Content uploaded by Ricardo Morán-López
Author content
All content in this area was uploaded by Ricardo Morán-López on Apr 19, 2018
Content may be subject to copyright.
ORIGINAL PAPER
Relative leaping abilities of native versus invasive cyprinids
as criteria for selective barrier design
R. Mora
´n-Lo
´pez .O. Uceda Tolosa
Received: 30 May 2016 / Accepted: 7 November 2016
ÓSpringer International Publishing Switzerland 2016
Abstract Restoration of river connectivity favors
the spread of native but also of exotic invasive
populations, mainly freshwater fish. This has rarely
been studied, and never between migrating cyprinids.
To harmonize both objectives, the feasibility of a
selective leaping barrier for migratory cyprinids is
studied through the measurement of fish leaping
capabilities while freely migrating at a weir in the
Guadiana River of southwest Spain. A cross-popula-
tion analysis provided the cost-benefit outcomes of
completely blocking the invasive Carassius auratus
while allowing negotiation for most of the native
Luciobarbus sp. populations. Larger fish reached
higher leap heights, the highest being attained by the
native barbel species. Barbels reached up to four times
its total length and a maximum height of approxi-
mately 150 cm, while the figures for the invasive
goldfish were double total length with a maximum of
approximately 80 cm. The selective obstacle height
(SOH) for goldfish was 81.2 cm, the estimated
maximum height that the tip of a jumping goldfish
could reach. Facility designs (whether culvert, weir or
fish pass) with this SOH criterion guarantees complete
migration failure of goldfish. Also, most (95–99%) of
the barbels migrating populations will surpass the
SOH regardless of body length; only the less capable
or those in poor condition will be blocked. Advice on
SOH utility to evaluate existing instream structures—
and their position within the cost-benefit balance—is
provided.
Keywords River connectivity Invasive species
Selective barrier Migratory fish Leap height
Facility design
Introduction
Intense modifications for water flow and longitudinal
connectivity have driven a sizeable and continuously
incrementing number of artificial barriers (Sabater
et al. 2009), a trend which climate change will most
probably exacerbate (Xenopoulos et al. 2005). Artifi-
cial barrier prevalence in turn stimulates the study of
the biological and ecological consequences of con-
nectivity loss (e.g. Mora
´n-Lo
´pez et al. 2012), which
have shown reduced recruitment success, altered fish
assemblages, and the local extinction of fish popula-
tions in Europe and elsewhere (Mueller et al. 2011;
R. Mora
´n-Lo
´pez (&)
Area of Zoology, Department of Anatomy, Cellular
Biology and Zoology, Science Faculty, University of
Extremadura, Avda. de Elvas s/n, 06071 Badajoz, Spain
e-mail: rmoran@unex.es
O. Uceda Tolosa
SIGnatur – Innovative Solutions for Nature Management,
C/Vicente Rodero 8, 5°B, 06300 Zafra (Badajoz), Spain
e-mail: ouceda@unex.es
URL: http://www.signatur.com.es
123
Biol Invasions
DOI 10.1007/s10530-016-1328-6
Branco et al. 2012). Also, invasive species represent
an important pressure that prevents the achievement of
good ecological status, through the alteration of biotic
communities and ecological function. Migration and
jumping behaviours are also present in introduced
species, and these abilities promote the dispersion of
invasive populations (Holthe et al. 2005). Restoring
connectivity for the native species while preventing
further spread of the exotics can be a difficult task, for
which new knowledge and technologies are in great
need.
Knowledge of the swimming and leaping capabil-
ities of fish can lead to the development of improved,
more permeable barriers and less selective fish
passage facilities for native species (Roscoe and
Hinch 2010; Bunt et al. 2012; Katopodis and Williams
2012; Nooman et al. 2012). But there remains
substantial research to be done. To date, most jumping
performance studies use hatchery-reared or wild
captured fish under laboratory conditions (Kondratieff
and Myrick 2005; Mueller et al. 2008; Lauritzen et al.
2010; Ficke et al. 2011); studies of fish in their natural
environments are conversely scarce (Ombredane et al.
1987; Lauritzen et al. 2005,2010). Laboratory control
and experimental designs provide high quality data,
but at the cost of using fish with potentially inferior
capabilities (McDonald et al. 1998; Kondratieff and
Myrick 2006; Pedersen et al. 2008; Bestgen et al.
2010). Also, it increases the uncertainty of whether the
results can be transferred to naturally occurring
conditions, which are more complex (Ovidio et al.
2007). Salmonids are the best known group studied; a
few studies include less economical or recreationally-
valued fish such as potamodromous cyprinids (Holthe
et al. 2005; Ficke et al. 2011). However, Cyprinidae is
the largest family of freshwater fishes with over 2000
species naturally distributed in Eurasia, Africa and
North America (Moyle and Cech 2004).
Invasive species superimpose an added threat factor
to habitat degradation for migratory and other fish
(Holthe et al. 2005; Northcote 2010). In these cases,
obstacles provide positive outcomes (Kondratieff and
Myrick 2005), with a number of obstacles being used
to restrict upstream movement of introduced or
undesirable native cyprinids such as the Asian carp
or minnows (see Table 4 in McLaughlin et al. 2013).
When compared to studies on non-jumping species
(Ve
´lez-Espino et al. 2011), even rarer are studies that
design selective dispersion barriers of invasive species
based on their leaping abilities (either by weirs or
fishways; Brandt et al. 2005; Holthe et al. 2005; Stuart
et al. 2006). To the best of our knowledge, no study
examines the differential leaping performance of
native versus invasive cyprinids, for the knowledge
and to the benefit (i.e. management and conservation)
of native cyprinids and other coarse fish species, in
their natural ranges.
