29.12.2012 Views

Chondrostoma nasus - Biology Centre of the Academy of Sciences of

Chondrostoma nasus - Biology Centre of the Academy of Sciences of

Chondrostoma nasus - Biology Centre of the Academy of Sciences of

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Ecology <strong>of</strong> Freshwater Fish 2008: 17: 502–514<br />

Printed in Malaysia Æ All rights reserved<br />

Functional role <strong>of</strong> environmental stimuli for <strong>the</strong><br />

spawning migration in Danube nase<br />

<strong>Chondrostoma</strong> <strong>nasus</strong> (L.)<br />

Rakowitz G, Berger B, Kubecka J, Keckeis H. Functional role <strong>of</strong><br />

environmental stimuli for <strong>the</strong> spawning migration in Danube nase<br />

<strong>Chondrostoma</strong> <strong>nasus</strong> (L.).<br />

Ecology <strong>of</strong> Freshwater Fish 2008: 17: 502–514. Ó 2008 The Authors.<br />

Journal compilation Ó 2008 Blackwell Munksgaard<br />

Abstract – In spring 2004 and 2005, an online monitoring by horizontal<br />

split-beam hydroacoustics complemented by trap net catches and electro<br />

fishing was conducted to investigate <strong>the</strong> spawning migration <strong>of</strong> nase<br />

<strong>Chondrostoma</strong> <strong>nasus</strong> (L.) into <strong>the</strong> Fischa River, a tributary <strong>of</strong> <strong>the</strong> Danube<br />

River east <strong>of</strong> Vienna, Austria. Upstream-moving adult nase could be<br />

counted hydroacoustically based on <strong>the</strong>ir dominance (93%) <strong>of</strong> fish in <strong>the</strong><br />

range 45–55 cm total length. Significant correlations were observed<br />

between <strong>the</strong> number <strong>of</strong> migrants and three environmental factors (water<br />

temperature, water level and turbidity), with special focus on <strong>the</strong>ir<br />

fluctuations. Thresholds <strong>of</strong> water temperature, water level and turbidity<br />

were revealed as general environmental cues on a seasonal scale. On a fine<br />

temporal scale, <strong>the</strong> fluctuations <strong>of</strong> <strong>the</strong> environmental factors over days,<br />

especially in <strong>the</strong> main river, represent relevant stimuli; <strong>the</strong>y initiate and<br />

control <strong>the</strong> fine-scaled temporal pattern as well as <strong>the</strong> intensity <strong>of</strong> <strong>the</strong><br />

spawning migration. In addition, a combination <strong>of</strong> increasing water<br />

temperatures and decreasing water levels, corresponding to <strong>the</strong> period after<br />

a flood peak, <strong>of</strong>fers high predictability <strong>of</strong> favourable spawning conditions<br />

at <strong>the</strong> spawning place and acts as a stimulus for <strong>the</strong> right timing for<br />

upstream migration in Danube nase.<br />

Introduction<br />

One <strong>of</strong> <strong>the</strong> most obvious changes in behaviour is that<br />

related to spawning activity, which has a marked<br />

impact on <strong>the</strong> migratory behaviour <strong>of</strong> many fish<br />

species (Lucas & Baras 2001). The initiation <strong>of</strong><br />

spawning in fish is controlled by an endogenous cycle<br />

<strong>of</strong> gonadal development and by an internal mechanism<br />

that synchronises this cycle with environmental cues<br />

(Vlaming 1974; Munro et al. 1990; Wootton 1990).<br />

Northcote (1969, 1998) and Westin (1977) proposed<br />

that temperature acts as a directional orientation cue or<br />

as a threshold stimulus for initiating migratory movement<br />

prior to spawning (Raymond 1979), particularly<br />

in combination with o<strong>the</strong>r environmental stimuli such<br />

as discharge and photoperiod, although distinguishing,<br />

Ó 2008 The Authors<br />

Journal compilation Ó 2008 Blackwell Munksgaard<br />

ECOLOGY OF<br />

FRESHWATER FISH<br />

G. Rakowitz 1 , B. Berger 1 ,<br />

J. Kubecka 2 , H. Keckeis 1<br />

1 Department <strong>of</strong> Freshwater Ecology, University <strong>of</strong><br />

Vienna, Vienna, Austria, 2 Hydrobiological Institute<br />

<strong>of</strong> <strong>Academy</strong> <strong>of</strong> <strong>Sciences</strong> <strong>of</strong> <strong>the</strong> Czech Republic,<br />

České Budějovice, Czech Republic<br />

Key words: freshwater fish; fish migration; environmental<br />

factors; hydroacoustics; multiple linear<br />

regression analysis<br />

G. Rakowitz, Department <strong>of</strong> Freshwater Ecology,<br />

University <strong>of</strong> Vienna, Althanstrasse 14, A-1090<br />

Vienna, Austria; e-mail: georg.rakowitz@univie.<br />

ac.at<br />

Accepted for publication February 29, 2008<br />

which are direct and which are indirect effects is<br />

difficult. Increased water temperature and increased<br />

water level are among <strong>the</strong> main factors effecting<br />

upstream spawning migration, especially in salmonids<br />

(Banks 1969; Northcote 1984; Jonsson 1991; Erkinaro<br />

et al. 1999). In some species, small changes in<br />

temperature may strongly affect <strong>the</strong> behaviour <strong>of</strong><br />

spawners, i.e., river lamprey Lampetra fluviatilis<br />

(Linnaeus) (Sjöberg 1977), especially when responses<br />

occur at precise temperature thresholds as, for example,<br />

in barbel Barbus barbus (Linnaeus) (Baras &<br />

Philippart 1999).<br />

Recently, more research has been carried out on <strong>the</strong><br />

movements <strong>of</strong> cyprinids (Mills 1991; Baras 1995;<br />

Slavik 1996; Clough et al. 1998; Lucas 2000), whereby<br />

migratory patterns clearly associated with spawning<br />

502 doi: 10.1111/j.1600-0633.2008.00302.x


(Rodriguez-Ruiz & Granado-Lorencio 1992; Penaz<br />

1996; Winter & Fredrich 2003) have been demonstrated.<br />

Several native cyprinids <strong>of</strong> <strong>the</strong> Danube undertake<br />

spawning migrations into tributaries (Schiemer &<br />

Waidbacher 1992), among which nase <strong>Chondrostoma</strong><br />

<strong>nasus</strong> (L.) is an indicator species <strong>of</strong> good river<br />

ecosystem health over <strong>the</strong> breadth <strong>of</strong> its distribution<br />

in Europe (Lelek 1987; Winfield & Townsend 1991;<br />

Schiemer et al. 2004). In many large rivers its abundance<br />

and distribution has dramatically declined<br />

(Penaz 1996) because <strong>of</strong> habitat fragmentation<br />

(Welcomme 1994; Ward & Tockner 2001). Studies<br />

about spawning behaviour (Zbinden & Maier 1996;<br />

Huber & Kirchh<strong>of</strong>er 1998), spawning sites composition<br />

(Lusk et al. 1995; Keckeis 2001) and large-scaled<br />

migrations (Steinmann et al. 1937; Povz 1988) have<br />

been carried out, but very little is known about <strong>the</strong> finescaled<br />

patterns and environmental triggers and cues for<br />

short-term variability <strong>of</strong> its spawning migration.<br />

In <strong>the</strong> present study, we used horizontal echo<br />

sounding for online monitoring <strong>of</strong> <strong>the</strong> spawning<br />

migration <strong>of</strong> nase into an Austrian tributary <strong>of</strong> <strong>the</strong><br />

Danube River and related <strong>the</strong> observed pattern to a<br />

range <strong>of</strong> abiotic parameters (i.e., water temperature,<br />

water level and turbidity) on a fine temporal scale.<br />

Beyond a general seasonal component, <strong>the</strong>se factors<br />

fluctuate considerably in natural systems. We provide<br />

evidence that <strong>the</strong> fluctuations <strong>of</strong> <strong>the</strong>se environmental<br />

cues – short-term increases or decreases within <strong>the</strong><br />

spawning season – are relevant stimuli for describing<br />

<strong>the</strong> magnitude and timing <strong>of</strong> <strong>the</strong> spawning migration<br />

in Danube nase.<br />

Fig. 1. The Danube River in Austria and its<br />

tributary, <strong>the</strong> Fischa River, showing spawning<br />

areas <strong>of</strong> nase, locations <strong>of</strong> <strong>the</strong> echo<br />

sounder and traps and where environmental<br />

variables were recorded.<br />

N<br />

Study area<br />

The Fischa River is a tributary <strong>of</strong> <strong>the</strong> Danube River<br />

east <strong>of</strong> Vienna, originating at an altitude <strong>of</strong> 240 m in<br />

sou<strong>the</strong>rn Lower Austria (Haschendorf ⁄ Ebenfurth) and<br />

flowing nor<strong>the</strong>ast for 35 km before joining <strong>the</strong> Danube<br />

(Fig. 1). Toge<strong>the</strong>r with its main tributary, <strong>the</strong> Piesting<br />

River, which originates at an altitude <strong>of</strong> 1200 m, <strong>the</strong><br />

total catchment area is 549.4 km 2 , with a pluvio-nival<br />

drain-regime (Mader et al. 1996) and a mean annual<br />

discharge <strong>of</strong> 7.5 m 3 Æ s )1 . Spring (March ⁄ April) and<br />

winter (November ⁄ December) floods are common<br />

(Kresser 1989). The water temperature does not<br />

exceed 16.0 °C during summer. Its hyporithral character<br />

changes into epi- and metapotamal within<br />

a transition area approximately 4.0 km upstream <strong>of</strong><br />

<strong>the</strong> confluence with <strong>the</strong> Danube, where discharge<br />

is mainly influenced by <strong>the</strong> Danube water level.<br />

In spring, <strong>the</strong> river is characterised by a sequence<br />

<strong>of</strong> single-species spawning events starting with pike<br />

Esox lucius Linnaeus and dace Leuciscus leuciscus<br />

(Linnaeus), followed by nase, barbel and ide Leuciscus<br />

idus (Linnaeus) (Keckeis & Rakowitz 2005). Two<br />

nase spawning sites are located 4.5 km upstream <strong>the</strong><br />

mouth, where <strong>the</strong> Fischa splits into several smaller<br />

branches, creating a local floodplain area (Fig. 1). Two<br />

<strong>of</strong> <strong>the</strong>se branches contain riffles with high current<br />

velocity and are regularly frequented by <strong>the</strong> same nase<br />

individuals from year to year (Keckeis 1999; Keckeis<br />

& Rakowitz 2006). Adults are present in this<br />

river system only during spawning season, from<br />

March to early June (Keckeis 1999).<br />

Spawning area<br />

Weir<br />

Echo sounder<br />

Traps 2004<br />

Traps 2005<br />

Water temperature<br />

Water gauge<br />

Turbidity<br />

Environmental cues in nase spawning migration<br />

Fischa river<br />

0 50 100 km<br />

Danube<br />

0 500 m<br />

503


Rakowitz et al.<br />

Materials and methods<br />

Hydroacoustic recording<br />

The hydroacoustic investigation <strong>of</strong> <strong>the</strong> spawning<br />

