Content uploaded by Miroslav Kubín
Author content
All content in this area was uploaded by Miroslav Kubín on Mar 14, 2018
Content may be subject to copyright.
J Appl Ichthyol. 201 8;1–9. wileyonlinelibrary.com/journal/jai
|
1
© 2018 Black well Verlag GmbH
1 | INTRODUCTION
Floods are an important phenomena in the freshwater stream eco-
system, as they form the stream habitat and connect streams with
the riparian zone. During floods, streambeds process and transfer
a large amount of sediment (Kondolf, 1997). This temporary distur-
bance of the bottom habitat is especially challenging for benthic
organisms and thus benthic animals display an array of adaptations
for high flow (Naiman, Decamps, & McClain, 2005). Alpine bullhead,
Cottus poecilopus, is a benthic species that is able to hold its position
in the fas t flowing water (Coom bs, Anderson, Braun, & G rosenbaugh,
2007). Adaptations of the Alpine bullhead for fast flowing water
include morphological traits such as a low body height, flat head and
body, as well as a large caudal peduncle depth (Kerfoot & Schaefer,
2006). Alpine bullhead also display behavioural adaptations like shel-
tering in the less exposed floodplain and tributaries during floods
(Bayley, 1991; Sato & Yoshimura, 2014).
Over the last hundred years, most European rivers have un-
dergone significant morphological changes associated with an-
thropogenic modifications of channels and floodplains, which has
resulted in serious habitat degradation (Kondolf, 1997). The artificial
changes have made streams straighter, incised, and insulated from
the floodplains. These changes in the morphology of stream chan-
nels, together with climate change, have increased the frequency
Received:4August2017
|
Accepted:7Februa ry2018
DOI: 10.1111/jai.13682
ORIGINAL ARTICLE
Habitat degradation and trout stocking can reinforce the
impact of flash floods on headwater specialist Alpine bullhead
Cottus poecilopus – A case study from the Carpathian
Mountains
M. Kubín1,2 | M. Rulík1 | S. Lusk3 | L. Závorka4
1Department of Ecology and Environmental
Sciences, Faculty of Science, Palacký
University Olomouc, Olomouc, Czech
Republic
2Nature Conservation A gency of the Czech
Republic, Chodov, Czech Republic
3BohuslavaMartinů,Brno,CzechRepublic
4CNRS, UMR 5174 EDB, Université Toulouse
3 Paul Sabatier, Toulouse, France
Correspondence
Miroslav Kubín, Department of Ecology and
Environmental Sciences, Faculty of Science,
Palacký University Olomouc, Olomouc,
Czech Republic.
Email: miroslav.kubin@email.cz
Funding information
IGA_PrF_2016_019 Project of Palacky
University in Olomouc
Summary
In freshwater streams, flooding is a typical source of natural disturbance that plays a
key role in the dynamics of animal populations and communities. However, habitat
degradation and fish stocking might increase the severity of its impact. We tested the
effects of a flash flood on the abundance of three size classes of headwater dwelling
Alpine bullhead, Cottus poecilopus, in the streams of the Carpathian Mountains in the
Czech Republic, that are stocked with hatchery- reared brown trout, Salmo trutta. We
showed that the overall abundance of Alpine bullhead was highest at the sites with
the least degraded habitat (i.e., natural habitat) and we caught almost no Alpine bull-
head at the sites with the most degraded habitat. The flash flood had a strong nega-
tive effect on the abundance of the largest individuals of Alpine bullhead. Abundance
of small and medium size Alpine bullhead was negatively affected by the abundance
of adult stocked brown trout before as well as after the flash flood. However, nega-
tive effect of adult brown trout abundance on abundance of large Alpine bullhead
was not significant before the flash flood, and it became significant after the flash
flood. This could indicate an accumulation of negative impacts of trout stocking and
flash flood on this size class. Overall, our results suggest that stocking of hatchery
trout and habitat degradation can reinforce the impact of flash floods on the popula-
tion of Alpine bullhead in the streams of the Carpathian Mountains.
2
|
KUBÍN et al.
and severity of sudden natural phenomena like flash floods (Brunetti
et al., 2006). Such events can arise due to heavy precipitation over
several hours, followed by a quick and usually short- term rise of the
flow (Doswell, Brooks, & Maddox, 1996). Flash floods are associ-
ated with fast flow rates and significant geomorphological changes
in the streambed, which can have a substantial negative effect on
fish communities in mountain streams (Foulds, Griffiths, Macklin, &
Brewer,2014;Lusk,Halačka,&Lusková,1998).
In addition to degradation of stream habitats, frequent stock-
ing of hatchery reared salmonids like brown trout, Salmo trutta, has
become a common practice in many streams worldwide since the
mid- 19th century (Stankovic, Crivelli, & Snoj, 2015). Stocking hatch-
ery brown trout is known to have a negative impact on the growth
and survival of various stream dwelling species (Buoro, Olden, &
Cucherousset, 2016). Alpine bullhead are less abundant at sites with
a high abundance of S. trutta (Baran et al., 2014). However, the ef fect
of habitat degradation and trout stocking on the capacity of Alpine
bullhead populations to resist natural disturbances like flash floods
has not been addressed.