In this study, we analyse the leaping capabilities of
three cyprinids—one invasive, and two native spe-
cies—freely migrating into the artificial waterfall of an
impassable weir in the Guadiana River of southwest
Spain. We hypothesized that there are differences in
maximum leaping capabilities associated to fish size.
We also hypothesized that there are differences in
maximum leaping capabilities associated to popula-
tions nativeness of these sympatric cyprinid taxa. Our
prediction is that these differences would allow to
determine the height of a waterfall barrier able to
block all invasive populations while allowing the
negotiation for most of the native jumping ones. This
new knowledge on the leaping capabilities and the
comparison between species and taxonomic groups is
discussed, along with its application to optimize weirs
and fish pass facility designs.
Materials and methods
Study area and species
The Granadilla weir is located about 6.5 km upstream
of the Spanish–Portuguese border in the middle
Guadiana (38°51042.500N7°0100200W), the main river
of the fourth largest basin in the Iberian Peninsula
(67,000 km
2
; Fig. 1). The Guadiana River has a low
gradient and a Mediterranean regime with most of the
precipitation occurring from November to March
(80% of total annual rainfall), then descending
throughout a dry and hot summer (Gasith and Resh
1999; Mora
´n-Lo
´pez et al. 2006). There are dams and a
number of differently designed weirs installed in the
main river and its tributaries to regulate this flow.
The Granadilla’s weir is more than 400 m long and
3 m high, and is located in the middle section of the
Guadiana River, at 166 m.a.s.l. Flow follows a semi-
natural regime with a mean annual discharge of
63.5 m
3
s
-1
, although the monthly discharge ranges
from 7.3 to 885.2 m
3
s
-1
(Guadiana Hydrographic
R. Mora
´n-Lo
´pez, O. Uceda Tolosa
123
Confederation). The weir was equipped with two
fishways prior to this study, both having very low
efficiency and being species- and size-selective for
both native and invasive species (GIC 2008).
As is typical for the Iberian Peninsula, the Guadi-
ana’s native freshwater fish community is species-
poor yet highly endemic with mostly cyprinids, and is
rife with invasive species (Cabral et al. 2005; Doadrio
et al. 2011). Concerning native cyprinids, only one out
of five small sized species migrate and none jump, all
these species being mostly or totally restricted to
tributaries (Doadrio et al. 2011; our own data). And
five out of six large sized species migrate and of these,
four out of five jump, all these species occupying the
Guadiana River and major tributaries (Doadrio et al.
2011; our own data). All native jumping migrants are
barbel (Luciobarbus sp.) species; the southern
straight-mouth nase (Pseudochondrostoma willkom-
mii) has never been observed displaying such
behaviour in the study area. We noted that while the
nase population is declining downstream of the
Granadilla weir, two barbel species still maintain
abundant migrating populations that try to negotiate
the Granadilla weir through leaping: the Iberian long-
snout barbel (Luciobarbus comizo) and the Iberian
small-head barbel (L. microcephalus).
Thirteen exotic species have been cited in the
Guadiana basin, with ten of them in the study area
(Doadrio 2001; Doadrio et al. 2011; preliminary
observations). Of those in our area of interest, three
cyprinids (Alburnus alburnus,Carassius auratus, and
Cyprinus carpio) and two non-cyprinids (Esox lucius
and Sander lucioperca) are partial or total migrants,
and we found only the cyprinids goldfish (C. auratus)
and common carp (C. carpio) displaying jumping
behaviour. We observed that while the latter have less
abundant migrant populations and only display occa-
sional jumping behaviour, the former has large
populations concentrated downstream of the Grana-
dilla weir that repeatedly try to negotiate the obstacle
through leaping.
Data collection and analysis
Data on jumping behaviour were collected through
video images taken downstream from the right margin
of the Granadilla weir (Fig. 1), in April 26, 2011 at
12:45 CET (Central European Time). Because the
weir is impassable by leaping and no juveniles were
observed with this behaviour, data correspond to adult
fish leaping to their maximum capabilities under
presumably optimal conditions of temperature (water
temperature: 21 °C). These data are within the max-
imum frequency of cyprinid jumping behaviour
observed in decadal (2007–2016), seasonal (January
to July), and daily (dawn to dusk) time scales: a long
term monitoring scheme showed that the seasonal
maximum number of barbels jumps per minute in the
Fig. 1 Inset of the south-
western Iberian Peninsula,
with the study area indicated
(arrow). The Granadilla
weir (in black) is retrofitted
with two fishways located at
the right (A) and left
(B) margins of the Guadiana
River (C). The study site
was located at the edge of
the right margin (D).
Direction of flow and
migrating fish are indicated
with black and white arrows,
respectively
Relative leaping abilities of native versus invasive cyprinids as criteria
123
37 m stretch nearest to the right margin of the
Guadiana River was 150 in 2011, while the decade
mean maximum was 55.1 ±93.3 (unpublished
results). Therefore, the data corresponds to carefully
selected conditions, where the selection of the year,
site, date, and horary of data collection represent fish
highly motivated to jump; such accessibility to high
quality data on free-ranging leaping fish is not known
elsewhere in the Guadiana basin. For comparative
purposes, we also collected capture data from the river
immediately downstream the weir using nets; total fish
length was measured on these migrating fish species.
