ArticlePDF Available

Habitat characteristics, hydrology and anthropogenic pollution as important factors for distribution of biota in the middle Paraná River, Argentina

Authors:
  • The National Institute of Limnology (INALI; CONICET-UNL), Santa Fe, Argentina
  • Instituto Nacional de Limnología (INALI-CONICET)

Abstract and Figures

The regulation of anthropogenic pollution inputs into large rivers is an important aspect of ecological resilience of aquatic systems and river pollution management. The current study examined the relationship between contamination loads, hydrological and morphological patterns and the distribution of macroinvertebrates and epipelic diatoms in the middle Paraná River system to form part of the development of a pollution monitoring framework. Seven sampling sites were selected over three main river areas predominantly impacted by sewage effluent and agriculture activities. The sampling areas were the Paraná Colastiné and Las Conchas rivers. In order to prevent dilution of pollutants and macroinvertebrate drift, sampling was performed during the base flow period of 2015 to determine pollution contaminated stretches of the river system. Results indicated that metals have been accumulated in river bottom sediments as a consequence of anthropogenic land use activities. Macroinvertebrate and epipelic diatom assemblages as bioindicators of anthropogenic pollution were evident downstream of urban sewage effluent discharges causing higher concentrations of Cr, As and Ni than the permitted threshold levels for bottom sediment. © 2018 European Regional Centre for Ecohydrology of the Polish Academy of Sciences
Content may be subject to copyright.
Original
Research
Article
Habitat
characteristics,
hydrology
and
anthropogenic
pollution
as
important
factors
for
distribution
of
biota
in
the
middle
Parana
´River,
Argentina
Martin
C.M.
Blettler
a
,
Paul
J.
Oberholster
b,c,f
,
Tebogo
Madlala
b,f,
*,
Eliana
G.
Eberle
a
,
Mario
L.
Amsler
a
,
Arno
R.
De
Klerk
d
,
Johannes
C.
Truter
c
,
Mercedes
R.
Marchese
a
,
Francisco
G.
Latosinski
e
,
Ricardo
Szupiany
e
a
National
Institute
of
Limnology
(INALI,
CONICET-UNL),
Ciudad
Universitaria,
3000
Santa
Fe,
Argentina
b
CSIR
Natural
Resources
and
the
Environment,
Jan
Cilliers
Street,
Stellenbosch
7599,
South
Africa
c
Department
of
Botany
and
Zoology,
University
of
Stellenbosch,
Private
Bag
X1,
Matieland,
Stellenbosch
7601,
South
Africa
d
CSIR
Natural
Resources
and
the
Environment,
P.O.
Box
395,
Pretoria
0001,
South
Africa
e
Faculty
of
Engineering
and
Water
Sciences,
Littoral
National
University,
Santa
Fe,
Argentina
f
Department
of
Earth
Sciences,
University
of
the
Western
Cape,
Private
Bag
X17,
Bellville
7535,
South
Africa
1.
Introduction
Numerous
biological
assessment
tools
have
been
developed
for
different
types
of
aquatic
ecosystems
and
for
various
monitoring
purposes
(Bain
et
al.,
2000;
Simon,
2000).
A
large
proportion
of
these
tools
are
aimed
at
monitoring
the
ecological
integrity
of
the
respective
systems
(Resh,
2008).
Aquatic
organisms,
like
macroin-
vertebrates
and
epipelic
diatoms,
have
been
used
as
monitoring
tools
due
to
the
fact
that
these
organisms
are
able
to
reflect
a
long-term
integrated
state
of
their
respective
ecosystems
(Rodrigues
Capı
´tulo
et
al.,
2001;
Matthews
et
al.,
2010;
Oberholster
et
al.,
2005,
2013;
Ecohydrology
&
Hydrobiology
xxx
(2018)
xxx–xxx
A
R
T
I
C
L
E
I
N
F
O
Article
history:
Received
11
March
2018
Accepted
14
August
2018
Available
online
xxx
Keywords:
Bioindicators
Anthropogenic
Biotopes
Hydrology
Large
rivers
A
B
S
T
R
A
C
T
The
regulation
of
anthropogenic
pollution
inputs
into
large
rivers
is
an
important
aspect
of
ecological
resilience
of
aquatic
systems
and
river
pollution
management.
The
current
study
examined
the
relationship
between
contamination
loads,
hydrological
and
morphological
patterns
and
the
distribution
of
macroinvertebrates
and
epipelic
diatoms
in
the
middle
Parana
´River
system
to
form
part
of
the
development
of
a
pollution
monitoring
framework.
Seven
sampling
sites
were
selected
over
three
main
river
areas
predominantly
impacted
by
sewage
effluent
and
agriculture
activities.
The
sampling
areas
were
the
Parana
´,
Colastine
´
and
Las
Conchas
rivers.
In
order
to
prevent
dilution
of
pollutants
and
macroinvertebrate
drift,
sampling
was
performed
during
the
base
flow
period
of
2015
to
determine
pollution
contaminated
stretches
of
the
river
system.
Results
indicated
that
metals
have
been
accumulated
in
river
bottom
sediments
as
a
consequence
of
anthropogenic
land
use
activities.
Macroinvertebrate
and
epipelic
diatom
assemblages
as
bioindicators
of
anthropogenic
pollution
were
evident
downstream
of
urban
sewage
effluent
discharges
causing
higher
concentrations
of
Cr,
As
and
Ni
than
the
permitted
threshold
levels
for
bottom
sediment.
ß
2018
European
Regional
Centre
for
Ecohydrology
of
the
Polish
Academy
of
Sciences.
Published
by
Elsevier
B.V.
All
rights
reserved.
*
Corresponding
author
at:
CSIR
Natural
Resources
and
the
Environ-
ment,
Jan
Cilliers
Street,
Stellenbosch
7599,
South
Africa.
E-mail
address:
TMadlala@csir.co.za
(T.
Madlala).
G
Model
ECOHYD-200;
No.
of
Pages
11
Please
cite
this
article
in
press
as:
Blettler,
M.C.M.,
et
al.,
Habitat
characteristics,
hydrology
and
anthropogenic
pollution
as
important
factors
for
distribution
of
biota
in
the
middle
Parana
´River,
Argentina.
Ecohydrol.
Hydrobiol.