This study compares the abundance of three size classes of
Alpine bullhead at 15 stretches of three streams in the Carpathian
Mountains in the Czech Republic, before and after a flash flood. The
aim of this study was to examine how the populations of Alpine bull-
head respond to flash floods across a range of habitats varying in
quality and abundance of stocked hatchery brown trout.
2 | MATERIALS AND METHODS
2.1 | Study locality
The headwaters of the Rožnovská Bečva River are located in
the Beskydy Landscape Protected area in the Czech Republic
(49°29′N,18°08′E;Figure1)atthewesternedgeoftheCarpathian
Mountains. It is a typical torrential gravel- bed stream with a length
of 37.6 km and basin area of 254.3 km2. The mean annual dis-
charge of the Rožnovská Bečvariversis2.72m3/sat Rožnovpod
Radhoštěm(datasource:CzechHydometeorologicalInstitute).The
bedrock of the study area is composed of flysch Cretaceous rocks
dominated by Quaternary sediment (Menčík & Tyráček, 1985).
The widespread vegetation in the upstream section is spruce for-
est, which alternates with willow, alder vegetation and meadows
downstream. Fifteen sampling stretches were chosen at the three
FIGURE1 MapshowingthewatershedoftheRožnovskáBečvaRiverandthreemaintributaries(Starozuberskýstream,Dolnopasecký
stream,Zákopeckýstream)with15samplingstretches(SZ1–5,DP1–5,ZP1–5)
|
3
KUBÍN e t al.
maintributariesoftheRožnovskáBečvaRiver:TheDolnopasecký,
StarozuberskýandZákopeckýstream(seeFigure1andTable1for
habitat description). The abundance of brown trout in the stud-
ied streams is maintained above its natural level due to periodical
stocking by the Czech angling association, which is performed in
2 year cycles. Adult brown trout (2+) are captured every second
year and juveniles (0+) are stocked of with about 3.5 thousand of
individuals per km. Torrential precipitation caused a flash flood in
alltheRožnovskáBečvacatchmentriversfrom16to19May2010
(precipitation of 145.4 mm was recorded on May 16). Consequently,
the water flow in the observed streams increased approximately a
hundred times within several hours (for example, the discharge in
Zákopeckýstreamduringthefloodpeakwas10.5m3/s, compared
to the average daily flow in May, which is 0.184 m3/s). The flash
flood rate reached the level of the 5–10 year flood record. In the
other streams, the flow reached approximately the 10- year flood
recordlevelaswell.TheRožnovskáBečvariverreachedavalueof
approximately the 50- year flood level (data source: Czech Hydro-
Meteorological Institute).
2.2 | Data sampling
We selected five 100 m long sampling stretches in each of three
study streams. The mean distance between the stretches within a
stream was 1 km. Sampling of the stretches was carried out before
(from September to November 20 09) and after (from July to October
2010) the flood. Alpine bullhead and brown trout were captured
by two- pass electrofishing using a backpack electroshocker (SEN
[200–240V, straight DC], B ednář, Czech Republic ). The upstream
and downstream edges of the sampling stretches were obstructed
by nets with a mesh size 5 × 5 mm, to prevent sampled fish escaping.
Every fish was measured (total length to the nearest mm). Following
the measurements, all specimens were placed in a recovery tank and
then released back to the place where they were caught. We did
not estimate densities of age 0+ juveniles (>35 mm total length, TL)
due to the low sampling efficiency of electrofishing for this age class
(Bohlin, Hamrin, Heggbergert, Rasmussen, & Saltveit, 1989). The
final estimate of the brown trout and Alpine bullhead abundance in
the investigated sampling sections was done in accordance with the
calculations in Seber and LeCren (1967). Alpine bullhead is defined
as an “endangered” species within the Czech national conservation
legislature (Law no. 114/1992 Coll. on the conservation of nature
and landscapes).
The level of habitat degradation within the sampled stretches
was evaluated before the flash flood following a method pro-
posed by Weiß, Matouskova, and Matschullat (2008) for EU- Water
Framework Directive. The method assesses the eco- hydrological
quality of rivers based on 32 parameters e.g., character of flow,
diversity of microhabitats, human- made changes in flow regime,
occurrence of artificial steps, vegetation, land- use in the flood-
plain, retention of the floodplain, flood protection measures, river
valley type, etc.). The aquatic ecosystem is understood as a terri-
tory formed by mutually interconnected zones: channel, riparian
belt and floodplain. Riverbank features, riparian belt and floodplain
are separately assessed during mapping, but the final assessment
is calculated jointly for both sides. Results in simple point classifi-
cation (points I–V), where (I) corresponds to the reference, i.e., the
natural habitat without human intervention, (II) Slightly changed, (III)
Moderately changed, (IV) Strongly changed, and (V) characterizes
a completely changed habitat. The average water depth was calcu-
lated based on measurements with a calibrated bar at five sites in
each section. The river substratum was sampled using a method de-
vised by Wolman (1954) and quantified according to Bain, Finn, and
Booke (1985). We used the stream substrate classification system:
boulder (>256 mm), cobble (64–256 mm), pebble (16–64 mm), gravel
(2–16 mm), sand, silt (<2 mm).