A SONY DCR-SR290E digital camera with a video
rate of 25 frames s
-1
, a focal distance of 25.9 mm
(35 mm equivalent ca. 200 mm), and a maximum
aperture of f/1.8 was used. Zoom level was adjusted to
cover a field of view of approximately 7 94 m, which
included the first six metres of water falling adjacent to
the margin; this was a compromise between horizontal
amplitude of spatial coverage of the obstacle and fish
size on the images. The field of view also covered parts
of the facility immediately adjacent to the waterfall;
their known dimensions allowed us to scale image
pixels to absolute metric measures in a Cartesian
coordinate system. The fall height (h) was measured
on the images as the vertical distance from the abrupt
drop of water at the crest of the weir to the downstream
pool water surface. As the latter level is somewhat
imprecise because of the fluctuating standing wave,
the mean level of emergence of a sample of leaping
fish was used (h =1.62 m, n =67). The depth of the
plunge pool within the video frame was measured at
six locations using a metre stick at the standing wave,
which varied from 0.85 to 1.05 m. As these values
were double the mean fish size of the larger migrating
populations (barbel mean ±S.D., 44.8 ±7.9 cm,
n=71), the depth of the water was considered
enough to not limit maximum leaping performance
of the fish (Powers and Osborn 1985; Lauritzen et al.
2005).
Individual video frames of 1024 9576 pixels were
extracted for complete fish trajectories using VLC
media player (version 2.1.5). ImageJ version 1.46
(Schneider et al. 2012) was used to set abscissa and
ordinate origins at pool water level and weir margin,
respectively, within which water develops the fall.
Microsoft Excel 97–2003 was used to convert the
spatial scale of images from pixels to centimetres.
Total fish length was measured in the image using the
upper height (when the fish was near zero velocity)
with the fish’s straightest posture (i.e. not C- or
S-bent); trajectories clearly departing from these
criteria were excluded from further analysis. In
addition, absolute horizontal and vertical coordinates
of fish were registered in the frame when and where it
reached its absolute maximum height. The tip of the
snout was used as it reached the maximum height in all
occasions, as it is a more accurate measure of fish
position in the videos (Shih and Techet 2010), and it is
a useful indicator of the height of an impassable
barrier.
Precision was estimated throughout repeated mea-
sures of a 500 mm static portion of the infrastructure
in a number of random video frames (CV =1.7%;
n=30); in addition, a free falling 239 mm diameter
ball was filmed and measured as aforementioned on
repeated occasions (CV =4.6%; n =68). These
measures indicated that a certain combination of the
objects’ size and velocity influenced the precision of
the measurements taken in the images, but remained
within a tolerable range of around 2.3% above or
below the true measure for moving fishes (approxi-
mately 1 cm in a barbel of average length). Measure-
ment errors for coordinates taken in the images have
no expected bias because their random nature causes
them to average. On the contrary, actual total fish
length measurements are prone to underestimation
bias, since fish may have a variable non-normal body
angle with respect to the camera. We used a barbel
total length (TL) conversion coefficient (Lauritzen
et al. 2005) of 1.22, developed from the size distribu-
tion of migrating barbel captured downstream the
weir, and applied it to the size distribution of a sample
of jumping fish on the video images (n =260); a
similar coefficient of variation for total fish length
from both data sources (respectively, 17.5% vs.
16.4%) support their use within the same migrating
population. The same method was used for goldfish
(conversion coefficient =1.02; respectively, n =93
vs. n =110 and CV =6.5% vs. 10.4%).
Sex determination was not possible from the videos
(e.g., Lauritzen et al. 2005), though sex differences in
leaping capabilities are not to be expected (Kondrati-
eff and Myrick 2006). Nor was it possible to distin-
guish between the two barbel species or to identify
individuals within the video frames (e.g., Mueller et al.
2008). Nevertheless, comparable contributions were
found between the two barbel species in migrating
R. Mora
´n-Lo
´pez, O. Uceda Tolosa
123
population size, individual size distribution, and sex
ratio in both data from captures downstream the weir
(L. comizo 46.4%; L. microcephalus 53.6%; n =71)
and photographs of leaping individuals taken at the
weir (L. comizo 57.4%; L. microcephalus 42.6%;
n=169; unpublished data). Therefore, hereafter we
refer to the barbel species group, which is still
representative of the native leaping fish community,
and remains useful for management and conservation
applications. Multispecies approaches have demon-
strated utility in quantitative assessments for conser-
vation management (Bonn and Schro
¨der 2001;
Dallimer et al. 2009; Schwenk and Donovan 2011;
Mora
´n-Lo
´pez et al. 2016).
Prior to the following analyses, normality of the
variables leap height (LH), fish TL, and relative LH to
total length (LH/TL) were checked using the Kol-
mogorov–Smirnov test. Relationships between LH
and TL were measured using Pearson product moment
correlation analyses by species. Within species, leap-
ing capabilities are described through basic statistics;
for comparative purposes, barbels were characterized
both along its own range of TL and within the range of
the goldfish TL. Between species comparisons of
height distribution were made through a Ttest both
within the same range and for the whole range of TLs.
The height of a barrier exceeding the invasive
goldfish, but not the native barbel leaping capabilities
(hereinafter called selective obstacle height, SOH)
was graphically determined using relative cumulative
frequency distribution curves (RCFD). These curves,
f(x
i
), were approximated by plotting the value of the
LH by species i(i=Cfor Carassius or i=Lfor
Luciobarbus) against the relative number of leaps up
to a given height xis reached. The ordinate axis of the
plot is given by:
yi¼LHxi
N
where LHxiis the number of leaps up to a given height
xis reached by species i, and Nis the total number of
cases. Therefore, the relative number of leaps is the
number of leaps up to a given height divided by the
total number of leaps of any height. The abscissa axis
of the plot is given by:
xi¼LH
where LH is the variable of interest, leap height. The
horizontal axis on the resulting RCFD curves covers
the entire range of LHs observed in any species, and
the vertical axis encompasses all negotiation attempts
(0–100%). Assuming invasive species have inferior
leaping capabilities compared to native species, the
vertical line located at the 100% limit of the goldfish
curve fulfils:
ðxCÞmax ¼ðxLÞSOH
where (X
L
)
SOH
is the SOH for goldfish that barbels
have to negotiate. This limit will show: (a) the height
of an obstacle that completely blocks negotiation
attempts by the invasive species [(x
C
)
max
]; and (b), the
fraction of unsuccessful leaps by the native species at
that height [f(x
L
)
SOH
]. For security reasons, the SOH
must add the aforementioned error term of 2.3%.
