
disturbance, and to repopulate areas if habitat conditions improve following
the disturbance. Furthermore, a better understanding of movement patterns can
lend insight into the degree of connectivity between seemingly isolated populations
within and among upper Cumberland River watersheds. Thus, the primary
objective of this study was to determine the frequency, spatial extent, directionality
(upstream vs. downstream), and environmental correlates of Blackside Dace
movement to aid in the conservation and management of the species.
Study Area
Movement patterns were monitored in two southeastern Kentucky watersheds,
Big Lick Branch (including an adjacent unnamed tributary, hereafter Dace
Branch) in Pulaski County and Rock Creek in southeastern McCreary County
(Fig. 1). Big Lick Branch is a small, second-order stream that is a direct tributary
to Lake Cumberland below Cumberland Falls. Big Lick Branch generally flows
southeastward to its confluence with Lake Cumberland, and there are no perennial
tributaries to the stream. The Blackside Dace population in Big Lick Branch is
presumably isolated from other substantial populations in the area because of the
reservoir. The Big Lick Branch watershed encompasses 7.1 km² and is comprised
of mostly Daniel Boone National Forest land and a smaller amount of privately
owned land. The majority of the watershed is undisturbed and well-forested. The
stream maintains good flow all year. A dirt and gravel forest road parallels the
lower two-thirds of the stream. Aside from the forest road, there is currently no
other development within the watershed.
Dace Branch is a small, first-order tributary to Lake Cumberland which was
a tributary to Big Lick Branch before the Cumberland River was impounded in
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist
66
Vol. 12, Special Issue 4
1952. After 1952, Lake Cumberland inundated the Big Lick Branch watershed
upstream of the historic confluence of these two streams, thus severing their lotic
connection. The entire watershed of this small stream is well-forested and lies
within Daniel Boone National Forest. A steep, 2-m-high cascade creates a barrier
near the mouth of Dace Branch when Lake Cumberland is drawn down to winter
pool, which likely restricts migration into this stream during the winter months.
Rock Creek is a moderate-sized, second-order stream that is a tributary of
the considerably larger third-order Jellico Creek (upstream from Cumberland
Falls), and its watershed drains 17 km² of both private and public land. A dirt
and gravel forest road parallels the majority of the stream. Other than the forest
road and a few seasonal cabins, there is currently no other development within
the watershed. Most of the floodplain and lower mountainsides on the upper 6.5
km of Rock Creek are privately owned, and timber harvest occurs there occasionally.
The ridges and upper mountainsides are mostly Daniel Boone National
Forest land. The majority of the lower 2.5 km of Rock Creek is on Daniel Boone
National Forest land. The upper portion of Rock Creek can become intermittent
during dry summers, but groundwater seepage appears to keep water temperatures
cool in the intermittent pools. The gradient is considerably higher in the
Figure 1. Map of the Big Lick Branch and Rock Creek watersheds in southeastern Kentucky
depicting locations of Blackside Dace tagging sites (black dots). Distances between
sites, study periods, number of tagged dace, and other details are provided in the text and
Table 1.
67
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist Vol. 12, Special Issue 4
lower 2 km of the stream, and a 3-m-high waterfall creates a barrier approximately
1 km upstream from its mouth, restricting fish migration into Rock Creek.
Stream-gradient profiles for Rock Creek and Big Lick Branch are provided by
Detar (2004).
Rock Creek has several relatively undisturbed first-order tributaries, four
flowing southward and one flowing northward (Fig. 1). During summer months,
discharge in these tributaries was substantially reduced, but some Blackside Dace
remained in these streams all year. Similar to the upper portion of Rock Creek,
the small tributaries apparently have sufficient groundwater connections during
summer to keep water cool enough in intermittent pools for dace to persist.
Methods
Tagging and trapping
Blackside Dace were initially captured in Big Lick Branch, Dace Branch, and
Rock Creek and its tributaries using a generator-powered variable-voltage AC
backpack electrofisher (in-house design similar to a Coffelt BP-1C). Seining was
relatively ineffective for capturing this species due to coarse substrate, and we
did not have access to a DC electrofisher when the study was initiated. Upon recovery
from electronarcosis, Blackside Dace were anesthetized using a 40-mg/L
concentration of clove oil as described by Detar and Mattingly (2005), measured
to the nearest mm total length (TL) , and tagged with a visible implant elastomer
(VIE; Northwest Marine Technologies, Inc., Shaw Island, WA) injection. Blackside
Dace were then placed in an aerated water bucket until fully recovered from
anesthesia and released back into one or two pools located approximately in the
middle of the site (sites are described below). Dace were batch-tagged with different
colors or tag locations at each site so that movement could be detected
among sites. The four possible locations used for tagging were on the dorsal
surface of the dace (i.e., left and right pre-dorsal-fin origin, left and right postdorsal-
fin origin).