(2018),
https://doi.org/10.1016/j.ecohyd.2018.08.002
Contents
lists
available
at
ScienceDirect
Ecohydrology
&
Hydrobiology
jo
u
rn
al
h
om
ep
age:
w
ww.els
evier.c
o
m/lo
c
ate/ec
oh
yd
https://doi.org/10.1016/j.ecohyd.2018.08.002
1642-3593/ß
2018
European
Regional
Centre
for
Ecohydrology
of
the
Polish
Academy
of
Sciences.
Published
by
Elsevier
B.V.
All
rights
reserved.
Cochero
et
al.,
2014,
2016).
These
tools
are
based
on
the
assumption
that
anthropogenic
impacts
will
result
in
changes
in
the
community
structure,
abundance
or
diversity
of
these
biota
(De
Klerk,
2016).
While
riverine
biomonitoring
studies
are
routine,
the
majority
of
them
are
focused
on
relatively
small
rivers
(Li
et
al.,
2010).
Similar
studies
in
large
rivers
are
still
scarce,
probably
for
the
reason
that
it
requires
a
great
technical
effort
due
to
the
strong
current
and
great
water
depths.
However,
there
are
certain
exceptions
(e.g.
Marchese
and
Ezcurra
de
Drago,
1999,
2006).
The
distinctive
hydrological
and
morphological
pat-
terns
of
rivers
create
specific
sites
or
habitats
for
biota
assemblages,
defining
habitat
patches
or
biotopes
at
different
scales
(Wadeson,
1994).
These
biotopes,
in
turn,
could
be
heavily
subjected
to
anthropogenic
pollutants
(Andre
`s
et
al.,
1999).
For
example,
several
studies
have
demonstrated
that
sandy
bedforms
act
as
sinks
for
contaminants
(e.g.
Ladd
et
al.,
1998;
Ciszewski,
2004).
Nevertheless,
further
research
is
needed
in
order
to
explore
the
relationships
between
hydro-morphological
variability
and
pollution
impacts
on
ecology
at
different
biotopes
in
large
rivers.
According
to
Marino
and
Ronco
(2005),
SAyDS-PNA-
UNLP
(2007),
Ronco
et
al.
(2008,
2016),
and
Peluso
et
al.
(2013)
certain
tributaries
of
the
Parana
´River
are
signifi-
cantly
contaminated
by
pollution
due
to
urban,
industrial
and
agricultural
land-use
activities
in
their
catchment
areas.
However,
metal
concentrations
in
the
bottom
sediments
of
the
Parana
´River
have
been
scantily
explored
(Pasquini
and
Depetris,
2012).
Additionally,
little
is
still
known
regarding
the
associations
between
contamination
loads,
hydrological
and
morphological
patterns,
and
ecology
of
organisms
inhabiting
the
Parana
´River.
In
this
regard,
the
aim
of
the
current
study
was
to
explore
a
diversity
of
sampling
sites
and
their
morphological
units
as
potential
sinks
for
contaminants
and
its
effect
on
the
ecology
of
macroinvertebrate
and
epipelic
diatom
assem-
blage
in
the
middle
Parana
´River
system.
The
objective
of
the
current
study
was
to
identify
polluted
stretches
in
the
middle
Parana
River
system
using
the
relationship
between
contamination
loads,
hydrological
and
morpho-
logical
patterns
and
the
distribution
of
macroinvertebrate
and
epipelic
diatom
assemblage
to
form
part
of
the
development
of
a
greater
pollution
monitoring
framework.
2.
Materials
and
methods
2.1.
Study
area
The
Parana
´River
is
a
mega-river
with
a
catchment
area
of
2.8
10
6
km
2
and
a
mean
discharge
of
18,000
m
3
s
1
(Latrubesse,
2008).
Its
sandy
mobile
bed,
where
fine
and
medium
grain
sizes
prevail,
is
transported
principally
by
saltation
and
in
suspension
(Drago,
2007).
The
main
channel
of
the
Middle
Parana
´River
is
typically
braided
and
composes
of
a
sequence
of
wide
segments
that
are
characterized
by
two
or
more
branches
with
lateral
erosion
and
sedimentation
activities
where
bars
and
dunes
are
usually
present
(Szupiany
et
al.,
2009).
Along
its
main
channel,
the
middle
Parana
´River
has
built
a
wide
floodplain
with
a
surface
area
of
about
20,000
km
2
(Iriondo
et
al.,
2007)
and
anabranching
planform
pattern
(Latru-
besse,
2008).
The
percent
rate
of
change
of
the
Parana
´discharges
was
amplified
when
compared
to
the
corresponding
rate
of
change
of
average
precipitation
on
the
basin
(Berbery
and
Barros,
2002).
This
feature
can
be
attributed
partially
to
deforestation
and
land
use
change
that
resulted
in
increased
runoff
(Collinshon
et
al.,
2001).
In
addition,
episodes
of
heavy
rainfall
are
becoming
more
frequent
as
a
result
of
the
regional
climate
trend
(Barros
et
al.,
2006).
In
this
study
area,
the
predominant
land
use
activities
were
agriculture
activities,
discharge
of
untreated
or
partially
treated
domestic
and
industrial
wastewater
(Fig.
1).
2.2.
Location
Seven
sampling
sites
were
selected
over
three
main
areas
with
sewage
effluent
and
agriculture
activities
as
the
main
source
of
point
and
diffuse
pollution.
These
areas
were
the
(1)
main
channel
of
the
Parana
´River;
(2)
the
Colastine
´River,
a
major
secondary
channel
of
the
Parana
´
River,
and
(3)
the
Las
Conchas
River,
a
minor
tributary
of
the
Parana
´River.
Although
it
is
ideal
to
sample
during
high
and
low
flows
to
capture
all
ecological
changes
in
the
river
system,
as
part
of
the
development
of
a
pollution
monitoring
framework
for
the
middle
Parana
´River,
the
current
study
was
performed
in
late
summer
2015
during
a
flow
period
with
a
mean-low
water
level
(3.2
m;
Parana
´
city
gauging
station).
This
period
was
chosen
to
avoid
dilution
of
pollutants
and
macroinvertebrate
drift.
Fig.
1
shows
the
study
area
and
the
location
of
the
defined
sampling
stations.
2.3.