The cost of the SOH to native species was studied
through two analyses. In the first, assuming that any
individual fish may reach the SOH in one out of
several negotiation attempts, TL distribution of bar-
bels reaching the SOH versus the whole jumping
population were compared using a Ttest. These data
do not correspond to individuals but to whole leaping
samples which were, thus, analysed at the population
level (similar to use-availability designs in habitat
selection studies; Manly et al. 1993).
In the second analysis, a cross-population approach
was used to estimate the fraction of native migrating
barbel populations that would be blocked by the SOH
(Kondratieff and Myrick 2006). A least square
regression model was developed to relate barbel LH
to TL. Then, the model was applied to the TL
distribution from capture data, and the number of
individuals and their sizes under the SOH were
calculated. The fraction of unsuccessful negotiation
attempts according to the SOH was also calculated.
Results
Kolmogorov–Smirnov tests showed LH (d =0.5429;
p[0.20), TL (d =0.4838; p[0.20), and LH/TL
(d =0.0553; p[0.20) to be normally distributed.
There was a general positive correlation between LH
and TL (r
p
=0.48; p\0.01), which was significant in
barbels (r
p
=0.30; p\0.01) but not in goldfish
(r
p
=0.28; p=0.06). Smaller barbels achieved
greater relative LH (r
p
=-0.28; p\0.01). There
was a significant difference in mean LH between
Relative leaping abilities of native versus invasive cyprinids as criteria
123
goldfish and barbels within the range of TL of the
former (tvalue =-5.10; p\0.00); where the gold-
fish reached LH =79.4 cm while the barbels reached
LH =138.7 cm. These differences increased when
the whole range of barbels TL was included in the
analysis (tvalue =21.24; p\0.01); where the leap-
ing capabilities were clearly superior in the barbels
than in the goldfish both in absolute and relative terms
(Table 1).
The RCFD curves for goldfish and barbels showed
that 50% of the leaps reached at most 32.2 and 94.1 cm
height, respectively (Fig. 2). At 100% RCFD, leaps
were (x
C
)
max
=79.4 cm and (x
L
)
max
=150.7 cm
(Fig. 2). Thus, the SOH for goldfish is 79.4 cm;
adjusting for errors in LH measurement results in a
SOH of 81.2 cm. The intersection between the SOH
for goldfish [(x
C
)
max
] and the cumulative leap fre-
quency curve of barbels [(x
L
)
SOH
] shows that f(x
L
)-
SOH
=34.3% of barbel leaps do not exceed the SOH
for goldfish. Note that this number of leaps corre-
sponds most probably to a smaller number of individ-
uals from the barbel populations, for it is reasonable to
expect that any individual will make repeated jumps
against an impassable obstacle.
There was no difference in TL between those
barbels exceeding the SOH compared to the whole
jumping population (tvalue =-1.5190; p=0.13).
Therefore, TL difference in barbels does not affect
their ability to reach the SOH. Our regression model
revealed that a TL of 34.0 cm is the estimated
minimum length required for barbels to cross the
SOH of goldfish, hereafter referred to as threshold size
(Fig. 3). However, this model explained very low
variance (R
2
=0.087), which indicates that there is a
large variation in LH for fish of the same TL. This is an
expected result when individuals make multiple
negotiation attempts, with some higher or lower than
others.
The very low variance explained by our regression
model is particularly informative for proper interpre-
tation of the cross-population estimate of the fraction
of barbel individuals that would be blocked by the
SOH for goldfish. The sample of n =71 barbels
captured contained 5.6% individuals under the thresh-
old size. These individuals were the youngest migrat-
ing, sexually mature (TL [30 cm; no juveniles in the
sample), and mainly male (sex ratio 3:1) barbels.
Based on the model, such fish would be blocked by the
SOH for goldfish. However, Fig. 3clearly shows that a
fraction of filmed barbels under the threshold size can
and do actually surpass the SOH. Therefore, the very
low variance explained by the regression model
indicates that the 34.3% of the native species leaps
(not individuals) estimated to be blocked by the SOH,
is a result that corresponds in terms of native
populations with a number well below the estimated
5.6% of the younger, mainly male, breeding individ-
uals (not leaps) while still maintaining 100% blockage
Table 1 Descriptive
statistics for fish leaping
capabilities in absolute (cm)
and relative (leap height to
total length) terms
Results are presented for all
sizes of the populations as
well as by comparable size
classes (note that the
goldfish do not reach the
barbels higher length class)
LH leap height, LH/TL
relative leap height to total
length
Species Valid N Mean Minimum Maximum SD
Absolute (LH)
Luciobarbus sp. 260 92.4 27.1 150.7 25.3
\35 cm 20 79.6 30.7 137.7 29.0
35–45 cm 114 86.4 27.1 144.2 25.6
[45 cm 126 99.8 44.6 150.7 22.1
Carassius auratus 110 36.8 10.1 79.4 16.4
\35 cm 27 45.1 26.7 79.4 14.5
35–45 cm 20 53.2 27.9 79.2 13.4
Relative (LH/TL)
Luciobarbus sp. 260 2.1 0.7 4.1 0.6
\35 cm 20 2.5 1.0 4.1 0.9
35–45 cm 114 2.2 0.7 3.5 0.6
[45 cm 126 2.0 1.0 3.0 0.5
Carassius auratus 47 1.4 0.7 2.4 0.4
\35 cm 27 1.4 0.8 2.4 0.4
35–45 cm 20 1.4 0.7 2.1 0.4
R. Mora
´n-Lo
´pez, O. Uceda Tolosa
123
of the invasive species. Moreover, any observed
(filmed or captured) native barbel individual could
exceed the SOH for the invasive goldfish according to
its size (TL [30 cm) and demonstrated capabilities
(Fig. 3). Therefore, the native population blocked with
the SOH for the invasive species approaches zero.