2.3 | Data handling and statistics
We used generalized linear models (GLM) for count data (i.e.,
Poisson distribution) with positive skewed distribution (i.e., log
link). All models contained a study stream as a covariate to ac-
count for variability between the sampled streams. The first
model was built to test the effect of flash flood and habitat degra-
dation across the size classes of Alpine bullhead. For this purpose,
individuals of Alpine bullhead were split in to three size classes:
small (body length range 40–65 mm), medium (body length range
70–90 mm), and large (body length range 95–125 mm). Alpine
bullhead can start to spawn when they reach a body length of
40 mm, but probability of maturity and number of eggs pro-
duced by females is positively correlated with body size (Freyhof,
Kottelat, & Nolte, 2005). The models contained abundance of
individuals as response variable and the year of sampling (two
levels), size class (three levels), level of habitat degradation (four
levels) as independent variables (see Table 2 for detailed model
structure). The effect of brown trout abundance on bullhead was
tested by models containing bullhead abundance as a response
and bullhead size class, initial abundance (i.e., before the flood) of
juvenile and adult brown trout, the year of sampling (two levels)
as independent variables (see Table 3 for detailed model struc-
ture). Based on a previous study describing growth of brown trout
inmountainousheadwatersincentralEurope(Závorka,Horký,&
Slavík, 2013), we considered brown trout smaller than 110 mm as
juvenile and brown trout larger than 110 mm as sub adult or adult.
The effect of the flash flood on the stream channel morphology
was tested using chi- square test for categorical variable (i.e., sub-
strate type) and pair- wise T- test for continuous variables (i.e.,
substrate homogeneity and water depth). To avoid the mistreat-
ment of covariate interaction terms, non- significant interactions
in GLM were removed from models and the model was run again
without them (Engqvist, 2005). The removal was performed by a
step- wise method of selection from the highest non- significant
interaction to the lowest. Model fit was assessed by visual inspec-
tion of distribution of model residuals and plots of fitted values
versus residuals. Differences among categories of fixed factors
were tested using Tukey’s HSD post- hoc test. Statistical analyses
4
|
KUBÍN et al.
TABLE1 Basic characteristics of each sampled river section. Ecological status of sampling sections following Weiß et al. (2008) was used to evaluate the level of habitat degradation
Study streams
Sampling
section ID
Width
(m)
Elevation
(ma.s.l.)
Habitat
degradation
Before flood (2009) After flood (2010)
Dominant
substrate (%)
Dominant
substrate
(%)
Water
depth (m)
Salmo trutta
(ind. per
100 m−1 )
Dominant
substrate
(%)
Dominant
substrate
(%)
Water
depth (m)
Salmo
trutta
(ind. per
100 m−1 )
Dolnopasecký DP 1 3.2 505 II Boulders 50 0.14 12 Boulders 41 0 .14 9
DP 2 4.3 455 III Pebbles 41 0.12 48 Gravel 55 0.13 19
DP 3 4.5 425 II Pebbles 50 0.15 131 Gravel 40 0.16 43
DP 4 4.2 410 III Pebbles 80 0.15 192 Gravel 45 0.17 33
DP 5 4.0 390 IV Gravel 40 0.17 122 Gravel 60 0.19 15
Starozuberský SZ 1 3.8 500 IPebbles 50 0.1 42 Pebbles 65 0.9 10
SZ 2 3.6 465 IPebbles 55 0.9 41 Pebbles 45 0.10 19
SZ 3 4.5 440 II Pebbles 70 0.11 14 Pebbles 60 0.12 9
SZ 4 4.2 405 III Pebbles 65 0.13 114 Pebbles 80 0.11 23
SZ 5 4.0 365 III Pebbles 60 0.15 88 Boulders 35 0.18 16
Zakopecký ZP 1 3.2 605 IBoulders 60 0.15 75 Boulders 45 0.21 106
ZP 2 4.3 545 III Pebbles 50 0.21 53 Boulders 40 0.23 136
ZP 3 4.5 525 II Pebbles 50 0.14 82 Pebbles 40 0.24 29
ZP 4 4.2 505 III Pebbles 45 0.25 40 Boulders 45 0.27 18
ZP 5 4.0 485 III Boulders 35 0.31 33 Pebbles 50 0.29 34
|
5
KUBÍN e t al.
were performed using R software v.3.2.2 (R Development Core
Team, 2015), and multcomp (Hothorn, Bretz, & Westfall, 2008),
car (Fox & Weisberg, 2015), and lme4 (Bates, Maechler, Bolker, &
Walker, 2015) packages.