The second component of monitoring Blackside Dace movement patterns
required that fish be captured multiple times; therefore, an efficient yet minimally
invasive technique was needed. A pilot study was conducted in July 2002
using minnow traps to capture Chrosomus erythrogaster (Rafinesque) (Southern
Redbelly Dace). Baited minnow traps produced higher catch rates than unbaited
traps, and thus baited traps were used for capturing Blackside Dace during our
movement study.
On each monitoring occasion, we counted the number of Blackside Dace
while also recording the presence of other fish species in the minnow traps. Number
of Blackside Dace per trap was divided by the number of hours the trap was
fished to calculate catch per unit effort (CPUE). Recaptured individuals with tags
were given a second VIE injection in March 2003 through January 2004 sampling
events. These uniquely tagged individuals provided the opportunity to document
individual movement histories.
Big Lick Branch watershed. In November 2002, Blackside Dace were collected
via backpack electrofishing in each of the four 200-m sites in Big Lick Branch
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist
68
Vol. 12, Special Issue 4
and tagged with a single VIE injection (Table 1, Fig. 1). Movement was monitored
in Big Lick Branch every 6 to 8 wk from February 2003 through March
2004. Twenty-four baited Gee’s model G-40 galvanized wire minnow traps
(420 x 190 mm, 20-mm opening, 6-mm mesh) were set throughout the length
of Big Lick Branch (two below, two within, and two above each of the four tagging
sites). Traps set below and above tagging sites were placed approximately
50–100 m from the downstream and upstream margins of the sites. Each trap was
baited with one slice of white bread enclosed in a 0.5-mm-mesh screen and one
medium dog biscuit (MILK-BONE ®) broken in half. The mesh screen was used
to protect the bread from consumption by fish and to minimize the particles floating
downstream of the trap which encouraged fish to enter the trap and increased
catch rates. Minnow traps were set starting at Site 1 working upstream to Site 4.
Traps were fished for 4.5 to 7.8 hr on each occasion and were checked in the same
order in which they were set.
In February 2003, Blackside Dace were collected via backpack electrofishing
from a 250-m reach of Dace Branch (Table 1, Fig. 1). Blackside Dace were
tagged in Dace Branch in an attempt to detect movement from this stream across
Table 1. Summary of Blackside Dace (dace) mark-recapture efforts at sites in the Big Lick Branch
and Rock Creek watersheds. Dace at each site were batch-marked with visible implant elastomer
tags using a different tag color or body location. Minnow traps were used to recapture dace unless
noted otherwise. Distance in stream kilometers from the mouth of each stream to mainstem tagging
sites or tributary confluences is provided in the third column (site locations are illustrated in
Fig. 1). Maximum movement distances (nearest 50 m) that could be detected in a downstream (-)
or upstream (+) direction from each site are given in the fourth column. In the last two columns are
the number of trapping events and number of dace initially tagged per site. Stream km = stream km
from mouth. Events = number of trapping events. Dace = number of dace tagged.
Max. detectable
Study site Study duration Stream km movement (m) Events Dace
Big Lick Branch watershed
Big Lick Branch 1 Nov 2002–Mar 2004 0.60 -100, +2400 8 50
Big Lick Branch 2 Nov 2002–Mar 2004 1.45 -950, +1550 8 50
Big Lick Branch 3 Nov 2002–Mar 2004 2.05 -1550, +950 8 50
Big Lick Branch 4 Nov 2002–Mar 2004 2.90 -2400, +100 8 50
Dace Branch* Feb 2003–Mar 2004 n/a n/a 8 30
Rock Creek watershed
Rock Creek 1 Feb 2003–Mar 2004 2.75 -100, +5950 7 24
Litton Branch Mar 2003–Mar 2004 3.05 -450, +5600 6 39
John Anderson Branch Feb 2003–Mar 2004 4.35 -1850, +4350 7 110
Rock Creek 2 Feb 2003–Mar 2004 5.75 -3100, +2950 7 50
Sid Anderson Branch Mar 2003–Mar 2004 6.20 -3600, +2550 6 115
Rock Creek 3 Feb 2003–Mar 2004 8.25 -5600, +450 7 50
Lot Hollow Branch Mar 2003–May 2003 8.40 -5800, +350 1 5
Rock Creek 4 Feb 2003–Mar 2004 8.60 -5950, +100 7 30
Total (both watersheds) 653
*Seining was used instead of minnow traps at Dace Branch; movements within Dace Branch itself
were not monitored.
69
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist Vol. 12, Special Issue 4
the embayment of Lake Cumberland and into Big Lick Branch or vice-versa. We
seined the study reach in Dace Branch each time minnow traps were set in Big
Lick Branch to recapture possible migrants from Big Lick Branch. We did not
monitor movement of dace within Dace Branch itself because the more remote
location of this stream precluded setting and checking minnow traps.
Rock Creek watershed. In February 2003, we used backpack electrofishing
to collect 154 Blackside Dace from the four Rock Creek mainstem sites plus 32
individuals from John Anderson Branch (Table 1, Fig. 1). In March 2003, we
collected an additional 159 Blackside Dace in the other three southward-flowing
first-order tributaries plus an additional 78 Blackside Dace in John Anderson
Branch. Following the methodology used in Big Lick Branch, dace were batchtagged
with VIE injections so that movement could be detected among sites in
Rock Creek and its tributaries.
