The longitudinal pattern of crustacean (Peracarida, Malacostraca) assemblages in a large south European river: bank reinforcement structures as stepping stones of invasion

Abstract

The spread of alien crustaceans has significantly contributed to the homogenization of macroinvertebrate fauna of European freshwaters. However, little is known about alien Peracarida crustaceans of the Sava River, which represents the most important corridor for the spread of invasive species into Dinaric rivers with highly endemic fauna. In this study, we investigated Peracarida (Amphipoda, Isopoda and Mysida) collected during three years (2011, 2012 and 2015) from a total of 61 sites along the entire course of the Sava River. Besides describing the longitudinal pattern of Peracarida assemblages, we tested the hypothesis that bank reinforcement structures facilitate peracarid invasions by comparing densities and assemblages on natural and artificial substrate at 15 sites. In a total, 14 peracarid
crustacean species (5 native and 9 alien) were recorded. The Upper third of Sava was inhabited by native peracarids only, while the Middle and Lower Sava were dominated by alien species. The invasive amphipods Dikerogammarus haemobaphes, Chelicorophium sowinskyi and Chelicorophium curvispinum, and invasive isopod, Jaera istri, were the most abundant species along the middle course. Densities of alien peracarids in the Middle Sava were the highest and their share in macroinvertebrate assemblages was very variable, while the Lower Sava had the highest number of alien species in low densities. The densities of alien amphipods and isopods were in most cases significantly higher on bank reinforcement structures than
on natural substrate. Therefore, artificial stony substrates act as stepping stones of invasion for alien peracarids and largely contribute to their success in large lowland rivers.

Full text

RESEARCH ARTICLE
The longitudinal pattern of crustacean (Peracarida, Malacostraca)
assemblages in a large south European river: bank reinforcement
structures as stepping stones of invasion
Kre
simir Žganec
1,*
, Renata Ćuk
2
, Jelena Tomović
4
, Jasna Lajtner
3
, Sanja Gottstein
3
, Simona
Kovačević
5
, Sandra Hudina
3
, Andreja Lucić
3
, Martina Mirt
6
, Vladica Simić
5
, Tatjana Simčič
7
and
Momir Paunović
4
1
University of Zadar, Department of Teacher Education Studies in Gospić, 53000 Gospić, Croatia
2
Hrvatske Vode, Central Water Management Laboratory, Ulica grada Vukovara 220, 10000 Zagreb, Croatia
3
University of Zagreb, Faculty of Science, Department of Biology, Rooseveltov trg 6, 10000 Zagreb, Croatia
4
Institute for Biological Research “Sini
sa Stanković”, University of Belgrade, Despota Stefana 142 Blvd, 11000 Belgrade, Serbia
5
Institute of Biology and Ecology, Faculty of Science, University of Kragujevac, 12 Radoja Domanovica str, 34000 Kragujevac, Serbia
6
University Clinical Centre Ljubljana, Zalo
ska 7, 1000 Ljubljana, Slovenia
7
National Institute of Biology, Večna pot111, 1000 Ljubljana, Slovenia
Received: 23 August 2017; Accepted: 26 February 2018
Abstract –The spread of alien crustaceans has significantly contributed to the homogenization of
macroinvertebrate fauna of European freshwaters. However, little is known about alien Peracarida
crustaceans of the Sava River, which represents the most important corridor for the spread of invasive
species into Dinaric rivers with highly endemic fauna. In this study, we investigated Peracarida
(Amphipoda, Isopoda and Mysida) collected during three years (2011, 2012 and 2015) from a total of 61
sites along the entire course of the Sava River. Besides describing the longitudinal pattern of Peracarida
assemblages, we tested the hypothesis that bank reinforcement structures facilitate peracarid invasions by
comparing densities and assemblages on natural and artificial substrate at 15 sites. In a total, 14 peracarid
crustacean species (5 native and 9 alien) were recorded. The Upper third of Sava was inhabited by native
peracarids only, while the Middle and Lower Sava were dominated by alien species. The invasive
amphipods Dikerogammarus haemobaphes,Chelicorophium sowinskyi and Chelicorophium curvispinum,
and invasive isopod, Jaera istri, were the most abundant species along the middle course. Densities of alien
peracarids in the Middle Sava were the highest and their share in macroinvertebrate assemblages was very
variable, while the Lower Sava had the highest number of alien species in low densities. The densities of
alien amphipods and isopods were in most cases significantly higher on bank reinforcement structures than
on natural substrate. Therefore, artificial stony substrates act as stepping stones of invasion for alien
peracarids and largely contribute to their success in large lowland rivers.
Keywords: invasive Amphipoda / Isopoda / Mysida / micro-distribution / Sava
1 Introduction
Large European rivers have been heavily affected by
combinations of anthropogenic factors, among which land-use
change, pollution, damming, hydromorphological changes and
biological invasions have had the greatest impact (Whitton,
1984;Petts et al., 1993;Tockner et al., 2009;Strayer et al.,
2014). Alien species that are usually better adapted to
disturbance (e.g. Den Hartog et al., 1992;Karatayev et al.,
2009) have replaced the native fauna of large European rivers
and spread through extensive trans-European inland channel
networks used for transportation (Bij de Vaate et al., 2002;
Leuven et al., 2009). In addition, freshwater introductions are
continuously on the rise, both in number of species and
pathways (Nunes et al., 2015). The end result of these
processes is a high level of biocontamination of most large
European rivers (Arbačiauskas et al., 2008). The number of
alien and invasive species in European waters is continually
increasing, with about 300 exotic freshwater invertebrates and
more than 130 fish recorded in pan-European lake and river
ecosystems (European Environment Agency, 2012).
*Corresponding author: kzganec@unizd.hr
Ann. Limnol. - Int. J. Lim. 2018, 54, No
©EDP Sciences, 2018
https://doi.org/10.1051/limn/2018008
Available online at:
www.limnology-journal.org