migration <strong>of</strong> nase into <strong>the</strong> Fischa River was conducted<br />

in <strong>the</strong> transition area at a site 4.0 km upstream <strong>of</strong> <strong>the</strong><br />

river mouth (Fig. 1). The cross-sectional riverbed<br />

topography here is well suited for fixed location<br />

horizontal beaming: a triangular cross-section with a<br />

gently sloping right shore and steeply sloping left<br />

shore. At average discharge <strong>the</strong> river cross-section at<br />

<strong>the</strong> sampling site is 15.0 m wide with a maximum<br />

depth <strong>of</strong> 2.15 m near <strong>the</strong> left shore. The bottom<br />

consists <strong>of</strong> gravel and has an even gradient with small<br />

obtrusions. The site is located about 500 m downstream<br />

<strong>of</strong> <strong>the</strong> spawning areas and about 4 km from <strong>the</strong><br />

confluence, a sufficient distance to minimise hydroacoustic<br />

recording <strong>of</strong> o<strong>the</strong>r Danubian fish species<br />

migrating in <strong>the</strong> lower part <strong>of</strong> <strong>the</strong> tributary. Also,<br />

water level fluctuations during spring floods are less<br />

extensive than in <strong>the</strong> area <strong>of</strong> <strong>the</strong> confluence without<br />

needing to dismantle <strong>the</strong> hydroacoustic setup because<br />

<strong>of</strong> high water levels.<br />

From 30 March to 27 May 2004 and from 12 March<br />

to 3 May 2005, hydroacoustic data were collected nearcontinuously<br />

with a portable 120 kHz digital echo<br />

sounder (Simrad, EK 60, Horten, Norway) and an<br />

elliptical (4.4° ·8.3°) splitbeam transducer. The transducer<br />

was mounted on a tetrapod-scaffold and aimed<br />

horizontally across <strong>the</strong> river from <strong>the</strong> right shore<br />

towards <strong>the</strong> left shore using a dual-axis remote<br />

controlled rotator (Sub-Atlantic, 1128-MAS, Aberdeen,<br />

Scotland). A panel fence was placed downstream<br />

<strong>of</strong> <strong>the</strong> transducer to guide upstream migrating fish out<br />

<strong>of</strong> <strong>the</strong> transducer’s nearfield. The panel fence was<br />

manufactured from solid plastic tubes with a lattice<br />

spacing <strong>of</strong> 1.5 cm, fixed on <strong>the</strong> ground and equipped<br />

with floating bodies on <strong>the</strong> top. Prior to continuous<br />

echo-sounding, beam mapping was conducted to<br />

adjust <strong>the</strong> horizontal beam as close to <strong>the</strong> bottom as<br />

possible in order to detect bottom-oriented upstream<br />

migrating fish. The beam was tilted downwards by 7.6°<br />

degrees and panned downstream with 13.0° degrees.<br />

Once properly oriented, <strong>the</strong> echo sounder system was<br />

calibrated in situ using <strong>the</strong> tungsten-carbide standard<br />

calibration sphere (Ø = 23.00 mm) with a nominal<br />

target strength (TS) value <strong>of</strong> TS = )40.40 dB and kept<br />

stationary for <strong>the</strong> whole sampling period. Basic<br />

information about <strong>the</strong> sound propagation in <strong>the</strong><br />

horizontal bottom-aligned beam is given by Rakowitz<br />

& Kubecka (2006). During monitoring, sound pulses<br />

were transmitted in average with <strong>the</strong> duration <strong>of</strong><br />

0.10 ms, <strong>the</strong> power <strong>of</strong> 63 W, <strong>the</strong> pulse width <strong>of</strong><br />

0.064 ms, <strong>the</strong> bandwidth <strong>of</strong> 11.80 kHz and <strong>the</strong> pulse<br />

repetition rate <strong>of</strong> 10 pingsÆs )1 . The sensitivity threshold<br />

504<br />

value <strong>of</strong> <strong>the</strong> detected echoes was set to )50.00 dB<br />

because <strong>of</strong> <strong>the</strong> level <strong>of</strong> <strong>the</strong> background noise.<br />

To increase <strong>the</strong> detectability in <strong>the</strong> horizontal bottomaligned<br />

beam, <strong>the</strong> echo length detector was set to 0.8–<br />

1.5 times <strong>the</strong> transmitted pulse, <strong>the</strong> maximum phase<br />

deviation was set to 1.2° and <strong>the</strong> maximum gain<br />

compensation was enhanced to 6 dB (one-way).<br />

Signals were automatically compensated for <strong>of</strong>f-axis<br />

distance. The echo sounder was in operation for an<br />

average 23.5 h a day for 59 days in 2004 and 53 days<br />

in 2005. Changing environmental conditions (water<br />

level, turbidity) did not alter <strong>the</strong> efficiency <strong>of</strong> <strong>the</strong><br />

hydroacoustic system. Analog data from <strong>the</strong> transducer<br />

were digitalised in <strong>the</strong> echo sounder’s GPT unit<br />

(General Purpose Transceiver), recorded on a laptop<br />

and stored on an external hard disk.<br />

Hydroacoustic analysis<br />

For this study, a total <strong>of</strong> 66 days – 33 days each year –<br />

were analysed with Sonar5-Pro version 5.9.1 postprocessing<br />

s<strong>of</strong>tware (Balk & Lindem 2000, 2002;<br />

Lindem Data Acquisition, Oslo, Norway), which was<br />

updated regularly (http://www.fys.uio.no/~hbalk).<br />

Data recorded in 2005 was additionally corrected for<br />

biased TS-values because <strong>of</strong> inhomogeneous sound<br />

propagation in <strong>the</strong> horizontal bottom-aligned beam<br />

(Rakowitz & Kubecka 2006). To obtain <strong>the</strong> total<br />

number <strong>of</strong> upstream migrating nase (Fig. 2e,f), <strong>the</strong><br />

remaining 26 and 23 days, respectively, were interpolated<br />

using <strong>the</strong> mean <strong>of</strong> upstream migrants <strong>of</strong> <strong>the</strong> day<br />

before and <strong>the</strong> day after. Linear regression analysis<br />

between daily counts and interpolated data from a<br />

continuous series <strong>of</strong> 11 days in 2005 revealed a highly<br />

significant relationship (R 2 = 0.91, P < 0.001). After<br />

conversion <strong>of</strong> raw data, <strong>the</strong> acoustic analysis was<br />

carried out by single echo detection (SED). For SED,<br />

we did not apply a traditional detector based on echo<br />

length (Soule et al. 1997), but <strong>the</strong> alternative Crossfilter<br />

detector (Balk 2001) found in Sonar5. This<br />

detector has been developed to improve track quality<br />

and to reduce erroneous detections in data with low<br />

signal-to-noise ratio (SNR). The detector contains a<br />

suggestor obtaining trace candidates, an evaluator<br />

sorting out unwanted candidates, a SED former which<br />

defines <strong>the</strong> individual echo detections within a trace<br />

and an echo quality estimator. The suggestor works<br />

with filters identifying <strong>the</strong> targets frequency response<br />

while <strong>the</strong> evaluator operates with a set <strong>of</strong> min and max<br />

trace feature tests. In 2004 and 2005, <strong>the</strong> cross-filter<br />

settings which provided best results for tracking fish in<br />

<strong>the</strong> Fischa were for <strong>the</strong> suggestor: foreground filter:<br />

height = 1 and 1, width = 1 and 1; background filter:<br />

height = 23 and 11, width = 7 and 13; <strong>of</strong>fset = +6 and<br />

+6 dB. For <strong>the</strong> evaluator, trace length and trace area<br />

were used. Minimum and maximum values for <strong>the</strong>


Fig. 2. Water temperature (a, b), water level<br />

and turbidity (c, d) at different sections <strong>of</strong><br />

<strong>the</strong> Fischa River and <strong>the</strong> Danube River and<br />

number <strong>of</strong> daily upstream migrating nase<br />

estimated from hydroacoustic records (e, f)<br />

in spring 2004 and 2005.<br />

Water temperature (°C)<br />

Water level (cm)<br />

Turbidity (ntu)<br />

No. <strong>of</strong> migranrts<br />

(Ind day –1 )<br />

16<br />

14<br />

12<br />

10<br />

8<br />

6<br />

4<br />

2<br />

0<br />

500<br />

400<br />

300<br />

200<br />

100<br />

track length were 1–250 pings and for <strong>the</strong> area 8–232<br />

number <strong>of</strong> samples in a detected region. The settings<br />

rejected unwanted single echoes from drifting<br />

debris, stones and from fluctuations in <strong>the</strong> background<br />

reverberation level to be included in <strong>the</strong> fish<br />

tracks. They remained fairly consistent over <strong>the</strong> study<br />

periods. Automatic tracking caused severe bias <strong>of</strong><br />

counts and sizing, we <strong>the</strong>refore applied manual<br />

tracking.<br />

Identifying acoustic nase targets using supplementary<br />

sampling<br />

To obtain qualitative information about migrating fish<br />

species, trap net catches and electro fishing were<br />

conducted, and sex, length (total length, cm) and<br />

weight (fresh weight, g) <strong>of</strong> <strong>the</strong> captured individuals<br />

were measured. In 2004, <strong>the</strong> traps were placed in <strong>the</strong><br />

left (in terms <strong>of</strong> flow direction) spawning arm 80 m<br />

downstream <strong>of</strong> <strong>the</strong> spawning place and 380 m<br />

upstream from <strong>the</strong> echo-sounding site and hence<br />

sampled only a part <strong>of</strong> <strong>the</strong> river. In 2005, <strong>the</strong> traps<br />

were put closer to <strong>the</strong> site <strong>of</strong> <strong>the</strong> echosounder and<br />

sampled <strong>the</strong> whole width (Fig. 1). They were built<br />

from iron bars with 4 m in length, 1.3 m wide and<br />

1.2 m high and <strong>the</strong> mesh size <strong>of</strong> <strong>the</strong> flute was 10 mm.<br />