Movement was monitored in the Rock Creek system every 6 to 8 wk from
March 2003 through March 2004. Following the same approach used in Big
Lick Branch, 24 baited minnow traps were set throughout the length of Rock
Creek (two below, two within, and two above each of the four sites) during each
sampling period. In addition, 3 to 4 traps were set in each of the four first order
tributaries to Rock Creek. However, during the first monitoring period (March
2003), traps were only set in the Rock Creek mainstem and John Anderson
Branch. Traps set below and above tagging sites were placed approximately
20–100 m from the downstream and upstream site margins. Traps were set starting
at Rock Creek Site 1 working upstream (including tributaries as they were
encountered) to Rock Creek Site 4, were fished for 4.6 to 8.3 hr on each occasion,
and were checked in the same order in which they were set.
Movement and environmental variables
All distances moved by Blackside Dace in Big Lick Branch, except for the single
documented intertributary movement, were measured to the nearest 1 m using
a 50-m fiberglass tape measure. The distance of the intertributary movement was
estimated to the nearest 10 m using Maptech Terrain Navigator digital topographic
map software. Similarly, distances ≤615 m in Rock Creek were measured to the
nearest m using the 50-m fiberglass tape measure, whereas distances >615 m were
estimated to the nearest 10 m using the Maptech software. For double-tagged individuals,
only the first movement distance was used for analysis.
Because Blackside Dace were released in one or two pools located approximately
in the middle of each 200-m site when they were originally tagged,
distance moved was measured from the upstream margin of each site for upstream
migrants and from the downstream margin of each site for downstream
movement. Therefore, movement measurements are conservative because a dace
would need to move a minimum of approximately 100 m in either direction from
the center of a site to be considered a migrant.
Six of the eight environmental variables were measured in only one location
of the watershed on each sampling date for monitoring Blackside Dace
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist
70
Vol. 12, Special Issue 4
movements (Table 2). Discharge was evaluated in a section of the stream that
best exhibited a uniform, bowl-shaped channel. A Marsh-McBirney Flo-Mate
model 2000 or model 201D portable flow meter and top-setting wading rod were
used to measure water velocity at 60% of stream depth at multiple points along
a transect perpendicular to flow. From these data, discharge was calculated using
the velocity-area method outlined by Gallagher and Stevenson (1999). Dissolved
oxygen, conductivity, and temperature were measured using a Yellow Springs
Instrument (YSI) Model 85 meter. Turbidity was measured using a HF Scientific
MicroTPI turbidimeter, and pH was documented using an Oakton Instruments
pHTestr 3+ meter. The remaining two environmental variables, days since tagging
and day length, were derived from basic calendar data.
Statistical analyses
Two nonparametric one-way ANOVAs (i.e., Kruskall-Wallace test; Kruskall
and Wallace 1952) were used to determine if there was a significant difference in
the distances moved upstream versus downstream in either Big Lick Branch or
Rock Creek. A nonparametric two-way ANOVA (i.e., Friedman Test; Friedman
1937) was used to determine if there was a significant difference in the mean
overall distance moved in Big Lick Branch versus Rock Creek. The two factors
in the Friedman Test were (1) stream (Big Lick Branch, Rock Creek) and (2)
movement direction (upstream, downstream).
Table 2. Variables used in Pearson correlation analyses to determine if environmental variables
were correlated with Blackside Dace (dace) movement variables in either the Big Lick Branch or
Rock Creek watersheds. Three variables were only evaluated in Rock Creek and are identified with
an asterisk (*).
Environmental variables Movement variables
Days since tagging Total number of dace moved
Discharge Number of dace moved upstream
Dissolved oxygen Number of dace moved downstream
pH Number of resident dace
Day length Movement ratio (number of migrants / number of residents)
Specific conductivity Dace captured per trap hr (Blackside Dace CPUE)
Turbidity Percentage of tagged dace in each sample (number of tagged dace
Water temperature captured / total number of dace captured)
Mean overall distance moved
Mean distance moved upstream
Mean distance moved downstream
Coefficient of variation (CV) of overall mean distance moved
CV of mean distance moved upstream
CV of mean distance moved downstream
*Number of dace moved into the first-order tributaries from Rock
Creek
*Number of dace moved out of the four first-order tributaries into
Rock Creek
*Tributary ratio (number of dace moved in or out of tributaries /
total number of dace moved)
71
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist Vol. 12, Special Issue 4
Pearson correlation analyses were used to determine if movement variables
were correlated with environmental variables in either of the Big Lick Branch (21
total variables) or Rock Creek (24 total variables) watersheds (Table 2). Significance
was determined at α = 0.05, and the sequential Bonferroni technique was
used to reduce experiment-wise error (Holm 1979).