In freshwaters, crustaceans are among the most successful
invaders (Holdich and Pöckl, 2007). The biomass, density and
lifespan of some crustacean species, as well as their functional
roles in ecosystem processes through feeding, bioturbation and
burrowing (e.g. Lodge, 1993;Statzner et al., 2003;Bernauer
and Jansen, 2006;Van Riel et al., 2006) make them a key
component of freshwater ecosystems. Holdich and Pöckl
(2007) listed 54 crustaceans, mainly from the superorder
Peracarida (Amphipoda, Cumacea, Isopoda, Mysida) and
order Decapoda, as invasive alien species in Europe. These
groups include some of the most notorious freshwater
invaders, such as the amphipods Dikerogammarus villosus
(Sowinsky, 1894) and Chelicorophium curvispinum (G.O.
Sars, 1895), crayfish Pacifastacus leniusculus (Dana, 1852)
and Procambarus clarkii (Girard, 1852), and mysids
Hemimysis anomala (G.O. Sars, 1907) and Limnomysis
benedeni Czerniavsky, 1882.
Invasive peracarid crustaceans can exhibit substantial
negative impacts on native crustaceans (Van den Brink et al.,
1993;Jazdzewski et al., 2002;Grabowski et al., 2007) and
other macroinvertebrates (Dick and Platvoet, 2000;Bernauer
and Jansen, 2006;Boets et al., 2010;Gergs and Rothhaupt,
2015). Moreover, alien peracarids have shown a greater
tolerance to severe environmental conditions, especially
elevated salinity (Devin and Beisel, 2006;Statzner et al.,
2008;Grabowski et al., 2009). Finally, invasive peracarids
exhibit relatively high dispersal rates in European rivers,
ranging from 14–461 km year
1
for amphipods, and 31–
185 km year
1
for the isopod Jaera istri Veuille, 1979 (Josens
et al., 2005;Leuven et al., 2009). Due to their high abundance,
pronounced negative impacts and relatively high dispersal
rates, alien and invasive peracarids and decapods have largely
contributed to the homogenization of macroinvertebrate
assemblages of European freshwater systems.
Despite their immense impacts, surprisingly few studies
report on the relative abundance of Peracarida in macro-
invertebrate assemblages (e.g. Van den Brink et al., 1993;Van
Riel et al., 2006), and studies examining variations of their
impact on macroinvertebrate composition and structures in
large rivers are scarce (e.g. Hellmann et al., 2017).
Furthermore, in most studies of alien Peracarida that have
been conducted thus far, the substrate types were either not
sampled separately or only artificial stony microhabitats were
sampled (e.g. Van den Brink et al., 1993;Josens et al., 2005),
and comparisons between the densities or assemblage
composition of Peracarida on natural and artificial substrates
in large rivers are lacking. This impedes our understanding of
invasion processes in freshwaters since the channel morphol-
ogy of large rivers has drastically changed due to straightening
and channelization, removal of large woody debris and the
construction of bank reinforcement structures such as ripraps,
groynes and embankments. Finally, a very limited number of
studies of alien macroinvertebrates in Western Balkan rivers
have been performed (Paunovićet al., 2005;Zorićet al., 2011;
Markovićet al., 2015). After the reconstruction of the Rhine-
Main-Danube canal in 1992, the Danube has assumed the role
of the southern invasion corridor through which Ponto-
Caspian invaders spread to central and Western Europe (Bij de
vaate et al., 2002). Although the Sava River, as the largest
tributary of the Danube in terms of discharge, is an important
international watercourse, biological invaders along its course
have been sporadically studied. Only a few recently published
articles have documented the macroinvertebrates of the Middle
and Lower Sava (Paunovićet al., 2012;Lucićet al., 2015),
whereas macroinvertebrate assemblages along the entire
course of the Sava were only examined once (Matoničkin
et al., 1975), and only two recent studies have documented
crustacean (Malacostraca) fauna of the Middle Sava (Žganec
et al., 2009;Maguire et al., 2011). The Sava River represents
the most important corridor for the spread of invasive species
into Dinaric rivers that harbor highly endemic fauna and are of
paramount importance for European freshwater biodiversity
(Schneider-Jacoby, 2005;Tierno de Figueroa et al., 2013;
Žganec et al., 2016). Thus, research into and the monitoring of
invasive alien species in the Sava River are urgently required to
increase our understanding of potential processes that facilitate
further invasions.
The aim of this study was to examine the longitudinal
distribution of native and alien Peracarida crustaceans along
the entire course of the Sava River and their relative share in
macroinvertebrate assemblages, as a measure of the impact of
alien peracarids. Records from previous studies were used to
assess upstream spread of invasive peracarids. In addition, the
impact of anthropogenic hydromorphological changes, i.e.
riprap stony bank reinforcement structures or artificial
substrate composed of meso-, macro- and megalithal at
reaches with finer natural sediment, was examined by
comparing the densities and peracarid assemblages on natural
and artificial substrates in order to test the hypothesis that bank
reinforcement structures facilitate Peracarida invasions.
2 Materials and methods
2.1 Study area Sava River
The Sava River is 926 km long (if considering the longer of
two source branches, the Sava Dolinka) (Schwarz, 2016) and is
the largest tributary of the Danube in volume (with
1572 m
3
s
1
average annual discharge at its mouth), and the
second largest river after the Tisza in terms of catchment area
(95 793 km
2
)(Sommerwerk et al., 2009). The Sava begins at
the confluence of two headwaters: the Sava Dolinka (source at
870 m a.s.l.) and Sava Bohinjka (526 m a.s.l.) in the Julian Alps
in Slovenia. At its spring area in the Julian Alps, the Sava River
flows through narrow river valleys or deep gorges, then passes
through more open valleys in Slovenia. In the middle course in
Croatia, it meanders along a wide valley, while in its lower
course, before it reaches the confluence with the Danube in
Belgrade (Serbia) it is 0.3–0.7 km wide. Based on previous
studies (Urbanič, 2008;Paunovićet al., 2012;Lucićet al.,
2015) and our own data, five major sections of the Sava can be
distinguished (Fig. 1): Alpine, Subalpine, Upper, Middle and
Lower Sava.
There are ten dams along the Sava in Slovenia (Fig. 1),
while the impoundment section of the Iron Gate I dam on the
Danube extends 100 km into the Lower Sava (ICPDR, 2005).
Although many towns, factories and power plants along its
course still represent sources of point pollution, the severe
pollution of the Middle Sava reported in previous studies
(Me
strov et al., 1978,1989) has been reduced, and water
quality has significantly improved due to the collapse of
industry after the war in the 1990s, and after Zagreb's
Page 2 of 12
K. Žganec et al.: Ann. Limnol. - Int. J. Lim. 2018, 54, No

wastewater treatment plant was put into operation in 2007
(Ogrinc et al., 2015;Andersen and Žganec, 2016). About two-
thirds of the Sava is navigable, up to the Kupa confluence in
Sisak (rkm 594). However, nowadays the Sava is hardly used
for transport, primarily due to a lack of maintenance and
investments (Komatina and Gro
selj, 2015). Flood protection,
navigation, hydropower and urbanization have been the main
drivers of morphological alterations of the Sava, of which the
most important include longitudinal and cross-section channel
changes, and the construction of bank reinforcement structures
(mainly stony ripraps), groynes, spikes and embankments.
2.2 Field sampling and laboratory analyses
Peracarida crustaceans and other macroinvertebrates were
collected during three years (2011, 2012 and 2015) using hand
net (aperture: 25 25 cm, mesh size: 500 mm). Sampling in
July 2011 and June 2012 along Croatian and Slovenian section
of Sava was preliminary and included either qualitative or
quantitative sampling to examine distribution and density of
native and alien Peracarida. (Fig. 1,Tab. 1). Sampling
campaign conducted at 15 sites (Fig. 1 black dots) in
September 2011 (9 sites) and in September 2012 (6 sites)
aimed to test the differences of Peracarida assemblages
(composition and density) between natural and artificial
substrates: ten replicate quantitative samples were collected
from each of two substrate types (10 Nþ10 A). Those
sites, located in the Upper (1 site), Middle (8) and Lower (6)
Sava, were selected to be approximately evenly spaced along
cca. 700 km long reach of Sava, if both substrate types were
close enough and accessible to sampling. Those 15 sites were
sampled in September during low water level, which usually
occurs in Sava at the end of summer. Each replicate sample
Table 1. Sampling time and methods used at all 61 sites at particular sections of the Sava River. Methods of sampling are indicated by bold, plain
text and italic site numbers: bold quantitative sampling (10 natural þ10 artificial substrate in 2011 and 2012 or 20 multihabitat-AQEM
sampling in 2015), plain five or ten replicate quantitative samples at different microhabitats, italic –qualitative multihabitat sampling during
preliminary field survey in 2011 and 2012 or in 2015.
Month/Year Alpine/Subalpine Upper Middle Lower
7/2011 ––25,28,33,34,37,38,41,48,54,55 –
9/2011 –21 29, 30, 35, 42, 50 56, 57, 59, 60
6/2012 1,2,5,3,4,6,7,8,9,10,11,12 14,15,16, 13, 17 ––
9/2012 ––23, 29, 32, 40, 47, 52 58, 61
5/2015 –18, 19, 20, 22 23, 31, 32, 37, 43, 44 –
6/2015 ––45, 46, 48, 49, 51, 53, 55 –
7/2015 ––24, 33, 35, 39 –
9/2015 ––25,26,27,28 –
Fig. 1. Sampling sites (61) along the entire course of the Sava River, with five major sections. Inset map shows eight very close sites at the
beginning of the Middle Sava section, with two sites marked by arrows where the upstream front of the invasive amphipod Dikerogammarus
haemobaphes was detected in 2011 and in 2015. Large black dots represent 15 sites were natural and artificial substrates were sampled
separately.
Page 3 of 12
K. Žganec et al.: Ann. Limnol. - Int. J. Lim. 2018, 54, No