50<br />

0<br />

140<br />

120<br />

100<br />

80<br />

60<br />

40<br />

20<br />

0<br />

2 9 / 2 / 0 4<br />

7 / 3 / 0 4<br />

1 4 / 3 / 0 4<br />

2 8 / 3 / 0 4<br />

4 / 4 / 0 4<br />

1 / 4 / 0 4<br />

1 8 / 4 / 0 4<br />

2 1 / 3 / 0 4<br />

Environmental cues in nase spawning migration<br />

2004<br />

2 5 / 4 / 0 4<br />

2 / 5 / 0 4<br />

9 / 5 / 0 4<br />

1 6 / 5 / 0 4<br />

2 3 / 5 / 0 4<br />

3 0 / 5 / 0 4<br />

(a)<br />

(c)<br />

(e)<br />

2 8 / 2 / 0 5<br />

7 / 3 / 0 5<br />

1 4 / 3 / 0 5<br />

2 1 / 3 / 0 5<br />

2 8 / 3 / 0 5<br />

4 / 4 / 0 5<br />

1 1 / 4 / 0 5<br />

2005<br />

1 8 / 4 / 0 5<br />

2 5 / 4 / 0 5<br />

2 / 5 / 0 5<br />

9 / 5 / 0 5<br />

1 6 / 5 / 0 5<br />

Spawning area<br />

Transition area<br />

Confluence<br />

Danube<br />

Turbidity<br />

2 3 / 5 / 0 5<br />

3 0 / 5 / 0 5<br />

They were set to catch upstream as well as downstream<br />

migrating fish. Detailed information on <strong>the</strong><br />

construction <strong>of</strong> <strong>the</strong> traps is described in Mühlbauer<br />

et al. (2003). The traps were emptied once a day in <strong>the</strong><br />

morning from 8 March to 28 May 2004 and from 25<br />

February to 6 May 2005, excluding 25 days when <strong>the</strong><br />

traps were flooded. Electro fishing was conducted four<br />

times in 2004 (21, 22, 26 April and 21 May) and three<br />

times in 2005 (7, 8 April and 3 May). We used a<br />

portable generator FEG 1000 (200 V, 3 A). Both<br />

spawning branches were sampled from <strong>the</strong> confluence<br />

<strong>of</strong> <strong>the</strong> two branches to <strong>the</strong> end <strong>of</strong> <strong>the</strong> spawning places<br />

(Fig. 1) within 1 day.<br />

From <strong>the</strong> species which undertake spawning migrations<br />

from <strong>the</strong> Danube into <strong>the</strong> tributary, 80% <strong>of</strong> <strong>the</strong><br />

total catch <strong>of</strong> migrants was comprised by nase, 9% by<br />

barbel, 5% by ide, 3% by white bream Abramis<br />

bjoerkna (Linnaeus) and <strong>the</strong> remaining 3% by pike,<br />

bream Abramis brama (Linnaeus), vimba Vimba<br />

vimba (Linnaeus) and burbot Lota lota (Linnaeus).<br />

Nase were separated from nontarget species based on<br />

<strong>the</strong> length distribution <strong>of</strong> captured individuals<br />

(Romakkaniemi et al. 2000; Lilja & Romakkaniemi<br />

2003). Although nase is breeding at a size above<br />

20 cm, 80% <strong>of</strong> <strong>the</strong> total catch <strong>of</strong> nase was comprised<br />

by nase between 45 and 55 cm. Fur<strong>the</strong>rmore, 93% <strong>of</strong><br />

(b)<br />

(d)<br />

(f)<br />

505


Rakowitz et al.<br />

all migrants in <strong>the</strong> size range <strong>of</strong> 45–55 cm were<br />

represented by nase. Thus a size range between 45 and<br />

55 cm was applied to <strong>the</strong> hydroacoustically recorded<br />

fish to separate nase from non-target fish.<br />

The relationship between TS and fish TL,<br />

TS = 24.71 · (log 10 · TL) ) 89.63, from Frouzova<br />

et al. (2005) was applied to convert <strong>the</strong> size <strong>of</strong><br />

captured individuals into TS values. Accordingly, <strong>the</strong><br />

total length <strong>of</strong> 45–55 cm corresponds to TS values <strong>of</strong><br />

)21.91 to )24.07 dB.<br />

Environmental parameters<br />

During <strong>the</strong> sampling period, water temperature (°C),<br />

water level (cm) and turbidity (ntu) were measured in<br />

daily intervals. Water temperature was measured at<br />

four and water level at three different sites, respectively<br />

along a longitudinal pr<strong>of</strong>ile (spawning area,<br />

transition area, confluence and Danube River), turbidity<br />

at one site (Fig. 1). The water level in <strong>the</strong> Fischa<br />

was read <strong>of</strong>f at a water level in <strong>the</strong> left spawning arm<br />

and at a water level near <strong>the</strong> echo-sounding site every<br />

day in <strong>the</strong> morning. Daily water level data and mean<br />

daily water temperature data <strong>of</strong> <strong>the</strong> Danube were<br />

provided from <strong>the</strong> nearest installed water level Orth by<br />

<strong>the</strong> Austria Water Authority Via-Donau. Turbidity was<br />

measured once a day using a portable turbidimeter<br />

(Turbiquant 1000 IR; Merck, Darmstadt, Germany) at<br />

<strong>the</strong> traps.<br />

Statistical analyses<br />

The Kolmogorov–Smirnov-test (K-S-test) was used to<br />

test <strong>the</strong> biotic and abiotic data set (fish passage rate,<br />

temperature, water level and turbidity) for normal<br />

distribution. Between year differences water temperature<br />

and average upstream passage rate were tested<br />

with <strong>the</strong> Mann–Whitney U-test. Between year differences<br />

<strong>of</strong> water temperature at <strong>the</strong> spawning site was<br />

tested with a t-test, as <strong>the</strong>se data revealed a normal<br />

distribution. Between site differences <strong>of</strong> water<br />

temperature and water level were tested with a<br />

Kruskal–Wallis Test. Spearman rank correlations<br />

between daily passage rate and time series <strong>of</strong><br />

environmental variables were estimated with time<br />

lags <strong>of</strong> up to ±7 days at intervals <strong>of</strong> 1 day for all<br />

variables. The time series were calculated according to<br />

Y=X t – X (t)1), where X is <strong>the</strong> measured value at <strong>the</strong><br />

point <strong>of</strong> time t, to render <strong>the</strong>m stationary (Lilja &<br />

Romakkaniemi 2003). For each time lag, <strong>the</strong> differences<br />

(1–7 days) were summed up. The significant<br />

correlation <strong>of</strong> each variable and <strong>the</strong> highest significant<br />

correlation <strong>of</strong> <strong>the</strong> time lags were used for fur<strong>the</strong>r<br />

statistical analyses. Regression models (peak, exponential)<br />

were used to analyse <strong>the</strong> mode <strong>of</strong> triggering.<br />

Finally, a multiple linear regression analysis (SPSS<br />

506<br />

15.0.1) was employed to analyse <strong>the</strong> pattern <strong>of</strong> <strong>the</strong><br />

fluctuations in <strong>the</strong> spawning runs.<br />

Results<br />

In 2004 and 2005, water temperatures were very<br />

similar and did not differ significantly between <strong>the</strong><br />

spawning and <strong>the</strong> transition area (Mann–Whitney<br />

U-Test, P > 0.05). At <strong>the</strong> spawning area, values<br />

ranged from 6.3 to 13.8 °C in 2004 and from 3.9 to<br />

14.5 °C in 2005. At this site highly significant<br />

differences (t-test, P < 0.001) were observed between<br />

<strong>the</strong> two investigation periods, <strong>the</strong> corresponding<br />

average (±SD) temperatures were 10.9 ± 2.0 and<br />

8.9 ± 2.7, respectively. During both years, temperature<br />

increased in an oscillating pattern with amplitudes <strong>of</strong><br />

±3.0–4.0 °C within a 5–10-day period (Fig. 2a,b). In<br />

general, <strong>the</strong> oscillation amplitudes in <strong>the</strong> Danube River<br />

were damped compared with <strong>the</strong> Fischa River. At <strong>the</strong><br />

start <strong>of</strong> <strong>the</strong> sampling period, <strong>the</strong> Danube was<br />

1.0–2.0 °C colder than <strong>the</strong> Fischa, followed by a<br />

period <strong>of</strong> nearly same temperature levels in both<br />

rivers. In late May, <strong>the</strong> Danube River was approximately<br />

2.0 °C warmer than <strong>the</strong> Fischa. The average<br />

temperatures between <strong>the</strong> two rivers within each<br />

investigation period were not significantly different<br />

(Kruskal–Wallis test, P > 0.05), but a significant linear<br />

relationship is evident between <strong>the</strong> water temperature<br />

<strong>of</strong> <strong>the</strong> Danube and that at <strong>the</strong> spawning site (Fig. 3a).<br />

The linear regression (R 2 = 0.87, a=3.27, b=0.70,<br />

P < 0.001) indicated significantly higher Fischa water<br />

temperatures at times <strong>of</strong> low temperatures in <strong>the</strong><br />

Danube, i.e., 3.00 °C in <strong>the</strong> Danube correspond with<br />

an average <strong>of</strong> 5.3 °C in <strong>the</strong> Fischa, and 6.0 °C in <strong>the</strong><br />

Danube correspond with 7.5 °C in <strong>the</strong> tributary.<br />

Two <strong>of</strong> <strong>the</strong> three recorded water levels revealed a<br />

similar pattern <strong>of</strong> water level fluctuations, resulting in<br />

a strong synchronisation (Spearman coefficient,<br />

r=0.88, P < 0.001) <strong>of</strong> water level in <strong>the</strong> transition<br />

area <strong>of</strong> <strong>the</strong> Fischa with water level in <strong>the</strong> Danube. This<br />

was not <strong>the</strong> case at <strong>the</strong> spawning site (Fig. 3b), except<br />

during several peak floods in 2005 when <strong>the</strong> hydrological<br />

situation <strong>of</strong> <strong>the</strong> spawning area was impacted.<br />

In 2004 no peak flood occurred in Danube, but 2005<br />

was characterised by water level fluctuations over<br />

several meters (Fig. 2c,d). Average water levels at <strong>the</strong><br />

three sites, as well as range and variations, are given in<br />

Table 1.<br />

Water level at all sites differed significantly between<br />

<strong>the</strong> 2 years (Kruskal–Wallis test, P < 0.001). In 2005,<br />

<strong>the</strong> variation <strong>of</strong> <strong>the</strong> Danube water level was significantly<br />

higher compared with 2004. Mean turbidity in<br />

<strong>the</strong> Fischa was twice as high in 2005 (tur 05 = 28.53<br />

ntu) than in 2004 (tur 04 = 12.59 ntu), which corresponds<br />

well with <strong>the</strong> observed water level differences<br />

between years. The higher turbidity in 2005 mainly


Water temperature (ºC)<br />

in Fischa river and Danuba river<br />

Water level Fischa river (cm)<br />

18<br />

16<br />

14<br />

12<br />

10<br />

8<br />

6<br />

4<br />

2<br />

0<br />

700<br />

600<br />

500<br />

400<br />

300<br />

200<br />

100<br />

Spawning area<br />

Danube<br />

1 : 1<br />

0 2 4 6 8 10 12 14 16 18<br />

Water temperature (°C)<br />

in <strong>the</strong> confluence Fischa - Danube<br />

Spawning area<br />

Transition area<br />

1 : 1<br />

0<br />

0 100 200 300 400 500 600 700<br />

Water level Danube river (cm)<br />

Fig. 3. Relationship between <strong>the</strong> water temperature in <strong>the</strong> Danube<br />

River and water temperature at <strong>the</strong> spawning area in <strong>the</strong> Fischa<br />