Results
Trapping
Catch per unit effort averaged 2.00 ± 0.83 (mean ± SD) Blackside Dace per
trap hour (range = 0.83–3.13 fish per trap hour) in Big Lick Branch and 2.82 ± 0.87
Blackside Dace per trap hour (range = 1.19–3.80 fish per trap hour) in Rock Creek
(Fig. 2). In Big Lick Branch, CPUE generally increased from February 2003 to
July 2003, when it reached a maximum and then began to decline, whereas CPUE
generally increased over time in Rock Creek. The percentage of tagged Blackside
Dace in each minnow-trapping event (i.e., number of tagged Blackside Dace captured
in the trapping event divided by the total number of Blackside Dace captured
in the trapping event) averaged 5.8 ± 1.6% (range = 4.0–7.9%) in Big Lick Branch
and 2.8 ± 0.5% (range = 1.75–3.3%) in Rock Creek.
Big Lick Branch
Most tagged dace (81%) were recaptured within the 200-m stream reach of
their original tagging (Fig. 3, Table 3). However, five individuals were recaptured
Figure 2. Summary of Blackside Dace (BSD) catch per unit effort (CPUE; number of fish
per trap hour) in the Big Lick Branch and Rock Creek watersheds from February 2003
to March 2004. Eight sampling trips were made in Big Lick Branch and seven in Rock
Creek using baited minnow traps to capture fish.
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist
72
Vol. 12, Special Issue 4
downstream and 17 were recaptured upstream of their original site of tagging.
Of these migrants, several individuals moved considerable distances from their
original site (Fig. 4). Three individuals moved >400 m upstream, while two
Figure 3. Summary of individual recapture locales of tagged Blackside Dace in the Big
Lick Branch and Rock Creek watersheds from February 2003 to March 2004. Recapture
locales were scored as either (1) upstream from the original tagging site (upstream
migrants; white stacked bars), (2) within the original 200-m-long tagging site (marked
residents; gray stacked bars), or (3) downstream from the original tagging site (downstream
migrants; black stacked bars).
73
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist Vol. 12, Special Issue 4
moved >100 m downstream. The mean distance moved upstream (148 ± 138 m)
in Big Lick Branch was more variable but not statistically different (H = 2.34,
P = 0.13) from the mean distance moved downstream (77 ± 29 m) (Fig. 4). The
longest upstream movement recorded was 465 m, and the longest downstream
movement was 104 m.
In addition to the within-stream movements, one individual apparently moved
downstream out of Dace Branch and traveled across the embayment of Lake
Cumberland and then upstream into Big Lick Branch where it was recaptured.
The total length of this movement was approximately 600 m. This observation
represents the first intertributary movement documented for this species.
Rock Creek
While greater movement was documented in the Rock Creek watershed, most
(58%) Blackside Dace were still recaptured within the 200-m stream reach of
their original tagging (Table 3, Fig. 3). Dace were observed moving in and out
of the four southward-flowing first-order tributaries to Rock Creek and nine individuals
migrated >1400 m upstream, while two individuals moved >1900 m
downstream. Twenty-three individuals were recaptured downstream and 27 were
captured upstream of their original site of tagging. The mean distance moved
upstream in Rock Creek (733 ± 1259 m) was not statistically different (H = 0.78,
P = 0.38) than the mean distance moved downstream (314 ± 617 m) (Fig. 4).
However, the mean overall distance moved was significantly greater (F = 17.03,
P = 0.0001) in Rock Creek than in Big Lick Branch. The longest upstream movement
recorded in Rock Creek was 3990 m, and the longest downstream movement
was 2400 m.
Movement and environmental variables
Blackside Dace movements were correlated with certain environmental
variables in both Big Lick Branch and Rock Creek as indicated by Pearson
correlation analyses. Six significant correlations (pre-correction P < 0.05) were
identified in Big Lick Branch, and 15 were identified in Rock Creek (Detar
2004). The sequential Bonferroni technique reduced the number of significant
Table 3. Number (n), percent (%), and location by distance category of tagged Blackside Dace
recaptured in minnow traps in the Big Lick Branch and Rock Creek watersheds during 2003–2004.
The original 200-m tagging site is represented by distance category “0 m”. Percentages for each
distance category were rounded to nearest whole numbers.
Big Lick Branch Rock Creek
Distance category (m) n Percent n Percent
0 94 81 70 58
10–100 12 10 30 25
101–500 10 9 8 7
501–1000 0 0 2 2
>1000 0 0 10 8
Totals 116 100 120 100
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist
74
Vol. 12, Special Issue 4
correlations from 6 to 2 in Big Lick Branch and from 15 to 3 in Rock Creek.
Post-correction significant correlations in Big Lick Branch were (1) mean overall
Figure 4. Summary of distances moved by Blackside Dace in Big Lick Branch and Rock
Creek watersheds from February 2003 to March 2004. White bars indicate upstream
migrants, gray bars represent individuals recaptured in their original 200-m-long tagging
site, and black bars indicate downstream migrants. Mean ± SD distance moved is shown
for each migrant group.