3 | RESULTS
The flash flood had a strong effect on the abundance of Alpine bull-
head in the studied streams. However this effect differed across the
size classes (χ2 = 12.1; p < .0001; Tables 2 and S1). Total abundance
of bullhead decreased from 781 to 684 individuals, which represents
a reduction of 12.4% of Alpine bullhead abundance before the flash
flood. There were only slight changes in the abundance of individuals
with small (χ2 = 4.0; p = .046) and medium (χ2 = 3.0; p = .084) body
sizes before and after the flash flood. However the number of large
individuals decreased substantially after the flash flood (χ2 = 37.6;
p < .0001; Figure 2).
The abundance of Alpine bullhead was highest at the sites with
the least degraded habitat (i.e., natural habitat) and we caught al-
most no Alpine bullhead at the sites with the most degraded hab-
itat (Figure 3; Table S1). However, abundance of Alpine bullhead at
the sites with natural habitat significantly decreased after the flash
flood, from 207 to 154 individuals (χ2 = 15.4; p < .0001; Figure 3).
In contrast, abundance of Alpine bullhead was not affected by flash
floods at the sites with degraded habitat (Habitat quality II: χ2 = 1.3;
p = .253; Habitat quality III and IV: χ2 = 2.0; p = .155; Figure 3). We
found no significant change in the type of the dominant substrate
(χ2 = 1.866; p = .3934) and substrate homogeneity (|t| = 0.8705;
p = .3987) and only a slight increment (2 cm in average) of the stream
channel depth (|t| = 2.2121; p = .0440; Table 1) after the flash flood.
Initial abundance of adult stocked brown trout (i.e., before
the flash flood) had a strong negative effect on the abundance of
small and medium sized Alpine bullhead, and this negative effect
was consistent between the years before and after the flash flood
(Tables 3 and S1; Figure 4). However, adult trout abundance also
had a significant additive negative effect on the abundance of large
Alpine bullhead following the flash flood. In particular, the nega-
tive slope of the relationship between initial abundance of adult
brown trout abundance and abundance of Alpine bullhead was not
significant before the flash flood (χ2 = 0.8; p = .363), but it became
significant after the flash flood (χ2 = 5.4; p = .020; Figure 4). In con-
trast to the impact of adult brown trout, the effect of abundance
of juvenile brown trout on Alpine bullhead was relatively weak
(Tables 3 and S1; Figure 4). Specifically, we found that there was a
negative relationship between abundance of juvenile brown trout
and abundance of Alpine bullhead only before the flash flood and
only in small (χ2 = 3.9; p = .048) and medium size classes (χ2 = 7. 0;
p = .008).
The overall abundance of stocked brown trout significantly de-
creased after the flash flood in most studied stretches (χ2 = 13.620;
p = .0086), from 1,087 to 519 individuals, which represents are duc-
tion of 52.3% of brown trout abundance before the flash flood.
4 | DISCUSSION
Several studies have demonstrated the negative effect of floods on
populations of sculpins. For example Cottus pullex in a Japanese river
TABLE2 Generalized linear models (GLM) testing the effects of
habitat degradation and flash flood on abundance of Alpine
bullhead
Overall model df χ2p
Intercept 1946. 2 <.0 01
Year 112.1 <.001
Stream 2353.8 <.001
Size group 2 75.1 <.0 01
Habitat quality 3107. 2 <.0 01
Size group: Year 2 39.4 <.001
Habitat quality: Year 312.6 .006
Size group 45–65 mm
Intercept 1554.6 <.0 01
Year 14.0 .046
Habitat quality 344.8 <.0 01
Stream 256.4 <.001
Size group 70–90 mm
Intercept 1873.7 <.001
Year 13.0 .084
Habitat quality 33 9. 3 <.0 01
Stream 2193 .5 <.001
Size group 95–125 mm
Intercept 1170.5 <.0 01
Year 137.6 <.001
Habitat quality 351 .9 <.001
Stream 21 57.5 <.001
Habitat quality I
Intercept 1536.9 <.001
Year 115.4 <.0 01
Size group 2 63.7 <.001
Stream 17 7. 3 <.001
Habitat quality II
Intercept 1305.7 <.001
Year 11.3 .253
Size group 2 80.3 <.001
Stream 261.6 <.001
Habitat quality III and IV
Intercept 1492.1 <.001
Year 12.0 .155
Size group 2 146.1 <.001
Stream 2201.1 <.0 01
Interactions in the overall model are decomposed in post- hoc models for
each size group and the level of habitat degradation. Report contains the
independent variables, degrees of freedom, χ2, and level of significance
described by p- value.
6
|
KUBÍN et al.