Physical
and
chemical
variables
In
order
to
characterize
the
water
quality,
the
following
physical
and
chemical
variables
were
recorded
in
situ:
transparency
(Secchi
disk;
m),
electric
conductivity
(
m
S
cm
1
),
water
temperature
(8C),
pH,
total
dissolved
solids
(mg
l
1
)
and
dissolved
oxygen
(mg
l
1
),
using
a
Hach
Sension+
MM156
Portable
Multi-Parameter
Meter
(Loveland,
USA).
Additional
duplicate
bed
sediment
samples
for
granulometric
analysis
(by
dry
sieving)
were
taken
at
the
same
sampling
sites.
2.4.
Benthic
macroinvertebrates
sampling
Triplicate
random
benthic
sediment
samples
were
taken
at
each
sampling
site
using
a
Tamura
TM
bottom
dredge
(Rigosha
&
Co.)
(322
cm
2
surface
of
extraction).
The
dredge
was
lowered
into
the
water
column
by
means
of
an
electric
winch
fitted
with
a
steel
cable.
Each
benthic
sample
was
filtered
through
a
200-
m
m
sieve
and
fixed
with
5%
formaldehyde
in
the
field.
In
the
laboratory
the
samples
were
stained
with
erythrosin
to
facilitate
the
sorting
of
the
macroinvertebrates.
The
macroinvertebrates
were
hand-
picked
in
the
laboratory
and
stored
in
a
70%
ethanol
solution.
All
benthic
taxa
were
identified
and
counted
under
an
upright
microscope
(Nikon,
Eclipse
E100-LED;
M.C.M.
Blettler
et
al.
/
Ecohydrology
&
Hydrobiology
xxx
(2018)
xxx–xxx
2
G
Model
ECOHYD-200;
No.
of
Pages
11
Please
cite
this
article
in
press
as:
Blettler,
M.C.M.,
et
al.,
Habitat
characteristics,
hydrology
and
anthropogenic
pollution
as
important
factors
for
distribution
of
biota
in
the
middle
Parana
´River,
Argentina.
Ecohydrol.
Hydrobiol.
(2018),
https://doi.org/10.1016/j.ecohyd.2018.08.002
40–60
magnification).
Taxonomical
determinations
were
made
at
species
and
genus
levels
using
the
following
taxonomic
keys:
Brinkhurst
and
Marchese
(1992)
for
Oligochaeta;
Paggi
(2001),
Trivinho-Strixino
(2011)
for
Diptera
Chironomidae;
Spinelli
and
Wirth
(1993)
for
Diptera
and
Ceratopogonidae;
Domı
´nguez
et
al.
(1994)
for
Ephemeroptera;
Domı
´nguez
and
Ferna
´ndez
(2009)
for
other
taxa.
2.5.
Diatom
sampling
Epipelic
diatoms
samples
were
taken
at
the
same
sampling
sites
used
to
sample
macroinvertebrates.
Tripli-
cate
bottom
sediment
samples
were
taken
with
a
Tamura
TM
bottom
dredge
(Rigosha
&
Co.),
while
the
first
3
cm
of
the
upper
bottom
sediment
of
each
sample
were
remove
and
preserved
with
2%
glutaraldehyde
solution
for
microscopic
analyses.
The
lack
of
cell
content
in
the
diatom
cells
was
determined
to
verify
the
ratio
of
dead
diatoms
in
the
samples
before
they
were
acid
cleaned
(Beninger
et
al.,
2008).
The
diatom
samples
were
cleared
of
organic
matter
by
heating
it
in
a
potassium
dichromate
and
sulphuric
acid
solution
and
the
cleared
material
was
rinsed,
diluted
and
mounted
in
Pleurax
medium
for
microscopic
examination.
Diatoms
were
identified
using
a
compound
microscope
at
1250
magnification
according
to
the
taxonomic
keys
of
Metzeltin
et
al.
(2005).
The
samples
were
sedimented
in
a
Sedgewick-Rafter
sedimentation
chamber
and
were
ana-
lysed
using
the
strip-count
method
(APHA,
AWWA,
WPCF,
1992).
The
concentration
(cell
cm
2
)
was
estimated
by
multiplying
the
number
of
valves
from
each
taxon
by
a
conversion
factor
according
to
the
methods
of
Hermany
et
al.
(2006).
The
numerical
indicators
for
general
grouping
of
the
abundance
of
diatom
taxa
at
each
sampling
site
were
categorised
as
follows:
1
=
50
(rare),
2
=
51–250
(scarce),
3
=
251–1000
(common),
4
=
1001–5000
(abun-
dant),
and
5
=
5001–25,000
(predominant)
cells
cm
2
.
Only
live
diatoms
containing
chlorophyll
were
counted.
2.6.
Bottom
sediment
metal
analyses
The
total
concentrations
of
a
selection
of
metals
and
other
elements
were
quantified
in
composite
sediment
samples,
consisting
of
three
random
grab
samples
collected
per
sampling
site.
The
samples
were
maintained
on
ice
and
at
4
8C
during
storage
and
transportation,
and
digested
in
a
nitric
acid:hydrogen
peroxide
solution
prior
to
analysis.
A
Thermo
ICAP
6300
ICP-AES
(Thermo
Scientific,
USA)
was
used
to
measure
Ca,
K,
Mg,
Na,
P
and
Si,
whereas
an
Agilent
7700
ICP-MS
(Agilent
Fig.
1.
Satellite
image
of
the
middle
Parana
´River
showing
the
sampling
areas
(circles;
a),
and
detailing
the
specific
sampling
sites
in
the
Colastine
´River
(b),
Parana
´main
channel
(c)
and
Las
Conchas
stream
(c
and
d).
Where,
PMC:
Parana
´River
main
channel;
C1-3:
Colastine
´River;
LC1-3:
Las
Conchas
River;
s.e:
sewage
effluent;
I.ch:
‘‘Iris’’
secondary
channel;
fd:
flow
direction.
M.C.M.
Blettler
et
al.
/
Ecohydrology
&
Hydrobiology
xxx
(2018)
xxx–xxx
3
G
Model
ECOHYD-200;
No.
of
Pages
11
Please
cite
this
article
in
press
as:
Blettler,
M.C.M.,
et
al.,
Habitat
characteristics,
hydrology
and
anthropogenic
pollution
as
important
factors
for
distribution
of
biota
in
the
middle
Parana
´River,
Argentina.