75
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist Vol. 12, Special Issue 4
distance moved * specific conductivity (r = 0.91, P = 0.0041) and (2) movement
ratio (number of migrants / number of residents) * day length (r = 0.84, P =
0.0095). Post-correction-significant correlations in Rock Creek were (1) number
of dace moved out of the four first-order tributaries into Rock Creek * days since
tagging (r = 0.94, P = 0.0017), (2) mean distance moved downstream * discharge
(r = 0.92, P = 0.0033), and (3) number moved downstream * days since tagging
(r = 0.92, P = 0.0034).
Discussion
Movement of individual fish maintains the connectivity of populations within
watersheds, reduces the potential for local extinctions, provides a means for recolonization,
and plays an important role in population genetics and community
structure (Jackson et al. 2001, Larson et al. 2002). Documentation of dispersal
and movement patterns of stream fishes (e.g., Fausch and Young 1995, Freeman
1995, Gowan and Fausch 1996, Gowan et al. 1994, Matheney and Rabeni 1995,
Skalski and Gilliam 2000, Smithson and Johnston 1999, Todd and Rabeni 1989,
Young 1994) has challenged the restricted-movement paradigm suggested by
Gerking (1959) (i.e., resident stream fish tend to be sedentary and have limited
home ranges). Several studies have shown that fish can be quite mobile in small
stream systems (Gowan and Fausch 1996, Gowan et al. 1994, Matheney and Rabeni
1995, Todd and Rabeni 1989), whereas others have indicated that stream fish
tend to have a small home range throughout the year (Chisholm et al. 1987, Freeman
1995, Heggenes et al. 1991, Hill and Grossman 1987, Skalski and Gilliam
2000, Smithson and Johnson 1999). Smithson and Johnston (1999) reported that
a small portion of four fish species—Creek Chub, Fundulus olivaceus (Storer)
(Blackspotted Topminnow), Lepomis megalotis (Rafinesque) (Longear Sunfish),
and Lepomis cyanellus Rafinesque (Green Sunfish)—in an Arkansas stream
completed regular roundtrip exploratory movements outside their original pool
of capture. Similiarly, Larson et al. (2002) reported that a small percentage of
five fish species—Rhinichthys atratulus (Hermann) (Blacknose Dace), Rhinichthys
cataractae (Valenciennes) (Longnose Dace), Cottus bairdi Girard (Mottled
Sculpin), Campostoma anomalum (Rafinesque) (Central Stoneroller), and Oncorhynchus
mykiss (Walbaum) (Rainbow Trout)—were captured upstream of
their home-range areas in a small Appalachian stream.
Our results for Blackside Dace also do not fully conform to Gerking’s (1959)
restricted-movement paradigm. Although the majority of dace were recaptured in
their original tagging sites (81% in Big Lick Branch, 58% in Rock Creek), several
individuals were recaptured considerable distances away. These results support
a body of fish-movement research suggesting that many stream fish populations
are comprised of a relatively large sedentary group and a smaller mobile group
(Breen et al. 2009, Freeman 1995, Gowan and Fausch 1996, Heggennes et al.
1991, Rodriguez 2002, Skalski and Gilliam 2000, Smithson and Johnston 1999),
although not all studies show support for two such distinct categories within
populations (e.g., Alldredge et al. 2011).
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist
76
Vol. 12, Special Issue 4
A common problem with mark-recapture movement studies is that recapture
sites are often placed in or near tagging sites, thereby making it easier to detect
shorter-range movements. Albanese et al. (2003) refer to this bias as “distance
weighting” and offer suggestions to reduce or eliminate such bias through better
study designs regarding the spatial extent and arrangement of recapture sites. We
analyzed trap arrangements for both watersheds to assess how our study design
might have produced bias toward detecting movements of different distances.
First, for any given trapping event, we determined the frequency of movement
distances in both upstream and downstream directions that could be detected from
each of the original four mainstem tagging sites in Big Lick Branch, as well as the
original seven tagging sites in Rock Creek (excluding Lot Hollow Branch). From
these distances, we then created a frequency histogram of detectable movement
distances to evaluate distance weighting in each of the two watersheds (Fig. 5).
We found that 62–66% of detectable movement distances in both streams were
located in the lower two distance quartiles (including the 0-m category), while
34–38% were in the upper two quartiles, indicating that we were more likely
to detect shorter-range than longer-range movements, particularly long-range
distances >1800 m in Big Lick Branch or >4500 m in Rock Creek (Table 4). For
short-range distances, detection of movements between 1–500 m in Rock Creek
represented the most heavily weighted category (Fig. 5). Interestingly, we had no
chance of detecting movements of certain distances in Big Lick Branch (201–400
m, 1001–1200 m, 1601–2000 m; Fig. 5), given the spatial arrangement of traps in
that stream. Our study design was not especially biased toward collecting resident
fishes in their original tagging sites, with only 5.6–8.3% of detectable movement
distances located in the 0-m category (Table 4, Fig. 5). In sum, our study did
contain distance-weighting bias, implying that we slightly overestimated the frequency
of shorter-distance movements (10–500 m), especially in Rock Creek, and
underestimated the frequency of longer-distance movements (>1000 m) reported
Table 4. Frequency distribution of movement distances that could be detected during Blackside
Dace recapture events. Distance categories are provided for original tagging sites (0 m) and streamspecific
distance quartiles determined by recapture section lengths in Big Lick Branch (2500 m)
and Rock Creek (6050 m). The frequency distribution for each watershed is illustrated at a finer
resolution in Figure 5.