covered 0.0625 m
2
of the river bottom area on a shallow bank.
The macroinvertebrate samples were collected from natural
substrate at microlocations that had not undergone any change
due to channelization and thus corresponded to natural
substrata along the assessed river reach. Natural substrate
samples were collected from every available substrate type
(mainly gravel, sand and mud), taking into consideration the
relative contribution to each microhabitat type (10% = 1
sample). On riprap stony artificial substrates (mainly macro-
and mesolithal) replicate samples were collected on the banks
(14 sites) or at the groyne (1 site). Differences between
densities at natural and artificial substrate types were examined
for eight taxa and two taxon groups. Taxa included seven
individual Peracarida species (two native and five alien) and
alien Chelicorophium sp. (two Chelicorophium species were
grouped together since in most cases unidentified juvenile
specimens had the highest share in the total abundance of
Chelicorophium), and two taxon groups native species
group and alien species group.
Sampling of macroinvertebrates in 2015 (May–July, Tab.
1) at 21 sites of the Croatian section of the Sava was the part of
the project aimed at assessing biocontamination of Croatian
large rivers (Žganec et al., 2016;Ćuk et al., unpublished) and
included standard multihabitat sampling according to AQEM
methodology (AQEM Consortium, 2002). Six of these sites
were the same as those sampled in 2011 or 2012. Additionally,
the distribution front change of the most notorious amphipod
invader Dikerogammarus haemobaphes (Eichwald, 1841) was
examined by multihabitat qualitative sampling at four sites at
the beginning of Middle Sava in September 2015 (Tab. 1,
Fig. 1). Macroinvertebrate samples were preserved on-site in
70% ethanol. In a total, 465 samples of macroinvertebrates
were collected at 61 sites along the entire course of the Sava,
from the site 360 m downstream of the source of Sava Dolinka
(site 1), and from the outflow of Sava Bohinjka at Lake Bohinj
(site 5) to the last site (61) 10 km upstream of the confluence
with the Danube in Belgrade. Site locations were recorded
using a GPS receiver and data were mapped using the Arc-GIS
10.1 program package.
In the laboratory, the macroinvertebrates were separated
from the sediment and organic detritus and stored in 70%
ethanol for later identification. Peracarid crustaceans were
identified using the following keys: Amphipoda Cărăusu
et al. (1955),Karaman and Pinkster (1977a,b), Pinkster
(1993),Eggers and Martens (2001); Isopoda Argano (1979),
Veuille (1979), Mysidacea Dobson (2012),Wittmann et al.
(2016). All other non-peracarid taxa were identified to a higher
taxonomic level, order (23), class (10) or phylum (2) at 53 sites
(at 8 sites only crustacean fauna was separated and identified).
2.3 Data analyses
Generalized linear models (GLMs) were used to test for
differences in taxon or taxon group abundance between natural
and artificial substrates within each sampling site and year. The
substrate effect was tested first as a main effect across sampling
sites, then again separately within each site for each taxon.
Because the response variable was count data with abundant
zeros, the best-fit model was a negative binomial model with
log link function. Model fit was tested assuming model
deviance was distributed as a chi-square variable with the
residual degrees of freedom. In these models, sampling site
was considered a fixed factor, because the individual substrate
effect at each site is of interest independently of all other sites.
Final model included site, year, substrate and site substrate
interaction for all GLM analyses. GLMs were also used to test
for differences in the number of species and the total share of
Peracarida between two substrate types. Poisson distribution
was the best-fit model (link function: log) for number of
species and quasibinomial distribution (link function: logit) for
proportion of Peracarida in total abundance of macroinverte-
brates. All these analyses were performed with the R 3.4.3. (R
Development Core Team, 2017) using packages MASS, pscl
and mgcv. Due to multiple testing, Bonferroni corrected p-
values were used for different number of valid tests in each of
three groups of GLM analyses (abundance: n= 40; species
number and Peracarida proportion: n= 15).
The spatial pattern of Peracarida assemblages along whole
Sava was analyzed using non-parametric multidimensional
scaling (NMDS) and PERMANOVA with software package
PRIMER Version 6.1.13. and PERMANOVAþVersion 1.0.3
(PRIMER-E Ltd 2009). Average abundance of all Peracarida
taxa at all site-year combinations was square-root transformed
to control the influence of dominant species and the Bray-
Curtis index of similarity was used to calculate the similarity
matrix. Differences between groups which were clearly
different in NMDS plots were tested using PERMANOVA
and SIMPER analysis. Two separate data sets (1. - average
abundance at 61 sites in all years, 2. - 15 sites where between
substrate differences were examined) where analyzed using
nested PERMANOVA (site-groups nested in years) to test
effects of both factors and differences between two distin-
guished site-groups identified in NMDS. SIMPER analysis
was used to identify taxa that contribute the most to
dissimilarity between tested groups. Differences in Peracarida
assemblages between natural and artificial substrates at 15
sites, for each of the two distinguished groups of sites
separately (native dominated-sites 21 and 23, and alien
dominated-other 13 sites), were tested using crossed design in
PERMANOVA. Sampling site, substrate and year were
considered fixed factors and after testing main effects and
site substrate interaction, the effect of substrate was tested
within each site and year separately using pair-wise tests in
PERMANOVA with Bonferroni correction for multiple testing
(10 valid tests in second-alien group of sites).
3 Results
3.1 Longitudinal pattern of Peracarida assemblages
in the Sava River
A total of 14 species of peracarid crustaceans (5 native and
9 alien) were recorded at 61 sites along the entire course of the
Sava River (Fig. 2): 11 amphipods (4 native and 7 alien), 2
isopods (1 native, 1 alien) and an alien species of Mysida
(L. benedeni). Multivariate analysis of Peracarida assemblages
(Fig. 3), revealed two major groups of sites: sites from the
Alpine to the beginning of the Middle Sava (Upper third of
Sava course) that contained only native species, and sites
dominated by alien species along the Middle and Lower Sava
sections. These two sections had clearly different Peracarida
assemblages (PERMANOVA, pseudo-F: 22.7, p= 0.0001),
Page 4 of 12
K. Žganec et al.: Ann. Limnol. - Int. J. Lim. 2018, 54, No

with non-significant effect of year of sampling (PERMA-
NOVA, pseudo-F: 0.05, p= 0.71). Average dissimilarity of
these two groups of sites was 98.6% (SIMPER analysis), with
native Gammarus fossarum Koch, 1836 specific for the first
and D. haemobaphes,Chelicorophium sowinskyi (Martynov,
1924) and J. istri for the second group.
Fig. 3. Non-parametric multidimensional scaling analysis (NMDS) of peracarid assemblages at 57 sites in the Sava (sites 1, 2, 5, 6 without
peracarid species were excluded); (labels: first number site codes as in Figure 1, second number the year of sampling, 1-2011, 2-2012,
5-2015).
Fig. 2. Distribution of 14 Peracarida species along the entire course of the Sava River with identified river sections (two sites 5 and 6 on the Sava
Bohinjka without peracarid crustaceans are not shown; N native, A alien species). Sites are arranged according to the distance from the
source of the Sava Dolinka and river kilometers (rkm) are shown above x-axis.
Page 5 of 12
K. Žganec et al.: Ann. Limnol. - Int. J. Lim. 2018, 54, No