Ecohydrol.
Hydrobiol.
(2018),
https://doi.org/10.1016/j.ecohyd.2018.08.002
Technologies,
USA)
was
used
to
measure
As,
Cd,
Cr,
Cu,
Pb,
Hg,
Ni,
Zn,
Al,
B,
Ba,
Be,
Co,
Fe,
Li,
Mn,
Mo,
Sb,
Se,
Sn,
Sr,
Ti
and
V.
2.7.
Bathymetric
and
three-dimensional
flow
The
river
bed
morphology
and
three-dimensional
flow
velocities
were
surveyed
using
a
1200
kHz
Teledyne
RDI
acoustic
Doppler
current
profiler
(ADCP),
coupled
to
a
global
positioning
system
(GPS)
deployed
on
the
boat.
At
each
selected
site,
flow
measurements
were
made,
providing
details
of
the
flow
structure
characteristics.
In
order
to
obtain
representative
values
of
the
time-averaged
three-dimensional
velocities
at
each
site,
a
series
of
4
replicate
transect
lines
were
performed
and
subsequent-
ly
averaged
according
to
Szupiany
et
al.
(2007).
The
Velocity
Mapping
Toolbox
(Parsons
et
al.,
2013)
was
employed
to
process
velocity
data.
Primary
(pv)
and
secondary
(sv)
flow
structures
around
the
studied
sites
were
analysed
according
to
the
Rozovskii
definition.
The
latter
method
essentially
identifies
the
primary
velocity
direction
for
each
profile
as
the
depth-integrated
flow
vector,
and
the
secondary
currents
were
then
obtained
by
the
differences
from
this
average
vector
within
the
profile.
This
procedure
effectively
identifies
individual
secondary
planes
at
each
vertical
profile
across
a
given
section,
thus
allowing
for
identification
of
helical
motion
within
a
section,
without
distorting
the
secondary
flow
results
(cf.
Szupiany
et
al.,
2007,
2009
for
methodological
details).
Following
this
methodology,
we
were
able
to
determine
discharge
(m
3
s
1
),
flow
direction
and
primary
and
secondary
velocities
(m
s
1
).
The
sampling
area
located
nearby
the
junction
between
the
Parana
´and
Las
Conchas
rivers
was
selected
for
bathymetric
and
three-dimensional
current
surveys
(see
Fig.
1c).
2.8.
Statistical
analysis
A
Canonical
Analysis
of
Principal
coordinates
(CAP;
Anderson,
2003)
was
employed
to
find
spatial
patterns
of
similarity
in
the
composition
of
diatoms
and
macroinverte-
brate
assemblages.
This
is
a
constrained
ordination
procedure
that
initially
calculates
unconstrained
principal
coordinate
axes
followed
by
canonical
discriminant
analy-
sis
on
the
principal
coordinates
to
maximize
the
separation
between
predefined
groups
(Anderson
and
Robinson,
2003;
Anderson
and
Willis,
2003).
Jaccard’s
similarity
matrix
and
999
permutations
were
the
parameters
selected
in
the
current
study.
A
one-way
analysis
of
variance
(ANOVA)
test
and
post
hoc
Fisher’s
test
(multiple
comparison
test)
were
applied
to
identify
where
significant
differences
occurred
between
scores
of
the
CAP
axis
1.
The
Principal
Components
Analysis
(PCA)
was
applied
to
identify
major
environmental
gradients,
including
metal
concentration,
in
relation
with
sampling
site
locations.
Lineal
correlations
were
performed
between
scores
of
the
PCA
axis
1
and
environmental
variables
to
identify
the
most
relevant
ones.
Statistical
analyses
were
carried
out
using
the
CAP
v1.0
(Anderson,
2003),
R
3.1
(R
Core
Team,
2012)
and
Statistica
(Tibco
Inc.,
2017)
software
programs.
For
all
analyses
the
statistical
significance
level
was
p
<
0.05.
3.
Results
3.1.
Water,
sediments
and
metal
analyses
The
average
surface
water
pH
values
at
all
sampling
sites
ranged
between
7.1
and
7.9,
while
the
highest
total
dissolved
solids
(150
mg
l
1
)
and
conductivity
(290
m
S
cm
1
)
were
measured
at
sampling
site
LC3
in
comparison
to
the
other
sampling
sites.
The
first
axis
of
the
PCA
explained
70.6%
of
the
environment
variations,
while
the
second
axis
explained
15%
(Fig.
3
and
Table
3).
According
to
this
analysis
sampling
sites
C1,
LC1
and
PMC
remained
clearly
separated
from
the
other
sampling
sites,
showing
the
lowest
concentration
of
metals
in
the
bottom
sediments.
On
the
contrary,
sites
C2
and
C3
(located
downstream
the
sewage
effluent
in
the
Colastine
´River)
were
associated
with
the
highest
concentration
of
metals.
LC1
was
located
in
between
both
previous
groups
of
sampling
sites
in
the
plot,
suggesting
intermediate
concentrations
of
bottom
sediment
metals.
The
latter
site
was
characterized
by
a
relatively
high
amount
of
fine
sediments
in
the
sampled
bedform
(particularly
clay).
According
to
Table
4,
sampling
sites
C2
and
C3
were
clearly
the
most
polluted
sites.
Particularly,
As,
Cr
and
Ni
were
recorded
at
much
higher
concentrations
at
these
sites
than
the
sediment
quality
guidelines
prescribed
by
Burton
(2002).
It
was
evident
that
the
urban
wastewater
effluent
from
the
city
of
Santa
Fe
into
the
Colastine
´River
increased
the
metal
concentrations
of
Cr,
As
and
Ni
at
these
sites
in
comparison
to
the
other
sampling
sites.
However,
as
noted
by
Sosa
and
Datta
(2015)
and
Sosa
et
al.
(2017),
As
concentrations
have
been
attributed
to
natural
geochemi-
cal
and
sedimentological
processes,
while
Cr
and
Ni
indicate
anthropogenic
inputs.
The
highest
concentrations
of
bottom
sediment
metals
were
recorded
at
sampling
stations
C2
and
C3
(Cu,
Pb,
Zn,
Al,
Ba,
B,
Co,
Fe,
K,
Li,
Mg,
Mn,
P,
Se,
Sr
and
V;
Fig.