(Natsumeda, 2003) or Cottus carolinae exposed to severe floods in
Arkansas (Matthews, 1986). In agreement, we found a decrease in
the abundance of Alpine bullhead following the flash flood, which
was especially pronounced in the largest individuals with the high-
est reproductive value (Freyhof et al., 2005). Large Alpine bull-
heads (95–125 mm) often occur in the middle of stream channels
with coarser substrata and deeper water (Davey, Hawkins, Turner,
& Doncaster, 2005; Liefferinge, Seeuws, Meire, & Verheyne, 2005).
These microhabitats provide shelters against predators and can pro-
tect bullhead against increased water flow. However, water energy
at the peak flow during the flash flood transports stream- bed ma-
terial like cobbles and boulders. Therefore, there is an increase in
TABLE3 Generalized linear models (GLM) testing the effects of
abundance of juvenile and adult brown trout and flash flood on
abundance of Alpine bullhead
Overall model df χ2p
Intercept 1715.9 <.001
Initial number of juvenile ST 1 26.0 <.001
Initial number of adult ST 1 21.3 <.001
Year 10.2 .618
Stream 2204.6 <. 001
Size group 2 3 9.4 <.001
Initial number of juvenile ST:
Year
11.2 .279
Initial number of juvenile ST:
Size group
230.8 <.001
Initial number of adult ST:
Year
11.8 .177
Initial number of adult ST:
Size group
21.7 .437
Size group: Year 2 4.7 .096
Initial number of juvenile ST:
Size group: Year
27.2 .026
Initial number of adult ST:
Size group: Year
29.5 .008
Size group 45–65 mm before flood
Intercept 1653.8 <.001
Initial number of juvenile
ST
13.9 .048
Initial number of adult ST 1 33.4 <.001
Stream 210.7 .005
Size group 45–65 mm after flood
Intercept 1671.4 <.001
Initial number of juvenile
ST
11.7 .188
Initial number of adult ST 1 28.1 <.001
Stream 236.0 <.001
Size group 70–90 mm before flood
Intercept 1896.4 <.001
Initial number of juvenile
ST
17.0 .008
Initial number of adult ST 1 27. 2 <.001
Stream 218.8 <.001
Size group 70–90 mm after flood
Intercept 1818.4 <.001
Initial number of juvenile
ST
11.3 .256
Initial number of adult ST 1 40.2 <.001
Stream 292.5 <.001
Size group 95–125 mm before flood
Intercept 195.1 <.001
Initial number of juvenile ST 11.1 .294
(Continues)
Overall model df χ2p
Initial number of adult ST 1 0.8 .363
Stream 2201.1 <.0 01
Size group 95–125 mm after flood
Intercept 12.2 .132
Initial number of juvenile
ST
10.1 .698
Initial number of adult ST 1 5.4 .020
Stream 2201.1 <.0 01
Interactions in the overall model are decomposed in post- hoc models for
each size group and the year before and after flash flood. Report con-
tains the independent variables, degrees of freedom, χ2, and level of sig-
nificance described by p- value.
TABLE3 (Continued)
FIGURE2 Barplot indicates number of Alpine bullhead
(mean ± SD) pooled by three size group (small 40–65 mm, medium
70–90 mm, and large 95–125 mm) before (year 2009) and after
(year 2010) the flash flood. Light grey fill columns represent
numbers of bullheads in 2009, black fill columns represent numbers
2010
|
7
KUBÍN e t al.
the mortality of benthic fishes like Alpine bullhead seeking shelter in
the riverbed during such events (Erman, Andrews, & Yoder- Williams,
1988). In contrast, smaller individuals usually live along the edges
of stream channels (Harvey & Stewart, 1991) and have a higher
chance of surviving the flood due to higher habitat complexity, less
water velocities (Leopold, Wolman, & Miller, 1964; Pearsons, Li, &
Lamberti, 1992).
Despite the fact that stream- bed morphology was only margin-
ally affected by flash flood, we cannot exclude the possibility that
the alternation of habitat during the flash flood may have caused
an outmigration of large individuals to less influenced stretches e.g.,
with relatively low water velocities along stream margins (Harvey,
Nakamoto, & White, 1999). We found that the degraded habitat sup-
ported an overall lower abundance of Alpine bullhead than natural
habitat. Their higher abundance in the more natural stretches was
probably due to the higher habitat complexity (Pearsons et al., 1992),
higherstabilityoftheriverbed(Lojkásek,Lusk,Halačka,Lusková,&
Drozd, 2005; Edwards & Cunjak, 2007), variability of flow velocities
(Grift et al., 2003) and refuge availability (Deboer, Ogren, Holtgren,
& Snyder, 2011).