The most widespread native species were G. fossarum (at
22 sites) and Asellus aquaticus (L. 1758) (14 sites) (Fig. 2). G.
fossarum had a very low abundance in the Alpine Sava, which
abruptly increased at the last site of this section (Fig. 4). No
peracarid crustaceans were found at the four most upstream
sites (sites 1 and 2 in the Sava Dolinka, and sites 5 and 6 in the
Sava Bohinjka). Only one specimen of Gammarus roeselii
Gervais, 1835 was recorded at site 29 and only in 2011. Of nine
alien Peracarida, only three species, D. haemobaphes
(identified at 31 sites), C. sowinskyi (27 sites) and J. istri
(22 sites) were omnipresent in the Middle and Upper section of
Lower Sava, while at sites 58–60 of Lower Sava alien
peracarids had very low densities. Some of the recorded
peracarid invasive species, Chelicorophium robustum (G.O.
Sars, 1895), Echinogammarus ischnus (Stebbing, 1899) and
Obesogammarus obesus (Sars G.O., 1894), were recorded for
the first time but only in the Lower Sava (Fig. 2). In the 37 sites
where Chelicorophium amphipods were found, C. sowinskyi
was more widespread (at 34 sites) than C. curvispinum (10
sites), while at most sites (32) C. sowinskyi was the more
abundant species.
During this study (2011–2015) an upstream range
expansion of the invasive amphipods D. haemobaphes and
D. villosus was observed. The upstream range extension of the
most upstream invader D. haemobaphes was slow (0.9 km
yr
1
)(Fig. 1). Although, the native species G. fossarum and A.
aquaticus, were more abundant than D. haemobaphes at site 29
in 2011, when G. roeselii was also found, only D.
haemobaphes was found in 2015 at this site. Further, site
27 was inhabited by G. fossarum and A. aquaticus in 2011,
whereas in 2015 only D. haemobaphes was found there. In
contrast to the slow spread of D. haemobaphes,D. villosus
which was found only in the Lower Sava (from 10 to 139 rkm)
in 2011 and 2012, in 2015 D. villosus suddenly appeared at
three sites in the Middle Sava (from 401 to 579 rkm). Thus, this
species has spread 440 km upstream and it seems that it has
already replaced the previous invader, D. haemobaphes, at site
33. The upper distribution limit of other widespread invasive
crustaceans, C. sowinskyi,C. curvispinum and J. istri, as well
as D. villosus, was at 579 rkm (site 32) and for the first three
species did not change from 2012 to 2015.
The total number of Peracarida species increased in a
downstream direction from one species (G. fossarum) in the
Alpine Sava, four and three in Subalpine and Upper Sava, 9 (3
native and 6 alien) in Middle Sava and 11 (2 native and 9 alien)
species in the Lower Sava. The relative proportion of native
species in the total abundance of collected macroinvertebrates
(Fig. 4), increased from less than 1% in the Alpine Sava to an
average of 40% in the Upper section. The Alpine section could
be distinguished by the low density of G. fossarum. This
species was predominant in terms of abundance in the
Subalpine and Upper Sava, where either S. ambulans or A.
aquaticus were the subdominant species. The proportion of
alien peracarids in total abundance of macroinvertebrates in
the Middle and Lower sections exhibited substantial variation
(range 0.1–82%), with higher average proportion in Middle
(24%) than Lower section (9%). The last four sites of Lower
Sava had the lowest average proportion of alien Peracarida
(3%). The high diversity of peracarid assemblages in the
Middle and Lower Sava, with different combinations of
dominant invasive species (Chelicorophium spp., D. haemo-
baphes and D. villosus), caused high overlap of sites at these
two sections in Figure 3. Only sites with low density of
invasive peracarids (three sites in Lower Sava) or sites with
only one species found could be clearly distinguished from the
main cluster of sites in these two sections.
3.2 Microdistribution of Peracarida on natural and
artificial substrates
The significant differences in abundance at two substrate
types across all sampling sites (where particular taxon or taxon
group was present) were established in most cases (GLM,
p<0.001), except for native amphipod G. fossarum and alien
Fig. 4. Relative abundance of native and alien peracarid crustaceans (Amphipoda þIsopoda þMysida) in macroinvertebrate assemblages along
the five sections of the Sava River; (x-axis labels: first number site codes as in Figure 1, second number the year of sampling, 1-2011,
2-2012, 5-2015).
Page 6 of 12
K. Žganec et al.: Ann. Limnol. - Int. J. Lim. 2018, 54, No

mysid L. benedeni (GLM, p>0.5; Tab. 1 in Supplementary
material). In all significant cases overall average abundance
was 2–96 times higher on artificial substrate. There was also
significant effect of site for all taxa/taxon groups, due to large
differences in abundance between particular sites, with
significant effect of year of sampling for most cases (6 of 9
tested cases) and significant interactions of site and substrate
factors (6 of 10). In four cases of significant site and substrate
interaction detected, direction of differences (artificial >natu-
ral) were the same across sites and only magnitudes differed.
For two taxa, D. haemobaphes and Chelicorophium sp., two
sites (i.e. one site for each taxa) with significantly different
direction (natural >artificial) were detected.
Of eight tests for native species for each site separately, two
cases of significantly higher abundance on artificial substrate
were established (GLM, p<0.001) (Fig. 5). For other six non-
significant cases, in four cases densities of two native species
were too low at both substrate types and tests were considered
unreliable (densities at both substrate types <2 ind. m
2
).
Of altogether 59 possible tests for six alien taxa and alien
taxon group, 23 were considered unreliable due to very low
abundance of particular taxon at both substrate types
(density <2 ind. m
2
) and were not done (Fig. 5: n.t.-not
tested). Therefore, of 36 reliable tests for alien taxa or taxon
group in most cases (28 or 78%) significantly higher (applying
Bonferroni correction) abundance was established on artificial
substrate (GLM, p<0.001), with only one case (3%) of
significantly higher abundance (GLM, p<0.001) on natural
substrate (D. haemobaphes at site 50) and seven cases (19%) of
non-significant differences between two substrate types
(GLM, p>0.05). Of altogether 27 species-site cases which
were not tested, in 21 (78%) cases density was higher at
artificial substrate.
Significantly higher average number of species on artificial
substrate was established for all sites combined together
(GLM, p<0.001). Of altogether 15 tests for between substrate
differences in average species number at each site separately,
there were seven cases (47%, with Bonferroni corection) of
significantly higher number of species on artificial substrate
(GLM, p<0.005). In all other cases difference between
substrates in average number of species per sample was not
significant (GLM, p>0.05).
Significantly higher average proportion of Peracarida in
total abundance of macroinvertebrates on artificial substrate
was established for all sites combined together (GLM,
p<0.001). There were altogether seven cases (47%, with
Bonferroni corection) of significantly higher proportion of
Peracarida in total abundance of macroinvertebrates on
artificial substrate (GLM, p<0.05), two significant cases of
natural >artificial (GLM, p<0.001), and six cases of non-
significant differences between substrates.
Preliminary check of data set (218 non-zero samples) used
for testing differences in Peracarida assemblages between two
substrate types at 15 sites revealed two groups of sites: (1) sites
21 and 23 where only native species occurred and (2) other 13
sites dominated by alien species. These two groups were
significantly different (PERMANOVA, pseudo-F= 33.4,
p= 0.0001) and subsequently analyzed separately. There
was significant site effect (pseudo-F= 45.7, p= 0.0001,
crossed PERMANOVA, year factor not included) in first
native-group, while effect of substrate was not significant
(pseudo-F= 0.96, p= 0.41). For second group dominated by
alien species, there were significant effects of all factors, site
(crossed PERMANOVA, pseudo-F= 11.1, p= 0.0001), sub-
strate (pseudo-F= 14.2, p= 0.0001) and year (pseudo-F= 6.1,
p= 0.0001), as well as significant site substrate interaction
(pseudo-F= 4.2, p= 0.0001). Pair-wise tests for between
substrate differences of Peracarida assemblages were reliable
(if nu. of permutations >10) for 9 of 13 sites dominated by
alien species (at sites 32, 52, 58, 60 tests were unreliable with
very small number of unique permutation due to only one or
two samples with peracarids on natural substrate). Significant
difference between substrates, due to higher densities of aliens
on artificial substrate, were established at four sites
(PERMANOVA pair-wise tests, pseudo-F= 2.6–3.2, p
<0.001), while at five sites these differences were non-
significant (pseudo-F= 1.16–1.50, p>0.05). Only at site 50
significant difference between peracarid assemblages on two
substrate types (pseudo-F= 4.0, p= 0.0002) were due to higher
densities of D. haemobaphes and Chelicorophium sp. on
natural substrate (Tab. 1 in Supplementary material).
4 Discussion
4.1 Distribution
Peracarid fauna in the Sava River, with 5 native and 9 alien
species, had following distribution pattern during the course of
this study (2011–2015): only native species inhabited Upper
third of Sava and aliens dominated the rest of the course.
Further, interesting distribution pattern of alien peracarid
crustaceans was observed: the highest densities and abundance
contamination of macroinvertebrate assemblages by alien
peracarids was not in the Lower Sava section as expected, but
in the Middle Sava section. The Lower Sava on the other hand
had the highest number of alien peracarid species (all 9
species), most of which were found in low densities. Since
many physicochemical parameters in the Middle and Lower
Sava sections did not differ significantly and as pollution at
some sites in the Middle Sava section was even higher than in
the Lower Sava (unpublished results), physicochemical
parameters were probably not responsible for the observed
differences. Higher propagule pressure due to more intensive
ship traffic and the proximity of the Danube as a source of new
alien species appears to be the main reason for the higher
number of alien species in the Lower Sava section. The higher
abundance contamination in the Middle Sava, on the other
hand, could be attributed to the higher share of coarse fractions
in the sediment (especially at the confluences of larger
tributaries), i.e. more muddy bottoms in the Lower Sava.
In comparison with the Danube, two of the seven
“ubiquitous”alien species recorded in the Danube (D.
haemobaphes and J. istri =J. sarsi in Borza et al., 2015),
were also recorded in the most of the sites examined in the
Sava River. Further, D. villosus, as the most widespread
species in the Danube, exhibited a disjunctive distribution in
the Sava: it was first recorded only in the Lower Sava (in 2011
and 2012) and in 2015 it appeared at three sites of the Middle
Sava. Dispersal of D. villosus from the Lower Sava or Danube
River to the Middle Sava section probably occurred by
transport on ships or on small fishing boats. Judging from its
invasion rates in other large European rivers (e.g. Bollache
Page 7 of 12
K. Žganec et al.: Ann. Limnol. - Int. J. Lim. 2018, 54, No