River during spawning season (a). Relationship between <strong>the</strong> water<br />

level in <strong>the</strong> Danube River and water level at <strong>the</strong> spawning area and<br />

at <strong>the</strong> transition area in <strong>the</strong> Fischa River during spawning season<br />

(b). Pooled data <strong>of</strong> two spawning seasons.<br />

Table 1. Average water level (cm) at three sites, as well as range and<br />

variation in 2004 and 2005.<br />

2004 2005<br />

Spawning<br />

area<br />

(cm)<br />

Transition<br />

area<br />

(cm)<br />

Danube<br />

River<br />

(cm)<br />

Spawning<br />

area<br />

(cm)<br />

Transition<br />

area<br />

(cm)<br />

reflected peak floods during that year. The main flood<br />

peak led to a steep increase in turbidity (to approximately<br />

10 times <strong>the</strong> average value) (Fig. 2d).<br />

(a)<br />

(b)<br />

Danube<br />

River<br />

(cm)<br />

Mean 35.9 64.0 230.7 35.9 100.6 255.0<br />

SD 3.7 18.3 30.5 19.6 53.8 110.4<br />

Median 35.0 55.0 228.0 31.0 80.0 274.0<br />

Min 25.0 40.0 179.0 27.1 48.0 79.0<br />

Max 48.0 118.0 298.0 159.4 312.0 506.0<br />

n 70.0 70.0 70.0 65.0 65.0 65.0<br />

mean ¼ average; SD ¼ standard deviation, Min ¼ minimum; Max ¼<br />

maximum, n ¼ number <strong>of</strong> observations.<br />

Environmental cues in nase spawning migration<br />

In both years <strong>the</strong>re was a continuous daily spawning<br />

migration in nase characterised by an oscillating<br />

pattern with several distinct peaks. In 2004, <strong>the</strong> first<br />

individuals were caught in <strong>the</strong> traps on 17 March, in<br />

2005 recorded on 12 March. The average passage rate<br />

(±SD) between <strong>the</strong> 2 years differed significantly<br />

(Mann–Whitney U-test, P < 0.05), being 22 (±22)<br />

individualsÆday )1 in 2004 and 31 (±16) individualÆday<br />

)1 in 2005. In 2004, <strong>the</strong> highest peak <strong>of</strong><br />

migrants was observed on 1 and 2 April (125 and 110<br />

nase, respectively), <strong>the</strong> second highest on 13 April<br />

(56) and <strong>the</strong> third highest on 25 April (25) (Fig. 2e).<br />

The spawning migration took place to a lesser extent<br />

during <strong>the</strong> remaining sampling period. In 2005, we<br />

also observed three main migration peaks. The first<br />

immigration peak was on 16 March (48 nase), <strong>the</strong><br />

second on 26 March (60) and <strong>the</strong> third on 7 April (81)<br />

(Fig. 2f).<br />

Highly significant correlations between daily<br />

upstream passage rate and environmental parameters<br />

like water temperature, water level and turbidity were<br />

detected (Fig. 4). In <strong>the</strong> Fischa spawning area, low<br />

water temperature (r =)0.266, P < 0.05), but especially<br />

increasing water temperature after 5 days<br />

(r = +0.430, P < 0.01), correlated significantly with<br />

<strong>the</strong> number <strong>of</strong> migrants (Fig. 4a). In 2004, a temperature<br />

increase from 5.1 to 7.0 °C in Danube and from<br />

7.3 to 9.8 °C in <strong>the</strong> Fischa within 5 days corresponded<br />

with <strong>the</strong> trap net catch <strong>of</strong> <strong>the</strong> first four mature nase<br />

individuals.<br />

Also in <strong>the</strong> confluence (r =)0.399, P < 0.01) and<br />

in <strong>the</strong> Danube (r =)0.395, P < 0.01) low water<br />

temperatures were most significantly correlated with<br />

<strong>the</strong> migration rate. Additionally, increasing water<br />

temperature after 1 day in <strong>the</strong> confluence (r =0.273,<br />

P < 0.05) and after 3 days in <strong>the</strong> Danube (r =0.270,<br />

P < 0.05) showed significant correlation with <strong>the</strong><br />

number <strong>of</strong> upstream migrants at <strong>the</strong> echosounder<br />

(Fig. 4b,c).<br />

At <strong>the</strong> transition area, an increased water level<br />

(r =0.424, P < 0.01) and a decreasing water level after<br />

6 days (r =)0.299, P < 0.05) showed <strong>the</strong> highest<br />

correlations with upstream-migrating nase (Fig. 4d). In<br />

<strong>the</strong> Danube, a significant correlation was observed<br />

between an increased water level and upstream-migrating<br />

spawners (r =0.277, P < 0.05), but decreasing<br />

water level after 6 days was even more highly correlated<br />

(r =)0.312, P < 0.05) (Fig. 4e). Decreasing<br />

turbidity in <strong>the</strong> Fischa with a time lag <strong>of</strong> )6 days<br />

(r = )0.448, P < 0.01), mainly caused by decreasing<br />

water level, showed <strong>the</strong> most significant correlation<br />

with <strong>the</strong> number <strong>of</strong> upstream migrants (Fig. 4f). The<br />

water level at <strong>the</strong> spawning place and number <strong>of</strong><br />

migrating nase were not significantly correlated.<br />

The correlation <strong>of</strong> each variable and <strong>the</strong> highest<br />

correlation <strong>of</strong> <strong>the</strong> time lags <strong>of</strong> each variable (white<br />

507


Rakowitz et al.<br />

Lag period (days)<br />

–8<br />

–6<br />

–4<br />

–2<br />

0<br />

2<br />

4<br />

6<br />

(a) (d)<br />

8<br />

–8<br />

–6<br />

–4<br />

–2<br />

0<br />

2<br />

4<br />

6<br />

(b)<br />

(e)<br />

8<br />

–8<br />

–6<br />

–4<br />

–2<br />

(c) (f) Fig. 4. Spearman correlation coefficients<br />

between number <strong>of</strong> upstream migrating nase<br />

and environmental variables at )7 to +7 day<br />

lag periods: water temperature (a) at <strong>the</strong><br />

spawning area, (b) in <strong>the</strong> confluence and (c)<br />

0<br />

in <strong>the</strong> Danube River; water level (d) at <strong>the</strong><br />

2<br />

transition area, (e) in <strong>the</strong> Danube River and<br />

4<br />

(f) turbidity in <strong>the</strong> Fischa River. Significant<br />

6<br />

8<br />

–0.50 –0.25 0.00 0.25 0.50 –0.50 –0.25 0.00 0.25 0.50<br />

Spearman correlation coefficient<br />

correlations (P =0.05) are indicated by<br />

dotted lines (……), most significant correlations<br />

(P =0.01) by dashed lines (- - - -).<br />

White bars indicate variables, which were<br />

used for fur<strong>the</strong>r analyses.<br />

bars in Fig. 4) were used to analyse <strong>the</strong> relationship<br />

between number <strong>of</strong> upstream migrating nase and<br />

environmental cues. A peak model [Gaussian,<br />

y = a · exp ( - 0.5 · ((x - x o) ⁄ b) 2 ] best explained<br />

<strong>the</strong> correlations <strong>of</strong> <strong>the</strong> discrete variables, whereas<br />

<strong>the</strong> best fit for <strong>the</strong> relationship <strong>of</strong> <strong>the</strong> time lags was<br />

achieved by exponential models [y = a · exp (b · x)<br />

for exponential growth and y = a · exp ( - b · x) for<br />

exponential decay]. At <strong>the</strong> spawning area (R 2 = 0.22,<br />

P < 0.001), highest nase abundance occurred at<br />

9.5 °C, in <strong>the</strong> confluence (R 2 = 0.29, P < 0.001)<br />

at 8.1 °C and in <strong>the</strong> Danube (R 2 = 0.27, P < 0.001)<br />

at 7.5 °C (Fig. 5a–c). Fur<strong>the</strong>rmore, a significant<br />

relationship between number <strong>of</strong> migrants and water<br />

temperature fluctuations was observed. Pooled data<br />

<strong>of</strong> water temperature <strong>of</strong> both years at <strong>the</strong> spawning<br />

area (R 2 = 0.24, P < 0.001) showed that a steep<br />

increase in water temperature <strong>of</strong> 2.5 °C with a time<br />

lag <strong>of</strong> )5 days was accompanied by <strong>the</strong> most<br />

upstream migrating nase (Fig. 5d). Less steep water<br />

temperature increases <strong>of</strong> 0.5 °C with a time lag <strong>of</strong><br />

1 day at <strong>the</strong> confluence (R 2 = 0.10, P < 0.01) and <strong>of</strong><br />

1.3 °C with a time lag <strong>of</strong> 3 days in <strong>the</strong> Danube<br />

(R 2 = 0.12, P < 0.01) corresponded with highest<br />

upstream passage rates <strong>of</strong> nase at <strong>the</strong> echo sounder<br />

(Fig. 5e,f).<br />

508<br />

Migration was also significantly related to absolute<br />

water levels and changes <strong>of</strong> <strong>the</strong> water level. The<br />

optimum water level in <strong>the</strong> transition area (R 2 = 0.16,<br />

P < 0.01) with respect to number <strong>of</strong> migrants was<br />

113 cm (Fig. 6a). In <strong>the</strong> Danube River, nase showed<br />

no significant relationship for a preferred water level<br />

(Fig. 6b). On <strong>the</strong> o<strong>the</strong>r hand, a decrease <strong>of</strong> water level<br />

with a time lag <strong>of</strong> )6 days, both in <strong>the</strong> Danube<br />

(R 2 = 0.17, P < 0.001) and in <strong>the</strong> transition area<br />

(R 2 = 0.10, P < 0.05) was correlated with <strong>the</strong> highest<br />

nase migration rate (Fig. 6c,d).<br />

Higher turbidity (28.15 ntu) corresponded with<br />

more migrants (R 2 = 0.13, P < 0.05). At <strong>the</strong> same<br />

time, a significant relationship (R 2 = 0.07, P < 0.05)<br />

was observed between decreasing turbidity, closely<br />

linked with <strong>the</strong> water level, with a time lag <strong>of</strong> )6 days<br />

in <strong>the</strong> Fischa, and an increasing number upstream<br />

migrating nase (Fig. 6e,f).<br />

A multiple linear regression analysis with <strong>the</strong> square<br />

root number <strong>of</strong> migrants per day as <strong>the</strong> dependent<br />

variable and <strong>the</strong> environmental variables [(a) water<br />

temperature at <strong>the</strong> spawning site, (b) at <strong>the</strong> confluence<br />

and (c) in <strong>the</strong> Danube River, (d) water level in <strong>the</strong><br />

transition area and (e) in <strong>the</strong> Danube River and (f)<br />

turbidity in <strong>the</strong> transition area] as well as <strong>the</strong>ir highest<br />

correlated time lags [(g) water temperature change at


Fig. 5. Relationship between daily number<br />

<strong>of</strong> migrants and water temperature in<br />

<strong>the</strong> Fischa River at <strong>the</strong> spawning area<br />

(a) (R 2 = 0.22, P < 0.001), in <strong>the</strong> confluence<br />

(b) (R 2 = 0.29, P < 0.001) and in <strong>the</strong><br />

Danube River (c) (R 2 = 0.27, P < 0.001)<br />

and between daily number <strong>of</strong> migrants and<br />

water temperature change at <strong>the</strong> spawning<br />

area (d) (R 2 = 0.24, P < 0.001), in <strong>the</strong><br />

confluence (e) (R 2 = 0.10, P < 0.01) and<br />

in <strong>the</strong> Danube (f) (R 2 = 0.12, P < 0.01).<br />

Pooled data <strong>of</strong> two spawning seasons. For<br />

fur<strong>the</strong>r explanation see text.<br />

y –1<br />

N o.<br />

<strong>of</strong><br />

migrants<br />

( Ind<br />

da<br />

)<br />

140 2004<br />

120 2005<br />

100<br />

80<br />

60<br />

40<br />

20<br />

0<br />

<strong>the</strong> spawning site with a time lag <strong>of</strong> )5 days, (h) in <strong>the</strong><br />

confluence with a time lag <strong>of</strong> 1 day and (i) in <strong>the</strong><br />