3
and
Table
4)
indicating
that
both
sites
were
impacted
by
metal
pollutants.
Furthermore,
sam-
pling
site
LC3
(Las
Conchas
River,
21
km
upstream
to
the
mouth)
showed
higher
concentrations
of
bottom
sediment
metals
in
comparison
to
sampling
sites
LC1,
LC2.
The
lowest
metal
concentrations
were
observed
at
sampling
sites
LC2,
C1
and
PMC
(Table
4).
19
of
the
27
metals
and
elements
tested
for
correlated
significantly
with
silt
content
when
the
study
area
was
considered
collectively
(p
<
0.05,
Spearman
Rank
Test).
Conversely,
not
a
single
metal
or
element
correlated
with
sand
or
clay
content
in
the
samples
analysed,
and
contaminant
loads
in
general
is
therefore
associated
with
silt
content.
Cd,
Hg,
Ni,
Sb,
Se
and
Sn
however
did
not
correlate
significantly
with
silt
content
(p
>
0.05,
Spear-
man
Rank
Test).
3.2.
Benthic
macroinvertebrates
and
diatoms
assemblage
The
results
obtained
by
CAP
indicated
that
several
sampling
sites
had
different
macroinvertebrate
assem-
blages
(p
=
0.001;
Fig.
2).
The
analysis
of
variances
(ANOVA),
and
ulterior
LSD
Fisher
test,
performed
with
scores
of
the
CAP
axis
1
confirmed
significant
differences
in
macroinvertebrate
composition
between
sampling
sites
M.C.M.
Blettler
et
al.
/
Ecohydrology
&
Hydrobiology
xxx
(2018)
xxx–xxx
4
G
Model
ECOHYD-200;
No.
of
Pages
11
Please
cite
this
article
in
press
as:
Blettler,
M.C.M.,
et
al.,
Habitat
characteristics,
hydrology
and
anthropogenic
pollution
as
important
factors
for
distribution
of
biota
in
the
middle
Parana
´River,
Argentina.
Ecohydrol.
Hydrobiol.
(2018),
https://doi.org/10.1016/j.ecohyd.2018.08.002
C1,
C2
and
C3
(Colastine
´River;
Table
1).
However,
analyses
did
show
that
sampling
sites
C2
and
C3
differ
in
their
macroinvertebrate
assemblages
between
each
other.
On
the
other
hand,
there
were
no
differences
observed
between
sampling
sites
C1,
LC2
and
PMC,
which
possibly
suggest
that
similar
benthic
macroinvertebrate
and
diatom
assemblages
inhabited
these
sites.
Furthermore,
sampling
site
LC3
did
show
dissimilarity
with
regard
to
all
the
other
sampling
sites.
Table
2
shows
that
the
diatom
Aulacoseira
granulata
was
absent
at
sampling
site
C2
and
C3,
where
higher
concentration
of
sediment
metals
were
recorded
in
the
current
study.
It
is
noteworthy
to
mention
that
Fallacia
monoculata,
Achnanthes
lanceolata,
Craticula
accomoda
and
Eolimna
minima
were
exclusively
present
at
sampling
site
C2
(Table
2).
The
macroinvertebrate
species
Narapa
bonettoi
and
Myoretronectes
paranaensis
were
present
in
high
densities
at
sampling
sites
C1,
LC2
and
PMC
(Fig.
2).
These
sites
were
characterized
by
sandy
bottom
sediments
and
very
low
concentration
of
bottom
sediment
metals
in
comparison
to
the
other
sampling
sites.
On
the
other
hand,
the
macroinvertebrate
species
Pristina
americana
(Naidi-
dae)
did
only
occurred
at
sampling
site
C2.
3.3.
Bathymetric
and
three-dimensional
flow
Fig.
4A
shows
the
bathymetry
and
flow
configuration
at
the
cross-section
of
sampling
site
LC1,
immediately
downstream
of
the
confluence
between
the
Las
Conchas
River
and
a
short
secondary
channel
of
the
Parana
´River
(herein
named
as
‘‘Iris’’
channel).
A
reduction
in
water
velocity
was
observed
where
both
currents
of
the
two
different
rivers
joined.
Together
with
this
velocity
reduc-
tion,
a
secondary
cell
movement
was
recorded
near
to
the
bottom
surface
(see
direction
and
magnitude
of
the
sv
vectors).
This
particular
hydraulic
configuration
did
create
a
depositional
area
characterized
by
fine
bed
material
(45%
silt
and
29%
clay).
Over
this
full
cross-section
a
discharge
of
363
m
3
s
1
and
a
mean
primary
velocity
(pv)
of
0.63
m
s
1
Fig.
2.
Ordination
plot
of
the
Canonical
Analysis
of
Principal
coordinates
(CAP)
showing
significant
differences
in
the
biota
composition.
The
plot
was
generated
with
the
first
two
principal
coordinate
axes
from
the
computed
results
(see
references
in
legend
of
Fig.
1).
Fig.
3.
Plot
of
scores
distribution
along
principal
component
analysis
(PCA)
axe
according
to
physical-chemical
and
metals
variables
recorded
at
sampling
stations.
Where:
Cond:
water
electrical
conductivity;
CS:
percentage
of
coarse
sand
in
the
bottom;
MS:
percentage
of
middle
sand
in
the
bottom;
FS:
percentage
of
fine
sand
in
the
bottom;
Clay:
percentage
of
bottom
clay;
TDS:
total
dissolved
solids;
Temp:
water
temperature;
O
2
:
water
dissolved
oxygen.
Short
vectors
were
removed.
Table
1
p-Values
of
the
ANOVA
result
performed
with
scores
of
the
CAP
axis
1,
based
on
diatom
and
benthic
macroinvertebrate
data.
Sampling
sites
C1
C2
C3
LC1
LC2
LC3
PMC
C1
1
C2
0.001
1
C3
0.001
0.001
1
LC1
0.936
0.001
0.001
1
LC2
0.120
0.001
0.001
0.104
1
LC3
0.001
0.001
0.001
0.001
0.001
1
PMC
0.158
0.001
0.001
0.180
0.007
0.001
1
M.C.M.
Blettler
et
al.