Distance category Frequency (%) distribution of detectable movement distances
Big Lick Branch watershed
0 m 8.3
1–600 m 25.0
601–1200 m 29.2
1201–1800 m 25.0
1801–2400 m 12.5
Rock Creek watershed
0 m 5.6
1–1500 m 29.3
1501–3000 m 31.0
3001–4500 m 19.0
4500–6000 m 15.1
77
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist Vol. 12, Special Issue 4
Figure 5. Frequency distribution of movement distances that could be detected during
trapping (recapture) events in the Big Lick Branch and Rock Creek watersheds. Distances
include both upstream and downstream directions. Distance categories from left are 0 m,
1–200 m, 201–400 m, …, and 2201–2400 m in Big Lick Branch, and 0 m, 1–500 m,
501–1000 m, …, and 5501–6000 m in Rock Creek. The 0-m category represents traps
within original tagging sites.
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist
78
Vol. 12, Special Issue 4
in Table 3 for both watersheds. We offer no corrections to Table 3, but encourage
the reader to be aware of this slight bias present in our study design.
Larson et al. (2002) used the small-mammal dispersal theory (Lidicker
and Stenseth 1992) to interpret the movement of individual fish outside of
their home-range areas in a small Appalachian stream. Under this theory, individuals
that leave and occupy a new home range are classified as dispersers,
whereas individuals that leave a home range and then return to it at a later time
are considered explorers. Stenseth (1983) noted that dispersers and explorers
are typically robust individuals but may be small animals that are healthy. In our
study, we only recaptured 3 double-tagged individuals, which limited our ability
to identify larger patterns of multiple movements by individuals. However,
behaviors consistent with both disperser and explorer categories were present in
our study. Blackside Dace that were recaptured long distances away from their
original sites (i.e., >1000 m) and the one dace that moved 600 m downstream out
of Dace Branch and up into Big Lick Branch could be considered dispersers. An
individual from Big Lick Branch that moved upstream 110 m from its original
tagging site and later was recaptured 200 m back downstream in its original site
could be considered an explorer.
The ratio of recaptured migrant to recaptured resident Blackside Dace was
positively related to conductivity and day length in Big Lick Branch. In a Virginia
stream network, Albanese et al. (2004) also noted a positive relation between the
number of Chrosomus oreas Cope (Mountain Redbelly Dace) movements and
day length, as well as temperature and flow events. Longer days generally are
associated with warmer temperatures and consequently greater metabolic activity
in ectotherms. However, we believe the relationship between Blackside Dace
migrant ratio and conductivity represents a spurious correlation because Big Lick
Branch had relatively low conductivity values that did not vary greatly over the
course of our study.
In Rock Creek, the number of Blackside Dace moving downstream and the
movement out of tributaries were each positively correlated with time since tagging.
Passage of time allows mobile individuals to spread away from tagging
sites, thereby representing a fairly intuitive relationship. The downstream directionality
of the movement is informed by the positive relation with discharge
given that flow events might displace or otherwise encourage individuals to move
in that direction. As mentioned above, Albanese et al. (2004) also observed a positive
relation between flow events and the number of moving Mountain Redbelly
Dace. Although we found a relation between the magnitude of flow and downstream
distance moved, we could not assess how increasing or declining flows
affected Blackside Dace movement. In short, our correlation analyses simply
suggest a greater likelihood of dace movement when day lengths are longer, and
further studies will be required to better understand other potentially important
environmental associations.
Density-dependent mechanisms as well as stochastic environmental events
undoubtedly influence fish movement patterns (Freeman 1995). Freeman et al.
(1988) suggested that periodic long-range movements may allow small stream
fishes to respond to variation in resources over a large area and across a variety
79
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist Vol. 12, Special Issue 4
of stream habitats. Migrant fish play important roles in adjusting the distributions
and abundances of individuals in response to extrinsic and intrinsic factors that
affect fitness levels and could be important to maintenance and recovery of imperiled
species (Larson et al. 2002).
The two Blackside Dace populations studied here, Big Lick Branch and
Rock Creek, are among the most robust across the species’ range in Kentucky
and Tennessee (Black et al. 2013 [this issue]). It remains uncertain whether the
movement patterns observed in these two watersheds would transfer or apply to
other populations because of differences in stream habitats among watersheds
and possible density-dependent influences on movement behavior. For example,
the greater movement distances we observed in Rock Creek might be expected
because it is a longer stream than Big Lick Branch. Our recapture section in Rock
Creek was 6.0 km compared to only 2.5 km in Big Lick Branch (Fig. 1).