We found that juvenile brown trout had a relatively low impact
on Alpine bullhead, but all size classes of Alpine bullhead were neg-
atively affected by adult brown trout, suggesting that predation of
juvenile bullheads and competition with adult bullhead can be the
main drivers of the species interaction. Brown trout can have a
negative effect on bullhead populations through interspecific com-
petition (i.e., diet overlap, Louhi, Mäki- Petäys, Huusko, & Muotka,
2014 or niche overlap, Hesthagen, Saksgård, Hegge, Dervo, &
Skurdal, 2004). Adult trout abundance also had a negative effect
on the abundance of large Alpine bullhead especially following the
flash flood. Therefore, we suggest that the overall decrease of large
Alpine bullhead after the flash flood can be explained by a combina-
tion of habitat loss and competition with stocked brown trout (Louhi
et al., 2014).
FIGURE3 Barplot indicates number of Alpine bullhead
(mean ± SD) pooled by level of habitat degradation at the sampling
site (see Table 1 for details) before (year 2009) and after (year
2010) the flash flood. Light grey fill columns represent numbers of
bullheads in 2009, black fill columns represent numbers 2010
FIGURE4 Relationship between the initial abundance of juvenile and adult brown trout (i.e., before the flash flood) and abundance of
Alpine bullhead before (black curves) and after (light grey curves) the flash flood predicted by GLM. Size groups are represented by different
line type: dotdash–small 40–65 mm, dashed –medium 70–90 mm, full line–large 95–125 mm
8
|
KUBÍN et al.
Abundance of brown trout itself was rapidly reduced after the
flash flood, likely because they are unable to keep their position
instreams during increased flow. This result is in agreement with
a previous study on hatchery reared brown trout, which demon-
strated that their abundance was significantly reduced following
massive floods in the Horokiwi stream, New Zealand (Allen, 1951).
Trout population decline during floods may be associated with
reduced heterogeneity of soil riverbed substrate or lack of sub-
mersed wood (Harvey et al., 1999) and available refuges (Deboer
et al., 2011). Likewise, the decrease of trout abundance can be re-
lated to the fact that the hatchery trout are insufficiently adapted
to natural environment (Brockmark, Adriaenssens, & Johnsson,
2010).
In conclusion, the results of the present study show that hab-
itat degradation and stocking of hatchery trout, above the nat-
ural occurring population size, imposes a negative effect on the
abundance of Alpine bullhead and reduces their capacity to mit-
igate the negative impacts of natural disturbances such as flash
floods. This was especially true for large bodied individuals with a
high reproductive capacity. Our findings imply that a reduction of
trout stocking and stream restoration may help to protect endan-
gered fish species and increase the stability of fish communities in
mountain headwaters.
ACKNOWLEDGEMENT
Wewouldliketothankthestudentsandco-workerJolanaJuřicová
fromthehighschoolofagricultureandenvironmentalinRožnovpod
Radhoštěm for assistancein thefield. Wearegratefulto Vlastimil
Kostkan,Evžen Tošenovský,VlastimilRaškaandJaromírKudlákfor
comments during the study. The study was supported by the IGA_
PrF_2016_019 project of Palacky University in Olomouc. All of the
experimental procedures comply with valid legislative regulations of
the Czech Republic (Law no. 246/1992, §19, art. 1, letter c and no.
114/1992, §56, art. 1).
ORCID
M. Kubín http://orcid.org/0000-0003-3662-9620
REFERENCES
Allen, K . R. (1951). The Horokiwi Stream: A study of a trout population.
The New Zealand Marine Department Fisheries Bulletin, 10, 238.
Bain, M . B., Finn, J. T., & Booke, H . E. (1985). Quanti fying stre am substrate
for habitat studies. North American Journal of Fisheries Management,
5, 499–50 0.
Baran, R., Kubecka, J., Kubin, M., Lojkasek, B., Mrkvicka, T., Ricard, D.,
& Rulik, M. (2014). Abundance of Cottus poecilopus is influenced by
O2saturation, food density and Salmo trutta in three tributaries of
theRožnovskáBečvaRiver,Czech Republic.Journal of Fish Biology,
86, 805–811.
Bates, D., Maechler, M., Bolker, B., & Walker, S. (2015). Fitting linear
mixed- effects models using lme4. Journal of Statistical Soft ware, 67,
1–4 8 .
Bayley, P. B. (1991). The flood pulse advantage and the restoration of
river–floodplain systems. Regulated Rivers Research & Management, 6,
75–86. https://doi.org/10.1002/(ISSN)1099-1646
Bohlin, T., Hamrin, S., Heggbergert, T. G., Rasmussen, G., & Saltveit, S.
J. (1989). Electrofishing—Theory and practice with special emphasis
on salmonids. Hydrobiologia, 173 , 9–43. https://doi.org/10.1007/
BF00008596
Brockmark, S., Adriaenssens, B., & Johnsson, J. I. (2010). Less is more:
Density influences the development of behavioural life skills in
trout. Procceding s of the Royal Society B, 277, 3035–3043. https://doi.
org/10.1098/rspb.2010.0561
Brunetti, M., Maugeri, M., Nanni, T., Auer, I., Bohm, R., & Schöner, W.