et al., 2004;Van Riel et al., 2006), as well in the Drava River in
Croatia (Žganec, unpublished), D. villosus spread through the
Middle Sava can be expected to be fast. Also, it could be
expected, similarly to other documented cases (review in
Rewicz et al., 2014) and situation in Drava (unpublished
results), that it will eliminate the previous invader, D.
Fig. 5. Density (log
10
-transformed) on natural (Nat.- above x-axis) and artificial substrate (Art.-below x-axis) of seven species, Chelicorophium
sp., and native and alien species grouped together at 15 sites where the density was established by replicate sampling (n= 10) on two types of
microhabitats. *Significant differences (GLM for count data with Bonferroni correction; n.s.-not significant, n.t.-not tested if density
<2 ind. m
2
).
Page 8 of 12
K. Žganec et al.: Ann. Limnol. - Int. J. Lim. 2018, 54, No

haemobaphes. Further, while some studies of the interaction of
D. villosus and D. haemobaphes showed that D. villosus is
stronger competitor (Kobak et al., 2016), others did not found
D. villosus to be the more resilient competitor/predator
(Kinzler et al., 2009). Also, since many field studies have
reported on the co-occurrence of these two species (e.g. Labat
et al., 2011;Borza et al., 2015), and since we found such co-
occurrence at site 57, it seems that in some situations long-term
coexistence of these two species is possible. Ongoing invasion
of the Middle Sava by D. villosus provides excellent
opportunity for the future studies that should examine
interactions of these two invaders in the Middle Sava.
Distributions and densities of other two most important
invasive amphipods, C. sowinskyi and C. curvispinum, were
different in Sava from those in Danube. In Sava C. sowinskyi
was more widespread and more abundant species than C.
curvispinum. These results suggest that either C. sowinskyi was
the first colonizer of the Sava, whereas C. curvispinum arrived
later, as suggested by Borza (2011) for Hungarian freshwater
or that C. curvispinum is less well adapted to conditions in
Sava and may be the weaker competitor. The finding of
invasive alien amphipod C. robustum at one site in the Lower
Sava should also be pointed out. This species probably recently
started to expand its range from the Danube into the Lower
Sava, since in Danube it is in the process of spreading
downstream and closing its distribution gap in the Middle
Danube (Borza et al., 2015).
The first records of alien peracarids in the Sava, two alien
amphipod taxa (Corophium sp. and Pontogammarus sp.) and
alien isopod Jaera sp., date back to the 1970s when only a
previous longitudinal study of macroinvertebrates along the
entire watercourse of the Sava (at 41 sites) was conducted
(Matoničkin et al., 1975). Despite misidentifications of some
species in this first study (D. haemobaphes probably
misidentified as Pontogammarus), using the upstream distri-
bution fronts of D. haemobaphes,Chelicorophium spp. and J.
istri from Matoničkin et al. (1975), as well as from study in
2004 (Žganec et al., 2009) and personal unpublished records
from 2009, upstream distribution range expansion rates of
three invasive species could be estimated: D. haemobaphes
(avg: 18.4, range: 0.9–54.5 km year
1
), Chelicorophium sp.
(avg: 15.8, range: 1.6–30.0 km year
1
) and J. istri (avg: 11.7,
range: 6.6–16.7 km year
1
). Similar estimates of upstream
range expansion rates for D. villosus (30–40 km year
1
) and C.
curvispinum (8–17 km year
1
) were reported for the Meuse
River in France (Josens et al., 2005), while much higher mean
dispersal rates but similar minimum dispersal rates for the
same two species and J. istri were observed in the Rhine
(Leuven et al., 2009). Our estimates are probably the lowest
documented values, especially in the case of D. haemobaphes,
which showed a slow upstream spread of 0.9 km year
1
during
study period (2011–2015). This could be because the reach of
the Sava where this was observed is not used for navigation
and probably only small fishing boats could facilitate the
spread. Another important factor is the distribution of bank
riprap reinforcement structures that are scattered between
locations with stronger bank erosion and long stretches of
natural banks in this Upper part of the Middle Sava. Therefore,
it can be assumed that more natural hydromorphology at this
part of Sava and absence of shipping traffic disables faster
upstream spread of invasive amphipod D. haemobaphes.
Furthermore, the upstream spread of two Chelicorophium
species and J. istri in Sava appears to have halted at site 32
(rkm 579), which is very close to Sisak (rkm 594) up to where
navigation is possible. This finding indicates that shipping
probably served as the main vector of spread of these three
species in the Sava. Also, since the first sites upstream of site
32 possess different physicochemical conditions indicative of
increased pollution (unpublished data), it is possible that
certain factors in this part of the Sava, beside lack of dispersal
vectors, have inhibited further upstream spread of three
invasive peracarids, C. sowinskyi,C. curvispinum and J. istri.
Finally, as range expansion of invaders might occur through
non-continuous processes (jump dispersal patterns) which can
also be human-mediated (Mineur et al., 2010), and as multiple
dispersal pathways might occur within the same ecosystem
(Suarez et al., 2001), long-term data-sets are required to
capture all patterns.
The upstream spread of D. haemobaphes appears to have
caused disappearance of the native species, G. fossarum,G.
roeselii and A. aquaticus at the beginning of Middle Sava. The
predatory behavior of this species was shown to be similar to that
of the better-known “killer shrimp”,D. villosus (Bacela-
Spychalska and Van der Velde, 2013). Accordingly, it can be
assumed that D. haemobaphes will slowly spread further
upstream in Upper Sava where it is expected to eliminate native
peracarids (G. fossarum,S. ambulans and A. aquaticus). Similar
cases of upstream spread of invasive Dikerogammarus amphi-
pods that has led to the disappearance of native species were
observed in many other large rivers (Bollache et al., 2004;Josens
et al.,2005;Grabowski et al., 2007;Borza et al., 2015). However,
Hellmann et al. (2017) showed that impact of D. villosus on
benthic assemblage was weak and differed between two studied
rivers. Hence, with this study as baseline, future studies of the
spread of invasive peracarid crustaceans in Sava should
concentrate on their impact on macroinvertebrate assemblages.
4.2 Artificial substrates as “exotic paradise”
microhabitats that facilitate invasion
In this study we observed strong preferences of artificial
coarser stony substrate (macro-, meso- and megalithal on
riprap embankments and groynes) over finer natural substrate
(gravel, sand and mud) by alien peracarids. At most sites, alien
species had higher densities on artificial substrate and often
these differences were significant (Fig. 5). Also, the Peracarida
assemblages on natural and artificial substrates significantly
differed at almost half of sites due to the much higher
abundance of Dikerogammarus spp., Chelicorophium spp. and
J. istri, on artificial substrates. In the literature these tree taxa
were described as “lithophilous dwellers”(Jazdzewski,
1980;Bij de Vaate et al., 2002). Further, numerous previous
field studies established high densities of alien amphipods and
isopods on artificial stony structures: for D. villosus (Devin
et al., 2003;Van Riel et al., 2006;MacNeil et al., 2008;Boets
et al., 2010;MacNeil and Platvoet, 2013), D. haemobaphes
(Environment Agency, 2012), E. ischnus (Van Overdijk et al.,
2003), C. curvispinum (Van der Velde et al., 2000) and isopod
J. istri (Kelleher et al., 2000). Here we also showed that
samples from natural substrates often did not contain alien
amphipods and isopods, although high densities of aliens were
found on artificial substrates. Also, we noted that the average
Page 9 of 12
K. Žganec et al.: Ann. Limnol. - Int. J. Lim. 2018, 54, No