Albanese et al. (2009) found movement rate and abundance were the strongest
predictors of recolonization and population recovery of fishes experimentally
removed from headwater and mainstem sections of a southwestern Virginia
stream. Certain Blackside Dace individuals exhibited exceptional movements
in our study, which suggests that populations with sufficient densities should be
competent colonists if suitable stream corridors are available (e.g., Roberts and
Angermeier 2007). We observed dace readily traveling into, from, and between
tributaries in both watersheds. Such tendencies are obvious prerequisites to reestablishment
of populations after local extinction events. On the other hand, many
individuals did not exhibit movement away from tagging sites. This sedentary
tendency, coupled with low densities in many populations (Black et al. 2013 [this
issue]), may render many populations susceptible to local extinction in the case
of a stochastic event, poor year-class strength, or habitat degradation. Therefore,
the sustainability of this species hinges on protecting stream habitat quality and
promoting connectivity of suitable habitats within and among watersheds harboring
Blackside Dace populations.
Acknowledgments
The US Fish and Wildlife Service provided research funding, and we are grateful for
additional assistance provided by the Cookeville Field Office staff. Supplemental support
was provided by the Center for the Management, Utilization, and Protection of Water
Resources and Department of Biology at Tennessee Technological University (TTU).
Completion of the manuscript was facilitated by a TTU Faculty Non-Instructional Assignment
during 2011–2012. We wish to thank numerous graduate and undergraduate
students that assisted with this project, especially Brena Jones and Anthony Smith. Christine
Peterson assisted with manuscript preparation, and Charles Sutherland constructed
Figure 1. The manuscript was improved by comments from D.L. Combs, S.B. Cook, the
guest editor, and two anonymous reviewers.
Literature Cited
Albanese, B., P.L. Angermeier, and C. Gowan. 2003. Designing mark-recapture studies
to reduce effects of distance weighting on movement-distance distributions of stream
fishes. Transactions of the American Fisheries Society 132:925–939.
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist
80
Vol. 12, Special Issue 4
Albanese, B., P.L. Angermeier, and S. Dorai-Raj. 2004. Ecological correlates of fish
movement in a network of Virginia streams. Canadian Journal of Fisheries and
Aquatic Sciences 61:857–869.
Albanese, B., P.L. Angermeier, and J.T. Peterson. 2009. Does mobility explain variation
in colonisation and population recovery among stream fishes? Freshwater Biology
54:1444–1460.
Alldredge, P., M. Gutierrez, D. Duvernell, J. Schaefer, P. Brunkow, and W. Matamoros.
2011. Variability in movement dynamics of topminnow (Fundulus notatus and F. olivaceus)
populations. Ecology of Freshwater Fish 20:513–521.
Black, T.R., J.E. Detar, and H.T. Mattingly. 2013. Population densities of the threatened
Blackside Dace, Chrosomus cumberlandensis, in Kentucky and Tennessee. Southeastern
Naturalist 12(Special Issue 4):6–26.
Breen, M.J., C.R. Ruetz, K.J. Thompson, and S.L. Kohler. 2009. Movements of Mottled
Sculpins (Cottus bairdii) in a Michigan stream: How restricted are they? Canadian
Journal of Fisheries and Aquatic Sciences 66:31–41.
Chisholm, I.M., W.A. Hubert, and T.A. Wesche. 1987. Winter stream conditions and use
of habitat by Brook Trout in high-elevation Wyoming streams. Transactions of the
American Fisheries Society 116:176–184.
Detar, J.E. 2004. Population densities and movement patterns of the threatened Blackside
Dace, Phoxinus cumberlandensis. M.Sc. Thesis. Tennessee Technological University,
Cookeville, TN. 75 pp.
Detar, J.E., and H.T. Mattingly. 2005. Response of Southern Redbelly Dace to clove oil
and MS-222: Effects of anesthetic concentration and water temperature. Proceedings
of the Annual Conference of Southeastern Fish and Wildlife Agencies 58:219–227.
Eisenhour, D.J., and R.M. Strange. 1998. Threatened fishes of the world: Phoxinus cumberlandensis
Starnes & Starnes, 1978 (Cyprinidae). Environmental Biology of Fishes
51:140.
Fausch, K.D., and M.K. Young. 1995. Evolutionary significant units and movement of
resident stream fishes: A cautionary tale. American Fisheries Society Symposium
17:360–370.
Freeman, M.C. 1995. Movements by two small fishes in a large stream. Copeia
1995:361–367.
Freeman, M.C., M.K. Crawford, J.C. Barrett, D.E. Facey, M.G. Flood, J. Hill, D.J. Stouder,
and G.D. Grossman. 1988. Fish assemblage stability in a southern Appalachian
stream. Canadian Journal of Fisheries and Aquatic Sciences 45:1949–1958.
Friedman, M. 1937. The use of ranks to avoid the assumption of normality implicit in
the analysis of variance. Journal of the American Statistical Association 32:675–701.