Discussion
Non-physical barriers to deter fish movements have
limited potential even when used as a component of an
integrated system for invasive species restriction
(Noatch and Suski 2012). Seasonally operated phys-
ical barriers and fishways may be an unsuccessful
approach when migration phenologies overlap
between the targeted and non-targeted native species
(Ve
´lez-Espino et al. 2011). A potentially useful
alternative is the use of leaping capabilities (Rahel
2013), as leaping barriers can be true selective
barriers. But empirical comparative studies of native
versus invasive species are extremely rare. In one
Fig. 2 Relative cumulative frequency distribution (RCFD)
curves of leap height for goldfish (grey) and barbels (black).
The abscissa axis represents the leap height in millimetres, and
the ordinate axis represents the relative cumulative fraction
(percentage) of leaps reaching that height. Vertical lines
indicate the height reached by 50% (dotted vertical lines) and
100% (dashed vertical lines) of leaps for each species. The
selective obstacle height (SOH) is indicated with the arrowed
line, and connects the maximum leap height of any goldfish
(100% cumulative frequency) with the cumulative frequency of
barbel leaps under this height (SOH =79.4 mm). Other
acronyms used are CARAUR for Carassius auratus, LUCSPP
for Luciobarbus sp., and LH for leap height
Fig. 3 Scatterplot of total length (TL) against leap height (LH)
for barbels (dots). The lines represent the regression model
relating LH to barbel TL (dotted line), and SOH for goldfish
(grey line). The vertical,black line indicates the TL where the
barbel model intersects the SOH for goldfish (TL =34.0 cm):
this line discriminates between the barbels not capable (left)
versus capable (right) of exceeding the SOH for goldfish due to
the barbels’ leaping abilities, which is associated with their total
length. Note that barbels smaller than 34 cm would not be
capable of exceeding the SOH in this model, but observed data
(LHobs) indicates that they indeed can and do
Relative leaping abilities of native versus invasive cyprinids as criteria
123
study, the leaping capabilities of the invasive Euro-
pean minnow (Phoxinus phoxinus) and the native
brown trout (Salmo trutta) were simultaneously mea-
sured with the aim of constructing suitable waterfall
barriers to prevent further dispersal of the invasive
while allowing passage for the native species (Holthe
et al. 2005). In another, the leaping behaviour of the
common carp (C. carpio) has been used to separate
this invasive species from native Australian fishes
lacking this behaviour (Stuart et al. 2006). To the best
of our knowledge, there are no more studies available
on selective leaping barriers for fish, and particularly
none about cyprinids.
In the taxa studied here, larger fish reached higher
LHs, with the highest being attained by the native
barbel species and the younger barbels showing
greater relative leaping capabilities. The invasive
goldfish clearly showed inferior leaping capabilities
compared to the barbels, even when the same size
range was considered. These differences provide an
opportunity to develop and design criteria for a
selective barrier to fully restrain further spread of the
invasive cyprinid and/or its population control—to the
extent that access is limited to habitats needed to
complete its life cycle (Rahel 2013)—while maintain-
ing most or all the connectivity for native cyprinids.
This SOH for goldfish was set to 81.2 cm, which
represents the estimated maximum height that the tip
of the snout of a goldfish could reach. Since the centre
of mass of the fish remains below the SOH, it
guarantees complete migration failure for goldfish
leaping under an obstacle of such vertical dimension.
Concerning the native barbels, the most of its
population has the capability to exceed the SOH for
goldfish, as most of the variation in LH was associated
with variables outside of TL. The large variability in
the LH compared to TL observed has several, non-
incompatible explanations: (a) the length–height rela-
tionship varies in a non-linear way, with smaller fish
having superior relative capabilities (as it was in the
case of Kondratieff and Myrick 2006); (b) two indi-
viduals of the same TL can have inherently different
stamina and/or conditions (Ovidio and Philippart
2002; Kondratieff and Myrick 2006); (c) the same
individual may reach different heights during separate
jumping instances due to occasionally underpowered
jumps, or simply because repeating a manoeuvre in a
plunge pool with mixed hydrodynamic signals from
multidirectional flows is very difficult (Lauritzen et al.
2005; Ovidio et al. 2007). Therefore, leap distance to
the weir, takeoff angle, and thrust probably vary both
within and between individuals of similar size,
suggesting that fish size is neither the only nor
necessarily the main factor determining LH. These
considerations promote support for the results
obtained in this study, in contrast to length–height
rigid capability rules (Kondratieff and Myrick 2006).