/
Ecohydrology
&
Hydrobiology
xxx
(2018)
xxx–xxx
5
G
Model
ECOHYD-200;
No.
of
Pages
11
Please
cite
this
article
in
press
as:
Blettler,
M.C.M.,
et
al.,
Habitat
characteristics,
hydrology
and
anthropogenic
pollution
as
important
factors
for
distribution
of
biota
in
the
middle
Parana
´River,
Argentina.
Ecohydrol.
Hydrobiol.
(2018),
https://doi.org/10.1016/j.ecohyd.2018.08.002
were
observed.
However,
the
local
pv
over
the
sampling
site
was
around
0.4
m
s
1
.
Over
the
cross-section
LC2
the
local
maximum
pv
ranged
between
0.5
and
0.55
m
s
1
,
being
smaller
than
at
sampling
site
LC1.
The
discharge
over
this
cross-section
(Las
Conchas
river)
was
only
50.5
m
3
s
1
,
with
a
mean
pv
of
0.39
m
s
1
(Fig.
4B).
A
large
sand
dune
with
its
trough
4–5
m
deeper
than
the
crest
was
recorded
and
sampled
at
PMC
station.
4.
Discussion
It
is
known
that
the
growth
of
diatoms
can
be
inhibited
by
a
low
supply
of
silica.
However,
the
Si
concentrations
recorded
in
the
current
study
were
significantly
above-
average
values
sufficient
for
diatom
reproduction
(Willen,
1991).
Therefore,
the
sparse
diatom
flora
observed
at
the
entire
sampling
sites
can
possibly
be
related
to
the
occurrence
of
suspended
solids
and
basic
inorganic
materials
that
decreased
the
water
column
transparency.
This
environment
condition
favoured
low
light
intensity
diatoms
like
A.
granulata
that
prevail
at
most
of
the
sampling
sites.
A.
granulata
was
widely
recorded
at
sampling
sites
C1,
LC1,
LC2,
LC3
and
PMC
and
was
only
absent
at
sites
C2
and
C3
(sites
located
downstream
of
the
urban
sewage
effluent;
Table
2).
This
species
is
an
r-
strategist
species
adapted
to
variations
in
turbulence
and
light
intensity,
typically
present
in
the
main
channel
of
the
Table
2
Most
abundant
species
of
macroinvertebrates
and
diatoms
recorded
at
each
sampling
sites
during
February
2015
(n
=
1).
Species
C1
C2
C3
LC1
LC2
LC3
PMC
Benthic
macroinvertebrates
(ind.
m
2
)
Limnoperna
fortune
18,000
11,000
Narapa
bonettoi
500
5600
1000
Tubificinae
sp.
1100
3250
300
Lopescladius
sp.
1600
700
Pristina
Americana
1000
Myoretronectes
paranaensis
900
50
Haplotaxis
sp.
300
250
Amphichaeta
sp.
100
Epipelic
diatoms
(Bacillariophyceae)
(cells
cm
2
)
Frustulia
vulgaris
31
Fallacia
monoculata
27
Achnanthes
lanceolata
35
Craticula
accomoda
48
Eolimna
minima
31
Aulacoseira
granulata
41
44
37
33
42
Cyclotella
meneghiniana
43
Fig.
4.
Primary
(color
scale;
cm
s
1
)
and
secondary
velocity
(vector
magnitude)
fields
in
the
surveyed
sections,
showing
the
cross-sections
located
over
the
LC1
(A),
LC2
(B).
M.C.M.
Blettler
et
al.
/
Ecohydrology
&
Hydrobiology
xxx
(2018)
xxx–xxx
6
G
Model
ECOHYD-200;
No.
of
Pages
11
Please
cite
this
article
in
press
as:
Blettler,
M.C.M.,
et
al.,
Habitat
characteristics,
hydrology
and
anthropogenic
pollution
as
important
factors
for
distribution
of
biota
in
the
middle
Parana
´River,
Argentina.
Ecohydrol.
Hydrobiol.
(2018),
https://doi.org/10.1016/j.ecohyd.2018.08.002
Parana
´River
with
relatively
good
water
quality
(Zalocar
de
Domitrovic
et
al.,
2007).
According
to
Zalocar
de
Domi-
trovic
and
Maidana
(1997)
and
Devercelli
(2010),
the
diatom
A.
granulata
is
the
most
important
species
in
the
Parana
River
system,
followed
by
the
diatom
C.
mene-
ghiniana
(only
presents
in
C3
station).
According
to
the
classification
by
Reynolds
et
al.
(2002),
A.
granulata
belongs
to
the
functional
group
P,
while
C.
meneghiniana
belong
to
the
functional
group
C,
which
is
a
colonizer
species,
sensitive
to
light
depletion,
with
a
rapid
reproduction
rate.
The
only
occurrence
of
the
diatom
Craticula
accomoda
at
sampling
site
C2
was
in
association
with
a
previous
report
by
Taylor
et
al.
(2007)
that
shows
that
this
species
is
found
in
strongly
organically
polluted
waters,
in
particular
effluent
from
sewage
treatment
works
with
a
scattered
occurrence
in
oligo-
to
eutrophic
water.
The
higher
bottom
sediment
metal
concentrations
recorded
in
the
Colastine
´River
(C2
and
C3),
were
possibly
related
to
the
point
source
pollution
downstream
of
the
urban
sewage
effluence
(Fig.
3
and
Table
3).
In
a
previous
study,
Oberholster
et
al.
(2013)
reported
that
the
ecological
impact
and
changes
in
phytoplankton
assem-
blage
downstream
of
the
Riverview
wastewater
treatment
plant
in
the
upper
Olifants
River
was
visible
over
a
distance
of
40
km.
In
the
current
study
ecological
indicator
differences
were
also
observed
among
the
latter
polluted
sites
(C2
and
C3),
(Fig.
2
and
Table
1).
Benthic
diatoms
were
clearly
more
abundant
and
diverse
at
site
C2
in
comparison
to
site
C3.
This
phenomenon
was
possibly
related
to
the
fact
that
certain
pollution
tolerant
epipelic
diatoms
become
more
abundant
where
water
systems
are
impact-
ed
by
anthropogenic
pollution
(Bate
et
al.,
2004).