numbers of alien peracarid species per sample at many sites
were significantly higher on artificial substrate, and the same
was observed for peracarid share in the entire benthos.
Obviously, alien amphipods and isopod J. istri show strong
preference for artificial substrate. This could be explained by
higher artificial substrate stability during high discharge events
and the fact that these microhabitats provide refugia not only
from strong current but also from fish predators. Furthermore,
large stones of artificial substrate probably act as traps for
coarse particular organic matter (CPOM), what makes them
even more attractive for alien gammarids, while for
Chelicorophium species they provide perfect stable surfaces
for attachment of mud tubes in which they live. Therefore,
similarly to conclusions of MacNeil and Platvoet (2013) for D.
villosus, this study showed that in order to detect amphipod and
isopod invaders, sampling protocols should be adapted to
always include any large stone or concrete structures on
artificial substrate. However, one case with higher density on
natural substrates was observed for D. haemobaphes at site
50). This site is hundred meters from the confluence of the
Bosna River, where the natural substrate contained a higher
proportion of a coarser substrate. It was shown that tributaries
exert a significant effect on the substrate and on the
physicochemical parameters in large rivers, thereby inducing
variability in longitudinal changes of macroinvertebrate
assemblages (Rice et al., 2001). Hence, future studies could
examine whether the confluences of larger tributaries, because
of higher proportion of coarser substrate components,
represent sites with higher densities of invasive pericarids.
Therefore, confluences of larger tributaries could represent
natural stepping stones that facilitate the invasion of peracarid
crustaceans. When different artificial structures, especially
riprap stony structures and groynes, are introduced to the main
watercourse, reaches on the river with a few favorable places
can easily be transformed into a chain of microhabitats where
lithophilous alien amphipods and isopods find their “paradise
microhabitats”. Since most large European rivers have been
transformed by channel modification structures in a similar
way to the Sava, and probably much more intensively, this type
of anthropomorphic pressure could be one of the main reasons
for the very rapid expansion of the range of most aquatic
invasive peracarids across Europe.
5 Conclusions
This study is the most detailed investigation of peracarid
assemblages along the entire course of the Sava River in the
last 40 years (since Matoničkin et al., 1975), and offers a
detailed insight into differences in composition and density of
peracarid crustaceans assemblages on natural and artificial
substrates along the Sava River. It represents a baseline for
future assessments of the impact of the spread of invasive
peracarids and other pressures on macroinvertebrate assemb-
lages. The Sava is the main corridor through which invasive
species could spread into Dinaric karst rivers that support
endemic macroinvertebrate fauna and other biota. Therefore, a
deeper understanding of dispersal patterns and pathways as
well as the changes in the macroinvertebrate fauna that inhabit
the Sava is of pivotal importance for the conservation of
freshwater biodiversity of Western Balkan rivers.
Supplementary Material
Table S1. Results of all tests for all taxa and taxon groups,
with all 15 sites combined and for each site separately: (1)
generalized linear models for abundance (density data
showed), (2) species number and (3) proportion of Peracar-
ida in total macroinvertebrate abundance, (4) crossed
PERMANOVA for each of two distinguished groups of sites
separately (native and alien site groups) with pair-wise tests
for between substrate differences at each site separately.
Bonferroni-corrected p-values were separately calculated for
abundance data (p= 0.05/40 = 0.00125), species number and
proportion of Peracarida (p= 0.05/15 = 0.0033) and PER-
MANOVA pair-wise tests (p= 0.05/10 = 0.005).
The Supplementary Material is available at https://www.
limnology-journal.org/10.1051/limn/2018008/olm.
Acknowledgements. This work is the result of two projects
supported by Ministry of Science, Education and Sports of the
Republic of Croatia, Ministry of Education, Science and
Technological Development of the Republic of Serbia and
Ministry of Higher Education, Science and Technology of the
Republic of Slovenia, project periods 2011–2012 and 2012–
2013. We also wish to thank for financial support by
GLOBAQUA project that has been supported by European
Communities 7th Framework Programme Funding under
Grant agreement no. 603629-ENV-2013-6.2.1 (Globaqua).
Sampling along Croatian section of Sava in 2015 was
supported by Croatian Water and Management Agency. We
thank Prof. Stewart Schultz, University of Zadar, for help with
GLM analyses in R and for linguistic improvements. Also,
three unknown referees are thanked for their critical remarks
that greatly improved earlier versions of the manuscript.
References
Andersen CB, Žganec K. 2016. Impact of Zagreb wastewater
treatment plant on the Sava River water quality. In: Annual Meeting
of the American Association of Geographers.
Arbačiauskas K, Semenchenko V, Grabowski M, Leuven RSEW,
PaunovićM, Son MO, Csányi B, GumuliauskaitėS, Konopacka A,
Nehring S, Van der Velde G, Vezhnovetz V, Panov VE. 2008.
Assessment of biocontamination of benthic macroinvertebrate
communities in European inland waterways. Aquat Invasions 3:
211–30.
Argano R. 1979. Guide per il riconoscimento delle specie animali
delle acque interne Italiane. 5. Isopodi (Crustacea, Isopoda).
Consiglio nazionale delle ricerche. Verona: Impresso della
stamperia Veldonega, 63 p.
AQEM Consortium. 2002. Manual for the application of the AQEM
system. A comprehensive method to assess European streams using
benthic macroinvertebrates, developed for the purpose of the Water
Framework Directive. Version 1, 202 p.
Bacela-Spychalska K, Van der Velde G. 2013. There is more than one
“killer shrimp”: trophic positions and predatory abilities of invasive
amphipods of Ponto-Caspian origin. Freshw Biol 58: 730–741.
Bernauer D, Jansen W. 2006. Recent invasions of alien macro-
invertebrates and loss of native species in the upper Rhine River,
Germany. Aquat Invasions 11: 55–71.
Bij de Vaate A, Jazdzewski K, Ketelaars HA, Gollasch S, Van der
Velde G. 2002. Geographical patterns in range extension of Ponto-
Page 10 of 12
K. Žganec et al.: Ann. Limnol. - Int. J. Lim. 2018, 54, No