Moreover, the lack of a rigid length–height rule is a
fortunate result as it reflects the barbels’ capability to
surpass the SOH for a wide range of fish sizes, which
likely encompasses the great majority ([99%) of
native jumping populations downstream the Grana-
dilla weir. Only the smallest (mainly non-migrant)
fish, the less capable or the ones in poor condition
would be blocked. Furthermore, the younger barbels
showed greater relative leaping capabilities—as it is
the case for Salvelinus fontinalis (Kondratieff and
Myrick 2006)—making them capable of exceeding
the SOH for goldfish from TL [30 cm onwards. This
guarantees that an obstacle using the SOH will not
impose substantial artificial selection pressures to the
native barbels.
Barbel leaping capabilities are comparable to those
observed in different salmonid species. Barbels
demonstrated inferior absolute leaping capabilities
compared to: larger species such as the Chinook
salmon (Oncorhynchus tshawytscha); species of sim-
ilar size such as steelhead (O. mykiss) or coho salmon
(O. kisutch); and with smaller species such as the
sockeye salmon (O. nerka; Table 3 in Reiser et al.
2006). In contrast, the capabilities of these barbels
exceeded those of the similarly-sized chum salmon (O.
keta) and the smaller pink salmon (O. gorbuscha;
Table 3 in Reiser et al. 2006). In relative terms
however, of all of these species, only O. mykiss is
superior to Luciobarbus sp. (4.79 vs. 4.08, respec-
tively). Their absolute capabilities are similar to Salmo
salar (1.5 m) but superior to S. trutta (1.1 m) or
Thymallus thymallus (0.85 m; Ovidio and Philippart
2002); Luciobarbus sp. have also superior relative
leaping capabilities to all these species along all the
length classes studied (Table 1). The variation in
leaping capabilities amongst species may provide
opportunities to control invasive populations through
selective barriers where cyprinids and salmonids
collide. For example, the brook trout (S. fontinalis)
is an invasive species introduced in several continents
and is also included in the Spanish catalogue of exotic
R. Mora
´n-Lo
´pez, O. Uceda Tolosa
123
invasive species (Royal Decree 630/2013 2013). In the
event that the barbels and the brook trout exist
sympatrically, Luciobarbus sp. surpass the leaping
capability of the salmonid, which can jump during
migration up to 4.7 body lengths when small in size
(approximately TL =15 cm), and 3–4 body lengths
(73.5 cm) when larger (TL C20 cm; Kondratieff and
Myrick 2006). This provides further applied examples
on the potential use of jumping abilities as design
criteria of selective barriers for deliberate fragmenta-
tion (Fausch et al. 2009).
To the best of our knowledge, there are no
published studies on the leaping performance of
cyprinid species comparable to the size of adult
Luciobarbus sp. Holthe et al. (2005) found European
minnows of TL =50–110 mm forced waterfall bar-
riers up to 27 cm at 14.0–16.5 °C. Ficke et al. (2011)
studied the North American cyprinid brassy minnow
(Hybognathus hankinsoni) and the common shiner
(Luxilus cornutus), and found leaping capabilities of
15 cm at 25 °C and 10 cm at 17.5 °C, respectively.
Geeraerts et al. (2007) cite the non-systematic obser-
vation of roach (Rutilus rutilus), an infrequent obstacle
leaper, jumping at least 15 cm—the species being
introduced in the Guadiana basin (Doadrio et al.
2011). All these species are small cyprinids
(TL \30 cm) with obvious, inferior leaping capabil-
ities than the larger Luciobarbus sp.
(TL =30–70 cm).
Conclusions
Our results indicate that a barrier designed to
prevent the upstream movement of the goldfish (C.
auratus), while allowing the negotiation of native
barbels (Luciobarbus sp.), should be at least
81.2 cm in height. An upward safety factor of
10% would likely be appropriate, setting the SOH
figure at approximately 90 cm. A lower height will
also meet the selective barrier criterion as long as a
lesser plunge pool depth limits the leaping capability
of the fishes, but larger native fishes like barbels
will also be negatively affected. The use of such a
design will provide population control of the
invasive species, by disrupting the connectivity
between habitats needed for the species to complete
its life cycle, and by preventing them from colonis-
ing further tributaries within the river network
where it is allochthonous (isolation management in
Rahel 2013). A selective barrier would also be
useful in the control of invasive exotics by incor-
porating a structure for the capture of jumping fish,
as it was in the case of traditional fishing (Brandt
1984). This selective capture system could be
retrofitted to the current weir, so the invasive fish
are guided by it to the margin where they can be
captured.
For non-jumping native species, a selective fish
pass should be retrofitted as a complementary mea-
sure. Data on Luciobarbus sp. leaping capabilities can
also be useful in optimising designs of the pool-and-
weir fishways, which are often used to mitigate the
effects of dams and weirs (Clay 1995), similar to
culverts retrofitted to road crossings (Maitland et al.
2015). Fish passes provide further opportunities for
the control of invasive species populations using them,
as it is in Australia for carp control (Stuart et al. 2006),
and could be the case of the goldfish in Iberia and
elsewhere.
Finally, the SOH for goldfish criterion found in this
study apply to the abiotic conditions (temperature,
flow, plunge pool depth, etc.) and taxa involved in the
study. The transferability of these results to other
native species or environments should be tested,
because of the limitations on the observation period,
spatial scale, and number of species studied. Future
studies including more years, more obstacles, and
more species will improve our understanding of the
differential leaping capabilities between native and
invasive cyprinids, and its utility for selective barrier
design.
Acknowledgements ChristopherMyrick and three anonymous
referees provided useful comments to the manuscript.
Compliance with ethical standards
Conflict of interest The authors declare that they have no
conflict of interest.