Danube River with a time lag <strong>of</strong> 3 days, (j) water level<br />

change in <strong>the</strong> transition area with a time lag <strong>of</strong><br />

)6 days, (k) <strong>of</strong> <strong>the</strong> Danube River with a time lag <strong>of</strong><br />

)6 days and (l) turbidity change in <strong>the</strong> transition area<br />

with a time lag <strong>of</strong> )6 days] as independent variables<br />

was applied. Stepwise backward selection provided a<br />

highly significant relationship (R 2 = 0.51, P < 0.001)<br />

and revealed <strong>the</strong> following set <strong>of</strong> six significant<br />

variables for <strong>the</strong> model Y = a + b · X 1 + c ·<br />

X2 + d · X3 + e · X4 + f · X5 + g · X6 (a =6.518,<br />

b=)0.594, c=0.489, d=1.887, e=)0.042,<br />

f=0.030, g=)0.007), where X 1 = water temperature<br />

at <strong>the</strong> spawning site (P < 0.001), X 2 = water<br />

temperature change at <strong>the</strong> spawning site with a time<br />

lag <strong>of</strong> )5 days (P < 0.001), X 3 = water temperature<br />

change in <strong>the</strong> confluence with a time lag <strong>of</strong><br />

1 day (P = 0.009), X 4 = water level in <strong>the</strong> transition<br />

area (P = 0.003), X5 = water level <strong>of</strong> <strong>the</strong> Danube<br />

(P < 0.001) and X6 = water level change <strong>of</strong> <strong>the</strong><br />

Danube with a time lag <strong>of</strong> )6 days (P = 0.015).<br />

0 2 4 6 8 10 12 14 16<br />

0 2 4 6 8 10 12 14 16<br />

Based on this set <strong>of</strong> environmental variables, a good<br />

concordance between observed and expected migration<br />

rates was observed (Fig. 7).<br />

Discussion<br />

Spawning area<br />

(a)<br />

Confluence<br />

140 (b)<br />

120<br />

100<br />

80<br />

60<br />

40<br />

20<br />

0<br />

140<br />

120<br />

100<br />

80<br />

60<br />

40<br />

20<br />

0<br />

Environmental cues in nase spawning migration<br />

0 2 4 6 8 10 12 14 16<br />

Water temperature (°C)<br />

Danube river<br />

(c)<br />

–4 –3 –2 –1 0 1 2 3 4 5<br />

Fluctuations <strong>of</strong> water temperature, water level and<br />

turbidity over several days were <strong>the</strong> main environmental<br />

stimuli triggering <strong>the</strong> onset, maintenance and<br />

temporal pattern <strong>of</strong> <strong>the</strong> spawning migration <strong>of</strong> Danube<br />

nase in <strong>the</strong> Fischa River. The significant relationships<br />

between spawning migration <strong>of</strong> nase, an early spring<br />

spawner, and thresholds <strong>of</strong> water temperature and<br />

water level were determined by season, but did not<br />

explain fine-scaled temporal patterns <strong>of</strong> <strong>the</strong> spawning<br />

migration. The spawning runs were characterised by<br />

an undulating pattern with several migration peaks. In<br />

early spring in <strong>the</strong> Danube, a slight increase <strong>of</strong> water<br />

temperature, which coincides with a steep increase in<br />

water temperature at <strong>the</strong> Fischa spawning site, initiated<br />

massive upstream spawning migration <strong>of</strong> nase, irrespective<br />

<strong>of</strong> absolute temperature. By way <strong>of</strong> contrast<br />

(d)<br />

(e)<br />

–0.6–0.4–0.2 0.0 0.2 0.4 0.6 0.8<br />

(f)<br />

–1.5–1.0–0.5 0.0 0.5 0.1 1.5 2.0<br />

Change in water temperature (°C)<br />

509


Rakowitz et al.<br />

y –1<br />

N o.<br />

<strong>of</strong><br />

migrants<br />

( Ind<br />

da<br />

)<br />

140 2004<br />

120<br />

2005<br />

100<br />

80<br />

60<br />

40<br />

20<br />

0<br />

0 100 200 300 400<br />

0 100 200 300 400 500 600<br />

Water level (cm)<br />

Transition area<br />

rapidly falling water temperatures stopped <strong>the</strong> spawning<br />

migration. Our results showed that a slight but<br />

quick increase <strong>of</strong> water temperature in <strong>the</strong> confluence<br />

and in <strong>the</strong> Danube corresponded with remarkably<br />

warmer temperatures over a longer period at <strong>the</strong><br />

spawning site. Spawning migrations into tributaries<br />

may <strong>the</strong>refore not only ensure that eggs are deposited<br />

in areas with improved oxygen supply compared with<br />

<strong>the</strong> main river (Wootton 1990) and decrease <strong>the</strong><br />

predation pressure on <strong>the</strong> embryos (Kedney et al.<br />

1987), but also provide an optimal temperature for<br />

spawning and development at times when temperatures<br />

in <strong>the</strong> main river, i.e., <strong>the</strong> main residence habitat<br />

(Lucas & Baras 2001), are still below <strong>the</strong> optimum.<br />

The undulating pattern <strong>of</strong> <strong>the</strong> nase spawning migration<br />

was characterised by migration peaks coinciding with<br />

<strong>the</strong> 5th day <strong>of</strong> continuous temperature increase at <strong>the</strong><br />

spawning site, irrespective <strong>of</strong> an immediate or delayed<br />

temperature decrease afterwards. In <strong>the</strong> early 1990s,<br />

Keckeis (2001) documented mass occurrence <strong>of</strong> nase<br />

within <strong>the</strong> spawning area when water temperature in<br />

510<br />

(a)<br />

Danube river<br />

140 (b)<br />

120<br />

100<br />

80<br />

60<br />

40<br />

20<br />

0<br />

140<br />

120<br />

100<br />

80<br />

60<br />

40<br />

20<br />

0<br />

0 25 50 75 250<br />

Turbidity (ntu)<br />

Transition area<br />

(e)<br />

(c)<br />

–200 –100 0 100 200 300<br />

(d)<br />

–200 –100 0 100 200 300 400 500<br />

Change in water level (cm)<br />

–300 –200 –100 0 100 200 300<br />

Change in turbidity (ntu)<br />

(f)<br />

Fig. 6. Relationship between daily number<br />

<strong>of</strong> migrants and water level in <strong>the</strong> Fischa<br />

River at <strong>the</strong> transition area (a) (R 2 = 0.16,<br />

P < 0.01) and in <strong>the</strong> Danube River (b) (non.<br />

sig.). Relationship between daily number <strong>of</strong><br />

migrants and water level change at <strong>the</strong> transition<br />

area (c) (R 2 = 0.10, P < 0.05) and in<br />

<strong>the</strong> Danube (d) (R 2 = 0.17, P < 0.001).<br />

Relationship between daily number <strong>of</strong><br />

migrants and turbidity at <strong>the</strong> transition area<br />

(e) (R 2 = 0.13, P < 0.05) and between daily<br />

number <strong>of</strong> migrants and turbidity change<br />

(f) (R 2 = 0.07, P < 0.05). Pooled data <strong>of</strong><br />

two spawning seasons. For fur<strong>the</strong>r explanation<br />

see text.<br />

<strong>the</strong> Fischa River rose from 7.5 °C to 11.5 °C, 9.0 °C<br />

to 13.0 °C and 10.0 °C to 14.0 °C within approximately<br />

4–5 days. Penaz et al. (1984) reported a<br />

preferred <strong>the</strong>rmal optimum for spawning nase between<br />

8.0 and 12.0 °C. Hladik & Kubecka (2003) presented<br />

similar results in <strong>the</strong>ir study on fish migration in <strong>the</strong><br />

tributary zone <strong>of</strong> <strong>the</strong> Rimov Reservoir (CZ): a<br />

temperature increase most significantly affected cyprinid<br />

spawning runs. Fish activity dropped after 2 days<br />

from <strong>the</strong> maximal number to nearly zero due to cold<br />

wea<strong>the</strong>r (Hladik & Kubecka 2004). During investigation<br />

<strong>of</strong> upstream migration <strong>of</strong> cyprinids and percids in<br />

<strong>the</strong> Äijälänsalmi channel, Lilja et al. (2003) observed<br />

that a +4.5 °C increase <strong>of</strong> water temperature was<br />

associated with <strong>the</strong> largest catch per unit effort<br />

(CPUE). Rodriguez-Ruiz & Granado-Lorencio<br />

(1992) identified water temperature as <strong>the</strong> dominant<br />

environmental factor influencing upstream migration<br />

and migration speed <strong>of</strong> cyprinids in <strong>the</strong> Guadalete<br />