The
genus
Achnanthes
observed
at
site
C2
has
previously
been
recorded
in
the
Parana
´River
under
environmental
condi-
tions
of
low
light
penetration,
almost
anoxic
conditions
and
high
nitrate
and
phosphate
concentrations
(Devercelli
et
al.,
2014).
According
to
these
authors,
species
of
this
genus
are
probably
mixotrophic
regarding
their
capacity
to
grow
in
extremely
poor
light
intensities
and
tolerate
very
low
oxygen
contents.
The
only
occurrence
of
the
macroinvertebrate
species
Pristina
americana
(Naididae)
at
sampling
site
C2
impacted
by
sewage
effluent
relates
with
findings
of
Petsch
et
al.
(2015),
who
reported
that
the
abundance
of
P.
americana
occurred
under
environmental
conditions
with
high
values
of
organic
matter
and
low
dissolved
oxygen,
which
is
symptomatic
of
surface
waters
impacted
by
sewage
effluent.
Although
literature
generally
suggests
that
metal
content
increases
when
sediment
grain
size
decreases,
this
relation
of
metal
content
to
particle
size
was
not
clear
in
the
current
study
(Fig.
3).
Spearman
Rank
correlations
however
indicated
silt
content
as
a
predictor
of
increased
metal
concentrations,
whereas,
sand
and
clay
content
were
not.
Du
Laing
et
al.
(2007)
performed
an
in-depth
assessment
of
metal
contamination
in
the
sediments
of
the
severe
polluted
freshwater
tidal
zone
system
of
the
Scheldt
River
(Belgium
and
the
Netherlands).
As
part
of
the
investigation,
the
silt,
clay,
sand,
organic
matter
among
other
factors
were
characterized
and
applied
in
predictive
modelling
using
stepwise
regression
equations.
Maslennikova
et
al.
(2012)
suggested
that
content
and
distribution
of
metals
and
organic
matter
were
defined
not
only
by
particle
size
of
sediments
but
also
by
conditions
of
its
accumulation,
particularly,
the
ecological
state
and
climatic
parameters.
Ciszewski
(2004)
found
changes
in
metal
concentrations
related
to
transport
of
river
bed
material,
dead
water
zones
and
bed
morphology
in
the
Mala
Panew
River.
Specifically,
in
this
study
the
confluence
zone
of
site
LC1
possibly
act
as
a
deposit
area
playing
an
important
role
as
sink
for
fine
and
metal
polluted
sediments
(Fig.
4A).
Although,
the
concentration
of
metals
at
this
sampling
site
was
below
the
threshold
level
according
to
Burton,
(2002),
it
was
higher
in
comparison
to
sampling
sites
PMC
and
LC2.
The
origin
of
the
polluted
sediment
from
site
LC1
is
not
clear
since
it
could
be
related
to
contaminants
coming
from
the
Parana
´River
main
channel
or,
on
the
contrary,
from
Las
Conchas
River.
The
bed
sediment
sizes
recorded
at
LC1
(45%
silt
and
around
30%
clay)
is
closely
related
to
that
of
the
fine
sediment
transported
in
suspension
in
the
Parana
´
Table
3
Eigenvalues
and
percentage
of
explanation
of
the
PCA
axe
1
and
2.
p
and
R
2
values
of
correlations
between
both
PCA
axe
and
each
physical–
chemical
and
metal
variables.
Axis
1
Axis
2
Eigenvalues
6.5
1.38
Cumulated
%
70.6
85.6
R
2
p
R
2
p
As
0.91
0.01
0.14
0.62
Cd
0.16
0.20
0.08
0.49
Cr
0.73
0.01
0.01
0.38
Cu
0.90
0.01
0.11
0.55
Pb
0.81
0.01
0.03
0.41
Hg
0.12
0.57
0.15
0.66
Ni
0.70
0.01
0.04
0.31
Zn
0.83
0.01
0.06
0.46
Al
0.79
0.01
0.05
0.43
B
0.93
0.01
0.17
0.76
Ba
0.84
0.01
0.06
0.46
Be
0.67
0.02
0.01
0.35
Ca
0.27
0.13
0.23
0.16
Co
0.88
0.01
0.12
0.58
Fe
0.86
0.01
0.10
0.53
K
0.80
0.01
0.03
0.40
Li
0.83
0.01
0.07
0.48
Mg
0.86
0.01
0.08
0.49
Mn
0.89
0.01
0.11
0.54
P
0.81
0.01
0.04
0.42
Sb
0.14
0.22
0.20
0.93
Se
0.84
0.00
0.05
0.44
Si
0.23
0.15
0.05
0.44
Sn
0.15
0.21
0.20
0.95
Sr
0.95
0.01
0.16
0.69
Ti
0.63
0.02
0.05
0.31
V
0.91
0.01
0.15
0.66
Sand
0.18
0.79
0.68
0.01
Clay
0.14
0.62
0.34
0.10
CS
0.51
0.04
0.19
0.89
MS
0.12
0.58
0.78
0.01
FS
0.25
0.14
0.13
0.60
TDS
0.11
0.56
0.38
0.08
Cond
0.13
0.61
0.85
0.01
Temp
0.08
0.49
0.33
0.11
pH
0.20
0.97
0.13
0.61
O
2
0.07
0.28
0.14
0.62
M.C.M.
Blettler
et
al.
/
Ecohydrology
&
Hydrobiology
xxx
(2018)
xxx–xxx
7
G
Model
ECOHYD-200;
No.
of
Pages
11
Please
cite
this
article
in
press
as:
Blettler,
M.C.M.,
et
al.,
Habitat
characteristics,
hydrology
and
anthropogenic
pollution
as
important
factors
for
distribution
of
biota
in
the
middle
Parana
´River,
Argentina.
Ecohydrol.
Hydrobiol.
(2018),
https://doi.org/10.1016/j.ecohyd.2018.08.002
River
(Amsler
and
Drago,
2009).
This
observation
suggests
two
possible
explanations.
Firstly,
that
the
suspended
sediments
coming
from
the
Parana
´River
main
channel
was
deposited
at
site
LC1
together
with
metals
transported
by
the
fine
suspended
sediments,
being
accumulated
gradu-
ally
at
relatively
low
concentrations.
This
possibility
conforms
with
Mugetti
(2004)
observations
who
reported
an
association
between
heavy
metals
and
suspended
material
in
the
Parana
´River.