(2006). Precipitation variability and changes in the greater Alpine re-
gion over the 1800–2003 period. Journal of Geophysical Research, 111,
https://doi.org/10.1029/2005JD006674.
Buoro, M., Olden, J. D., & Cucherousset, J. (2016). Global Salmonidae
introductions reveal stronger ecological effects of changing
intraspecific compared to interspecific diversity. Ecology Letters, 19,
1363–1371. htt ps://doi.org/10 .1111/ele.12673
Coombs, S., Anderson, E., Braun, C. B., & Grosenbaugh, M. (2007). The
hydrodynamic footprint of a benthic, sedentary fish in unidirectional
flow. The Journal of the Acou stical Society of A merica, 122, 1227–1237.
https://doi.org/10.1121/1.2749455
Davey, A. J. H., Hawkins, S. J., Turner, G. F., & Doncaster, C. P. (2005).
Size- dependent microhabitat use and intraspecific competition
in Cottus gobio. Journal of Fish Biology, 67, 428–443. https://doi.
org /10.1111/j.0 022-1112. 20 05.0 0736.x
Deboer, J. A., Ogren, S. A., Holtgren, J. M., & Snyder, E. B. (2011). A 100-
year flood in a low- gradient stream: Response of the resident and
non- resident fish assemblages. The American Midland Naturalist, 166,
446–452. https://doi.org/10.1674/0003-0031-166.2.446
Doswell, Ch. A., Brooks, H. B., & Maddox, R. A. (1996). Flash flood fore-
casting: An ingredients- based methodology. Weather and Forecasting,
11, 56 0–581. https://doi.org/10.1175/1520-0 434(1996)011<056
0:FFFAIB>2.0.CO;2
Edwards, P. A., & Cunjak, R. A. (2007). Influence of temperature and
streambed stability on the abundance and distribution of slimy scul-
pin (Cottus cognatus). Environmental Biology of Fish, 80, 9–22. https://
doi.org/10.10 07/s10641-0 06-9102-8
Engqvist, L. (2005). The mistreatment of covariate interaction terms
in linear model analyses of behavioural and evolutionary ecology
studies. Animal Behaviour, 70, 967–971. https://doi.org/10.1016/j.
anbehav.2005.01.016
Erman, D. C., Andrews, E. D., & Yoder-Williams, M. (1988). Effects
of winter floods on stream fishes in the Sierra Nevada. Canadian
Journal of Fisheries and Aquatic Sciences, 45, 2195–2200. https://doi.
or g/10 .113 9/f 8 8-255
Foulds, S. A., Griffiths, H. M., Macklin, M. G., & Brewer, P. A. (2014).
Geomorphological records of extreme floods and their relationship
to decadal- scale climate change. Geomorphology, 216, 193–207.
https://doi.org/10.1016/j.geomorph.2014.04.003
Fox, J., & Weisberg, S. (2015). Functions and datasets to accompany, an
R companion to applied regression (2nd ed.). Thousand Oaks, CA:
Sage.
Freyhof, J., Kottelat, M., & Nolte, A. (2005). Taxonomic diversity of
European Cottus with description of eight new species (Teleostei:
Cottidae). Ichthyological Exploration of Freshwaters, 16(2), 107–172.
Grift, R. E., Buijse, A. D., Van Dense, W. L. T., Machiels, A. A. M.,
Kranenbarg, J., Klein Breteler, J. G. P., & Backx, J. J. G. M. (2003).
Suitable habitats for 0- group fish in rehabilitated floodplains along
the lower river Rhine. River Research and Applications, 19, 353–374.
https://doi.org/10.1002/(ISSN)1535-1467
Harvey, B. C., Nakamoto, R. J., & White, J. L. (1999). Influence of large
woody debris and a bankfull flood on movement of adult resident
coastal cutthroat trout (Oncorhynchus clarki) during fall and winter.
|
9
KUBÍN e t al.
Canadian Journal of Fisheries and Aquatic Sciences, 56, 2161–2166.
https://doi.org/10.1139/f99-154
Harvey, B. C., & Stewar t, A. J. (1991). Fish size and habitat depth rela-
tionships inheadwater streams. Oecologia, 87, 336–342. https://doi.
org/10.1007/BF0063458 8
Hesth agen, T., Sak sgår d, R. , Hegg e, O., Der vo, B. K ., & S kurda l, J. (20 04). N iche
overlap between young brow n trout (Salmo trutta) and Siberian sculpin
(Cottus poecilopus) in a subalpine Norwegian river. Hydrobiologia, 521,
117–125. https://doi.org/10.1023/B:HYDR.0000026354.22430.17
Hothorn, T., Bretz, F., & Westfall, P. (2008). Simultaneous inference in
general parametric models. Biom Journal, 50, 346–363. https://doi.
org/10.1002/(ISSN)1521-4 036
Kerfoot, J. R., & Schaefer, J. F. (2006). Ecomorphology and habitat uti-
lization of Cottus species. Environmental Biology of Fishes, 76 , 1–13.
https://doi.org/10.1007/s10641-006-9000-0
Kondolf, G. M. (1997). Hungry water: Effects of dams and gravel mining
on river channels. Environmental Management, 21, 533–551. https://
doi.org/10.1007/s002679900048
Leopold, L. B., Wolman, M. G., & Miller, J. P. (1964). Fluvial processes in
geomorfology (522 pp.). San Francisco, CA: W. H. Freeman.