River in Spain. Steinmann et al. (1937) concluded<br />

from mark-recapture experiments in <strong>the</strong> Danube River,


Sqrt-lnd day –1<br />

Sqrt-lnd day –1<br />

12<br />

10<br />

8<br />

6<br />

4<br />

2<br />

0<br />

0 1 . 0 3 . 0 4<br />

0 8 . 0 3 . 0 4<br />

12<br />

10<br />

8<br />

6<br />

4<br />

2<br />

0<br />

0 1 . 0 3 . 0 5<br />

0 8 . 0 3 . 0 5<br />

Observed (a)<br />

Expected<br />

Upper and lower 95% CL<br />

1 5 . 0 3 . 0 4<br />

2 2 . 0 3 . 0 4<br />

2 9 . 0 3 . 0 4<br />

0 5 . 0 4 . 0 4<br />

1 2 . 0 4 . 0 4<br />

1 9 . 0 4 . 0 4<br />

2 6 . 0 4 . 0 4<br />

0 3 . 0 5 . 0 4<br />

1 0 . 0 5 . 0 4<br />

1 7 . 0 5 . 0 4<br />

2 4 . 0 5 . 0 4<br />

3 1 . 0 5 . 0 4<br />

<strong>the</strong> Rhine River and its tributaries in Austria, Germany<br />

and Switzerland from 1923 to 1934 that cyprinid<br />

migration was mainly influenced by increasing water<br />

temperature, with migration dropping considerably<br />

during cold spells. The triggering effect <strong>of</strong> changing<br />

<strong>the</strong>rmal conditions was also shown by Baril &<br />

Magnan (2002) for lacustrine brook charr Salvelinus<br />

fontinalis (Mitchill): migration peaked when water<br />

temperature suddenly fell from 11.0 to 7.0 °C.<br />

Although spawning migrations <strong>of</strong> cyprinids are stimulated<br />

by increasing water temperature, <strong>the</strong> opposite<br />

effect also exists. A drop in water temperature is a<br />

stimulus for certain salmonid species, although in all<br />

(b)<br />

1 5 . 0 3 . 0 5<br />

2 2 . 0 3 . 0 5<br />

2 9 . 0 3 . 0 5<br />

0 5 . 0 4 . 0 5<br />

1 2 . 0 4 . 0 5<br />

1 9 . 0 4 . 0 5<br />

2 6 . 0 4 . 0 5<br />

0 3 . 0 5 . 0 5<br />

1 0 . 0 5 . 0 5<br />

1 7 . 0 5 . 0 5<br />

2 4 . 0 5 . 0 5<br />

3 1 . 0 5 . 0 5<br />

Fig. 7. Observed and expected number <strong>of</strong> migrants per day by <strong>the</strong><br />

multiple regression analyses <strong>of</strong> six environmental abiotic variables<br />

(R 2 = 0.51, P < 0.001; Y = a + b · X 1 + c · X 2 + d · X 3 + e ·<br />

X 4 + f · X 5 + g · X 6; a=6.518, b=)0.594, c=0.489,<br />

d=1.887, e=)0.042, f=0.030, g=)0.007) for <strong>the</strong> two<br />

investigated spawning seasons. (a) 2004 and (b) 2005. X 1 = water<br />

temperature at <strong>the</strong> spawning site (P < 0.001), X 2 = water temperature<br />

change at <strong>the</strong> spawning site with a time lag <strong>of</strong> )5 days<br />

(P < 0.001), X 3 = water temperature change in <strong>the</strong> confluence with<br />

a time lag <strong>of</strong> 1 day (P =0.009), X4 = water level in <strong>the</strong> transition<br />

area (P =0.003), X 5 = water level <strong>of</strong> <strong>the</strong> Danube (P < 0.001) and<br />

X 6 = water level change <strong>of</strong> <strong>the</strong> Danube with a time lag <strong>of</strong> )6 days<br />

(P =0.015).<br />

Environmental cues in nase spawning migration<br />

cases <strong>the</strong> fluctuation within a certain period seems to<br />

be <strong>the</strong> prevailing factor.<br />

In our study, water level fluctuations significantly<br />

affected <strong>the</strong> spawning migration in Danube nase.<br />

A progressing decrease in water level over a 6-day<br />

period in <strong>the</strong> Danube prompted nase to migrate to <strong>the</strong>ir<br />

distinct spawning areas in <strong>the</strong> Fischa. Nase preferred<br />

flow conditions corresponding to <strong>the</strong> hydrological<br />

situation shortly after a flood peak; this seemed to<br />

guarantee an appropriate water level in <strong>the</strong> tributary in<br />

terms <strong>of</strong> reachability and spawning conditions. Several<br />

authors (Northcote 1984; Jonsson 1991) concluded<br />

that many temperate fish, especially salmonids, are<br />

stimulated to move upstream on <strong>the</strong>ir spawning<br />

migrations during or following periods <strong>of</strong> high flow.<br />

In Table 2, we integrated <strong>the</strong> available published<br />

literature on <strong>the</strong> relationship between water level and<br />

spawning migration. It gives evidence that in nine <strong>of</strong><br />

<strong>the</strong> overall 13 cases (69%), fish <strong>of</strong> five different<br />

families undertook <strong>the</strong>ir spawning migration after<br />

floods and ⁄ or during falling water level. In three<br />

studies (23%), fish moved during increasing water<br />

level and only in one study (8%) migration during<br />

high flow is mentioned.<br />

In line with <strong>the</strong>se reports, our detailed analyses<br />

based on daily water level fluctuations indicate that <strong>the</strong><br />

period after a flood peak <strong>of</strong>fers favourable conditions<br />

for upstream movement. One explanation is that <strong>the</strong><br />

water level is still high enough to overcome obstacles<br />

and to reach appropriate spawning habitats, and it<br />

might be less energy-demanding to get <strong>the</strong>re.<br />

Hence, appropriate timing <strong>of</strong> <strong>the</strong> spawning migration,<br />

with respect to day-to-day fluctuations in flow<br />

and turbidity, may reduce predation risk, enhance<br />

access to spawning grounds in suitable conditions and<br />

minimize energy wastage for spawning in areas <strong>of</strong><br />

strong water current over several weeks (Keckeis<br />

2001). During <strong>the</strong> spawning season, turbidity fluctuations,<br />

which were closely correlated with <strong>the</strong> water<br />

level, showed a weaker relationship with <strong>the</strong> spawning<br />

migration <strong>of</strong> nase than <strong>the</strong> o<strong>the</strong>r investigated variables.<br />

Correlation between decreasing turbidity within<br />

6 days and increasing passage rates was mainly caused<br />

by <strong>the</strong> synchronous decreasing water level. Nase<br />

seemed to prefer nei<strong>the</strong>r very highly turbid conditions,<br />

as usual during flood events, nor very clear water<br />

conditions for <strong>the</strong>ir spawning migration; perhaps <strong>the</strong>y<br />

take advantage <strong>of</strong> slightly increased turbidity, which<br />

decreases predation pressure, such as from herons<br />

Ardea cinerea (Linnaeus).<br />

Environmental unpredictability stimulates fish to<br />

trade <strong>of</strong>f reproduction against refuge, feeding and<br />

growth. Reliance on combined stimuli is apparently a<br />

more efficient strategy than responding to a single<br />

cue. Such single cues can occur on several occasions<br />

outside <strong>the</strong> breeding season, potentially reducing <strong>the</strong><br />

511


Rakowitz et al.<br />

Table 2. Summary <strong>of</strong> published literature on <strong>the</strong> relationship between water level and spawning migration <strong>of</strong> different fish species in nor<strong>the</strong>rn temperate regions.<br />

Species Taxa Family<br />

fishes’ fitness (Ovidio et al. 1998). Our results<br />

demonstrate that an optimal stimulus for <strong>the</strong> spawning<br />

migration in nase is a combination <strong>of</strong> variables,<br />

namely a slight increase <strong>of</strong> water temperature in <strong>the</strong><br />

main habitat (<strong>the</strong> Danube River) synchronized with a<br />

steep increase <strong>of</strong> water temperature at <strong>the</strong> spawning<br />

site in <strong>the</strong> tributary, in combination with a falling<br />

water level after flood peaks. During spawning<br />

season <strong>the</strong> warmer water <strong>of</strong> <strong>the</strong> tributary causes a<br />

small, constant temperature difference between <strong>the</strong><br />

confluence and <strong>the</strong> Danube, and might also act as a<br />

stimulus to maintain spawning migration into <strong>the</strong><br />

tributary. Our results <strong>the</strong>refore also demonstrate that<br />

fluctuations <strong>of</strong> <strong>the</strong> environmental conditions within<br />

<strong>the</strong> main river itself represent relevant cues which<br />

initiate and control <strong>the</strong> temporal pattern as well as <strong>the</strong><br />

intensity <strong>of</strong> nase spawning migrations. For nase, <strong>the</strong><br />

predictability <strong>of</strong> favourable spawning conditions at<br />

<strong>the</strong> spawning place is high when a combination <strong>of</strong><br />

increasing water temperatures and decreasing water<br />

levels (with certain time lags) acts as a stimulus for<br />

<strong>the</strong> right timing. During periods <strong>of</strong> high environmental<br />

unpredictability, as occur in spring in nor<strong>the</strong>rn<br />

and temperate latitudes, fish have to take advantage<br />

<strong>of</strong> short periods <strong>of</strong> high environmental predictability,<br />

especially in terms <strong>of</strong> spawning migrations. This no<br />

doubt enables <strong>the</strong>m to achieve high reproductive<br />

success and to increase <strong>the</strong>ir fitness.<br />

To grasp <strong>the</strong> functional role <strong>of</strong> <strong>the</strong> environmental<br />

factors, discrete thresholds have impact on <strong>the</strong> migratory<br />

behaviour on a seasonal scale, but fail to explain<br />

short-term initiation <strong>of</strong> migration and migration patterns<br />

within spawning seasons. Our results revealed <strong>the</strong><br />

important functional role <strong>of</strong> <strong>the</strong> fluctuations <strong>of</strong> variables<br />

within <strong>the</strong> season. They serve as a time- (in reference<br />

to season) and threshold-independent environmental<br />

stimulus for migratory behaviour in general and for <strong>the</strong><br />

spawning migration <strong>of</strong> nase specifically.<br />

Acknowledgements<br />

This project was funded by <strong>the</strong> European Fund for Regional<br />

Development (EFRE), <strong>the</strong> City <strong>of</strong> Vienna (MA 45), <strong>the</strong><br />

Austrian Ministry <strong>of</strong> <strong>Sciences</strong> and <strong>the</strong> Regional Government<br />

<strong>of</strong> Lower Austria within <strong>the</strong> Interreg III A project FIDON. We<br />

thank our colleagues for <strong>the</strong>ir valuable assistance in <strong>the</strong> field.<br />

Special thanks to Helge Balk (University <strong>of</strong> Oslo) for<br />

developing <strong>the</strong> s<strong>of</strong>tware tool to analyse hydroacoustically<br />

recorded fish in rivers. Special thanks also to Franz Langenberger<br />

for his enthusiasm and support concerning our scientific<br />

work. We acknowledge <strong>the</strong> Municipality <strong>of</strong> Fischamend for <strong>the</strong><br />

logistic support during <strong>the</strong> fieldwork. The data <strong>of</strong> <strong>the</strong> water<br />

level and <strong>the</strong> water temperature <strong>of</strong> <strong>the</strong> Danube Rive were<br />

provided by <strong>the</strong> Austrian river authority (Via-Donau). We are<br />

grateful to Michael Stachowitsch (University <strong>of</strong> Vienna) for his<br />

comments on <strong>the</strong> draft manuscript.<br />

References<br />

Water level<br />

Increasing Peak Decreasing<br />

Razorback sucker 1,2<br />

Xyrauchen texanus (Abbott) Catostomidae X<br />

American shad 3<br />

Alosa sapidissima (Wilson) Clupeidae X<br />

Allis shad 4<br />

Alosa alosa (Linnaeus) Clupeidae X<br />

Whitefish 5<br />

Coregonus lavaretus (Linnaeus) Coregonidae X<br />

Colorado pikeminnow 6<br />

Ptychocheilus lucius (Girard) Cyprinidae X<br />

White bream 7<br />

Blicca bjoerkna (Linnaeus) Cyprinidae X<br />

Roach 8<br />

Rutilus rutilus (Linnaeus) Cyprinidae X<br />

Salmonids 9,10,11,12<br />

Salmonidae Salmonidae X X<br />

Brown trout 13<br />

Salmo trutta Linnaeus Salmonidae X<br />

Atlantic salmon 14<br />

Salmo salar Linnaeus Salmonidae X<br />

Atlantic salmon 15<br />

Salmo salar Linnaeus Salmonidae X<br />

Atlantic salmon 16<br />

Salmo salar Linnaeus Salmonidae X X<br />

Sum 5 3 1 9<br />

Percent 23 8 69<br />

1 Tyus & Karp (1990); 2 Modde & Irving (1998); 3 Leggett & Whitney (1972); 4 Bellariva & Belaud (1998); 5 Laineetal. (1998); 6 Tyus (1990); 7 Molls (1999); 8 Vollestad<br />