Secondly,
it
is
recognized
that
metals
have
natural
or
anthropogenic
sources
in
aquatic
environments.
Based
on
this
fact,
it
may
be
inferred
that
metals
deposited
at
LC1
originated
in
sites
upstream
the
sampling
area
from
both
types
of
sources.
On
one
hand,
the
Parana
´River
exhibits
thick
beds
of
marine
and
continental
sedimentary
rocks
as
well
as
outcrops
of
metamorphic
rocks
along
with
volcanic
rocks
of
Quater-
nary
age,
which
could
be
a
source
of
metals
(Pasquini
and
Depetris,
2012).
On
the
other
hand,
metals
induced
by
human
activities
is
a
true
possibility
in
the
Parana
´River
Peluso
et
al.,
2013),
since
large
human
settlements
and
factories
of
Argentina
and
Paraguay
extend
along
the
banks
and
nearby
areas
of
its
main
channel
and
tributaries,
discharging
point
and
nonpoint
sources
of
waste
and
sewage
effluent
from
domestic
as
well
as
industrial
origin.
High
concentrations
of
ammonium,
nitrates
and
phos-
phates,
as
well
as
metals,
agrochemicals
and
biphenyl
poly-chlorines
have
also
been
found
near
to
urban
areas
of
this
river
system
(UN-Water/WWAP,
2007;
Peluso
et
al.,
2013;
Ronco
et
al.,
2016).
In
contrast,
distinctive
ecological
features
(macroin-
vertebrate
composition
and
relative
abundance)
were
observed
at
sampling
sites
C1,
LC2
and
PMC
(Fig.
2
and
Table
1).
These
sites
were
predominantly
colonized
by
N.
bonettoi,
M.
paranaensis,
and
Lopescladius
sp.
(Table
2).
According
to
Ezcurra
de
Drago
et
al.,
2007
and
Blettler
et
al.
(2012),
the
presence
of
these
species
which
integrate
the
so
called
‘‘active
bed
assemblage’’,
indicate
good
water
quality
conditions
and
a
bottom
substrate
composed
essentially
by
medium
and
coarse
sands
(Fig.
3
and
Table
3).
Macroinvertebrate
species
from
this
assemblage
typically
inhabiting
interstitial
spaces
between
sand
grains
of
sandy
rivers
(Blettler
et
al.,
2008;
Amsler
et
al.,
2009),
have
a
very
low
biomass
which
results
in
poor
energy
resources
for
higher
trophic
levels
(Ezcurra
de
Drago
et
al.,
2007).
It
should
be
noted
that
the
dune
trough
at
site
PMC
did
not
accumulate
pollutants
as
expected
(Fig.
3).
Even
when
a
reduction
of
bed
hydraulic
forces
occur
in
alluvial
dunes
as
is
widely
known
(e.g.
Kostaschuk
et
al.,
2004;
Best,
2005;
Amsler
et
al.,
2009),
it
was
not
enough
as
to
generate
a
sink
area
for
metal
polluted
sediments
in
the
main
channel
in
this
case.
Two
possible
explanations
for
this
fact
are:
(i)
the
current
in
the
central
strip
of
the
main
channel
was
too
strong
as
to
cause
any
deposition
of
pollutants
in
the
dune
trough,
and
(ii)
the
sandy
sediment
of
the
dune
(93.8%
sand)
prevented
metal
accumulation
due
to
vertical
dune
packing
and
porosity.
This
finding
is
of
ecological
relevance
because
dune
troughs
(where
a
reduced
disturbance
of
bed
Table
4
Concentration
of
metals
and
other
elements
in
bottom
sediments
recorded
at
each
sampling
site.
Bold
type
indicates
values
above
the
threshold
level
according
to
Burton
(2002).
Sampling
sites
Metals
(mg
kg
1
)
C1
C2
C3
LC1
LC2
LC3
PMC
C.
Std
Accuracy
(%)
As
0.95
8.78
7.68
2.33
0.61
4.49
0.39
98.1
Cd
0.27
0.09
0.08
0.03
0.10
0.03
0.16
96.4
Cr
2.58
35.02
51.36
<0.01
<0.01
13.78
<0.01
98.3
Cu
2.51
25.06
27.80
7.05
1.98
12.46
1.00
97.7
Pb
2.87
19.65
20.22
3.44
1.24
7.44
1.11
93.6
Hg
0.06
0.03
0.04
0.01
0.00
0.02
0.01
90.1
Ni
1.82
16.71
21.92
0.005
0.005
4.76
0.005
98.9
Zn
11.8
76.41
88.54
14.03
3.09
33.64
3.75
93.7
Al
1784.7
48263.5
68638
7820.4
1803.6
20262.3
1150.6
97.8
B
7.7
29.47
30.59
12.02
0.5
16.39
0.87
97.6
Ba
34.7
361.0
448.7
74
49.4
180.3
25
97.7
Be
0.02
1.39
2.20
0.37
0.05
0.29
0.05
92.5
Ca
680.7
3740.9
4034
2736.5
735.4
10979.6
479
97.4
Co
1.52
12.40
16.21
4.19
0.88
6.59
0.59
96.5
Fe
3287.8
32469.3
41569.6
7485.7
759
16240.1
1331.8
97.3
K
1216.5
17603.7
20514.3
2040.9
1528.7
7524.1
939.2
97.1
Li
9.86
40.9
50.9
19.9
12.9
23.1
11.5
97.8
Mg
139.9
7358.3
8180.1
1283.5
96.4
3168.2
58.2
97.4
Mn
97.6
703.8
826.1
196.5
68.7
359.3
25.2
96.7
Mo
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
94.4
Na
<25
<25
<25
<25
<25
<25
<25
98.1
P
33
151.2
173.9
38
27.05
75.2
25.8
94.0
Sb
24.6
0.60
0.9
0.07
3.55
0.33
6.49
96.6
Se
1.5
9.75
9.91
2.59
2.03
4.88
1.54
98.2
Si
3554.6
5061.2
4933.2
6745
3765.9
4475.3
2386.1
92.7
Sn
230.3
5.3
3.5
0.9
31.8
2.6
60.9
98.7
Sr
5.11
70.8
80.4
27.98
8.82
43
3.8
96.1
Ti
1046.5
1352.4
1408
1219.5
148.7
1622.5
319.5
V
12
78.2
102.1
28.4