Caspian macroinvertebrate species in Europe. Can J Fish Aquat Sci
59: 1159–1174.
Boets P, Lock K, Messiaen M, Goethals PLM. 2010. Combining data-
driven methods and lab studies to analyse the ecology of
Dikerogammarus villosus.Ecol Inform 5: 133–9.
Bollache L, Devin S, Wattier R, Chovet M, Beisel J-N, Moreteau JC,
Rigaud T. 2004. Rapid range extension of the Ponto-Caspian
amphipod Dikerogammarus villosus in France: potential con-
sequences. Arch für Hydrobiol 160: 57–66.
Borza P. 2011. Revision of invasion history, distributional patterns,
and new records of Corophiidae (Crustacea: Amphipoda) in
Hungary. Acta Zool Acad Sci Hung. 57: 75–84.
Borza P, Csányi B, Huber T, Leitner P, PaunovićM, Remund N,
Szekeres J, Graf W. 2015. Longitudinal distributional patterns of
Peracarida (Crustacea, Malacostraca) in the River Danube.
Fundam Appl Limnol 187: 113–26.
Cărăusu S, Dobreanu E, Manolache C. 1955. Amphipoda forme
salmastre si de apădulce. Fauna Republ pop Rom Crustac 4: 1–
407.
Den Hartog C, Van den Brink FWB, Van der Velde G. 1992. Why was
the invasion of the river Rhine by Corophium curvispinum and
Corbicula species so successful? J Nat Hist 26: 1121–1129.
Devin S, Beisel JN. 2006. Biological and ecological characteristics of
invasive species: a gammarid study. Biol Invasions 9: 13–24.
Devin S, Piscart C, Beisel JN, Moreteau JC. 2003. Ecological traits of
the amphipod invader Dikerogammarus villosus on a mesohabitat
scale. Arch Fur Hydrobiol 158: 43–56.
Dick JTA, Platvoet D. 2000. Invading predatory crustacean
Dikerogammarus villosus eliminates both native and exotic
species. Proc. R. Soc. Lond. Ser. B 267: 977–983.
Dobson M. 2012. Identifying invasive freshwater shrimps and
isopods. Ambleside: Freshwater Biological Association, 29 p.
Eggers TO, Martens A. 2001. Bestimmungsschlüssel der Sübwasser-
Amphipoda (Crustacea) Deutschlands. Laurterbornia 42: 1–68.
Environment Agency. 2012. Invasive shrimp (D. haemobaphes)
interim briefing note, Environment Agency UK, www.nonnatives
pecies.org.
European Environment Agency. 2012. European waters assessment
of status and pressures, Copenhagen: European Environment
Agency, 96 p.
Gergs R, Rothhaupt K-O. 2015. Invasive species as driving factors for
the structure of benthic communities in Lake Constance, Germany.
Hydrobiologia 746: 245–254.
Grabowski M, Bacela K, Konopacka A. 2007. How to be an invasive
gammarid (Amphipoda: Gammaroidea) comparison of life
history traits. Hydrobiologia 590: 75–84.
Grabowski M, Bacela K, Konopacka A, Jazdzewski K. 2009.
Salinity-related distribution of alien amphipods in rivers provides
refugia for native species. Biol Invasions 11: 2107–2117.
Hellmann C, Schöll F, Worischka S, Becker J, Winkelmann C. 2017.
River-specific effects of the invasive amphipod Dikerogammarus
villosus (Crustacea: Amphipoda) on benthic communities. Biol
Invasions 19: 381–398.
Holdich DM, Pöckl M. 2007. Invasive crustaceans in European inland
waters. In: Gherardi F, ed. Biological invaders in inland waters:
profiles, distribution, and threats. Dordrecht: Springer, pp. 29–75.
ICPDR. 2005. The Danube River Basin District. River basin
characteristics, impact of human activities and economic analysis.
Part A Basin-wide overview. International Commission for the
Protection of the Danube River (ICPDR), Vienna, Austria.
Jazdzewski K. 1980. Range extensions of some gammaridean species
in European inland waters caused by human activity. Crustaceana
6: 84–107.
Jazdzewski K, Konopacka A, Grabowski M. 2002. Four Ponto-
Caspian and one American gammarid species (Crustacea,
Amphipoda) recently invading Polish waters. Contrib Zool 71:
115–122.
Josens G, De Vaate AB, Usseglio-Polatera P, Cammaerts R, Chérot F,
Grisez F, Verboonen P, Vanden Bossche JP. 2005. Native and exotic
Amphipoda and other Peracarida in the River Meuse: new
assemblages emerge from a fast changing fauna. Hydrobiologia
542: 203–220.
Karaman GS, Pinkster S. 1977a. Freshwater Gammarus species from
Europe, North Africa and adjacent regions of Asia (Crustacea 
Amphipoda) Part I. Gammarus pulex group and related species.
Bijdr Dierkd 47: 1–97.
Karaman GS, Pinkster S. 1977b. Freshwater Gammarus species from
Europe, North Africa and adjacent regions of Asia (Crustacea 
Amphipoda) Part II. Gammarus roeseli group and related
species. Bijdr Dierkd 47: 165–96.
Karatayev AY, Burlakova LE, Padilla DK, Mastitsky SE, Olenin S.
2009. Invaders are not a random selection of species. Biol Invasions
11: 2009–2019.
Kelleher B, Bij de Vaate A, Swarte M, Klink AG, Van der Velde G.
2000. Identification, invasion and population development of the
Ponto-Caspian isopod Jaera istri Veuille (Janiridae) in the Lower
Rhine, the Netherlands. Beaufortia Bull Zool Mus Univ Amst 50:
89–94.
Kinzler W, Kley A, Mayer G, Waloszek D, Maier G. 2009. Mutual
predation between and cannibalism within several freshwater
gammarids: Dikerogammarus villosus versus one native and three
invasives. Aquat Ecol 43: 457–464.
Kobak J, Rachalewski M, Bacela-Spychalska K. 2016. Conquerors or
exiles? Impact of interference competition among invasive Ponto-
Caspian gammarideans on their dispersal rates. Biol Invasions 18:
1953–1965.
Komatina D, Gro
selj S. 2015. Transboundary water cooperation for
sustainable development of the Sava River Basin. In: MilačičR,
Ščančar J, PaunovićM, eds. The Sava River. The handbook. Berlin/
Heidelberg: Springer, pp. 1–28.
Labat F, Piscart C, Fontan B. 2011. First records, pathways and
distributions of four new Ponto-Caspian amphipods in France.
Limnologica 41: 290–295.
Leuven RSEW, Van der Velde G, Baijens I, Snijders J, van der Zwart
C, Lenders HJR, Bij de Vaate A. 2009. The river Rhine: a global
highway for dispersal of aquatic invasive species. Biol Invasions
11: 1989–2008.
Lodge DM. 1993. Biological invasions: lessons for ecology. Trends
Ecol Evol 8: 133–137.
LucićA, PaunovićM, TomovićJ, KovačevićS, ZorićK, SimićV,
AtanackovićA, MarkovićV, Kračun-KolarevićM, Hudina S,
Lajtner J, Gottstein S, Milo
sevićĐ,Anđus S, Žganec K, JakličM,
SimčičT,Vilenica M. 2015. Aquatic macroinvertebrates of the Sava
River. In: MilačičR, Ščančar J, PaunovićM, eds. The Sava River.
The handbook. Berlin/Heidelberg: Springer, pp. 335–359.
Macneil C, Platvoet D. 2013. Could artificial structures such as fish
passes facilitate the establishment and spread of the “killer shrimp”
Dikerogammarus villosus (Crustacea: Amphipoda) in river
systems? Aquat Conserv 23: 667–77.
MacNeil C, Platvoet D, Dick JT. 2008. Potential roles for differential
body size and microhabitat complexity in mediating biotic
interactions within invasive freshwater amphipod assemblages.
Fundam Appl Limnol 172: 175–182.
Maguire I, JelićM, Klobučar G. 2011. Update on the distribution of
freshwater crayfish in Croatia. Knowl Manag Aquat Ecosyst 401:
1–10.
Page 11 of 12
K. Žganec et al.: Ann. Limnol. - Int. J. Lim. 2018, 54, No