& L’Abee-Lund (1987); 9 Banks (1969); 10 Northcote (1984); 11 Jonsson (1991); 12 Jensen etal. (1998); 13 Svendsen etal. (2004); 14 Lilja & Romakkaniemi (2003);<br />

15 Trepanier et al. (1996); 16 Erkinaro etal. (1999).<br />

512<br />

Balk, H. 2001. Development <strong>of</strong> hydroacoustic methods for fish<br />

detection in shallow water. Ph.D. <strong>the</strong>sis. Oslo, Norway:<br />

University <strong>of</strong> Oslo. 348 pp.<br />

Balk, H. & Lindem, T. 2000. Improved fish detection in data<br />

from split beam transducers. Aquatic Living Resources 13:<br />

297–303.<br />

Balk, H. & Lindem, T. 2002. A new method for single target<br />

detection in sonar data characterised with low signal-to-noise<br />

ratio. Proceedings <strong>of</strong> 6th ICES Symposium on Acoustics<br />

in Fisheries and Aquatic Ecology. Montpellier, France. 135 pp.<br />

Banks, J.W. 1969. A review on <strong>the</strong> literature on <strong>the</strong> upstream<br />

migration <strong>of</strong> salmonids. Journal <strong>of</strong> Fish <strong>Biology</strong> 1: 85–136.<br />

Baras, E. 1995. Seasonal activity <strong>of</strong> Barbus barbus: effect <strong>of</strong><br />

temperature on time-budgeting. Journal <strong>of</strong> Fish <strong>Biology</strong> 46:<br />

806–818.


Baras, E. & Philippart, J.C. 1999. Adaptive and evolutionary<br />

significance <strong>of</strong> a reproductive <strong>the</strong>rmal threshold in Barbus<br />

barbus. Journal <strong>of</strong> Fish <strong>Biology</strong> 55: 354–375.<br />

Baril, M. & Magnan, P. 2002. Seasonal timing and diel activity<br />

<strong>of</strong> lacustrine brook charr, Salvelinus fontinalis, spawning in a<br />

lake outlet. Environmental <strong>Biology</strong> <strong>of</strong> Fishes 64: 175–181.<br />

Bellariva, J.L. & Belaud, A. 1998. Environmental factors<br />

influencing <strong>the</strong> passage <strong>of</strong> allis shad Alosa alosa at <strong>the</strong><br />

Golfech fish lift on <strong>the</strong> Garonne River, France. In: Jungwirth,<br />

M., Schmutz, S. & Weiss, S., eds. Fish migration and fish<br />

bypasses. Oxford: Fishing News Books, Blackwell Science,<br />

pp. 171–179.<br />

Clough, S., Garner, P., Deans, D. & Ladle, M. 1998.<br />

Postspawning movements and habitat selection <strong>of</strong> dace in<br />

<strong>the</strong> River Frome, Dorset, sou<strong>the</strong>rn England. Journal <strong>of</strong> Fish<br />

<strong>Biology</strong> 53: 1060–1070.<br />

Erkinaro, J., Okland, F., Moen, K., Niemelä, E. & Rahiala, M.<br />

1999. Return migration <strong>of</strong> Atlantic Salmon in <strong>the</strong> River Tana:<br />

<strong>the</strong> role <strong>of</strong> environmental factors. Journal <strong>of</strong> Fish <strong>Biology</strong> 55:<br />

506–516.<br />

Frouzova, J., Kubecka, J., Balk, H. & Frouz, J. 2005. Target<br />

strength <strong>of</strong> some European fish species and its dependence on<br />

fish body parameters. Fisheries Research 75: 86–96.<br />

Hladik, M. & Kubecka, J. 2003. Fish migration between a<br />

temperate reservoir and its main tributary. Hydrobiologia<br />

504: 251–266.<br />

Hladik, M. & Kubecka, J. 2004. The effect <strong>of</strong> water level<br />

fluctuation on tributary spawning migration <strong>of</strong> reservoir fish.<br />

Ecohydrology & Hydrobiology 4: 449–457.<br />

Huber, M. & Kirchh<strong>of</strong>er, A. 1998. Radio telemetry as a tool to<br />

study habitat use <strong>of</strong> nase (<strong>Chondrostoma</strong> <strong>nasus</strong> L.) in<br />

medium-sized rivers. Hydrobiologia 371 ⁄ 372: 309–319.<br />

Jensen, A., Hvidsten, A.N. & Johnsen, B.O. 1998. Effects <strong>of</strong><br />

temperature and flow on <strong>the</strong> upstream migration <strong>of</strong> adult<br />

Atlantic salmon in two Norwegian Rivers. In: Jungwirth, M.,<br />

Schmutz, S. & Weiss, S., eds. Fish migration and fish<br />

bypasses. Oxford: Fishing News Books, Blackwell Science,<br />

pp. 45–54.<br />

Jonsson, N. 1991. Influence <strong>of</strong> water flow, water temperature<br />

and light on fish migration in rivers. Nordic Journal <strong>of</strong><br />

Freshwater Research 66: 20–35.<br />

Keckeis, H. 1999. Fortpflanzungsbiologie und Ökologische<br />

Kennzeichnung der Laichgebiete von Flussfischen in Fließgewässern<br />

verschiedener Größenordnung. In: Keckeis, H.,<br />

ed. Endbericht Österreichische-Polnische-Slowenische-<br />

Tschechische Forschungskooperation GZ 45.189 ⁄ 2-27b ⁄ 91.<br />

Vienna: Austrian Ministry <strong>of</strong> <strong>Sciences</strong>. 97 pp.<br />

Keckeis, H. 2001. Influence <strong>of</strong> river morphology and current<br />

velocity conditions on <strong>the</strong> spawning site selection <strong>of</strong><br />

<strong>Chondrostoma</strong> <strong>nasus</strong> (L.). Archiv für Hydrobiolologie Supplement<br />

135: 341–356.<br />

Keckeis, H. & Rakowitz, G. 2005. Fish migration in <strong>the</strong> free<br />

flowing section <strong>of</strong> River Danube east <strong>of</strong> Vienna. The<br />

functional role <strong>of</strong> tributaries and sidearms. In: Goldschmid,<br />

U., ed. Annual report <strong>of</strong> FIDON. Vienna: MA45-Wasserbau,<br />

pp. 27–47.<br />

Keckeis, H. & Rakowitz, G. 2006. Fish migration in <strong>the</strong> free<br />

flowing section <strong>of</strong> River Danube east <strong>of</strong> Vienna. The<br />

functional role <strong>of</strong> tributaries and sidearms. In: Goldschmid,<br />

U., ed. Annual report <strong>of</strong> FIDON. Vienna: MA45-Wasserbau,<br />

pp. 40–70.<br />

Environmental cues in nase spawning migration<br />

Kedney, G.I., Boule, V. & FitzGerald, G.J. 1987.<br />

The reproductive ecology <strong>of</strong> threespine sticklebacks breeding<br />

in fresh and brackish water. American Fisheries Society<br />

Symposium 1: 151–161.<br />

Kresser, W. 1989. Hydraulik Offener Gerinne. In: Kresser, W.,<br />

ed. Symposium Hydraulik Offener Gerinne. Vienna: Wiener<br />

Mitteilungen, 79. Österreichischer Wasserwirtschaftsverband.<br />

375 pp.<br />

Laine, A., Ylinärä, T., Heikkilä, J. & Hooli, J. 1998. Behaviour<br />

<strong>of</strong> upstream migrating whitefish Coregonus lavaretus in <strong>the</strong><br />

Kukkolankoski rapids, Nor<strong>the</strong>rn Finland. In: Jungwirth, M.,<br />

Schmutz, S. & Weiss, S., eds. Fish migration and fish<br />

bypasses. Oxford: Fishing News Books, Blackwell Science,<br />

pp. 33–44.<br />

Leggett, W.C. & Whitney, R.R. 1972. Water temperature<br />

and <strong>the</strong> migration <strong>of</strong> American shad. Fishery Bulletin 70:<br />

659–670.<br />

Lelek, A. 1987. The freshwater fishes <strong>of</strong> Europe, Vol. 9.<br />

Threatened fishes <strong>of</strong> Europe. Wiesbaden: Aula-Verlag.<br />

Lilja, J. & Romakkaniemi, A. 2003. Early-season river entry <strong>of</strong><br />

adult Atlantic salmon: its dependency on environmental<br />

factors. Journal <strong>of</strong> Fish <strong>Biology</strong> 62: 41–50.<br />

Lilja, J., Keskinen, T., Marjomäki, T.J., Valkeajärvi, P. &<br />

Karjalainen, J. 2003. Upstream migration <strong>of</strong> cyprinids and<br />

percids in a channel, monitored by horizontal split-beam<br />

echosounder. Aquatic Living Resources 16: 185–190.<br />

Lucas, M. 2000. The influence <strong>of</strong> environmental factors on <strong>the</strong><br />

movements <strong>of</strong> lowland-river fish in <strong>the</strong> Yorkshire Ouse<br />

system. The Science <strong>of</strong> <strong>the</strong> Total Environment 251 ⁄ 252:<br />

223–232.<br />

Lucas, M.C. & Baras, E. 2001. Migration <strong>of</strong> freshwater fishes.<br />

Oxford: Blackwell Science.<br />

Lusk, S., Jurajda, P. & Penaz, M. 1995. Age structure in<br />

spawning shoals <strong>of</strong> <strong>Chondrostoma</strong> <strong>nasus</strong>. Folia Zoologica<br />