References
Bestgen KR, Mefford B, Bundy JM, Walford CD, Compton RI
(2010) Swimming performance and fishway model passage
success of Rio Grande silvery minnow. Trans Am Fish Soc
139:433–448
Bonn A, Schro
¨der B (2001) Habitat models and their transfer for
single and multi species groups: a case study of carabids in
an alluvial forest. Ecography 24:483–496
Relative leaping abilities of native versus invasive cyprinids as criteria
123
Branco P, Segurado P, Santos JM, Pinheiro P, Ferreira MT
(2012) Does longitudinal connectivity loss affect the dis-
tribution of freshwater fish? Ecol Eng 48:70–78
Brandt MM, Holloway JP, Myrick CA, Kondratieff MC (2005)
Effects of waterfall dimensions and light intensity on age-0
brook trout jumping performance. Trans Am Fish Soc
134:496–502
Bunt CM, Castro-Santos T, Haro A (2012) Performance of fish
passage structures at upstream barriers to migrations. River
Res Appl 28:457–478
Cabral MJ, Almeida J, Almeida PR, Dellinger T, Ferrand de
Almeida N, Oliveira ME, Palmeirim JM, Queiroz AI,
Rogado L, Santos-Reis M (eds) (2005) Livro Vermelho dos
Vertebrados de Portugal. Instituto de Conservac¸a
˜oda
Natureza, Lisboa
Clay CH (1995) Design of fishways and other fish facilities.
Lewis, Boca Rato
´n
Dallimer M, King T, Atkinson RJ (2009) Pervasive threats
within a protected area: conserving the endemic birds of
Sa
˜o Tome
´, West Africa. Anim Conserv 12:209–219
Doadrio I (ed) (2001) Atlas y Libro Rojo de los Peces Conti-
nentales de Espan
˜a. Direccio
´n General de Conservacio
´nde
la Naturaleza, Museo Nacional de Ciencias Naturales,
Madrid
Doadrio I, Perea S, Garzo
´n-Heydt P, Gonza
´lez JL (2011)
Ictiofauna continental espan
˜ola: Bases para su segui-
miento. Direccio
´n General de Medio Natural y Polı
´tica
Forestal, Ministerio de Medio Ambiente Rural y Marino,
Madrid
Fausch KD, Rieman BE, Dunham JB, Young MK, Peterson DP
(2009) Invasion versus isolation: trade-offs in managing
native salmonids with barriers to upstream movement.
Conserv Biol 23:859–870
Ficke AD, Myrick CA, Jud N (2011) The swimming and
jumping ability of three small great Plains fishes: impli-
cations for fishway design. Trans Am Fish Soc
140:1521–1531
Gasith A, Resh VH (1999) Streams in Mediterranean climate
regions: abiotic influences and biotic responses to pre-
dictable seasonal events. Annu Rev Ecol Syst 30:51–81
Geeraerts C, Ovidio M, Verbiest H, Buysse D, Coeck J, Belpaire
C, Philippart J-C (2007) Mobility of individual roach
Rutilus rutilus (L.) in three weir-fragmented Belgian rivers.
Hydrobiologia 582:143–153
GIC (Grupo de Investigacio
´n en Biologı
´a de la Conservacio
´n)
(2008) Ictiofauna: efectividad de escalas y movilidad de
especies en la cuenca media del Guadiana. Confederacio
´n
Hidrogra
´fica del Guadiana, Ministerio de Medio Ambiente
y Medio Rural y Marino (in Spanish)
Holthe E, Lund E, Finstad B, Thorstad EB, McKinley RS (2005)
A fish selective obstacle to prevent dispersion of an
unwanted fish species, based on leaping capabilities. Fish
Manag Ecol 12:143–147
Katopodis C, Williams JG (2012) The development of fish
passage research in a historical context. Ecol Eng 48:8–18
Kondratieff MC, Myrick CA (2005) Two adjustable waterfalls
for evaluating fish jumping performance. Trans Am Fish
Soc 134:503–508
Kondratieff MC, Myrick CA (2006) How high can brook trout
jump? a laboratory evaluation of brook trout jumping
performance. Trans Am Fish Soc 135:361–370
Lauritzen DV, Hertel F, Gordon MS (2005) A kinematic
examination of wild sockeye salmon jumping up natural
waterfalls. J Fish Biol 67:1010–1020
Lauritzen DV, Hertel FS, Jordan LK, Gordon MS (2010) Sal-
mon jumping: behavior, kinematics and optimal condi-
tions, with possible implications for fish passageway
design. Bioinspir Biomim 5:035006
Maitland BM, Poesch M, Anderson AE, Pandit SN (2015)
Industrial road crossings drive changes in community
structure and instream habitat for freshwater fishes in the
boreal forest. Freshw Biol 61:1–18
Manly BFJ, McDonald LL, Thomas DL (1993) Resource
selection by animals: statistical design and analysis for
field studies. Chapman & Hall, London
McDonald DG, Milligan CL, McFarlane WJ, Croke S, Currie S,
Hooke B, Angus RB, Tufts BL, Davidson K (1998) Con-
dition and performance of juvenile Atlantic salmon (Salmo
salar): effects of rearing practices on hatchery fish and
comparison with wild fish. Can J Fish Aquat Sci
55:1208–1219
McLaughlin RL, Smyth ERB, Castro-Santos T, Jones ML,
Koops MA, Pratt TC, Ve
´lez-Espino L-A (2013) Unin-
tended consequences and trade-offs of fish passage. Fish
Fish 14:580–604
Mora
´n-Lo
´pez R, Da Silva E, Pe
´rez-Bote JL, Corbacho Amado C
(2006) Associations between fish assemblages and envi-
ronmental factors for Mediterranean-type rivers during
summer. J Fish Biol 69:1552–1569
Mora
´n-Lo
´pez R, Pe
´rez-Bote JL, Da Silva E, Perales Casildo AB
(2012) Hierarchical large-scale to local-scale influence of
abiotic factors in summer-fragmented Mediterranean rivers:
structuring effects on fish distributions, assemblage compo-
sition and species richness. Hydrobiologia 696:137–158
Mora
´n-Lo
´pez R, Sa
´nchez Guzma
´n JM, Perales Casildo AB,
Uceda Tolosa O (2016) Extended negative impact of sec-
ondary infrastructure on the high conservation values of
sparsely developed areas. Oryx. doi:10.1017/
S0030605316000338
Moyle PB, Cech JJ (2004) Fishes: an introduction to ichthyol-
ogy. Pearson Prentice Hall, Englewood Cliffs
Mueller RP, Southard SS, May CW, Pearson WH, Cullinan VI
(2008) Juvenile coho salmon leaping ability and behavior
in an experimental culvert test bed. Trans Am Fish Soc
137:941–950
Mueller M, Pander J, Geist J (2011) The effects of weirs on
structural stream habitat and biological communities.