MarkovićV, TomovićJ, AtanackovićA, Kračun M, IlićM, NikolićV,
PaunovićM. 2015. Macroinvertebrate communities along the
Velika Morava River. Turkish J Zool 39: 210–224.
Matoničkin I, PavletićZ, Habdija I, StilinovićB. 1975. A contribution
to the valorisation of waters of the ecosystem of river Sava.
Sveučili
sna naklada Liber, Zagreb, 96 p.
Me
strov M, De
skovićI, Tavčar V. 1978. Pollution of the river Sava 
accordingto severalyears of ecologicalresearch. Ekologija 13: 61–79.
Me
strov M, Habdija I, StilinovićB, Maloseja Ž,Tavčar V, Kerovec
M, Primc B, FutačN. 1989. Biolo
sko-ekolo
ska valorizacija
kvalitete vode rijeke Save. In: Me
strov M, ed. Rijeka Sava-za
stita i
kori
stenje voda. Vol. 14, Zagreb: JAZU, 290–308.
Mineur F, Davies AJ, Maggs CA, Verlaque M, Johnson MP. 2010.
Fronts, jumps and secondary introductions suggested as different
invasion patterns in marine species, with an increase in spread rates
over time. Proc R Soc Lond B 277: 2693–2701.
Nunes AL, Tricarico E, Panov VE, Cardoso AC, Katsanevakis S.
2015. Pathways and gateways of freshwater invasions in Europe.
Aquat Invasions 10: 359–370.
Ogrinc N, KandučT, Kocman D. 2015. Integrated approach to the
evaluation of chemical dynamics and anthropogenic pollution
sources in the Sava River Basin. In: MilačičR, Ščančar J, Paunović
M, eds. The Sava River. The hanbook. Berlin/Heidelberg: Springer,
pp. 75–94.
PaunovićM, MiljanovićB, SimićV, CakićP, DjikanovićV, Jakovčev-
TodorovićD, StojanovićB, VeljkovićA. 2005. Distribution of non-
indigenous tubificid worm Branchiura sowerbyi (Beddard, 1892)
in Serbia. Biotechnol Biotechnol Equip 19: 91–97.
PaunovićM, TomovićJ, KovačevićS, ZorićK, Žganec K, SimićV,
AtanackovićA, MarkovićV, Kračun M, Hudina S, Lajtner J,
Gottstein S, LucićA. 2012. Macroinvertebrates of the natural
substrate of the Sava River preliminary results. Water Res Manag
2: 33–39.
Petts GE, Möller H, Roux AL. 1993. Historical change in large
alluvial rivers: Western Europe, Chichester: Wiley, 355 p.
Pinkster S. 1993. A revision of the genus Echinogammarus Stebbing,
1899 with some notes on related genera (Crustacea, Amphipoda).
Mem del Mus Civ Stor Nat 10: 1–185.
R Development Core Team. 2017. A language and environment for
statistical computing, Vienna, Austria: R Foundation for Statistical
Computing,http://www.R-project.org.
Rewicz T, Grabowski M, MacNeil C, Bącela-Spychalska K. 2014.
The profile of a “perfect”invader the case of killer shrimp,
Dikerogammarus villosus.Aquat Invasions 9: 267–288.
Rice SP, Greenwood MT, Joyce CB. 2001. Tributaries, sediment
sources, and the longitudinal organisation of macroinvertebrate
fauna along river systems. Can J Fish Aquat Sci 58: 824–840.
Schneider-Jacoby M. 2005. The Sava and Drava floodplains.
Threatened ecosystems of international importance. Arch für
Hydrobiol Supp. 158: 249–288.
Schwarz U. 2016. Sava White Book. The River Sava: threats and
restoration Potenti al. Radolfzell/Wien: EuroNatur/Riverwatch, 144 p.
Sommerwerk N, Hein T, Schneider-Jacoby M, Baumgartner C,
OstojićA, Siber R, Bloesch J, PaunovićM, Tockner K. 2009. The
Danube River Basin. In: Tockner K, Uehlinger U, Robinson CT,
eds. Rivers of Europe. London: Academic Press, pp. 59–112.
Statzner B, Peltret O, Tomanova S. 2003. Crayfish as geomorphic
agents and ecosystem engineers: effect of a biomass gradient on
baseflow and flood-induced transport of gravel and sand in
experimental streams. Freshw Biol 48: 147–163.
Statzner B, Bonada N, Dolédec S. 2008. Biological attributes
discriminating invasive from native European stream macro-
invertebrates. Biol Invasions 10: 517–530.
Strayer DL, Cole JJ, Findlay SEG, Fischer DT, Gephart JA, Malcom
HM, Pace ML, Rosi-Marshall EJ. 2014. Decadal-scale change in a
large-river ecosystem. Bioscience 64: 496–510.
Suarez AV, Holway DA, Case TJ. 2001. Patterns of spread in
biological invasions dominated by long-distance jump dispersal:
Insights from Argentine ants. PNAS 98: 1095–1100.
Tierno de Figueroa JM, López-Rodríguez MJ, Fenoglio S, Sánchez-
Castillo P, Fochetti R. 2013. Freshwater biodiversity in the rivers of
the Mediterranean Basin. Hydrobiologia 719: 137–186.
Tockner K, Uehlinger U, Robinson CT. (eds.) 2009. Rivers of Europe.
London: Academic Press, 728 p.
UrbaničG. 2008. Redelineation of European inland water ecoregions
in Slovenia. Rev Hydrobiol 1: 17–25.
Van den Brink FWB, Van der Velde G, Bij de Vaate A. 1993. Ecological
aspects, explosive range extension and impact of a mass invader,
Corophium curvispinum Sars, 1895 (Crustacea: Amphipoda), in the
Lower Rhine (The Netherlands). Oecologia 93: 224–232.
Van der Velde G, Rajagopal S, Kelleher B, Muskó IB, Bij de Vaate A.
2000. Ecological impact of crustacean invaders: general consider-
ations and examples from the River Rhine. In: The biodiversity
crisis and Crustacea: Proc 4th Int Crustacean Congress. Vol.
2, Brill, Leiden, Amsterdam, pp. 3–33.
Van Overdijk CDA, Grigorovich IA, Mabee T, Ray WJ, Ciborowski
JJH, Macisaac HJ. 2003. Microhabitat selection by the invasive
amphipod Echinogammarus ischnus and native Gammarus
fasciatus in laboratory experiments and in Lake Erie. Freshw Biol
48: 567–578.
Van Riel MC, Van der Velde G, Rajagopal S, Marguillier S, Dehairs F,
Bij de Vaate A. 2006. Trophic relationships in the Rhine food web
during invasion and after establishment of the Ponto-Caspian
invader Dikerogammarus villosus.Hydrobiologia 565: 39–58.
Veuille M. 1979. Ĺévolution du genre Jaera Leach (Isopodes;
Asellotes) et ses rapports avec ĺhistoire de la Méditerranée. Sci Ser
49: 195–217.
Whitton BA. (ed.) 1984. Ecology of European Rivers, London:
Blackwell Scientific Publications, 584 p.
Wittmann KJ, Ariani AP, Daneliya M. 2016. The Mysidae (Crustacea:
Peracarida: Mysida) in fresh and oligohaline waters of the
Mediterranean. Taxonomy, biogeography, and bioinvasion.
Zootaxa 4142: 1–70.
ZorićK, Jakovčev-TodorovićD, DikanovićV, VasiljevićB, Tomović
J, AtanackovićA, SimićV, PaunovićM. 2011. Distribution of the
Ponto-Caspian polychaeta Hypania invalida (Grube, 1860) in
inland waters of Serbia. Aquat Invasions 6: 33–38.
Žganec K, Gottstein S, Hudina S. 2009. Ponto-Caspian amphipods in
Croatian large rivers. Aquat Invasions 4: 327–335.
Žganec K, Ćuk R, DekićS, Mili
sa M. 2016. Biocontamination of
benthic macroinvertebrate communities of four major large rivers
in Croatia. In: Móra A, Csabai Z, eds. Book of abstracts and
programme. 2nd Central European Symposium for Aquatic
Macroinvertebrate Research (CESAMIR). Mohács, Pécs: Car-
pathes Nature Foundation, p. 123.
Žganec K, Lunko P, Stroj A, Mamos T, Grabowski M. 2016.
Distribution, ecology and conservation status of two endemic
amphipods, Echinogammarus acarinatus and Fontogammarus
dalmatinus, from the Dinaric karst rivers, Balkan Peninsula. Ann
Limnol 52: 13–26.
Cite this article as:Žganec K, Ćuk R, TomovićJ, Lajtner J, Gottstein S, KovačevićS, Hudina S, LucićA, Mirt M, SimićV, SimčičT,
PaunovićM. 2018. The longitudinal pattern of crustacean (Peracarida, Malacostraca) assemblages in a large south European river: bank
reinforcement structures as stepping stones of invasion. Ann. Limnol. - Int. J. Lim. 54–
Page 12 of 12
K. Žganec et al.: Ann. Limnol. - Int. J. Lim. 2018, 54, No