Liefferinge, C. V., Seeuws, P., Meire, P., & Verheyne, R. F. (2005).
Microhabitat use and preferences of the endangered Cottus gobio in
the River Voer, Belgium. Journal of Fish Biology, 67, 897–909. https://
doi.org/10.1111/j.0022-1112 .2005.00782.x
Lojkásek,B.,Lusk,S.,Halačka,K.,Lusková,V.,&Drozd,P.(2005).Theim-
pact of the extreme floods in July 1997 on the ichthyocenosis of the
Oder Catchment area (Czech Republic). Hydrobiologia, 548, 11–22 .
Louhi, P., Mäki-Petäys, A ., Huusko, A., & Muotka, T. (2014). Resource use
by juvenile brown trout and Alpine bullhead: Influence of interspe-
cific versus intraspecific competition. Ecology of Freshwater Fish, 23,
234–243. https://doi.org/10.1111/eff.12072
Lusk,S.,Halačka,K.,&Lusková,V.(1998).Theeffectofanextremeflood
onthefishcommun itiesintheupperreachesoftheT icháOrliceRiver
(The L abe drainage ar ea). Czech Journal of Animal Science, 43, 531–536.
Matthews, W. (1986). Fish faunal structure in an Ozark stream: Stability,
persistence and a catastrophic flood. Copeia, 2, 388–397. https://doi.
org /10. 2307/1444997
Menčík, E., & Tyráček, J. (1985). Beskydy a Podbeskydská pahorka-
tina (Geological map 1: 100000). Pr ague, Czech Rep ublic: Ústře dní
Ústavgeologický.
Naiman, R. J., Decamps, H., & McClain, M. E. (2005). Riparia: Ecology,
conservation, and management of streamside communities (448 pp.).
London, UK: Elsevier Academic Press.
Natsumeda, T. (2003). Effects of a severe flood on the movements of
Japanese fluvial sculpin. Environmental Biology of Fishes, 68, 417–424.
https://doi.org/10.1023/B:EBFI.0000005777.90560.90
Pearsons, T. N., Li, H. W., & Lamberti, G. A. (1992). Influence of hab-
itat complexity on resistance to flooding and resilience of stream
fish assemblages. Transactions of the American Fisheries Society, 121,
427–436. https://doi.org/10.1577/1548-8659(1992)121<0 427:IO
HCOR>2.3.CO;2
R Development Core Team. (2015). A language and environment for
statistical computing. Vienna, Austria: R Foundation for Statistical
Computing. Retrieved from http://www.R-project.org/.
Sato, T., & Yoshimura, M. (2014). Fish assemblages in headwater tributar-
ies of the Agano River System, Japan. Fisheries Science, 80, 493–498.
https://doi.org/10.1007/s12562-014-0730-1
Seber, F., & LeCren, E. D. (1967). Estimating population parameters from
large catches relative to the population. Journal of Animal Ecology, 36,
631–643. https://doi.org/10.2307/2818
Stankovic, D., Crivelli, A. J., & Snoj, A. (2015). Rainbow trout in Europe:
Introduction, naturalization, and impacts. Reviews in Fisheries Science
& Aquaculture, 23, 39–71. https://doi.org/10.1080/23308249.2015.
1024825
Weiß, A., Matouskova, M., & Matschullat, J. (20 08). Hydromorphological
assessment within the EU- Water Framework Directive—Trans-
boundary cooperation and application to different water basins.
Hydrobiologia, 603, 53–72.
Wolman, M. G. (1954). A method of sampling coarse riverbed material.
Transactions of the American Geophysical Union, 35, 951–956. https://
doi.org/10.1029/TR035i006p00951
Závorka , L., Hork ý,P., & Slaví k, O. (2013). Dist ribution a nd growth of
brown trout in pristine headwaters of Central Europe. Central
European Journal of Biology, 8, 263–271. https://doi.org/10.2478/
s11535-013-0133-1
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the sup-
porting information tab for this article.
How to cite this article:KubínM,RulíkM,LuskS,ZávorkaL.
Habitat degradation and trout stocking can reinforce the
impact of flash floods on headwater specialist Alpine
bullhead Cottus poecilopus – A case study from the
Carpathian Mountains. J Appl Ichthyol. 2018;00:1–9. ht tps://
doi.org/10.1111/jai.13682