J Appl Ecol 48:1450–1461
Noatch MR, Suski CD (2012) Non-physical barriers to deter fish
movements. Environ Rev 20:71–82
Nooman MJ, Grant JWA, Jackson CD (2012) A quantitative
assessment of fish passage efficiency. Fish Fish
12:450–464
Northcote TG (2010) Controls for trout and char migratory/
resident behavior mainly in stream systems above and
below waterfalls/barriers: a multidecadal and broad geo-
graphical review. Ecol Freshw Fish 19:487–509
Ombredane D, Fontenelle G, Ohresser H, Rochedpeau S (1987)
Dam overpassing by migratory adult salmonids: an anal-
ysis of the leaping behavior with the view of a better
management. BFPP Bull Fr Peche Prot Milieux Aquat
305:67–80
R. Mora
´n-Lo
´pez, O. Uceda Tolosa
123
Ovidio M, Philippart J-C (2002) The impact of small physical
obstacles on upstream movements of six species of fish.
Hydrobiologia 483:55–69
Ovidio M, Capra H, Philipart J-C (2007) Field protocol for
assessing small obstacles to migration of brown trout
Salmo trutta, and European grayling Thymallus thymallus:
a contribution to the management of free movement in
rivers. Fish Manag Ecol 14:41–50
Pedersen LF, Koed A, Malte H (2008) Swimming performance
of wild and F1-hatchery-reared Atlantic salmon (Salmo
salar) and brown trout (Salmo trutta). Ecol Freshw Fish
17:425–431
Powers PD, Osborn JF (1985) Analysis of barriers to upstream
fish migration: an investigation of the physical and bio-
logical conditions affecting fish passage success at culverts
and waterfalls. US Department of Energy, Boneville Power
Administration, Division of Fish and Wildlife, Final Pro-
ject Report Part 4 of 4 n DOE/BP-36523-1, Project No.
198201400
Rahel FJ (2013) Intentional fragmentation as a management
strategy in aquatic systems. Bioscience 63:362–372
Reiser DW, Huang C-M, Beck S, Gagner M, Jeanes E (2006)
Defining flow windows for upstream passage of adult
anadromous salmonids at cascades and falls. Trans Am
Fish Soc 135:668–679
Roscoe DW, Hinch SG (2010) Effectiveness monitoring of fish
passage facilities: historical trends, geographic patterns
and future directions. Fish Fish 11:12–33
Royal Decree 630/2013 (2013) Por el que se regula el Cata
´logo
espan
˜ol de especies exo
´ticas invasoras, Ministerio de
Agricultura, Alimentacio
´n y Medio Ambiente, BOE nu
´m.
185, pp 56764–56786
Sabater S, Mun
˜oz I, Feio MJ, Romanı
´AM, Grac¸a MAS (2009)
The Iberian rivers. In: Tockner K, Robinson CT, Uehlinger
U (eds) Rivers of Europe. Academic Press, London,
pp 113–149
Schneider CA, Rasband WS, Eliceiri KW (2012) NIH image to
ImageJ: 25 years of image analysis. Nat Methods
9:671–675
Schwenk WS, Donovan TM (2011) A multispecies framework
for landscape conservation planning. Conserv Biol
25:1010–1021
Shih AM, Techet AH (2010) Characterization of the jumping
behavior of archer fish, Toxotes Microlepis. In: 6th world
congress of biomechanics (WCB 2010). Singapore, 1–6
Aug, pp 48–51
Stuart IG, Williams A, McKenzie J, Holt T (2006) Managing a
migratory pest species: a selective trap for common carp.
N Am J Fish Manag 26:888–893
Ve
´lez-Espino LA, McLaughlin RL, Jones ML, Pratt TC (2011)
Demographic analysis of trade-offs with deliberate frag-
mentation of streams: control of invasive species versus
protection of native species. Biol Conserv 144:1068–1080
von Brandt A (1984) Fish catching methods of the world, 3rd
edn. Fishing News Books, Farnham
Xenopoulos MA, Lodge DM, Alcamo J, Ma
¨rker M, Schulze K,
Van Vuuren DP (2005) Scenarios of freshwater fish
extinctions from climate change and water withdrawal.
Glob Change Biol 11:1557–1564
Relative leaping abilities of native versus invasive cyprinids as criteria
123
- A preview of this full-text is provided by Springer Nature.
- Learn more
Preview content only
Content available from Biological Invasions
This content is subject to copyright. Terms and conditions apply.