Gallagher, A.S., and N.J. Stevenson. 1999. Streamflow. Pp. 149–157, In M.B. Bain and
N.J. Stevenson (Eds.). Aquatic Habitat Assessment: Common Methods. American
Fisheries Society, Bethesda, MD. 224 pp.
Gerking, S.D. 1959. The restricted movements of fish populations. Biological Review of
the Cambridge Philosophical Society 34:221–242.
Gowan, C., and K.D. Fausch. 1996. Mobile Brook Trout in two high-elevation Colorado
streams: Re-evaluating the concept of restricted movement. Canadian Journal of Fisheries
and Aquatic Sciences 53:1370–1381.
Gowan, C., M.K. Young, K.D. Fausch, and S.C. Reilly. 1994. Restricted movement
in resident stream salmonids: A paradigm lost? Canadian Journal of Fisheries and
Aquatic Sciences 51:2626–2637.
Heggenes, J.T., G. Northcote, and A. Peter. 1991. Spatial stability of Cutthroat Trout
(Oncorhynchus clarki) in a small coastal stream. Canadian Journal of Fisheries and
Aquatic Sciences 48:1364–1370.
81
J.E. Detar and H.T. Mattingly
2013 Southeastern Naturalist Vol. 12, Special Issue 4
Hill, J., and G.D. Grossman. 1987. Home-range estimates of three North American
stream fishes. Copeia 1987:376–380.
Holm, S. 1979. A simple sequentially rejective multiple test procedure. Scandinavian
Journal of Statistics 6:65–70.
Jackson, D.A., P.R. Peres-Neto, and J.D. Olden. 2001. What controls who is where in
freshwater fish communities –- the roles of biotic, abiotic, and spatial factors. Canadian
Journal of Fisheries and Aquatic Sciences 58:157–170.
Kruskall, W.H., and W.A. Wallis. 1952. Use of ranks in one-criterion analysis of variance.
Journal of the American Statistical Association 47:583–621.
Larson, G.L., R.L. Hoffman, and S.E. Moore. 2002. Observations of the distributions of
five fish species in a small Appalachian stream. Transactions of the American Fisheries
Society 131:791–796.
Lidicker, W.Z., Jr., and N.C. Stenseth. 1992. To disperse or not to disperse: Who does it
and why? Pp. 21–36, In N.C. Stenseth and W.Z. Lidicker, Jr. (Eds.). Animal Dispersal:
Small Mammals as a Model. Chapman and Hall, London, UK. 365 pp.
Matheney, M.P., IV, and C.F. Rabeni. 1995. Patterns of movement and habitat use by
Northern Hog Suckers in an Ozark stream. Transactions of the American Fisheries
Society 124:886–897.
O’Bara, C.J. 1990. Distribution and ecology of the Blackside Dace Phoxinus cumberlandensis
(Osteicthyes: Cyprinidae). Brimleyana 16:9–15.
Roberts, J.H., and P.L. Angermeier. 2007. Movement responses of stream fishes to introduced
corridors of complex cover. Transactions of the American Fisheries Society
136:971–978.
Rodriguez, M.A. 2002. Restricted movement in stream fish: The paradigm is incomplete,
not lost. Ecology 83(1):1–13.
Skalski, G.T., and J.F. Gilliam. 2000. Modeling diffusive spread in a heterogeneous
population: A movement study with stream fish. Ecology 81(6):1685–1700.
Smithson, E.B., and C.E. Johnston. 1999. Movement patterns of stream fishes in a
Ouachita Highlands stream: An examination of the restricted-movement paradigm.
Transactions of the American Fisheries Society 128:847–853.
Starnes, L.B., and W.C. Starnes. 1981. Biology of the Blackside Dace, Phoxinus cumberlandensis.
The American Midland Naturalist 106:360–370.
Starnes, W.C. 1981. Listing package for the Blackside Dace, Phoxinus cumberlandensis.
US Fish and Wildlife Service, Asheville, NC.
Starnes, W.C., and L.B. Starnes. 1978. A new cyprinid of the genus Phoxinus endemic to
the upper Cumberland River drainage. Copeia 1978:508–516.
Stenseth, N.C. 1983. Causes and consequences of dispersal in mammals. Pp. 63–102, In
I.R. Swingland and P.J. Greenwood (Eds.). The Ecology of Animal Movement, Clarendon
Press, Oxford, UK. 311 pp.
Todd, B.L., and C.F. Rabeni. 1989. Movement and habitat use by stream-dwelling Smallmouth
Bass. Transactions of the American Fisheries Society 118:229–242.
US Fish and Wildlife Service (USFWS). 1988. Recovery plan for Blackside Dace (Phoxinus
cumberlandensis). Atlanta, GA. 23 pp.
Woolnough, D.A., J.A. Downing, and T.J. Newton. 2009. Fish movement and habitat use
depends on water body size and shape. Ecology of Freshwater Fish 18:83–91.
Young, M. 1994. Mobility of Brown Trout in south-central Wyoming streams. Canadian
Journal of Zoology 72:2078–2083.