Supplement 44: 25–34.<br />

Mader, H., Steidl, Th. & Wimmer, R. 1996. Abflussregime<br />

österreichischer Fliessgewässer. In: Umweltbundesamt, ed.<br />

Monographien, Vol. 82. Vienna: Umweltbundesamt. 192 pp.<br />

Mills, C.A. 1991. Reproduction and life history. In: Winfield,<br />

I.J. & Nelson, J.S., eds. Cyprinid fishes. Systematics,<br />

biology and exploitation. London: Chapman & Hall, pp.<br />

483–508.<br />

Modde, T. & Irving, D.B. 1998. Use <strong>of</strong> multiple spawning sites<br />

and seasonal movement by razorback suckers in <strong>the</strong> Middle<br />

Green River, Utah. North American Journal <strong>of</strong> Fisheries<br />

Management 18: 318–326.<br />

Molls, F. 1999. New insights into <strong>the</strong> migration and habitat use<br />

by bream and white bream in <strong>the</strong> floodplain <strong>of</strong> <strong>the</strong> River<br />

Rhine. Journal <strong>of</strong> Fish <strong>Biology</strong> 55: 1187–1200.<br />

Mühlbauer, M., Traxler, E., Zitek, A. & Schmutz, S. 2003. Das<br />

dynamische Fischwehr – Ein hochwassersicheres Fischwehr<br />

zur Untersuchung der Fischwanderung an kleinen bis mittelgrossen<br />

Flüssen. Österreichs Fischerei 56: 136–148.<br />

Munro, A.D., Scott, A.P. & Lam, T.J. 1990. Reproductive<br />

seasonality in teleosts: environmental influences. Boca<br />

Raton, FL: CRC Press Inc., pp. 264.<br />

Northcote, T.G. 1969. Patterns and mechanisms in <strong>the</strong> lakeward<br />

migratory behaviour <strong>of</strong> juvenile trout. In: Northcote, T.G., ed.<br />

Symposium on Salmon and Trout in Streams. Vancouver,<br />

Canada: H.R. MacMillan Lectures in Fisheries, The University<br />

<strong>of</strong> British Columbia, pp. 183–203.<br />

513


Rakowitz et al.<br />

Northcote, T.G. 1984. Mechanisms <strong>of</strong> fish migration in rivers.<br />

In: McCleave, J.D., Arnold, G.P., Dodson, J.J. & Neill, W.H.,<br />

eds. Mechanisms <strong>of</strong> migration in fishes. New York: Plenum<br />

Press, pp. 317–355.<br />

Northcote, T.G. 1998. Migratory behaviour <strong>of</strong> fish and its<br />

significance to Movement through Riverine fish passage<br />

facilities. In: Jungwirth, M., Schmutz, S. & Weiss, S., eds.<br />

Fish migration and fish bypasses. Oxford: Fishing News<br />

Books, Blackwell Science, pp. 3–18.<br />

Ovidio, M., Baras, E., G<strong>of</strong>faux, D., Birtles, C. & Philippart,<br />

J.C. 1998. Environmental unpredictability rules <strong>the</strong> autumn<br />

migration <strong>of</strong> brown trout (Salmo trutta L.) in <strong>the</strong> Belgian<br />

Ardennes. Hydrobiologia 371 ⁄ 372: 263–274.<br />

Penaz, M. 1996. <strong>Chondrostoma</strong> <strong>nasus</strong>: its reproductive strategy<br />

and possible reason for widely observed population decline.<br />

Review. In: Kirchh<strong>of</strong>er, A. & Hefti, D., eds. Conservation <strong>of</strong><br />

endangered freshwater fish in Europe. Basel, Switzerland:<br />

Birkhäuser Verlag, pp. 279–285.<br />

Penaz, M., Lusk, S., Szabo, S., Lusha, R., Smirnova, E.N.,<br />

Storozhenko, S.S. & Tchepurnova, L.V. 1984. Spawning. In:<br />

Volskis, R.S., ed. The undermouth. Vilnius, Lithuania:<br />

Mokslas Publisher, pp. 34–51 (in Russian).<br />

Povz, M. 1988. Migration <strong>of</strong> nase carps (<strong>Chondrostoma</strong> <strong>nasus</strong><br />

L. 1758) in <strong>the</strong> River Sava. Journal <strong>of</strong> Aquatic Production 2:<br />

149–163.<br />

Rakowitz, G. & Kubecka, J. 2006. Sound propagation in a<br />

horizontal bottom aligned beam. In: Jesus, S.M. & Rodriguez,<br />

O.C., eds. Proceedings <strong>of</strong> <strong>the</strong> Eighth European<br />

Conference on Underwater Acoustics, 8th ECUA. Faro,<br />

Portugal: CINTAL, University <strong>of</strong> Algarve, pp. 249–254.<br />

Raymond, H.L. 1979. Effects <strong>of</strong> dams and impoundments on<br />

migration <strong>of</strong> juvenile Chinook salmon and steelhead from <strong>the</strong><br />

Snake River, 1966 to 1975. Transactions <strong>of</strong> <strong>the</strong> American<br />

Fisheries Society 108: 505–529.<br />

Rodriguez-Ruiz, A. & Granado-Lorencio, C. 1992. Spawning<br />

period and migration <strong>of</strong> 3 species <strong>of</strong> cyprinids in a stream<br />

with Mediterranean regimen (SW Spain). Journal <strong>of</strong> Fish<br />

<strong>Biology</strong> 41: 545–556.<br />

Romakkaniemi, A., Lilja, J., Nykänen, M., Marjomäki, T.J. &<br />

Jurvelius, J. 2000. Spawning run <strong>of</strong> Atlantic salmon (Salmo<br />

salar) in <strong>the</strong> River Tornionjoki monitored by horizontal splitbeam<br />

echosounding. Aquatic Living Resources 13: 349–354.<br />

Schiemer, F. & Waidbacher, H. 1992. Strategies for conservation<br />

<strong>of</strong> a Danubian fish fauna. In: Boon, P.J., Calow, P. &<br />

Petts, G.E., eds. River conservation and management.<br />

Chichester: John Wiley & Sons Ltd, pp. 363–382.<br />

Schiemer, F., Guti, G., Keckeis, H. & Staras, M. 2004.<br />

Ecological status and problems <strong>of</strong> <strong>the</strong> Danube River and its<br />

fish Fauna: a review. In: Welcomme, R.L. & Petr, T., eds.<br />

Proceedings <strong>of</strong> <strong>the</strong> Second International Symposium on <strong>the</strong><br />

Management <strong>of</strong> Large Rivers for Fisheries, Vol. I. Bangkok,<br />

Thailand: FAO Regional Office for Asia and <strong>the</strong> Pacific, pp.<br />

273–300.<br />

Sjöberg, K. 1977. Locomotor activity <strong>of</strong> river lamprey,<br />

Lampetra fluviatilis L., during spawning season. Hydrobiologia<br />

55: 265–270.<br />

514<br />

Slavik, O. 1996. The migration <strong>of</strong> fish in <strong>the</strong> Elbe River below<br />

Strekow. Ziva 4: 179–180.<br />

Soule, M., Barange, M., Solli, H. & Hampton, I. 1997.<br />

Performance <strong>of</strong> a new phase algorithm for discriminating<br />

between single and overlapping echoes in a split-beam echo<br />

sounder. ICES Journal <strong>of</strong> Marine Science 54: 934–938.<br />

Steinmann, P., Koch, W. & Scheuring, L. 1937. Die Wanderung<br />

der Süßwasserfische dargestellt auf Grund von Markierungsversuchen.<br />

Zeitschrift für Fischerei 35: 369–467.<br />

Svendsen, J.C., Koed, A. & Aarestrup, K. 2004. Factors<br />

influencing <strong>the</strong> spawning migration <strong>of</strong> female anadromous<br />

brown trout. Journal <strong>of</strong> Fish <strong>Biology</strong> 64: 528–540.<br />

Trepanier, S., Rodriguez, M.A. & Magnan, P. 1996. Spawning<br />

migrations in landlocked Atlantic salmon: time series modelling<br />

<strong>of</strong> river discharge and water temperature effects.<br />

Journal <strong>of</strong> Fish <strong>Biology</strong> 48: 925–936.<br />

Tyus, H.M. 1990. Potamodromy and reproduction <strong>of</strong> Colorado<br />

squawfish in <strong>the</strong> Green River Basin, Colorado and Utah.<br />

Transactions <strong>of</strong> <strong>the</strong> American Fisheries Society 119: 1035–<br />

1047.<br />

Tyus, H.M. & Karp, A.C. 1990. Spawning and movements<br />

<strong>of</strong> razorback sucker Xyrauchen texanus in <strong>the</strong> Green River<br />

basin <strong>of</strong> Colorado and Utah. Southwestern Naturalist 35:<br />

427–433.<br />

Vlaming, R.J. de 1974. Environmental and endocrine control <strong>of</strong><br />

teleost reproduction. In: Schreck, C.B., ed. Control <strong>of</strong> sex in<br />

fishes. Blacksburg: Virginia Polytechnic and State University<br />

Press, pp. 18–83.<br />

Vollestad, L.A. & L’Abee-Lund, J.H. 1987. Reproductive<br />

biology <strong>of</strong> stream spawning roach, Rutilus rutilus. Environmental<br />

<strong>Biology</strong> <strong>of</strong> Fishes 18: 219–227.<br />

Ward, J.V. & Tockner, K. 2001. Biodiversity: towards a unifying<br />

<strong>the</strong>me for river ecology. Freshwater <strong>Biology</strong> 46: 807–819.<br />

Welcomme, R.L. 1994. The status <strong>of</strong> large river habitats. In:<br />

Cowx, I.G., ed. Rehabilitation <strong>of</strong> freshwater fisheries.<br />

Oxford: Blackwell Scientific Publications, pp. 11–20.<br />

Westin, L. 1977. Temperature as orientation cue in migrating<br />

silver eels, Anguilla anguilla (L.). In: University <strong>of</strong><br />

Stockholm, ed. Contribution 17, Asko, Sweden: Asko<br />

Laboratory.<br />

Winfield, I.J. & Townsend, C.R. 1991. The role <strong>of</strong> cyprinids in<br />

ecosystems. In: Winfield, I.J. & Nelson, J.S., eds. Cyprinid<br />

fishes: systematics, biology and exploitation. London: Chapman<br />

& Hall, pp. 552–571.<br />

Winter, H.V. & Fredrich, F. 2003. Migratory behaviour <strong>of</strong> ide: a<br />

comparison between <strong>the</strong> lowland rivers Elbe, Germany,<br />

and Vecht, The Ne<strong>the</strong>rlands. Journal <strong>of</strong> Fish <strong>Biology</strong> 63:<br />

871–880.<br />

Wootton, R.J. 1990. Ecology <strong>of</strong> teleost fishes. London:<br />

Chapman & Hall.<br />

Zbinden, S. & Maier, K.J. 1996. Contribution to <strong>the</strong> knowledge<br />

<strong>of</strong> <strong>the</strong> distribution and spawning grounds <strong>of</strong> <strong>Chondrostoma</strong><br />

<strong>nasus</strong> and <strong>Chondrostoma</strong> toxostoma (Pisces: Cyprinidae)<br />

in Switzerland. In: Kirchh<strong>of</strong>er, A. & Hefti, D., eds.<br />

Conservation <strong>of</strong> endangered freshwater fish in europe. Basel,<br />

Switzerland: Birkhäuser Verlag, pp. 287–297.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!